Percutaneous Penetration Enhancers

Second Edition

PercutaneousPenetrationEnhancers

© 2006 by Taylor & Francis Group, LLC

A CRC title, part of the Taylor & Francis imprint, a member of theTaylor & Francis Group, the academic division of T&F Informa plc.

Second Edition

Edited by

Eric W. SmithHoward I. Maibach

PercutaneousPenetrationEnhancers

Boca Raton London New York


Published in 2006 byCRC PressTaylor & Francis Group 6000 Broken Sound Parkway NW, Suite 300Boca Raton, FL 33487-2742

© 2006 by Taylor & Francis Group, LLCCRC Press is an imprint of Taylor & Francis Group

No claim to original U.S. Government worksPrinted in the United States of America on acid-free paper10 9 8 7 6 5 4 3 2 1

International Standard Book Number-10: 0-8493-2152-2 (Hardcover) International Standard Book Number-13: 978-0-8493-2152-8 (Hardcover) Library of Congress Card Number 2005050563

This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted withpermission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publishreliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materialsor for the consequences of their use.

No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, orother means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any informationstorage or retrieval system, without written permission from the publishers.

01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. Fororganizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only foridentification and explanation without intent to infringe.

Library of Congress Cataloging-in-Publication Data

Percutaneous penetration enhancers / edited by Eric W. Smith and Howard I. Maibach.-- 2nd ed.p. cm.

Includes bibliographical references and index.ISBN 0-8493-2152-21. Transdermal medication. 2. Skin absorption. I. Maibach, Howard I.

RM151.P474 2005615'.6--dc22 2005050563

Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com

and the CRC Press Web site at http://www.crcpress.com

Taylor & Francis Group is the Academic Division of T&F Informa plc.

2152_Discl.fm Page 1 Tuesday, June 14, 2005 9:59 AM


For permission to photocopy or use material electronically from this work, please access www.copyright.com(http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC) 222 Rosewood Drive, Danvers, MA

http://www.copyright.com

http://www.copyright.com

http://www.taylorandfrancis.com

http://www.crcpress.com

Contents

I. INTRODUCTION............................................................................................................... 1

1. Penetration Enhancer Classification.......................................................................................... 3Brian W. Barry

2. Structure–Activity Relationship of Chemical Penetration Enhancers..................................... 17Narayanasamy Kanikkannan, R. J. Babu, and Mandip Singh

3. Quantitative Structure–Enhancement Relationship and the Microenvironmentof the Enhancer Site of Action ................................................................................................ 35S. Kevin Li and William I. Higuchi

4. The Role of Prodrugs in Penetration Enhancement............................................................... 51Kenneth B. Sloan and Scott C. Wasdo

II. VEHICLE EFFECTS IN PENETRATION ENHANCEMENT .................................................. 65

5. Penetration Enhancement by Skin Hydration ........................................................................ 67Jin Zhang, Carryn H. Purdon, Eric W. Smith, Howard I. Maibach,and Christian Surber

6. Enhancement of Delivery with Transdermal Sprays .............................................................. 73Barrie C. Finnin and Jonathan Hadgraft

7. Hydrogel Vehicles for Hydrophilic Compounds .................................................................... 83Teresa Cerchiara and Barbara Luppi

8. Enhanced Skin Permeation Using Ethosomes........................................................................ 95Elka Touitou and Biana Godin

9. Microemulsions in Topical Drug Delivery............................................................................ 109Sari Pappinen and Arto Urtti

10. Nanoparticles as Carriers for Enhanced Skin Penetration.................................................... 117Shozo Miyazaki

11. Solid Lipid Nanoparticles for Topical Delivery..................................................................... 125Zhinan Mei

Smith and Maibach / Percutaneous Penetration Enhancers 2nd edn 2152_C000 Final Proof page v 26.9.2005 8:05pm


12. Fatty Alcohols and Fatty Acids .............................................................................................. 137R. J. Babu, Mandip Singh, and Narayanasamy Kanikkannan

13. Essential Oils and Terpenes .................................................................................................. 159Rashmi A. Thakur, Yiping Wang, and Bozena B. Michniak

III. PHYSICAL METHODS OF PENETRATION ENHANCEMENT.........................................175

14. Iontophoresis: Clinical Applications and Future Challenges ............................................... 177Nada Abla, Aarti Naik, Richard H. Guy, and Yogeshvar N. Kalia

15. Electroporation ...................................................................................................................... 221Babu M. Medi and Jagdish Singh

16. Microneedles .......................................................................................................................... 239Mark R. Prausnitz, John A. Mikszta, and Jennifer Raeder-Devens

17. Vesicles under Voltage........................................................................................................... 257Michael C. Bonner and Brian W. Barry

IV. ASSESSMENT OF PENETRATION ENHANCEMENT.......................................................269

18. Mechanistic Studies of Permeation Enhancers ..................................................................... 271S. Kevin Li and William I. Higuchi

19. Penetration Enhancer Assessment by Corneoxenometry .................................................... 293Claudine Pierard-Franchimont, Frederique Henry, Emmanuelle Uhoda,Caroline Flagothier, and Gerald E. Pierard

20. Assessment of Vehicle Effects by Skin Stripping .................................................................. 299Carryn H. Purdon, Carolina Pellanda, Christian Surber, and Eric W. Smith

21. The Use of Skin Alternatives for Testing Percutaneous Penetration ................................... 311Charles Scott Asbill, Gary W. Bumgarner, and Bozena B. Michniak

22. High Throughput Screening of Transdermal Penetration Enhancers:Opportunities, Methods, and Applications........................................................................... 319Amit Jain, Pankaj Karande, and Samir Mitragotri

23. Confocal Laser Scanning Microscopy: An Excellent Tool for TrackingCompounds in the Skin......................................................................................................... 335Daya D. Verma and Alfred Fahr

V. THE RETARDATION OF PERCUTANEOUS PENETRATION ...........................................359

24. Fundamentals of Retarding Penetration................................................................................ 361Jonathan Hadgraft and Barrie C. Finnin

25. Retardation Strategies for Sunscreen Agents ........................................................................ 373Carryn H. Purdon, Eric W. Smith, and Christian Surber

26. Military Perspectives in Chemical Penetration Retardation.................................................. 385Ernest H. Braue, Jr., Bryce F. Doxzon, Horace L. Lumpkin, Kelly A. Hanssen,Robert S. Stevenson, Robin R. Deckert, and John S. Graham

VI. COMMERCIAL APPLICATIONS OF PENETRATION ENHANCERS ................................399

27. Preclinical and Clinical Development of a Penetration Enhancer SEPA 0009 .................... 401Thomas C. K. Chan

Smith and Maibach / Percutaneous Penetration Enhancers 2nd edn 2152_C000 Final Proof page vi 26.9.2005 8:05pm

vi & Contents


Contributors

Nada AblaSchool of PharmacyUniversity of GenevaGeneva, Switzerland

Charles Scott AsbillMcWhorter School of PharmacySamford UniversityBirmingham, Alabama

R. J. BabuCollege of Pharmacy and Pharmaceutical

SciencesFlorida A&M UniversityTallahassee, Florida

Brian W. BarrySchool of PharmacyUniversity of BradfordBradford, West Yorkshire, U.K.

Michael C. BonnerSchool of PharmacyUniversity of BradfordBradford, West Yorkshire, U.K.

Ernest H. Braue, Jr.U.S. Army Medical Research Institute

of Chemical DefenseAberdeen Proving Ground, Maryland

Gary W. BumgarnerMcWhorter School of PharmacySamford UniversityBirmingham, Alabama

Teresa CerchiaraDepartment of ChemistryCalabria UniversityArcavacata di Rende (CS), Italy

Thomas C. K. ChanMacroChem CorporationLexington, Massachusetts

Robin R. DeckertU.S. Army Medical Research Institute


Bryce F. DoxzonU.S. Army Medical Research Institute


Alfred FahrLehrstuhl fur Pharmazeutische

TechnologieFriedrich Schiller Universitat JenaJena, Germany

Smith and Maibach / Percutaneous Penetration Enhancers 2nd edn 2152_C000 Final Proof page vii 26.9.2005 8:05pm


Caroline FlagothierDepartment of DermatopathologyUniversity Hospital of LiegeLiege, Belgium

Barrie C. FinninMonash UniversityParkville, Victoria, Australia

Biana GodinDepartment of PharmaceuticsThe Hebrew University of JerusalemJerusalem, Israel

John S. GrahamU.S. Army Medical Research Institute


Richard H. GuyDepartment of Pharmacy and

PharmacologyUniversity of BathBath, U.K.

Jonathan HadgraftThe School of PharmacyUniversity of LondonLondon, U.K.

Kelly A. HanssenU.S. Army Medical Research Institute


Frederique HenryDepartment of DermatopathologyUniversity Hospital of LiegeLiege, Belgium

William I. HiguchiCollege of PharmacyUniversity of UtahSalt Lake City, Utah

Amit JainDepartment of Chemical EngineeringUniversity of CaliforniaSanta Barbara, California

Yogeshvar N. KaliaSchool of PharmacyUniversity of GenevaGeneva, Switzerland

Narayanasamy KanikkannanPaddock Laboratories, Inc.Minneapolis, Minnesota

Pankaj KarandeDepartment of Chemical EngineeringUniversity of CaliforniaSanta Barbara, California

S. Kevin LiCollege of PharmacyUniversity of UtahSalt Lake City, Utah

Horace L. LumpkinU.S. Army Medical Research Institute


Barbara LuppiDepartment of Pharmaceutical SciencesUniversity of BolognaBologna, Italy

Howard I. MaibachDepartment of DermatologyUniversity of CaliforniaSan Francisco, California

Babu M. MediDelSite Biotechnologies, Inc.Irving, Texas

Zhinan MeiCollege of Life ScienceSouth-Central University for NationalitiesWuhan, PR China

Bozena B. MichniakDepartment of Pharmacology &

PhysiologyUniversity of Medicine and Dentistry

of New JerseyNewark, New Jersey

Smith and Maibach / Percutaneous Penetration Enhancers 2nd edn 2152_C000 Final Proof page viii 26.9.2005 8:05pm

viii & Contributors


John A. MiksztaBD TechnologiesResearch Triangle Park, North Carolina

Samir MitragotriDepartment of Chemical EngineeringUniversity of CaliforniaSanta Barbara, California

Shozo MiyazakiFaculty of Pharmaceutical SciencesHealth Sciences University of HokkaidoHokkaido, Japan

Aarti NaikSchool of PharmacyUniversity of GenevaGeneva, Switzerland

Sari PappinenDepartment of PharmaceuticsUniversity of KuopioKuopio, Finland

Carolina PellandaDepartment of PharmacyUniversity of BaselBasel, Switzerland

Gerald E. PierardDepartment of DermatopathologyUniversity Hospital of LiegeLiege, Belgium

Claudine Pierard-FranchimontDepartment of DermatopathologyUniversity Hospital of LiegeLiege, Belgium

Mark R. PrausnitzSchool of Chemical and Biomolecular

EngineeringGeorgia Institute of TechnologyAtlanta, Georgia

Carryn H. PurdonCollege of PharmacyUniversity of South CarolinaColumbia, South Carolina

Jennifer Raeder-Devens3M CenterSt. Paul, Minnesota

Jagdish SinghDepartment of Pharmaceutical

SciencesNorth Dakota State UniversityFargo, North Dakota

Mandip SinghCollege of Pharmacy and

Pharmaceutical SciencesFlorida A&M UniversityTallahassee, Florida

Kenneth B. SloanDepartment of Medicinal

ChemistryUniversity of FloridaGainesville, Florida

Eric W. SmithCollege of PharmacyUniversity of South CarolinaColumbia, South Carolina

Robert S. StevensonU.S. Army Medical Research Institute

of Chemical DefenseAberdeen Proving Ground,

Maryland

Christian SurberDepartments of Dermatology

and PharmacyUniversity of BaselBasel, Switzerland

Rashmi A. ThakurErnest Mario School of PharmacyRutgers UniversityPiscataway, New Jersey

Elka TouitouSchool of PharmacyThe Hebrew University

of JerusalemJerusalem, Israel

Smith and Maibach / Percutaneous Penetration Enhancers 2nd edn 2152_C000 Final Proof page ix 26.9.2005 8:05pm

Contributors & ix


Emmanuelle UhodaDepartment of DermatopathologyUniversity Hospital of LiegeLiege, Belgium

Arto UrttiDepartment of PharmaceuticsUniversity of KuopioKuopio, Finland

Daya D. VermaBouve College of Health SciencesNortheastern UniversityBoston, Massachusetts

Yiping WangErnest Mario School of PharmacyRutgers UniversityPiscataway, New Jersey

Scott C. WasdoDepartment of Medicinal ChemistryUniversity of FloridaGainesville, Florida

Jin ZhangCollege of PharmacyUniversity of South CarolinaColumbia, South Carolina

Smith and Maibach / Percutaneous Penetration Enhancers 2nd edn 2152_C000 Final Proof page x 26.9.2005 8:05pm

x & Contributors


Editors

Eric W. Smith, Ph.D., is currently an associate professor of pharmaceutics in the Collegeof Pharmacy at the University of South Carolina in Columbia. Dr. Smith is a graduate ofRhodes University, South Africa, and pursued a postdoctoral fellowship in dermatology atthe University of California, San Francisco. He has served on the pharmacy faculties ofRhodes University, Ohio Northern University, and the University of South Carolina.Among other learned associations, Dr. Smith is a member of the American Associationof Pharmaceutical Scientists and the South Carolina Cancer Center. Dr. Smith has been therecipient of various fellowship and research grants throughout his career. He has numer-ous publications, conference presentations, and a reference text covering aspects oftransdermal drug delivery. His research interests are the in vitro and in vivo assessmentand optimization of drug delivery from topical formulations, topical delivery vehicledesign, penetration enhancement and retardation, and topical formulation bio-equivalence testing.

Howard I. Maibach, M.D., is a professor of dermatology at the University of California,San Francisco, and has been a long-term contributor to experimental research in derma-topharmacology and to clinical research on contact dermatitis, contact urticaria, andother skin conditions.

Smith and Maibach / Percutaneous Penetration Enhancers 2nd edn 2152_C000 Final Proof page xi 26.9.2005 8:05pm


Preface

It has been ten years since the first edition of Percutaneous Penetration Enhancers waspublished. At that time we expected to see an explosion in the number of chemicalenhancers researched and developed for commercial formulations. Surprisingly, this hasnot been the case; at this point there are still only a handful of chemical entities that areclose to realizing this goal. In the first edition we suggested that the full impact ofpenetration enhancer species on transdermal delivery may not become evident simplybecause of the costs associated with regulatory registration formalities. It now appearsthat this suggestion may have held more validity than we initially believed. This theorymay be corroborated by the evidence of dramatic growth and innovation in the field ofphysical (rather than chemical) penetration enhancement systems. Several commercialunits utilizing physical enhancement mechanisms, spanning the full spectrum fromiontophoresis to microneedle devices, are in the final stages of development and testing.On the other hand, there is some renewed interest in transdermal penetration retardationto limit the absorption of chemicals through the skin. These retardation systems are basedon the biochemical groundwork established by enhancer studies in the past. To assistwith all these research efforts, our analytical, bioengineering, and predictive systemscontinue to become ever more sophisticated to the point that much laboratory wet-workcan now be replaced by computer-assisted systems. The field is clearly evolving andredefining itself, and it is therefore timely to attempt to summarize our current know-ledge. To this end we have assembled a list of researchers who are authorities in theirrespective disciplines — this volume is an elegant summary of their recent researchefforts in the ever-broadening fields of topical penetration, enhancement, and retard-ation. We are thankful to each author for their individual contributions to this volume. Wehope that readers will find these chapters useful in establishing the broad framework forthe topic and a stimulant for continued research in the diverse areas of percutaneouspenetration enhancement.

Eric W. SmithUniversity of South Carolina

Columbia, South CarolinaHoward I. Maibach

University of California — San FranciscoSan Francisco, California

Smith and Maibach / Percutaneous Penetration Enhancers 2nd edn 2152_C000 Final Proof page xiii 26.9.2005 8:05pm


INTRODUCTION I

Smith and Maibach / Percutaneous Penetration Enhancers 2nd edn TF2152_c001 Final Proof page 1 13.9.2005 2:57pm


Chapter 1

Penetration EnhancerClassification

Brian W. Barry

CONTENTS

Introduction ........................................................................................................................................ 4Drug Transport Routes through Human Skin ................................................................................... 4Enhancing Transdermal Drug Delivery ............................................................................................. 5

Interactions between Drug and Vehicle ........................................................................................ 5Selection of Correct Drug or Prodrug........................................................................................ 5Chemical Potential Adjustment .................................................................................................. 6Ion Pairs and Complex Coacervates .......................................................................................... 7Eutectic Systems.......................................................................................................................... 7

Vesicles and Particles...................................................................................................................... 7Liposomes and Analogs.............................................................................................................. 7High-Velocity Particles................................................................................................................ 8

Stratum Corneum Modified............................................................................................................ 8Hydration .................................................................................................................................... 8Chemical Enhancers ................................................................................................................... 9

Stratum Corneum Bypassed or Removed.................................................................................... 11Microneedle Array .................................................................................................................... 11Stratum Corneum Ablated ........................................................................................................ 11Follicular Delivery..................................................................................................................... 13

Electrically Assisted Techniques................................................................................................... 13Ultrasound (Phonophoresis, Sonophoresis)............................................................................ 13Iontophoresis ............................................................................................................................ 13Electroporation ......................................................................................................................... 13Magnetophoresis....................................................................................................................... 14Radio Waves.............................................................................................................................. 14Photomechanical Wave ............................................................................................................ 14

References......................................................................................................................................... 14


3


Introduction

This chapter presents a brief overview of the topics dealt with in more detail in sub-sequent contributions to this book, that is, the major ways by which scientists attemptto overcome the highly impermeable nature of human skin so as to deliver drugs atclinically active body concentrations. The organization of the material essentially followsthe review published in 2001.1 To conserve space, most of the 200 references from thatwork will not be repeated here, nor those listed in subsequent chapters in this book.

Drug Transport Routes through Human Skin

Human skin selectively and effectively inhibits chemical penetration.2 The most import-ant control element is generally the stratum corneum and accelerant techniques usuallytry to reduce this barrier’s hindrance so as to maximize drug flux, although occasionallythe follicular route may also be relevant.

At the skin surface, a molecule has three possible routes to reach the viable tissue: viahair follicles with their sebaceous glands, through eccrine sweat ducts, or across thecontinuous horny layer (Figure 1.1). Because of the low fractional appendageal area(about 0.1%), except for ions and highly polar molecules that struggle to cross intact

appendages may function as shunts, which may be important at short times prior tosteady-state diffusion. Additionally, polymers and colloidal particles can target thefollicle.

The main barrier is thus the intact horny layer with its ‘‘brick and mortar’’ structure3

Figure 1.1 Simplified diagram of skin structure and macroroutes of drug penetration: (1) via thesweat ducts, (2) across the continuous stratum corneum, or (3) through the hair follicles withtheir associated sebaceous glands.


4 & Percutaneous Penetration Enhancers


stratum corneum, this pathway usually adds little to steady-state drug flux. However,

(Figure 1.2). The ‘‘bricks’’ of hydrated keratin in the corneocytes distribute in a ‘‘mortar,’’

consisting of lipid bilayers of ceramides, fatty acids, cholesterol, and cholesterol esters.Most transdermal molecules penetrate through this intercellular microroute and thereforemany accelerant methods disrupt or bypass these crystalline, semicrystalline, gel, andliquid crystal domains.

Enhancing Transdermal Drug Delivery

stratum corneum.

Interactions between Drug and Vehicle

Selection of Correct Drug or Prodrug

If at all possible, we choose a drug possessing the optimal physicochemical properties totranslocate well across skin, and our transdermal problems essentially evaporate. Thesimple equation for steady-state flux is useful when considering factors controlling stratumcorneum permeation rates (Equation (1.1)). When we plot the cumulative mass of diffu-sant, m, passing per unit area through a membrane, at long times the graph approacheslinearity and its slope yields the steady flux, dm/dt, as in the following Equation:

dm

dt¼ DC0K

h(1:1)

where C0 represents the constant donor drug concentration; K, the partition coefficient ofsolute between membrane and bathing solution; D, the diffusion coefficient; and h, themembrane thickness.

Figure 1.2 Simplified diagram of stratum corneum and two microroutes of drug penetration.


Penetration Enhancer Classification & 5


Figure 1.3 summarizes some techniques for overcoming the barricade offered by an intact

From Equation (1.1), we can assess the ideal properties needed for a molecule topenetrate stratum corneum well. These are: low molecular mass, solubility in oil andwater, high but balanced (optimal) partition coefficient, and a low melting point, correl-ating with good solubility as predicted by ideal solubility theory.

However, saturated systems of most drugs fail to provide adequate topical bioavail-abilities, and then we must have recourse to other approaches.

Chemical Potential Adjustment

An alternative form of Equation (1.1) uses thermodynamic activities4

dm

dt¼ aD

gh(1:2)

where a is the thermodynamic activity of penetrant in its vehicle and g is its effectiveactivity coefficient in the skin membrane. For the greatest flux, the drug should operate atits maximum thermodynamic activity. Dissolved molecules in saturated solution equili-brate with pure solid (defined as maximum activity for an equilibrated system) and theyare also thus at maximum activity. Therefore, all vehicles containing drug as a finelyground suspension should produce the same penetration rate, provided that the systemsbehave ideally, that is, D, g, and h remain constant.

Supersaturated solutions may form, either by design or by uncontrolled evaporationon the skin; in either situation, the theoretical maximum stratum corneum uptake andflux may increase many-fold compared to a stable system.5 The practical problem withusing this approach is, of course, how do we maintain a suitable period of metastabilityon storage?

EutecticSystems

Ion Pairs orCoacervates

ChemicalPotential

Drug or Prodrug

Drug or VehicleInteractions

High VelocityParticles

Liposomes andAnalogs

Vesicles & Particles

Chemical Enhancers

Hydration

Stratum CorneumModified

FollicularDelivery

Ablation

MicroneedleArray

Stratum CorneumBypassed or Removed

PhotomechanicalWaves

Radio waves

Magnetophoresis

Electroporation

Iontophoresis

Ultrasound

Electrically AssistedMethods

ENHANCING TRANSDERMAL DRUG DELIVERY

Figure 1.3 Some methods for enhancing transdermal drug therapy.




Ion Pairs and Complex Coacervates

Charged species do not readily penetrate lipid membranes. One enhancement approachuses an oppositely charged species to form a lipophilic ion pair. As charges temporarilyneutralize, the complex partitions into the stratum corneum lipids. The ion pair diffusesin to the interface between the horny layer and viable epidermis, dissociates into itscharged species, which partition into the aqueous epidermis and diffuse onward. Asimilar process, complex coacervation, is the phenomenon whereby oppositely chargedions separate into an oil phase, rich in ionic complex. The coacervate partitions intohorny layer, where it behaves as ion pairs, diffusing, dissociating, and passing into viabletissues. Generally, for either process, any enhancement derived is rather modest.

Eutectic Systems

The eutectic mixture of lidocaine and prilocaine in EMLA cream provided formulationadvantages6 for a successful product that encouraged the study of such systems forother drugs, such as ibuprofen and propranolol (as well as lidocaine) interacting withterpenes.

Vesicles and Particles

Liposomes and Analogs

Most early reports on traditional liposomes when applied to skin propose a localizingeffect; the vesicles deposit their enclosed drugs in the upper layers of the stratumcorneum or pilosebaceous unit. Generally, liposomes were not expected to penetrateinto viable skin. How well vesicles transport drugs through the skin is still the subject ofconsiderable debate.

The introduction by Cevc of Transfersomes1 (recently reviewed7) that incorporate‘‘edge activators’’ excited much interest. Their inventor argues that such ultradeformablevesicles squeeze through pores in stratum corneum that are less than one-tenththe liposome’s diameter. Two features are claimed to be important. Transfersomesrequire a hydration gradient to encourage skin penetration (nonoccluded conditions);the gradient operating from the (relatively) dry skin surface towards waterlogged viable

under in vivo conditions. Data indicate that as much as 50% of a topical dose of a proteinor peptide (such as insulin) penetrate skin in vivo in 30 min.

Other investigators, such as Barry and his colleagues, investigated drug delivery fromultradeformable liposomes and traditional vesicles, using open and occluded conditionsin vitro. Both types raised maximum flux and skin deposition compared to saturatedaqueous drug solution (maximum thermodynamic control) under a nonoccluded envir-onment, but results were not as dramatic as detailed in earlier work. Five potentialmechanisms of action of these liposomes were assessed:

1. A free drug process — the liposome releases the drug, which independentlypermeates skin.

2. Vesicles release their lipids which then act as penetration enhancers with respectto the skin lipids.

3. Improved skin uptake of drug.




tissues drives Transfersomes through the horny layer (Figure 1.4). They also work best

4. The different entrapment efficiencies of the liposomes control drug input.5. Deep penetration of stratum corneum by intact liposomes.

As developed by Touitou, ethosomes (liposomes with a high ethanol content) pene-trate and deliver compounds into the skin as the alcohol fluidizes both ethosomal lipids

penetrate through the disorganized lipid bilayers.Niosomes use nonionic surfactants to form vesicles. Flexible ones, as investigated by

the Bouwstra group, consist of a mixture of a bilayer-forming molecule (stabilizer) anda micelle-forming component (destabilizer) and penetrate to the deeper layers of thestratum corneum.

High-Velocity Particles

The PowderJect system fires solid nanoparticles through the horny layer into viable tissues,driven by a supersonic shock wave of helium. Although many advantages were claimed forthis delivery system (e.g., freedom from pain and needle phobia, improved efficacy andbioavailability, targeting, controlled release, accurate dosing, and safety), there have beenproblems with bruising and particles bouncing off skin surfaces. Regulatory authoritiesmay be concerned by the damage caused by high-velocity particles breaking throughthe horny layer (Figure 1.2) and also allowing extraneous contaminants such as bacteriato enter into living tissues. Commercial work is now concentrating on vaccine delivery.8

The Intraject is a development of the vaccine gun designed to deliver liquids throughskin without using needles.9 It is surprising that, after the intensive use of similar devicesfor vaccination, such as by the U.S. military during the Vietnam conflict, it was not earlierdeveloped for drug delivery.

Stratum Corneum Modified

Hydration

Most (but not all) substances penetrate better through hydrated stratum corneum; wateropens up its compact structure of horny layer. Moisturizing factors, occlusive films, andpatches, together with hydrophobic ointments, all enhance topical bioavailability.

Figure 1.4 Ultradeformable Transfersome squeezing through minute pores in the stratumcorneum, driven by the water concentration gradient. The liposome with edge-activator thuspenetrates from the horny layer surface (relatively dry) to the aqueous viable tissues.




and those in the intercellular bilayers (Figure 1.2). The soft, malleable vesicles then

Chemical Enhancers

Substances that temporarily reduce skin resistance (also known as accelerants or sorptionpromoters) thereby enhance drug passage. Examples include water, hydrocarbons, sulf-oxides (especially dimethylsulfoxide [DMSO]) and their analogs, pyrrolidones, fatty acids,esters and alcohols, azone and its derivatives, surfactants (anionic, cationic, and non-ionic), amides (including urea and its derivatives), polyols, essential oils, terpenes andderivatives, oxazolidines, epidermal enzymes, polymers, lipid synthesis inhibitors, bio-degradable enhancers, and synergistic mixtures.

For safety, effectiveness, and cheapness, the best penetration enhancer is water. Anychemical that is nondamaging, pharmacologically inactive, and which promotes stratumcorneum hydration is a penetration enhancer. Examples include the natural moisturizingfactor and urea.

One simple classification of enhancers is through the lipid–protein partitioning (LPP)concept that provides an easy way both to categorize chemical accelerants and torationalize their different modes of action.10,11 This hypothesis proposes that promotersoperate in one or more of three main ways (see Figure 1.5).

FissureIntact stratum corneum

Split stratum corneum

(b)

Keratin fibers Keratin denatured

Vacuoles

Enhancerentry

(c)

Lipidenhancer

Lipid extraction

Invertedmicelle

Phase separation

Waterpool

Polarenhancer

Polarity alteration

Polarheadgroups

Lipidtails

Fluidization

Phase separation

Fluidization

(a)

Intercellularlipid bilayer

Figure 1.5 Some actions of penetration enhancers on human stratum corneum: (a) Action atintercellular lipids. Some of the ways by which chemical penetration enhancers attack andmodify the structured intercellular lipid domain of the stratum corneum. (b) Action at desmo-somes and protein structures. Such dramatic disruption by accelerants (particularly potentsolvents) as they split the stratum corneum into additional squames and individual cells wouldbe clinically inappropriate. (c) Action within corneocytes. Swelling, further keratin denaturation,and vacuolation within individual horny layer cells would not be so drastic but would usually becosmetically challenging. (Reprinted with permission from Barry, B.W., Nature Biotechnology,22, 165, 2004.)




Lipid action. The enhancer disrupts lipid organization within the stratum corneum,making it more permeable and increasing the penetrant’s diffusion coefficient (Equation(1.1)). Many enhancers operate mainly in this fashion (e.g., azone, terpenes, fatty acids,DMSO, and alcohols). They may or may not mix homogeneously with the endogenouslipids.

Good solvents such as DMSO and ethanol, as well as micellar solutions, may alsoextract lipids, making the horny layer more permeable as aqueous channels form.

Protein alteration. Compounds such as ionic surfactants, decylmethylsulfoxide, andDMSO can open up the dense keratin structure in corneocytes, increasing its permeabilityby again raising the appropriate diffusion coefficient (Equation (1.1)). Such moleculesmay also modify peptide or protein material in the bilayer domain, and even split the

Partitioning promotion. The entry of solvent into the stratum corneum alters thechemical environment and thus may increase partitioning of a companion molecule(drug, coenhancer, or cosolvent) into the horny layer (i.e., raise K in Equation (1.1)).

Many chemical enhancers combine these three LPP mechanisms. Thus, DMSO (above60%) disturbs intercellular organization, extracts lipids, interacts with keratin, and pro-motes partitioning of lipid drugs.

As for other routes of drug delivery, researchers have investigated structure–activityrelationships. Terpenes and sesquiterpenes have been investigated and other attemptswere based on factors such as chain length, polarity, unsaturation, and the presence ofspecial groups. Another technique uses a conceptual diagram of three areas based on theaccelerants’ organic and inorganic characteristics — first region for solvents, the secondfor hydrophilic drugs, and the third for lipophilic compounds.

Unfortunately, despite various in silico attempts, we still cannot predict theoreticallywhat safe enhancer to use with a particular drug to achieve a satisfactory clinical result.Many potent enhancers irritate tissues as they interfere with viable cell membranes.Formulators therefore often limit their choice of a suitable enhancer to materialsknown to be gentle to the skin, for example, generally regarded as safe (GRAS) sub-stances. The metered-dose transdermal spray adopts this approach while incorporatingsunscreens as enhancers in a volatile:nonvolatile vehicle that provides accurate andprecise dosing.12 However, multiple time-consuming skin experiments are still necessaryto develop suitable formulations that will satisfy drug regulatory bodies. The challengethen is: how can we screen many possibilities within a reasonable time? We do know that,in general, enhancer mixtures are more efficient than single chemicals. Karande andcoworkers13 therefore recently introduced the concept of an in vitro process for skinimpedance high-throughput screening. The technique claims to be more than 100-foldmore efficient than current screening methods; it provides what the authors term assynergistic combinations of penetration enhancers (SCOPE) formulations. They selected32 enhancers from 100 chemicals reported in the literature. They then assessed 5040binary formulations in 50% ethanol/buffer, four times each, using conductivity measure-ments in vitro with porcine skin, yielding more than 20,000 measurements. (Note thatethanol itself can be an enhancer,14 but this was allowed for in the control.) The leadinghits were then evaluated for their irritation potential using Epiderm cell culture. Potentand safe enhancer mixtures (SCOPE formulations) were selected for flux measurementswith candidate drugs. Finally, the best formulations were assessed for bioavailability andsafety in vivo in hairless rats.

Ninety-eight percent of candidate formulations were eliminated based on poor po-tency, 99.5% were discarded after irritation studies, the remaining 0.5% was tested for flux




stratum corneum, a clinically unacceptable process (Figure 1.5).

enhancement, and 0.02% was finally assessed for bioavailability. The investigatorsdiscovered rare mixtures of enhancers that increased the skin permeability to macromol-ecules, such as heparin, leutenizing hormone releasing hormone, and an oligonucleo-tide, by up to 100-fold, without irritating the skin. The two most successful SCOPEformulations were a mixture of sodium laureth sulfate with phenyl piperazine and a

A challenge for the future would be to elucidate why the areas of potency hot spotswere so restricted, and the fundamental molecular mechanisms producing the enhance-ment. Examination of the molecular structures of the most successful SCOPE mixtures, asillustrated in Figure 1.6, suggests that surface-active phenomena may play a crucial role.

In recent years, investigators have combined chemical enhancers with other promot-ing techniques, such as ultrasound, iontophoresis, and electroporation.

Metabolic interventions use strategies that interfere with barrier homeostasis.15,16

They attack the processes of synthesis, assembly, secretion, activation, processing, orassembling or disassembling of the extracellular lamellar membranes in the stratumcorneum. However, the idea of challenging barrier homeostasis for a significant timebrings in many clinical considerations and possible regulatory problems, as stratumcorneum is a ‘‘smart’’ material that responds to the environment.17

Stratum Corneum Bypassed or Removed

Microneedle Array

The stratum corneum can be bypassed by injection, and many years ago attempts weremade to develop devices based on multiple tiny needles, but these were abandonedbecause of breakage in the skin. More recently, as fabrication techniques and materialshave improved, a similar approach has developed a device of 400 microneedles thatinsert drug just below the horny layer. The solid silicon needles (coated with drug) orhollow metal needles (filled with drug solution) penetrate the stratum corneum; thefeeling is rather like sharkskin, or a cat’s tongue, rubbing against the skin. Drug fluxincreases up to 100,000-fold are claimed. The Macroflux1 technology of the Alza Cor-poration similarly uses a thin titanium screen with precise microprojections (approxi-mately 200 mm long) to transport macromolecules into the skin; the technique may alsobe combined with electrotransport. Microneedles have been used to insert moleculessuch as oligonucleotides, insulin, and protein and DNA vaccines.18

Stratum Corneum Ablated

We could consider simply removing the horny layer. Chemical peels operate at differenttissue layers, dermabrasion employs a motor-driven abrasive fraise or cylinder andmicrodermabrasion uses a stream of aluminum oxide crystals. A new development ofthis technique (termed microscission) drives the aluminum oxide crystals in a stream ofnitrogen into the stratum corneum through a mask, to form microconduits that are 100 to250 mm in diameter and between 50 and 200 mm deep.19 A somewhat different approachemploys high-powered laser pulses to vaporize sections of the horny layer, producingpermeable regions.

Adhesive tape can remove stratum corneum prior to drug application. Tape strippingis also now popular for assessing bioavailability by measuring drug uptake into skin.




combination of N-lauroyl sarcosine with sorbitan monolaurate (Figure 1.6a and b).

A microinfusor device has been proposed to deliver peptides, proteins, and othermacromolecules. Another method forms a suction blister, an epidermatome removesthe raised tissue, and then a morphine solution delivered directly to the exposed dermisquickly relieves pain.

Na+

SO O

O

O−

O

sodium laureth (1 mol) sulfate

NHN

phenyl piperazine

SCOPE formulation: sodium laureth sulfate + phenyl piperazine(a)

N

O

OH

O

N-lauroyl sarcosine

HOOH

OHO

O O

sorbitan monolaurate

SCOPE formulation: N-lauroyl sarcosine + sorbitan monolaurate(b)

Figure 1.6 The two most successful SCOPE formulations13: (a) a mixture of sodium laurethsulfate with phenyl piperazine and (b) a combination of N-lauroyl sarcosine with sorbitanmonolaurate.




Follicular Delivery

The pilosebaceous unit provides a route that bypasses the intact horny layer, representinga target for drug delivery. Even topical application of a macromolecule such as ‘‘naked’’DNA can immunize, and the employment of the hair follicle as a gene therapy targetseems promising. Colloidal particles, such as liposomes and analogs, together with smallcrystals, may target the follicle. In general, particles greater than 10 mm remain on the skinsurface, those that are approximately 3 to 10 mm concentrate in the follicle and thoselesser than 3 mm, penetrate follicles and stratum corneum alike.

Electrically Assisted Techniques

Ultrasound (Phonophoresis, Sonophoresis)

This technique, used originally in physiotherapy and sports medicine, massages a topicalpreparation with an ultrasound source. The low-frequency ultrasonic energy (~20 kHz)

of collapsing vacuum cavities increase free volume space in bimolecular leaflets and thusenhance drug penetration into the horny layer by a thousand-fold.20,21

Investigations have probed many aspects: a possible deactivation of skin enzymes byultrasound, effects of pulsed delivery, synergistic cooperation of ultrasound with ionto-phoresis, penetration enhancers, and electroporation, phonophoresis used to probe therelative contribution of the follicular route to the penetration of hydrophilic permeants,and its potential for the transdermal extraction of blood and tissue analytes.

Iontophoresis

Iontophoresis passes a small direct current (approximately 0.5 mA/cm2) through a drug-containing electrode in contact with the skin; a grounding electrode completes thecircuit. Three main mechanisms promote drug entry: (a) charged species are drivenmainly by electrical repulsion from the driving electrode; (b) the electric current mayincrease the permeability of skin; and (c) electroosmosis may promote passage ofuncharged molecules and large polar peptides. Efficiency of transport depends mainlyon polarity, valency, and mobility of the charged species, as well as electrical duty cyclesand formulation components.

Considerable interest is now being shown in transdermal delivery of therapeuticpeptides, proteins, and oligonucleotides, as well as many other drugs such as lidocaineand fentanyl.

A lidocaine–epinephrine (adrenaline) device for local anaesthesia is now available(the Vyteris system22) and work proceeds on the development of iontophoretic patchsystems, such as the E-Trans1 technology of Alza.23

An interesting development is reverse iontophoresis for clinical sampling. A moleculein the systemic circulation (such as glucose) can be extracted at the skin surface using theelectroosmotic effect; thus the GlucoWatch Biographer monitors blood glucose concen-trations in diabetics using this procedure.

Electroporation

Skin electroporation or electropermeabilization applies short (micro to millisecond)electrical pulses of approximately 100 to 1000 V/cm to generate transient aqueous




disrupts the lipid packing in stratum corneum (see Figure 1.2) by cavitation. Shock waves

corneum, providing pathways for drug delivery. Molecules transport via iontophoresis orelectroosmosis or both while the pulse is on. Between pulses, simple diffusion can allow

resistance. The in vivo application of electroporation is claimed to be well tolerated,although the process usually induces muscle contractions.24,25

Fluxes can increase 10- to 10,000-fold for neutral and highly charged molecules of upto 40 kDa. The process may also transport vaccines, liposomes, nanoparticles, and micro-spheres. Macromolecules and small molecules may sterically stabilize pores created inskin, and thus enhance electroporation flux.

Electroporation may combine with iontophoresis to enhance the penetration ofpeptides such as vasopressin, neurotensin, calcitonin, and LHRH. The combinationhas recently been applied to ultradeformable liposomes.26 Electroporation has alsobeen combined with ultrasound.

Magnetophoresis

Magnetic fields can move diamagnetic materials through skin, and some work hasinvestigated this process.

Radio Waves

A recent technique (the Viaderm device) uses the energy of radiofrequency waves toform micro-channels through the stratum corneum, with the possibility of feedbackcontrol.27 A densely spaced array of microelectrodes takes microseconds to form theholes; applied drug then easily passes into the skin.

Photomechanical Wave

In this procedure, a laser pulse irradiates a black polystyrene target on the skin coveringa drug solution. The resulting photomechanical wave stresses the horny layer andpromotes drug delivery. A single pressure wave can permeabilize the stratum corneumso that macromolecules can penetrate into the deeper skin tissues.28

References1. Barry, B.W., Novel mechanisms and devices to enable successful transdermal drug delivery,

Eur. J. Pharm. Sci., 14, 101, 2001.2. Barry, B.W., Dermatological Formulations: Percutaneous Absorption, Marcel Dekker, New

York and Basel, 1983.3. Barry, B.W. and Williams, A.C., Permeation enhancement through skin, in Encyclopedia of

Pharmaceutical Technology, Vol. 11, Swarbrick J. and Boylan, J.C. (Eds), Marcel Dekker, NewYork and Basel, 449, 1995.

4. Higuchi, T., Physical chemical analysis of percutaneous absorption process, J. Soc. Cosm.Chem., 11, 85, 1960.

5. Pellett, M. et al., The application of supersaturated systems to percutaneous delivery, inTransdermal Drug Delivery, Guy R.H. and Hadgraft, J. (Eds), 2nd ed., Marcel Dekker, NewYork and Basel, 305, 2004.




additional movement as relatively persistent changes in the stratum corneum lower its

pores in the lipid bilayers (Figure 1.2). These pores travel straight through the stratum

6. Nyqvist-Mayer, A.A., Brodin, A.F., and Frank, S.G., Drug release studies on an oil–wateremulsion based on a eutectic mixture of lidocaine and prilocaine as the dispersed phase,J. Pharm. Sci., 75, 365, 1986.

7. Cevc, G., Lipid vesicles and other colloids as drug carriers on the skin, Adv. Drug Deliv. Rev.,56, 675, 2004.

8. Bellhouse, B.J. and Kendall, M.A.F., Dermal PowderJect device, in Modified Release DrugDelivery Technology, Rathbone, M.J., Hadgraft, J., and Roberts, M.S. (Eds), Marcel Dekker,New York and Basel, 607, 2003.

9. Levy, A., Intraject: prefilled, disposable, needle-free injection of liquid drugs and vaccines, inModified Release Drug Delivery Technology, Rathbone, M.J., Hadgraft, J., and Roberts, M.S.(Eds), Marcel Dekker, New York and Basel, 619, 2003.

10. Williams, A.C. and Barry, B.W., Penetration enhancers, Adv. Drug Deliv. Rev., 56, 603, 2004.11. Barry, B.W., Breaching the skin’s barrier to drugs, Nat. Biotechnol., 22, 165, 2004.12. Morgan, T.M., Reed, B.L., and Finnin, B.C., Metered-dose transdermal spray, in Modified

Release Drug Delivery Technology, Rathbone, M.J., Hadgraft, J., and Roberts, M.S. (Eds), MarcelDekker, New York and Basel, 523, 2003.

13. Karande, P., Jain, A., and Mitragotri, S., Discovery of transdermal penetration enhancers byhigh-throughput screening, Nat. Biotechnol., 22, 192, 2004.

14. Megrab, N.A., Williams, A.C., and Barry, B.W., Estradiol permeation through human skin andsilastic membrane: effects of propylene glycol and supersaturation, J. Control. Rel., 36, 277,1995.

15. Elias, P.M. et al., The potential for metabolic interventions to enhance transdermal drugdelivery, JID Symp. Proc., 7, 79, 2002.

16. Elias, P.M. et al., Metabolic approach to transdermal drug delivery, in Transdermal DrugDelivery, Guy R.H. and Hadgraft, J. (Eds), 2nd ed., Marcel Dekker, New York and Basel, 285,2004.

17. Menon, G.K., New insights into skin structure: scratching the surface, Adv. Drug Deliv. Rev., 54,S3, 2002.

18. Prauznitz, M.R., Microneedles for transdermal drug delivery, Adv. Drug Deliv. Rev., 56, 581,2004.

19. Herndon, T.O. et al., Transdermal microconduits by microscission for drug delivery andsample acquisition, BMC Med., 2, 12, 2004.

20. Mitragotri, S. and Kost, J., Low-frequency sonophoresis. A review, Adv. Drug Deliv. Rev., 56,589, 2004.

21. Meiden, V., Sonophoresis: ultrasound-enhanced transdermal drug delivery, in TransdermalDrug Delivery, Guy R.H. and Hadgraft, J. (Eds), 2nd ed., Marcel Dekker, New York and Basel,255, 2004.

22. Kalia, Y.N. et al., Iontophoretic drug delivery, Adv. Drug Deliv. Rev., 56, 619, 2004.23. Phipps, J.B. et al., E-Trans technology, in Modified Release Drug Delivery Technology, Rath-

bone, M.J., Hadgraft, J., and Roberts, M.S. (Eds), Marcel Dekker, New York and Basel, 499,2003.

24. Preat, V. and Vanbever, R., Skin electroporation for transdermal and topical delivery, inTransdermal Drug Delivery, Guy R.H. and Hadgraft, J. (Eds), 2nd ed., Marcel Dekker, NewYork and Basel, 227, 2004.

25. Denet, A.-R., Vanbever, R., and Preat. V., Skin electroporation for transdermal and topicaldelivery, Adv. Drug Deliv. Rev., 56, 659, 2004.

26. Essa, E.A., Bonner, M.C., and Barry, B.W., Electrical enhancement of transdermal delivery ofultradeformable liposomes, in Percutaneous Absorption, Bronaugh, R.L. and Maibach, H.I.(Eds), 4th ed., Marcel Dekker, New York and Basel, 2005.

27. Sintov, A.C. et al., Radiofrequency-driven skin microchanneling as a new way for electricallyassisted transdermal delivery of hydrophilic drugs, J. Control. Rel., 89, 311, 2003.

28. Doukas, A.G. and Kollias, N., Transdermal drug delivery with a pressure wave, Adv. DrugDeliv. Rev., 56, 559, 2004.




Chapter 2

Structure–ActivityRelationship of ChemicalPenetration Enhancers

Narayanasamy Kanikkannan, R. J. Babu, and Mandip Singh

CONTENTS

Introduction ...................................................................................................................................... 17Fatty Acids......................................................................................................................................... 18

Effect of Carbon Chain Length..................................................................................................... 18Saturated and Unsaturated Fatty Acids ........................................................................................ 20Branched versus Unbranched Fatty Acids................................................................................... 20Position of Double Bond ............................................................................................................. 21Geometric Isomers........................................................................................................................ 21Number of Double Bonds............................................................................................................ 21

Fatty Alcohols ................................................................................................................................... 22Fatty Acids versus Fatty Alcohols................................................................................................. 24

Terpenes ........................................................................................................................................... 24Pyrrolidones...................................................................................................................................... 26Surfactants......................................................................................................................................... 28Conclusions ...................................................................................................................................... 29References......................................................................................................................................... 29

Introduction

Transdermal drug delivery offers many advantages over the conventional routes ofadministration. Elimination of hepatic first-pass effects, reduced side effects through


17


optimization of the blood concentration profile, and extended duration of activity aresome of the advantages of transdermal delivery. However, the highly organized structureof the stratum corneum forms an effective barrier to the penetration of a diverse rangeof agents, which must be modified if poorly penetrating drugs are to be administered.The stratum corneum consists of dead, anucleate, keratinized cells embedded in a lipidmatrix. The drug molecules have two major routes of passage through the stratumcorneum, passage between the cells (intercellular route) and passage across the corneo-cytes (transcellular route).

The use of chemical penetration enhancers would significantly enhance the numberof candidates suitable for transdermal delivery. According to the lipid protein partitioning(LPP) theory,1 chemical penetration enhancers would act by one or more of three majormechanisms: (a) disruption of the stratum corneum lipid matrix; (b) interaction withintracellular protein; (c) improvement in partitioning of a drug or solvent into the stratumcorneum. The LPP theory was recently extended to recognize: (d) disruption of thecorneocyte envelope by compounds such as phenol, in high concentrations and insome vehicles and hydrocarbons; (e) effects on proteic junctions, such as desmosomes;(f) change in the partitioning between stratum corneum components and the lipid in thediffusion pathway.2,3

Compounds with a wide variety of chemical structures have been evaluated as skinpenetration enhancers. These compounds include fatty acids, fatty alcohols, terpenes,pyrrolidones, surfactants, amides, azone and its derivatives, urea and its derivatives,sulfoxides, alkanes, esters, and cyclodextrins. The differences in the structure and phy-sicochemical properties among each class of the enhancers accounted for their penetra-tion enhancement potencies. Structure–activity relationship (SAR) represents an attemptto correlate the structure or physicochemical property of a compound with its enhance-ment activity. The physicochemical descriptors include molecular shape, size, lipophili-city, hydrophilicity, molecular geometry, and electronic and steric effects, which havestrong influence in the biological activity of the compounds. SAR is currently beingapplied in many disciplines pertaining to drug design, proteomics, and environmentalrisk assessment. In this chapter, the relationship between the chemical structure and skinpermeation enhancement effect of some of the extensively studied chemical penetrationenhancers such as fatty acids, fatty alcohols, terpenes, pyrrolidones, and surfactantshas been discussed.

Fatty Acids

Saturated and unsaturated fatty acids have been established as effective enhancers fortransdermal permeation of drugs.4–7 The SAR of fatty acids is covered in detail in thissection.

Effect of Carbon Chain Length

There are several reports on the effect of carbon chain length of fatty acids on thepercutaneous permeation enhancement of drugs. Aungst et al.4 investigated the effectof carbon chain length of saturated fatty acids (C7–C18) on the penetration of naloxonethrough human skin. As the carbon chain length increased from C7 to C12, there wasan increase in the permeation of naloxone. An increase in the carbon chain length




beyond C12 decreased the flux of naloxone. Maximum permeation was observed withC9–C12.

Ogiso and shintani 8 studied the effect of a series of saturated fatty acids on thepermeation of propranolol through rabbit skin using gel formulations. Lauric acid andmyristic acid were the most effective agents among the fatty acids used in increasing thepermeation of propranolol and the enhancement was significantly larger than those inshort and long chain fatty acids. Lee et al.9 investigated the effect of a series of saturatedfatty acids (C6–C18) and unsaturated fatty acids (oleic and linoleic acid) on the perme-ation of Tegafur across hairless mouse skin. These enhancers were studied using Ethanol/Panasate 800 (40/60) and Ethanol/Water (60/40) systems as vehicles. The fatty acidsenhanced the skin permeation of Tegafur in the Ethanol/Panasate 800 (60:40) binaryvehicle in the following order: oleic acid>C12> linoleic acid>C10>C8>C6>no fattyacid >C14>C16>C18. All fatty acids increased the skin permeation of Tegafur in theEthanol/Water (60:40) binary vehicle. The skin permeation of Tegafur decreased in thefollowing order: C12>C10> linoleic acid>oleic acid>C8>C6>no fatty acid. Theseresults suggest that vehicle plays an important role in the skin permeation enhancementeffect of fatty acids.

The skin permeation enhancement and the skin perturbation effects of a number offatty acids, namely, straight chain saturated, monounsaturated and polyunsaturated acids,were evaluated using human stratum corneum.5 Saturated fatty acids with 6 to 12 carbonsshowed a parabolic correlation between enhancement effect and chain length, with amaximum at nonanoic–decanoic acids (with 9 and 10 carbons). A parabolic relationshipbetween carbon chain length of fatty acids and skin permeation enhancement was alsoobserved with thiamine disulfide,10 testosterone,11 and indomethacin.12

Kandimalla et al.13 investigated the effect of saturated fatty acids (C9–C14) on thepermeation of melatonin across excised rat skin. A sharp increase in the permeation ofmelatonin was observed, as the fatty acid chain length increased from 9 to 10 carbons

chain length was increased to 11. However, the permeation of melatonin decreasedwhen the chain length was increased beyond 11 carbons. It can be observed that thepermeation of melatonin has a parabolic relationship with the chain length of thesaturated fatty acids. In general, medium chain fatty acids have showed greater perme-ation enhancement effect compared to short or long chain fatty acids.

It has been proposed that acids with a certain chain length, that is, around 12 carbons,possess an optimal balance between partition coefficient or solubility parameter andaffinity to skin.8 Shorter chain fatty acids would have insufficient lipophilicity for skinpermeation, whereas longer chain fatty acids would have much higher affinity to lipids instratum corneum and thereby retard their own permeation and that of other permeants.The parallel effect with the permeation enhancement suggests that the mode of actionof saturated fatty acids as enhancers is dependent on their own permeation across thestratum corneum or skin.5

The mechanism by which fatty acids increase skin permeability appears to involvedisruption of the densely packed lipids that fill the extracellular spaces of the stratumcorneum.14,15 The change in the physical structure of stratum corneum lipids has beenassessed using differential scanning calorimetric (DSC) and infrared spectroscopic tech-niques.16,17 Treatment of rabbit stratum corneum with various unsaturated fatty acidsresulted in a shift to higher frequency for the CH asymmetric stretch peak near 2920 cm–1

on FTIR Spectra, which primarily results from the acyl chains of intercellular lipid in thestratum corneum lipid.17


SAR of Chemical Penetration Enhancers & 19


(Figure 2.1). A further increase in the permeation of melatonin was observed when the

Saturated and Unsaturated Fatty Acids

The application of saturated long chain fatty acids (stearic acid [C18], myristic acid [C14],and lauric acid [C12]) as enhancers was studied on the percutaneous transport ofthiamine disulphide from propylene glycol through excised rat skin.10 The permeationof thiamine disulphide was enhanced 31 times by C12 and 1.4 times by C14 andsuppressed to 80% of its original value by C18. However, with unsaturated fatty acids,the permeation of indomethacin was enhanced in the following order:C20>C22>C18¼C16>C14 and the flux values were correlated well with the uptakeof these compounds into the stratum corneum.17 Oleic acid (C18, unsaturated) has beenshown in several studies to be an effective skin permeation enhancer, whereas stearicacid (C18, saturated) is not a good skin permeation enhancer. Chi et al.18 reported anincrease of 6.5- to 17.5-fold in the permeation rate of flurbiprofen by unsaturated fattyacids, while no significant increase was observed with saturated fatty acids. Thus satur-ated and unsaturated fatty acids behave differently on the skin permeation enhancement.

Branched versus Unbranched Fatty Acids

Aungst14 reported that maximum flux of naloxone was observed with C9–C12-branchedand unbranched fatty acids across human skin. The branched and unbranched isomers ofC5–C14 fatty acids showed similar effects. However, isostearic acid [(CH3)2CH(CH2)14

COOH] was a more effective permeation enhancer than stearic acid. The higher

00

400

800

1200

1600

2000

2400

2800

10 20 30 40 50

Time (h)

Cum

ulat

ive

amou

nt p

erm

eate

d (µ

g cm

−2)

Figure 2.1 Effect of saturated fatty acids (5% w/v) on the permeation profile of melatoninthrough rat skin. & Control, ^ nonanoic acid, D decanoic acid, ~ undecanoic acid, ^ lauricacid, & myristic acid. Control is the permeation profile of melatonin from the vehicle withoutenhancer. Data are means+ SE (n 5 3). (From Kandimalla, K., Kanikannan, N., Ardega, S., andSingh, M., J. Pharm. Pharmacol., 51, 783, 1999. With permission.)




permeation enhancement effect of isostearic acid than stearic acid was attributed to itslower melting point and greater solubility in propylene glycol.19

Position of Double Bond

Tanojo et al.5 investigated the effect of position of double bond on the percutaneousabsorption of para amino benzoic acid with human stratum corneum using cis-octadecenoic acid with a double bond at 6th, 9th, 11th, or 13th position counted fromthe carboxyl head group. There was no significant difference in the effect of these acidson the permeation of para amino benzoic acid. Morimoto et al.17 studied the effect ofdouble bond positions of unsaturated fatty acids (C18) on the permeation of indometha-cin through rat skin. The permeation of indomethacin with oleic acid (cis-9), asclepic acid(cis-11), petroselinic acid (cis-6) was not affected by the position of the double bonds.

Geometric Isomers

The effect of geometric isomers of unsaturated fatty acids on the permeation ofindomethacin through rat skin was investigated.17 The indomethacin flux with elaidicacid (trans-9-octadecenoic acid) was significantly lower than that of oleic acid(cis-9-octadecenoic acid). The flux of salicylic acid enhanced by trans-isomers of9-octadecenoic acid was lower than that of their cis-isomers.20 However, there was nosignificant difference between cis- and trans- unsaturated C16–C18 fatty acid isomers intheir effects on naloxone flux across human skin.14 The discrepancy in these results maybe due to the difference in the properties of drugs employed and the variation in the skinspecies used for the studies.

Number of Double Bonds

As the number of double bonds in the C18 fatty acid increased from one (oleic acid)to two (linoleic acid), a significant increase in the flux of naloxone was observed.4 Anincrease in the number of double bonds to three (linolenic acid), however, did notincrease the flux further. Tanojo et al.5 investigated the effect of number of doublebonds (in cis-conformation) in straight chain polyunsaturated acids on the permeationof para amino benzoic acid in human stratum corneum. Polyunsaturated fatty acids suchas linoleic, linolenic, and arachidonic acid with, respectively, two, three, and four doublebonds produced a significantly higher permeation of para amino benzoic acid than themonounsaturated fatty acid. However, there was no significant difference in the perme-ation enhancement effects among the polyunsaturated fatty acids. Carelli et al.21 alsoreported that the enhancement of flux of alprazolam by linoleic acid was greater than thatof oleic acid through hairless mouse skin. However, the flux of indomethacin was notaffected by the number of double bonds.17

Kandimalla et al.13 studied the effect of oleic acid, linoleic, and linolenic acid on the

bonds increased, there was a slight increase in the permeation of melatonin. The flux ofmelatonin with linolenic acid was significantly higher than that of oleic acid (P < 0.05).However, there was no significant difference in the flux values of linoleic acid andlinolenic acid (P> 0.05). Recently, Fang et al.22 studied the effect of oleic acid, linoleic




permeation of melatonin across excised rat skin (Figure 2.2). As the number of double

acid, and linolenic acid on the permeation of flurbiprofen through mouse skin. Thepermeation of flurbiprofen increased with an increase in the number of double bonds inthe fatty acid.

Oleic acid has been reported to be an effective skin penetration enhancer for polar andnonpolar drugs.23–26 Cis-unsaturated fatty acids (e.g., oleic acid, linoleic acid, and linolenicacid) have been reported to form separate domains within stratum corneum lipids thateffectively decrease the diffusional path length or the resistance.27,28 The formation ofseparate domains would provide permeability defects within the bilayer lipids andfacilitate the permeation of hydrophilic permeants. The presence of double bonds in thestructure has been proposed to cause the formation of kinks in the lipid matrix to allowwater permeation across the skin.29 An increase in the number of double bonds increasesthe flux of drugs, possibly by causing more kinks in the lipid structure of skin.

Fatty Alcohols

The effect of saturated alcohols (C8-OH to C18-OH) on the flux of naloxone in propyleneglycol was studied through human skin.4 A parabolic effect of alkyl chain length wasobserved with C10-OH and C12-OH being most effective. The effect of a series of straightchain alkanols on the transdermal delivery of levonorgestrel through excised rat andhuman cadaver skin was investigated by Friend et al.30 The flux of levonorgestrelincreased as the alkyl chain increased from C2 to C4, but decreased as the chain lengthincreased above 1-butanol.

00

500

1000

2000

1500

2500

10 20 30 40 50

Time (h)

Cum

ulat

ive

amou

nt p

erm

eate

d (m

g cm

−2)

Figure 2.2 Effect of unsaturated fatty acids (5% w/v) on the permeation profile of melatoninthrough rat skin. & Control, ^ oleic acid, ~ linoleic acid, & linolenic acid. Control is thepermeation profile of melatonin from the vehicle without enhancer. Data are means+ SE (n 5 3).(From Kandimalla, K., Kanikannan, N., Ardega, S., and Singh, M., J. Pharm. Pharmacol. 51, 783,1999. With permission.)




Lee et al.9 studied the effect of a series of fatty alcohols in Ethanol/Panasate 800 andEthanol/Water on the permeation of Tegafur across hairless mouse skin. All fattyalcohols, except the C18-OH, increased the skin permeation of Tegafur in the Ethanol/Panasate 800 (60:40) binary vehicle. The degree of permeation percentage of Tegafurobtained was same at 12 h (64.1 to 67.9% of dose) in all cases, and no significantdifference between them was observed. However, all fatty alcohols significantly en-hanced the skin permeation of Tegafur with Ethanol/Water (60:40) binary vehicle. Theflux of Tegafur increased with an increase in alkyl chain length, reached a maximumpermeation in C12-OH, then decreased as the alkyl chain length increased further. Theskin permeability of Tegafur was in the following order: C12-OH>C10-OH>C9-OH>C8-OH>C14-OH>C16-OH>C18-OH>no fatty alcohol. Fatty alcohols with9, 10, and 12 carbon atoms showed the greatest permeation percentage of Tegafur at 12 hin the Ethanol/Water (60:40) binary vehicle. These results suggest that vehicle plays animportant role in the permeation enhancement effect of fatty alcohols.

The effect of n-alkanols on the permeation of a polar, nonelectrolyte penetrant,nicotinamide through hairless mouse skin was studied by Kai et al.31 The enhancementversus alkanol chain length profile was parabolic, C6-OH being the maximum. The alkanolflux after a 6-h contact period, versus carbon number, was also a parabolic function.Alkanol uptake on the other hand increased with increasing chain length. The authorssuggested that the primary mechanism by which alkanols increase percutaneous absorp-tion is extraction of stratum corneum intercellular lipids. Sloan et al.32 studied the fluxes oftheophylline through hairless mouse skin from suspensions in straight alkyl chain alka-nols. The flux of theophylline was the lowest from methanol (C1-OH), increased by almost100-fold from pentanol (C5-OH), hexanol (C6-OH), heptanol (C7-OH), octanol (C8-OH)and nonanol (C9-OH), then decreased tenfold from undecanol (C11-OH).

In our laboratory, we studied the effect of saturated fatty alcohols (C8-OH to C14-OH)on the permeation of melatonin across excised hairless rat skin.33 All saturated fattyalcohols increased the permeation of melatonin through hairless rat skin and the perme-ation of melatonin was found to be related to the carbon chain length of the fattyalcohols. An increase in the flux of melatonin was observed when the fatty alcoholchain length increased from 8 to 10 carbons. However, the flux of melatonin decreasedwhen the chain length was increased beyond ten carbons. The maximum permeation ofmelatonin was observed with decanol. The parabolic relationship between carbon chainlength of fatty alcohol and skin permeation enhancement was also observed for testos-terone11 and indomethacin.12

The effect of number of double bonds in the C18 fatty alcohol on the permeation ofnaloxone across human skin was investigated.4 The permeation of naloxone was in-creased with an increase in the number of double bonds. Like fatty acids, fatty alcoholsalso act by disrupting the stratum corneum lipid matrix.15 Recently, the influence ofhydrocarbon chain branching on the effectiveness of alkanol skin permeation enhancershas been investigated using corticosterone as a model drug across hairless mouse skin.34

The branched-chain alkanols showed lower enhancer potency than the 1-alkanols ofthe same molecular formula; the potency decreases as the hydroxyl group moves fromthe end of the chain towards the center of the enhancer alkyl chain. The authors alsoreported that the intrinsic potencies of the 1-alkyl enhancers (1-alkanols, 1-alkyl-2-pyrrolidones, and 1-alkyl-2-azacycloheptanones) are essentially the same and independ-ent of their alkyl chain length at their isoenhancement concentrations.34–36

It has been reported that the most effective chain lengths (C10–C12) correspond tothe length of the steroid nucleus of cholesterol, suggesting that these may act by




disrupting ceramide–cholesterol or cholesterol–cholesterol interaction.37 Ackermannet al.38 studied the permeation of a series of alkanols (C1-OH to C8-OH) across thenude mouse skin. The permeability coefficients of alkanols increased linearly as the chainlength increases. Further, the permeability coefficients of n-alkanols correlated well withtheir ether–water partition coefficients. These results could be used to explain thepermeation enhancement effect of different alkanols. The increase in the enhancementeffect of lower alkanols with increase in the alkyl chain length may be attributed to theincreased permeation of alkanols through the skin.

Fatty Acids versus Fatty Alcohols

Fatty acids have a higher melting point than their corresponding fatty alcohols, but lowersolubility parameters. If the enhancement by these fatty acids and alcohols was solely dueto solubility effects, then it would be expected that the alcohols would be more effectivethan the acids, whereas the reverse is true for alkyl chains up to C18. This suggests thatmore specific interactions must occur.37 Introduction of double bonds into long alkylchains modifies the effect significantly and, for the C18 compounds, there was littledifference between the corresponding fatty acids and alcohols. There was a greaterconcentration dependence of permeation enhancement for lauric acid than laurylalcohol.4

Terpenes

Terpenes are naturally occurring compounds, which consist of isoprene (C5H8) units.Terpenes are classified according to the number of isoprene units they contain: mono-terpenes (C10) have two isoprene units, sesquiterpenes (C15) have three, and diterpenes(C20) have four. The structural formulae of different types of terpenes (hydrocarbon,ketone, alcohol, oxide, and cyclic ether terpenes) evaluated as skin penetration en-

Terpenes have been widely studied as skin penetration enhancers for variousdrugs.1,39–41 Okabe et al.39 studied ten cyclic monoterpenes as penetration enhancersfor lipophilic drug indomethacin in rats. The absorption of indomethacin from gelointment was substantially enhanced by hydrocarbon terpenes such as d-limonene.However, the oxygen containing terpenes did not affect the permeation of indomethacin.The authors concluded that cyclic monoterpenes with lipophilic indices greater than 0were most effective for indomethacin. But the alcohol and ketone terpenes were lesseffective for lipophilic drugs such as diazepam42 and estradiol.43

Williams and Barry1 evaluated a series of terpenes as skin penetration enhancers forthe hydrophilic drug 5-fluorouracil in human skin. Cyclic terpenes were chosen from thechemical classes of hydrocarbons, alcohols, ketones, and oxides. Of the terpenes studied,hydrocarbons were poor enhancers and alcohols and ketones were more effective. Theepoxides showed mild enhancing activity, whereas the cyclic ethers were very effective;ascaridole, 7-oxabicyclo[2.2.1]heptane, and 1,8-cineole all induce a near 90-fold increasein the permeability coefficient of 5-fluorouracil. The five-membered cyclopentene oxideshowed higher enhancing activity than the six-membered cyclohexene oxide.

The effect of 12 sesquiterpenes on the permeation of 5-fluorouracil was evaluatedacross human skin.44 Pretreatment of epidermal membranes with sesquiterpene oils or




hancers are shown in Figure 2.3.

using solid sesquiterpenes saturated in dimethyl isosorbide enhanced the absorption of5-fluorouracil. Enhancers containing polar functional groups were generally more effect-ive than pure hydrocarbons and enhancers with the least ‘‘bunched’’ structures were themost active.

O

Fenchone

O

Pulegone

O

Piperitone

O

Menthone

O

Cyclohexene oxide

O

Limonene oxide

O

Pinene oxide

O

Cyclopentene oxide

OO

Ascaridole

O

7-Oxabicylo-(2-2-1)heptane

O

1,8-Cineole

alpha-Pinene

d-Limonene

3-Carene

OH

alpha-Terpineol

OH

Menthol

OH

Terpinen-4-ol

OH

Carveol

O

Carvone

Figure 2.3 Structural formulae of various types of terpenes (hydrocarbon, ketone, alcohol,oxide, and cyclic ether terpenes) assessed as skin penetration enhancers.




Obata et al.45 reported that percutaneous absorption of hydrophilic diclofenacsodium was substantially enhanced in the presence of l-menthol and dl-menthone,while it was little enhanced by d-limonene and p-menthane. Overall, the skin permeationenhancing effect of terpenes depends on the physicochemical properties of the drugs. Ingeneral, hydrocarbon terpenes are effective for lipophilic drugs and oxygen containingterpenes are effective for hydrophilic drugs.

Okamoto et al.46,47 evaluated the compounds containing azacyclo ring and acyclicterpene hydrocarbon chains as enhancers for a variety of drugs. These studies demon-strated that azacyclo ring size has little effect on the potency of the enhancers, whereasthe length of hydrophobic terpene chain has a significant effect; a chain length of 12carbons provided maximum effect.

El-Kattan et al.48 investigated the effect of terpene lipophilicity (log P 1.06 to 5.36)(terpene-4-ol, verbenone, fenchone, carvone, menthone, alpha-terpineol, cineole, ger-aniol, thymol, cymene, d-limonene, and nerolidol) on the percutaneous absorption ofhydrocortisone from hydroxypropyl methyl cellulose gel formulations using hairlessmouse skin in vitro. A linear relationship was found between the log P of terpene andthe cumulative amount of hydrocortisone in the receptor compartment after 24 h. Anincrease in terpene lipophilicity was associated with an increase in the cumulativeamount of hydrocortisone transported.

The effects of terpene enhancers (fenchone, thymol, d-limonene, and nerolidol) onthe percutaneous absorption of drugs with different lipophilicities (nicardipine hydro-chloride, hydrocortisone, carbamazepine, and tamoxifen) were studied.49 Nerolidol(highest lipophilicity) provided the highest increase in the flux of the model drugs. Thelowest increase in the flux was observed with fenchone (lowest lipophilicity). The resultsindicated that these four enhancers were more effective at enhancing the penetration ofhydrophilic drugs rather than lipophilic drugs.

The synergism of ethyl alcohol and limonene on the permeation enhancement ofindomethacin was examined and it was found to be significant.50 The combined effect ofmenthol and ethanol as skin penetration enhancers was also studied by Kobayashiet al.51 Addition of ethyl alcohol to water and 5% menthol enhanced the drug solubilityin the vehicle, decreased skin polarity, and increased the role of pore pathway to wholeskin permeation. Synergistic action was also observed with terpene or propylene glycolmixture as evaluated by DSC and x-ray diffraction.52,53 The terpenes act mainly bydisrupting the lipid matrix of the stratum corneum.1 Spectroscopic studies havealso suggested that terpenes could exist within separate domains in stratum corneumlipids.52

Pyrrolidones

Pyrrolidones and their derivatives have been investigated as potential skin penetrationenhancers.54–56 2-Pyrrolidone and N-methyl-2-pyrrolidone (NMP) have been evaluatedas penetration enhancers for a variety of drugs.57–59

structures of some pyrrolidones, which have been evaluated as skin penetrationenhancers. Aoyagi et al.60 synthesized a new group of 2-pyrrolidone enhancerscontaining a short alkyl group, such as methyl, ethyl, propyl, or butyl group, at the1-position and a dodecyl group at the 3-position of a 2-pyrrolidone ring. The enhancingeffect of these compounds was studied using indomethacin as a model drug. The lengthof the short alkyl group at the 1-position greatly impacted the enhancing activity of the




Figure 2.4 presents the chemical

2-pyrrolidone derivatives. 1-Propyl and 1-butyl-3-dodecyl-2-pyrrolidone showed thegreatest permeation enhancement effect of indomethacin through the skin.

The skin permeation enhancement activity of a series of alkyl substituted pyrrolidoneswas studied using phenol red as a model drug across rat skin in vitro and in vivo.61–64

A correlation between the flux of phenol red and partition coefficient of the pyrrolidoneswas observed. The percutaneous penetration enhancement of 6-mercaptopurine by nineazacycloalkanone derivatives with an alkyl or terpene chain was studied using excisedguinea pig skin.47 The number of carbonyl groups in the chain influenced the enhancingactivity more effectively than the ring size.

It has been reported that pyrrolidone derivatives alter the liposomal membrane madewith stratum corneum lipid.65 Yoneto et al.66 studied the effects of 1-ethyl, 1-butyl,1-hexyl, 1-octyl-2-pyrrolidones on the transport of beta-estradiol, hydrocortisone, andcorticosterone across hairless mouse skin. The results showed a 3.5-fold increase inenhancement potency per methylene group introduced at the 1-N position. The authorsreported that the 1-alkyl-2-pyrrolidones may act via the intercalation of the alkyl groupof the enhancer into the highly ordered interfacial region of the lipid bilayers, inducingsignificant disorder and enhancing microenvironmental fluidity. The authors studied thefluidizing effects of alkyl pyrrolidones upon the stratum corneum lipid liposome bilayerusing steady-state anisotropy and fluorescence lifetime studies.67 The results suggestedthat the alkyl pyrrolidones might induce a general fluidizing effect upon the lipid bilayer.

NO

2-Pyrrolidone

NO

N-Methyl-2-pyrrolidone

CH3

NO

1-Ethyl-2-pyrrolidone

C2H5

NO

5-Methyl-2-pyrrolidone

H3C H3CN

O

1,5-Dimethyl-2-pyrrolidone

CH3

NO

2-Pyrrolidone-5-carboxylic acid

HOOC

NO

(CH2)4CH3 (CH2)11CH3

(CH2)11CH3

CH3

N-Hexyl-2-pyrrolidone

NO

N-Lauryl-2-pyrrolidone

NO

1-Methyl-3-dodecyl-2-pyrrolidone

Figure 2.4 Structural formulae of pyrrolidone enhancers.




As a continuing effort to understand the mechanism of action, the authors studied theinfluence of the alkyl pyrrolidones on permeant partitioning into hairless mouse stratumcorneum under the isoenhancement concentration conditions using beta-estradiol as themodel drug.68 The results suggested that inducing a higher partitioning tendency forbeta-estradiol into the lipoidal pathway of hairless mouse stratum corneum is a principalmechanism of action of the alkyl pyrrolidones in enhancing percutaneous absorption.

Surfactants

Surfactants generally consist of a lipophilic alkyl or aryl chain with a hydrophilic headgroup. Surfactants may be classified according to the nature of the head group as anionic,cationic, nonionic, or zwitterionic. Surfactants have been used as skin permeation en-hancers in several studies.69–72 In general, cationic surfactants cause greater increase inthe flux of drugs than anionic surfactants, which, in turn, produce greater increasesin flux than nonionic surfactants. Ashton et al.73 compared the effects of dodecyltri-methylammonium bromide (DTAB), sodium lauryl sulfate (SLS), and polyoxyethylenefatty ether (Brij 36Te

ˆ) on the in vitro flux of methyl nicotinamide across excised human

skin. The permeation enhancement of methyl nicotinamide was in the following order:DTAB> SLS>Brij 36T. However, Brij 36T exhibited a smaller but more immediate effecton the permeation of methyl nicotinate, resulting in the highest degree of flux enhance-ment over the first 24-h period.

The effects of various cationic surfactants (alkyl trimethylammonium halides, alkyldimethylbenzylammonium halides, and alkyl pyridinium halides) on the permeation ofradiolabeled water and lidocaine through excised human epidermis have been studied.74

All surfactants increased the mean steady-state flux of water and lidocaine by two tofourfold compared to the initial control period. However, there was no significantdifference in the enhancing effects of these three hexadecyl derivatives. The maximumflux enhancement was observed from those derivatives with an alkyl chain length of 12 to14 carbons. Cooper and Berner75 reported that the optimal chain length for skin barrierimpairment might be attributed to the factors such as solubility of the surfactant in thedonor vehicle, the critical micellar concentration, the stratum corneum–hyphen;vehiclepartition coefficient, and the binding affinity of the surfactant for epidermal keratin. Anoptimum chain length of 12 to 14 carbons may represent compromise between watersolubility and lipophilic character. Furthermore, stratum corneum keratin may bindpreferentially with carbon chains of specific length.

Cappel and Kreuter76 compared the enhancement potential of polysorbates 20, 21, 80,and 81. The results of these studies showed that polysorbates had a lesser effect on thetransdermal permeation of methanol. Maximum permeation enhancement was achievedin the presence of polysorbates 21 and 81 enhanced the permeation of methanol of twoto threefold, indicating that the more lipophilic polysorbates alter the barrier properties ofthe skin to a greater extent than their hydrophilic analogs.

Lopez et al.76 studied the influence of the polar functional group on the skin perme-ation enhancement effects of nonionic surfactants. Their results indicated that the natureof the enhancer head group greatly influences cutaneous barrier impairment. Span120showed greater permeation enhancement of all compounds compared to Tween120.Ionic surfactants interact well with keratin filaments in the corneocytes and make it morepermeable and increase the diffusion coefficient of the drug.15 Surfactants may alsomodify peptide or protein material in the bilayer domain.1




Conclusions

Extensive research has been undertaken to study the effects of a variety of chemicalcompounds as skin penetration enhancers. The list of potential drugs that can be effect-ively delivered via transdermal route is increasing. Structure–permeation enhancementrelationship studies have increased our understanding of the effect of penetration en-hancers for different types of drugs. The chemical structures and physicochemical prop-erties of penetration enhancers play an important role in their permeation enhancementeffects. In general, a parabolic relationship between the carbon chain length of fatty acidsand fatty alcohols and skin permeation enhancement has been observed. The unsaturatedfatty acids have shown a greater permeation enhancement effect compared to theircorresponding saturated fatty acids. The hydrocarbon terpenes have been found to bemore effective for lipophilic drugs and oxygen containing terpenes are more effective forhydrophilic drugs. The chain length of the alkyl pyrrolidone enhancers plays an importantrole in their skin permeation enhancement potencies. In general, ionic surfactants showeda greater flux of drugs than nonionic surfactants. The permeation enhancement effectof enhancers is also greatly influenced by the physicochemical properties of the drug.Unfortunately, many of the chemical penetration enhancers that showed good permeationenhancement effect also cause skin irritation.33,77 The practical use of chemical penetrationenhancers requires careful balancing of their benefits and risks, that is, penetration ratesand irritation. Further studies are needed in the areas of evaluation of skin permeationenhancement vis-a-vis skin irritation in order to choose penetration enhancers, whichpossess optimum enhancement effect with no skin irritation. Further studies are alsoneeded to understand the mechanism of action of chemical penetration enhancers.

References1. Williams, A.C. and Barry, B.W., Terpenes and the lipid–protein partitioning theory of skin

penetration enhancement, Pharm. Res., 8, 17, 1991.2. Menon, G.K., Lee, S.H., and Roberts, M.S., Ultrastructural effects of some solvents and vehicles

on the stratum corneum and other skin components: evidence for an ‘‘extended mosaicpartitioning model of the skin barrier,’’ in Dermal Absorption and Toxicity Assessment, Roberts,M.S. and Walters, K.A. (Eds), Marcel Dekker, New York, 1998, p. 727.

3. Magnusson, B.M., Walters, K.A., and Roberts, M.S., Veterinary drug delivery: potential for skinpenetration enhancement, Adv. Drug Del. Rev., 50, 205, 2001.

4. Aungst, B.J., Rogers, N.J., and Shefter, E., Enhancement of naloxone penetration throughhuman skin in vitro using fatty acids, fatty alcohols, surfactants, sulfoxides and amides.Int. J. Pharm., 33, 225, 1986.

5. Tanojo, H., Bouwstra, J.A., Junginger, H.E., and Bodde, H.A., In-vitro human skin barriermodulation by fatty acids: skin permeation and thermal analysis studies, Pharm. Res., 14, 42,1997.

6. Oh, H.J., Oh, Y.K., and Kim, C.K., Effects of vehicles and enhancers on transdermal delivery ofmelatonin, Int. J. Pharm., 212, 63, 2001.

7. Thomas, N.S. and Panchagnula, R., Combination strategies to enhance transdermal permeationof zidovudine (AZT), Pharmazie, 58, 895, 2003.

8. Ogiso, T. and Shintani, M., Mechanism for the enhancement effect of fatty acids on thepercutaneous absorption of propranolol, J. Pharm. Sci., 79, 1065, 1990.

9. Lee, C.K., Uchida, T., Noguchi, E., Kim, N-S., and Goto, S., Skin permeation enhancement oftegafur by ethanol/panasate 800 or ethanol/water binary vehicle and combined effects of fattyacids and fatty alcohols, J. Pharm. Sci., 82, 1155, 1993.




10. Komata, Y., Inaoka, M., Kaneko, A., and Fujie, T., In vitro percutaneous absorption of thiaminedisulfide from a mixture of propylene glycol and fatty acid, J. Pharm. Sci., 81, 744, 1992.

11. Yu, J., Chien, T., and Chien, Y.W., Transdermal dual-controlled delivery of testosterone andestradiol: (II) enhanced skin permeability and membrane moderated delivery, Drug Dev. Ind.Pharm., 17, 1905, 1991.

12. Chien, Y.W., Xu, H., Chiang, C.C., and Huang, Y.C., Transdermal controlled administration ofindomethacin. I. Enhancement of skin permeability, Pharm. Res., 5, 103, 1988.

13. Kandimalla, K., Kanikkannan, N., Andega, S., and Singh, M., Effect of fatty acids on thepermeation of melatonin across rat and pig skin in vitro and on the transepidermal waterloss in vivo, J. Pharm. Pharmacol., 51, 783, 1999.

14. Aungst, B.J., Structure/effect studies of fatty acid isomers as skin penetration enhancers andskin irritants, Pharm. Res., 6, 244, 1989.

15. Barry, B.W., Novel mechanisms and devices to enable successful transdermal drug delivery,Eur. J. Pharm. Sci., 14, 101, 2001.

16. Golden, G.M., Guzek, D.B., Harris, R.R., McKie, J.E., and Potts, R.O., Lipid thermotropictransitions in human stratum corneum, J. Invest. Dermatol., 86, 255, 1986.

17. Morimoto, K., Tojima, H., Haruta, T., Suzuki, M., and Kakemi, M., Enhancement effects ofunsaturated fatty acids with various structures on the permeation of indomethacin through ratskin, J. Pharm. Pharmacol., 48, 1133, 1996.

18. Chi, S.-C., Park, E.-S., and Kim, H., Effect of penetration enhancers on flurbiprofen permeationthrough rat skin, Int. J. Pharm., 126, 267, 1995.

19. Aungst, B.J., Fatty acids as skin permeation enhancers, in Percutaneous Penetration En-hancers, Smith, E.W. and Maibach, H.I. (Eds), CRC Press, New York, NY, 1995, p. 277.

20. Golden, G.M., McKie, J.E., and Potts, R.O., Role of SC lipid fluidity in transdermal drugflux, J. Pharm. Sci., 76, 25, 1987.

21. Carelli, V, Di Colo, G, Nannipieri, E, and Serafini, M.F., Enhancement effects in the permeationof alprazolam through hairless mouse skin, Int. J. Pharm., 88, 89, 1992.

22. Fang, J.Y., Hwang, T.L., and Leu, Y.L., Effect of enhancers and retarders on percutaneousabsorption of flurbiprofen from hydrogels, Int. J. Pharm., 50, 313, 2003.

23. Narishetty, S.T. and Panchagnula, R., Transdermal delivery system for zidovudine: in vitro, exvivo and in vivo evaluation, Biopharm. Drug Dispos., 25, 9, 2004.

24. Alberti, I., Kalia, Y.N., Naik, A., Bonny, J.D., and Guy, R.H., In vivo assessment of enhancedtopical delivery of terbinafine to human stratum corneum, J Control. Rel., 71, 319, 2001.

25. Takahashi, K. and Rytting, J.H. Novel approach to improve permeation of ondansetron acrossshed snake skin as a model membrane, J. Pharm. Pharmacol., 53, 789, 2001.

26. Goodman, M. and Barry, B.W., Action of penetration enhancers on human skin as assessed bythe permeation of model drugs 5-fluorouracil and estradiol. I. Infinite dose technique, J. Invest.Dermatol., 91, 323, 1988.

27. Ongpipattanakul, B., Burnette, R., and Potts, R.O., Evidence that oleic acid exists as a separatephase within stratum corneum, Pharm. Res., 8, 350, 1991.

28. Tanojo, H., BosvanGeest, A., Bouwstra, J.A., Junginger, H.E., and Bodde, H.A., In-vitro humanskin barrier perturbation by oleic acid: thermal analysis and freeze fracture electron micro-scopy studies, Thermochim. Acta, 293, 77, 1997.

29. Potts, R.O. and Francoeur, M.L., Lipid biophysics of water loss through the skin, Proc. Natl.Acad. Sci. USA, 87, 3871, 1990.

30. Friend, D., Catz, P., Heller, J., Reid, J., and Baker, R., Transdermal delivery of levonorgestrel I:Alkanols as permeation enhancers in vitro, J. Control. Rel., 7, 243, 1988.

31. Kai, T., Mak, V.H.W., Potts, R.O., and Guy, R.H., Mechanism of percutaneous penetra-tion enhancement: effect of n-alkanols on the permeability barrier of hairless mouseskin, J. Control. Rel., 12, 103, 1990.

32. Sloan, K.B., Beal, H.D., Taylor, H.E., Getz, J.J., Villaneuva, R., Nipper, R., and Smith, K.,Transdermal delivery of theophylline from alcohol vehicles, Int. J. Pharm., 171, 185, 1998.




33. Kanikkannan, N. and Singh, M., Skin permeation enhancement effect and skin irritation ofsaturated fatty alcohols, Int. J. Pharm., 248, 219, 2002.

34. Chantasart, D., Li, S.K., He, N., Warner, K.S., Prakongpan, S., and Higuchi, W.I., Mechanisticstudies of branched-chain alkanols as skin permeation enhancers, J. Pharm. Sci., 93, 762,2004.

35. Warner, K.S., Li, S.K., and Higuchi, W.I., Influences of alkyl group chain length and polar headgroup on chemical skin permeation enhancement, J. Pharm. Sci., 90, 1143, 2001.

36. Warner, K.S., Li, S.K., He, N., Suhonen, T.M., Chantasart, D., Bolikal, D., and Higuchi, W.I.,Structure–activity relationship for chemical skin permeation enhancers: Probing the chemicalmicroenvironment of the site of action, J. Pharm. Sci., 92, 1305, 2003.

37. Brain, K.R., and Walters, K.A., Molecular modeling of skin permeation enhancement bychemical agents, in Pharmaceutical Skin Penetration Enhancement, Walter, K.A. and Had-graft, J. (Eds), Marcel Dekker, New York, 1993, p. 389.

38. Ackermann, C., Flynn, G.L., and Smith, W.M., Ether–water partitioning and permeabilitythrough nude mouse skin in vitro. II. Hydrocortisone 21–n-alkyl esters, alkanols and hydro-philic compounds, Int. J. Pharm., 36, 67, 1987.

39. Okabe, H., Takayama, K., Ogura, A., and Nagai, T., Effect of limonene and related compoundson the percutaneous absorption of indomethacin, Drug Design Del., 4, 313, 1989.

40. Gao, S. and Singh, J., In vitro percutaneous absorption enhancement of a lipophilic drugtamoxifen by terpenes, J. Control. Rel., 51, 193, 1998.

41. Godwin, D.A. and Michniak, B.B., Influence of drug lipophilicity on terpenes as penetrationenhancers, Drug Dev. Ind. Pharm., 25, 905, 1999.

42. Hori, M., Satoh, S., Maibach, H.I., and Guy, R.H., Enhancement of propranolol hydrochlorideand diazepam skin absorption in vitro: effect of enhancer lipophilicity, J. Pharm. Sci., 80, 32,1991.

43. Williams, A.C. and Barry, B.W., The enhancement index concept applied to penetrationenhancers for human skin and model lipophilic (estradiol) and hydrophilic (5-fluorouracil)drugs, Int. J. Pharm., 74, 157, 1991.

44. Cornwell, P.A. and Barry, B.W., Sesquiterpene components of volatile oils as skin penetrationenhancers for the hydrophilic permeant 5-fluorouracil, J. Pharm. Pharmacol., 46, 261, 1994.

45. Obata, Y., Takayama, K., Okabe, H., and Nagai, T., Effect of cyclic monoterpenes on percu-taneous absorption in the case of a water-soluble drug (diclofenac sodium), Drug Design Del.,6, 319, 1990.

46. Okamoto, H., Ohyabu, M., Hashida, M., and Sezaki, H., Enhanced penetration of mitomycinC through hairless mouse and rat skin by enhancers with terpene moieties, J. Pharm.Pharmacol., 39, 531, 1987.

47. Okamoto, H., Hashida, M., and Sezaki, H., Structure activity relationship of 1-alkyl- or1-alkenylazacycloalkanone derivatives as percutaneous penetration enhancers, J. Pharm.Sci., 77, 418, 1988.

48. El-Kattan, A., Asbill, C.S., and Michniak, B.B., The effect of terpene enhancer lipophilicityon the percutaneous permeation of hydrocortisone formulated in HPMC gel systems,Int. J. Pharm., 198, 179, 2000.

49. El-Kattan, A., Asbill, C.S., Kim, N., and Michniak, B.B., The effects of terpene enhancers on thepercutaneous permeation of drugs with different lipophilicities, Int. J. Pharm., 215, 229, 2001.

50. Takayama, K., Kikuchi, K., Obata, Y., Okabe, H., Machida, Y., and Nagai, T., Terpenes aspercutaneous absorption promoters, S.T.P. Pharm. Sci., 1, 83, 1991.

51. Kobayashi, D., Matsuzawa, T., Sugibayashi, K., Morimoto, Y., and Kimura, M., Analysis of thecombined effect of 1-menthol and ethanol as skin permeation enhancers based on a two-layerskin model, Pharm. Res., 11, 96, 1994.

52. Cornwell, P.A., Barry, B.W., Bouwstra, J.A., and Gooris, G.S., Modes of action of terpenepenetration enhancers in human skin; differential scanning calorimetry, small-angle x-raydiffraction and enhancer uptake studies, Int. J. Pharm., 127, 9, 1996.




53. Yamane, M.A., Williams, A.C., and Barry, B.W., Terpene penetration enhancers in propyleneglycol/water co-solvent systems: effectiveness and mechanism of action, J. Pharm. Pharma-col., 47, 978, 1995.

54. Naito, S.I., Nakamori, S., Awataguchi, M., Nakajima, T., and Tominaga, H., Observations andpharmacokinetic discussion of percutaneous absorption of mefenamic acid, Int. J. Pharm., 24,127, 1985.

55. Mollgaard, B., Hoelgaard, A., and Baker, E., Vehicle effect on topical drug delivery-effect ofN-methylpyrrolidone, polar lipids and Azone on percutaneous drug transport, Proc. Intern.Symp. Control. Rel. Bioact. Mater., 15, 209, 1988.

56. Williams, A.C. and Barry, B.W., Penetration enhancers, Adv. Drug Del. Rev., 56, 603, 2004.57. Southwell, D. and Barry, B.W., Penetration enhancement in human skin: effect of 2-pyrroli-

done, dimethylformamide and increased hydration on finite dose permeation of aspirin andcaffeine, Int. J. Pharm., 22, 291, 1984.

58. Hoelgaard, A., Mollgaard, B., and Baker, E., Vehicle effect on topical drug delivery. IV. Effect ofN-methylpyrrolidone and polar lipids on percutaneous drug transport, Int. J. Pharm., 43, 233,1988.

59. Bhatia, K.S. and Singh, J., Percutaneous absorption of LHRH through porcine skin: effect ofN-methyl 2-pyrrolidone and isopropyl myristate, Drug Dev. Ind. Pharm., 23, 1111, 1997.

60. Aoyagi, T., Yamamura, M., Suzuki, N., Matsui, K., and Nagase, Y., Preparation of substitutedpyrrolidone derivatives and their evaluation as transdermal penetration enhancers, Drug Des.Discov., 8, 37, 1991.

61. Sasaki, H., Kojima, M., Mori, Y., Nakamura, J., and Shibasaki, J., Enhancing effect of pyrroli-done derivatives on transdermal drug delivery, I., Int. J. Pharm., 44, 15, 1988.

62. Sasaki, H., Kojima, M., Nakamura, J., and Shibasaki, J., Enhancing effect of pyrrolidonederivatives on transdermal penetration of phenolsulfonphthalein and indomethacin fromaqueous vehicle, Chem. Pharm. Bull. (Tokyo), 38, 797, 1990.

63. Sasaki, H., Kojima, M., Nakamura, J., and Shibasaki, J., Enhancing effect of pyrrolidonederivatives on transdermal penetration of sulfaguanidine, aminopyrine, Sudan III, J. Pharma-cobiodyn., 13, 200, 1990.

64. Sasaki, H., Kojima, M., Mori, Y., Nakamura, J., and Shibasaki, J., Enhancing effect of pyrroli-done derivatives on transdermal penetration of 5-fluorouracil, triamcinolone acetate, indo-methacin, and flurbiprofen, J. Pharm. Sci., 80, 533, 1991.

65. Kim, C.K., Hong, M.S., Kim, Y.B., and Han, S.K., Effect of penetration enhancers (pyrrolidonederivatives) on multilamellar liposomes of SC lipid: a study by UV spectroscopy and differen-tial scanning calorimetry, Int. J. Pharm., 95, 43, 1993.

66. Yoneto, S.K., Ghanem, A-H., Higuchi, W.I., Peck, K.D., and Li, S.K., Mechanistic studies of the1-alkyl-2-pyrrolidones as skin permeation enhancers, J. Pharm. Sci., 84, 312, 1995.

67. Yoneto, K., Li, S.K., Higuchi, W.I., Jiskoot, W., and Herron, J.N., Fluorescent probe studies ofthe interactions of 1-alkyl-2-pyrrolidones with stratum corneum lipid liposomes, J. Pharm.Sci., 85, 511, 1996.

68. Yoneto, K., Li, S. K., Higuchi, W.I., and Shimabayashi, S., Influence of the permeationenhancers 1-alkyl-2-pyrrolidones on permeant partitioning into the stratum corneum,J. Pharm. Sci., 87, 209, 1998.

69. Lopez, A., Llinares, F., Cortell, C., and Herraez, M., Comparative enhancer effects of Span120with Tween120 and Azone1 on the in vitro percutaneous penetration of compounds withdifferent lipophilicities, Int. J. Pharm., 202, 133, 2000.

70. Park, E.S, Chang S.Y, Hahn, M, and Chi, S.C., Enhancing effect of polyoxyethylene alkyl etherson the skin permeation of ibuprofen, Int. J. Pharm., 218, 167, 2001.

71. Mutalik, S. and Udupa, N., Transdermal delivery of glibenclamide and glipizide: in vitropermeation studies through mouse skin, Pharmazie, 57, 838, 2002.

72. Nokhodchi, A., Shokri, J., Dashbolaghi, A., Hassan-Zadeh, D., Ghafourian, T., and Barzegar-Jalali, M., The enhancement effect of surfactants on the penetration of lorazepam through ratskin, Int. J. Pharm., 250, 359, 2003.




73. Ashton, P., Walters, K.A., Brain, K.R., and Hadgraft, J., Surfactant effects in percutaneousabsorption. I. Effects on the transdermal flux of methyl nicotinate, Int. J. Pharm., 87, 261, 1992.

74. Kushla, G.P. and Zatz, J.L., Correlation of water and lidocaine flux enhancement by cationicsurfactants in vitrom J. Pharm. Sci., 80, 1079, 1991.

75. Cooper, E.R. and Berner, B., Interactions of surfactants with epidermal tissues — physico-chemical aspects, in Surfactants in Cosmetics, Rieger, M.M. (Ed.), Marcel Dekker, New York,1984, p. 185.

76. Cappel, M.J. and Kreuter, J., Effect of nonionic surfactants on transdermal drug delivery. I.Polysorbates, Int. J. Pharm., 69, 143, 1991.

77. Barry, B.W. and Williams, A.C., Permeation enhancement through skin, in Encyclopedia ofPharmaceutical Technology, Swarbrick, J. and Boylon, J.C. (Eds), Vol. 11, Marcel Dekker, NewYork, 1995, p. 449.




Chapter 3

Quantitative Structure–Enhancement Relationshipand the Microenvironmentof the Enhancer Site of Action

S. Kevin Li and William I. Higuchi

CONTENTS

Introduction ...................................................................................................................................... 35Methods ............................................................................................................................................ 36Quantitative Structure–Enhancement Relationships Based on Aqueous

Concentrations of the Enhancer.................................................................................................. 38Quantitative Structure–Enhancement Relationship Based on Enhancer

Concentrations in the Stratum Corneum Intercellular Lipids .................................................... 42Microenvironment of the Site of Enhancer Action.......................................................................... 45Transdermal Drug Delivery.............................................................................................................. 46Acknowledgment ............................................................................................................................. 47References......................................................................................................................................... 47

Introduction

Over the past 30 years, numerous studies on the influence of chemical permeation en-hancers upon drug permeation across skin have been performed. The literature is abun-dantwithhundredsof articles, patents, and reviewson this topic (e.g., reviewed in Leeet al.,1991; Williams and Barry, 1992; Smith and Maibach, 1995; Walters and Hadgraft, 1993; Pottsand Guy, 1997). It is generally believed that the mechanisms of action for most transdermalenhancers are through physical enhancer–membrane interactions (e.g., intercalation and


35


perturbation) and by fluidizing the lipids in the stratum corneum (SC) (e.g., Barry, 1987;Yoneto et al., 1996). Some enhancers can also act by SC lipid solubilization and lipidextraction (e.g., Goates and Knutson, 1994; Ogiso et al., 1995). Despite continuingadvances in the knowledge of transdermal absorption and enhancer mechanisms, theeffects of enhancers upon drug permeation have been quite unpredictable, and themechanism of action of permeation enhancers and their site of action in the SC are notfully understood. A clear quantitative structure–enhancement relationship (relationshipbetween the enhancer molecular structures and their enhancing potencies) for predictingthe effectiveness of enhancers is not available. Arbitrary screening remains a commonapproach in industrial practice to identify effective permeation enhancers for improvingpercutaneous absorption as opposed to a rational design approach. A screening approachis generally ineffective, and therefore obtaining a quantitative structure–enhancementrelationship for permeation enhancers based on a clear understanding of the mechanismof action of enhancers and the nature of the microenvironment of the enhancer site ofaction is important. If one does not understand the structure–enhancement relationship,the rational design of an effective enhancer and the prediction of the enhancer effects willremain a difficult task.

This chapter reviews our recent findings and will address the following questions.How does the nature of the enhancer polar head group and the hydrocarbongroup contribute to the potency of a chemical permeation enhancer? What is therelationship between the physicochemical properties of an enhancer such as lipophilicity(e.g., octanol–water partition coefficient) and its potency? What is the nature ofthe microenvironment of the enhancer site of action? In this chapter, we will establish aquantitative structure–enhancement relationship of the studied enhancers for the lipoidalpathway of the SC. Such a structure–enhancement relationship would providebasic insights into the mechanism of action of chemical permeation enhancers.These insights can aid pharmaceutical scientists in employing physical–chemicalprinciples rather than trial-and-error screening methods in the search for effectiveenhancers. The structure–enhancement relationship, moreover, will provide directinformation regarding the nature of the microenvironment of the enhancer site of actionin the SC.

Methods

A review of the percutaneous absorption literature has indicated that most transdermalabsorption studies for the purpose of enhancer evaluation have been conducted in anasymmetric enhancer configuration: the test enhancer solution with the drug is appliedon the SC while the dermis side remains in contact with the receiver solution that isusually an aqueous buffer solution (Warner et al., 2003). Although this experimentalsetup is adequate in assessing the effects of permeation enhancers upon the delivery ofa particular drug and mimics the practical situation of a transdermal delivery system,there are potential fundamental problems. For example, the asymmetric setup can resultin an enhancer concentration gradient across the skin membrane. This leads to acomplex situation in which the local permeation enhancement varies with the positionwithin the membrane, making mechanistic analysis of the data difficult (Liu et al., 1992).Another shortcoming in some past studies is that the importance of different pathways inthe SC (i.e., parallel lipoidal and pore pathways) has not been recognized even though itis a general view that permeation enhancers can affect either or both the SC lipoidal and




pore pathways to enhance the permeation of lipophilic and polar compounds. A parallellipoidal and pore pathway SC transport model should be utilized to delineate theenhancement effects upon the lipoidal pathway induced by the enhancers (e.g., Warneret al., 2001). Other studies have not corrected for the changes in permeant thermo-dynamic activities in the presence of enhancers and cosolvents in the donor chamberin data analysis. From a mechanistic point of view, the effects of the enhancers andcosolvents upon permeant activities that affect permeant transport should be properlytaken into account in assessing the potencies of the enhancers.

Studies in our laboratory have involved the use of a different experimental approachto establish a quantitative structure–enhancement relationship for the lipoidal pathway ofthe SC and to address the question of the nature of the microenvironment of the enhancersite of action (Kim et al., 1992; Yoneto et al., 1995; Warner et al., 2003; He et al., 2004). Inthese studies, permeation experiments were conducted under symmetric and equilibriumconditions (i.e., aqueous enhancer solution in both the donor and receiver chambers inequilibrium with the SC). The permeability enhancement factor, E, the ratio of thepermeant flux with the enhancer solution to that with the control phosphate bufferedsaline solution (PBS), was determined in these experiments with a model lipophilicpermeant. The enhancement factor was corrected for any changes in the chemicalpotential of the permeant in the enhancer solution with respect to that in PBS; thisallowed the comparison of enhancement factors at the same permeant thermodynamicactivity. Corticosterone (CS) was selected as the model permeant because the lipoidalpathway of the SC is the main transport rate-determining pathway for this permeant.Transport experiments were also conducted with a polar permeant to correct for possibleeffects of the enhancers upon the pore pathway. The equilibrium concentration of theenhancer in the SC intercellular lipid domain was determined in separate enhanceruptake experiments with SC and delipidized SC. In these experiments, a sample of SCor delipidized SC was weighed and soaked in the enhancer solution until the membranewas in equilibrium with the enhancer solution. If depletion of the enhancer in theenhancer solution was observed, the solution was replaced to maintain a constantenhancer concentration. After equilibration, the SC was removed from the solution andweighed. Then, the enhancer in the SC was extracted with 100% ethanol and assayed.The details of the experimental procedure for both the transport and uptake experiments

and in the references (Yoneto et al., 1995; Warner et al., 2003; Chantasart et al., 2004; Heet al., 2004).

concentration for CS permeation across the skin lipoidal pathway with 1-butyl-, 1-hexyl-,and 1-octyl-2-pyrrolidones as the enhancers. From plots such as those in Figure 3.1, thecorresponding enhancer aqueous concentrations for E¼ 10 (which are in thermo-dynamic equilibrium with the skin) were determined for all the enhancers studied.These concentrations are defined as the aqueous isoenhancement concentrations ofE¼ 10 and will be used as a means to evaluate the potencies of the enhancers. Similarly,the aqueous isoenhancement concentration of E¼ 4 can be obtained. The relativepotencies of the enhancers were assessed and compared based on (a) their aqueousisoenhancement concentrations and (b) their SC intercellular lipid concentrations.

homologous series of enhancers with different polar head groups and enhancers witha carbon–carbon double bond substituting for a single bond in their hydrocarbon chain(the lipophilic moiety).


Quantitative Structure–Enhancement Relationship & 37


Figure 3.1 shows representative plots of enhancement factor vs. aqueous enhancer

Figure 3.2 shows all the enhancers studied to date in our laboratory. They include

can be found in Chapter 18, ‘‘Mechanistic studies of permeation enhancers,’’ this volume

Quantitative Structure–Enhancement RelationshipsBased on Aqueous Concentrations of the Enhancer

concentrations of E¼ 10 and the octanol/PBS partition coefficients (Koctanol/PBS) for the

interpolated from the E vs. aqueous enhancer concentration plots similar to those inFigure 3.1 and the n-octanol/PBS partition coefficients were determined at the E¼ 10conditions. In Figure 3.3, the data points of all enhancers essentially fall on the same linewith a slope of –1. While there are a few moderate outliers, the overall correlation is quite

octanol/PBS partition coefficients. Similar to Figure 3.3, there is again a good correlationbetween the enhancer potency and the n-octanol/PBS partition coefficient. Two majorconclusions can be deduced from Figure 3.3 and Figure 3.4. First, the correlations in theenhancer isoenhancement concentration vs. enhancer partition coefficient plots with aslope of around –1 suggest that water saturated n-octanol may represent well thechemical microenvironment of the studied enhancers at their site of action in the SClipid lamellae. Second, these data demonstrate a structure–enhancement relationship ofthe enhancers, in which the potencies of the enhancers for transport enhancement acrossthe SC lipoidal pathway are related to the enhancer lipophilicities. These results alsosuggest that the n-octanol/PBS partition coefficient is an excellent predictor of enhancerpotency (as expressed in terms of the aqueous enhancer concentration) for the largegroup of skin permeation enhancers studied to date.

Three other partition coefficient systems (n-hexanol/PBS, n-decanol/PBS, andn-hexane/PBS) were investigated for their ability to provide correlations with the

0

5

10

15

0.01 0.1 1 10

Enhancer Concentration (%w/w)

Enh

ance

men

t F

acto

r, E

BP

HP

OP

Figure 3.1 Representative plots of permeability enhancement factor across the SC lipoidalpathway (E) vs. aqueous enhancer concentration for corticosterone with enhancers 1-butyl-2-pyrrolidone (BP), 1-hexyl-2-pyrrolidone (HP), and 1-octyl-2-pyrrolidone (OP).




(Warner et al., 2003; He et al., 2003). Figure 3.3 presents the aqueous isoenhancement

good. Figure 3.4 presents the plot of the E¼ 4 isoenhancement concentrations vs. the n-

enhancers shown in Figure 3.2. The isoenhancement concentrations in the plot were

The enhancers in Figure 3.2 were examined employing the above experimental strategy

potencies of the enhancers. Here, a main purpose was to further characterize the micro-environment of the enhancer site of action in the transport rate-limiting domain in the SCintercellular lipid lamellae. First, the n-hexanol/PBS and n-decanol/PBS systems were totest for the degree of selectivity of n-octanol among the n-alkanols as the organic phase.At the other extreme, the n-hexane/PBS system was to test how a pure hydrocarbonenvironment would represent the microenvironment of the enhancer site of action in

concentration of E¼ 10 vs. the n-hexanol/PBS, n-decanol/PBS, and n-hexane/PBS par-tition coefficients, respectively. The correlations between enhancer potencies and the

systems (within the scatter of the data) are quite good and comparable to that with the

1-alkyl-2-pyrrolidones(AP) 1-alkyl-2-piperidinones (API) 1-alkyl-2-azacycloheptanones (AZ)

OH

OH

OH

R

1,2,3-alkanetriols (AT) 1,2-alkanediols (AD) n-alkyl-b -D-glucopyranosides (AG)

R

OO

2-(1-alkyl)-2-methyl-1,3-dioxolanes (MD) 1-alkanols (AL) N,N-dimethylalkanamides (AM)

OH

OHO

O

R

R OHOHR

1,2-dihydroxypropyl alkanoates (MG) trans-3-alken-1-ols (TAL) cis-3-alken-1-ols (CAL)

NO

R

O

NH

OH

trans-hydroxyproline-N-alkanamide-C-ethylamide (HAE)

HO R

HO

OH

R

N

O

R

O

O R

HO

HO

HO HO

N

O

R

N RN R

O O

Figure 3.2 Chemical structures of the permeation enhancers, R 5 alkyl-chain.




the SC transport rate-limiting domain. Figure 3.5–Figure 3.7 show the isoenhancement

partition coefficients of the n-hexanol/PBS (Figure 3.5) and n-decanol/PBS (Figure 3.6)

n-octanol/PBS system. This suggests that the chemical environment differences amongthe three n-alkanols are probably not large enough for choosing a system among thethree to best mimic the chemical microenvironment of the enhancer in the lipid lamellaeof the SC. In contrast to the correlation found between the enhancer potencies and thepartition coefficients in the studied n-alkanol/PBS systems, the correlation betweenenhancer potency and the partition coefficient with the n-hexane/PBS system is poor

−5

−4

−3

−2

−1

0

0.5 1.5 2.5 3.5 4.5 5.5

Log K octanol/PBS

Log

(aq

ueou

s co

ncen

trat

ion

[M ])

AP AL API AZ

AD AM AG MDCAL TAL MG

Figure 3.3 Correlation between aqueous E 5 10 isoenhancement concentration and octanol/PBS partition coefficient (Koctanol/PBS). The slope of the line in the figure 5 –1. Each data pointrepresents the average value without showing the standard deviation because the error bargenerally lies within the symbol in the log–log plot. Enhancer abbreviations are provided in

−4

−3

−2

−1

0

0.5 1.5 2.5 3.5 4.5

Log K octanol/PBS

Log

(aq

ueou

s co

ncen

trat

ion

[M])

AP API AZ AD AM

AG MD MG AT HAE

Figure 3.4 Correlation between the aqueous isoenhancement concentration at E 5 4 and octa-nol/PBS partition coefficient (Koctanol/PBS). The slope of the line in the figure 5 –1. Each data pointrepresents the average value without showing the standard deviation because the error bargenerally lies within the symbol in the log–log plot. Enhancer abbreviations are provided in




Figure 3.2.

Figure 3.2.

lipid lamellae is quite different from a pure hydrocarbon environment. It is also note-

leftward displacement of the data point from the correlation line in Figure 3.7, and themore nonpolar the enhancer, the more is the data displacement in the rightward direc-tion. This pattern is consistent with the view that the data scatter seen in Figure 3.7 is

−4

−3

−2

−1

0

0.5 1.5 2.5 3.5 4.5

Log K hexanol/PBS

Log

(aq

ueou

s co

ncen

trat

ion

[M])

AP AL API AZ AD

AM AG MD MG

Figure 3.5 Correlation between the aqueous isoenhancement concentration at E 5 10 andhexanol/PBS partition coefficient (Khexanol/PBS). The slope of the line in the figure 5 –1. Eachdata point represents the average value without showing the standard deviation because theerror bar generally lies within the symbol in the log–log plot. Enhancer abbreviations are

−4

−3

−2

−1

0

0.5 1.5 2.5 3.5 4.5

Log K decanol/PBS

Log

(aq

ueou

s co

ncen

trat

ion

[M])

AP AL API AZ AD

AM AG MD MG

Figure 3.6 Correlation between the aqueous isoenhancement concentration at E 5 10 anddecanol/PBS partition coefficient (Kdecanol/PBS). The slope of the line in the figure 5 –1. Eachdata point represents the average value without showing the standard deviation because theerror bar generally lies within the symbol in the log–log plot. Enhancer abbreviations areprovided in Figure 3.2.




provided in Figure 3.2.

(Figure 3.7) suggesting that the microenvironment of the enhancer site of action in the SC

worthy from Figure 3.3 and Figure 3.7 that the more polar the enhancer, the greater the

primarily related to the lack of intermolecular hydrogen bonding occurring between theenhancer molecule and the components of the organic phase. The difference between

enhancer with the components of the actual semipolar microenvironment of the SClipid lamellae matching rather well that of the intermolecular hydrogen bonding betweenthe enhancer and the n-alkanol molecule. This match is absent when hexane is theorganic phase.

In summary, comparisons of the enhancer potencies with the n-hexanol/PBS, n-octanol/PBS, n-decanol/PBS, and n-hexane/PBS partition coefficients showed goodcorrelations for the n-alkanol solvents but not for n-hexane. A quantitative structure–enhancement relationship has been established with the enhancers. This result supportsthe hypotheses that (a) the effectiveness of the permeation enhancers is related to theirlipophilicity and their ability to partition into the transport rate-limiting domain in the SCintercellular lipid lamellae and (b) the transport rate-limiting domain has a microenviron-ment with polarity similar to the polarity of water-saturated bulk n-hexanol, n-octanol,and n-decanol.

Quantitative Structure–Enhancement RelationshipBased on Enhancer Concentrations in the StratumCorneum Intercellular Lipids

The foregoing discussion has provided new insight into the physicochemical factorsinfluencing the potencies of chemical permeation enhancers (their effectiveness aspermeation enhancers). A general quantitative structure–enhancement relationship

−4

−3

−2

−1

0

-3 -2 -1 0 1 2 3 4

Log K hexane/PBS

Log

(aq

ueou

s co

ncen

trat

ion

[M])

AP AL API

AZ AD AM

AG MD MG

Figure 3.7 Correlation between the aqueous isoenhancement concentration at E 5 10 and

hexane/PBS). The slope of the line in the figure 5 –1. Eachdata point represents the average value without showing the standard deviation because theerror bar generally lies within the symbol in the log–log plot. Enhancer abbreviations are




hexane/PBS partition coefficient (K

provided in Figure 3.2.

the poor correlation found in Figure 3.7 and the good correlations seen in Figure 3.3–Figure 3.6 is an indication of the extent of intermolecular hydrogen bonding of the

based on the enhancer aqueous isoenhancement concentration data and their n-alkanol/PBS partition coefficients has been established. The discussion up to this point hasinferred that, because of the good correlation between the aqueous isoenhancementconcentration and the n-alkanol/PBS partition coefficient, the enhancer target site micro-environment can be well mimicked by liquid n-alkanols. However, it would be inappro-priate to make any definite conclusions regarding this without information concerning theconcentration of the enhancer at the site of action. An independent set of experiments wastherefore conducted to quantify enhancer potencies by determining the enhancer con-centration in the SC intercellular lipids under the isoenhancement E¼ 10 conditions (Heet al., 2004). The experimental results obtained from this approach allow for a morecritical examination than is possible based on the aqueous isoenhancement concentrationdata alone. As will be discussed in some detail, the data of the aqueous isoenhancementenhancer concentrations and the corresponding equilibrium enhancer concentrations inthe SC intercellular lipids, taken together, will provide a more complete understanding ofthe enhancer molecular factors associated with enhancer potency as well as furtherinsight into the nature of the enhancer target site microenvironment.

Figure 3.8 presents the data on enhancer uptake into the intercellular lipid domain ofthe SC under the E¼ 10 isoenhancement conditions for 18 of the enhancers studied in theprevious section. The SC intercellular lipid enhancer concentration determined under theisoenhancement condition may be considered as a measure of the intrinsic enhancerpotency; the lower this concentration, the higher the intrinsic enhancer potency. Basedon the results presented in Figure 3.8, the potencies of the enhancers are seen to berelatively independent of their octanol/PBS partition coefficients. Note that the x- and

enhancer concentrations in the SC intercellular lipid in Figure 3.8 are in contrast to thestrong dependence between the lipophilicities of the enhancers and their potencies basedon the enhancer concentrations in the aqueous phase in Figure 3.3. This result is quite

0.001

0.01

0.1

1

10

100

0.5 1.5 2.5 3.5 4.5 5.5

Log K octanol/PBS

Mem

bran

e co

ncen

trat

ion

(µm

ol/m

g m

embr

ane)

AP AZ AL

AG AD AMAPI CALT AL

Figure 3.8 Relationship between enhancer uptake into SC intercellular lipid domain per mg drySC (mmol/mg) under isoenhancement E 5 10 and octanol/PBS partition coefficient (Koctanol/PBS).Each data point represents the average value. The standard deviations of the log Koctanol/PBS dataare not shown because the error bars generally lie within the symbols in the plot. Enhancer




y-axis of Figure 3.8 have the same scales as those in Figure 3.3. The relatively constant

abbreviations are provided in Figure 3.2.

surprising as one would have expected some enhancer alkyl chain length and/or polarhead group dependency on the potencies of the enhancers (based on their concentration

that (a) the intrinsic potency of the enhancer at its site of action is relatively independent ofits alkyl group chain length and the nature of its polar head group, (b) the lipophilicity ofthe enhancer (both the alkyl chain and the polar head moieties) mainly assists thetranslocation of the enhancer to its site of action through a free energy of transfer fromthe bulk aqueous phase to the site, and (c) permeation enhancement is related to the abilityof the permeation enhancer to partition into the transport rate-limiting domain, whichseems to be well-represented by the intercellular lipid ‘‘phase.’’

We will now take a closer look at the data in Figure 3.8, which support the view thatall the studied enhancers (at E¼ 10) exhibit essentially the same intrinsic potency. First,the E¼ 10 intrinsic potency of n-octyl-b-D-glucopyranoside is seen to be essentially thesame as that of n-octanol. Considering the large size and polarity of the glucopyranosidegroup, this suggests that the free volume of the microenvironment of the enhancer site ofaction in the SC transport rate-limiting domain is insensitive to the differences in the sizesof the enhancer polar head group. This also suggests further that, under the conditions ofthe present study, the SC transport rate-limiting domain may not behave as that of anordered lipid lamellae of the lipid domain in the intercellular region of SC (Bouwstra et al.,2002b; Kuempel et al., 1998; Norlen, 2001; White et al., 1988) but more like a conven-tional, homogeneous, bulk liquid ‘‘phase.’’ Second, on the matter of n-alkyl group chainlength effect, it has been suggested that there may be an optimum chain length (in therange of C9 to C12) for skin permeation enhancers (e.g., Aungst et al., 1986; Lee et al.,1991); yet there is little or no indication of this with the 1-alkyl-2-pyrrolidones (C4 to C12)in the present study. More studies with longer chain 1-alkyl-2-pyrrolidones would be ofinterest. Finally, it has been hypothesized that an enhancer with unsaturated alkyl chainsuch as unsaturated alcohols is more potent than an enhancer with saturated alkyl chain,based on the molecular geometry and the presence of kinks in the alkenyl chain of theunsaturated enhancer (Cooper, 1984; Aungst et al., 1986; Aungst, 1989; Brain and Walters,1993). For a similar reason, the unsaturated enhancer with a cis conformation carbon–carbon double bond has been expected to be more potent than the enhancer with a transdouble bond. However, the results of the present study suggest no influence of a carbon–carbon double bond (in the lipophilic moiety) upon the enhancement effects of theenhancers. Here again, more studies with longer chain enhancers with a double bondlocated at different positions along the carbon chain would be of interest.

Although the intrinsic potencies of the studied enhancers have been found to beessentially the same and relatively independent of the enhancer molecular structures asrevealed by the essentially same SC intercellular lipid ‘‘phase’’ enhancer concentrations, itshould be pointed out that enhancer lipophilicity is still an important factor becauselipophilicity is essential to the translocation of the enhancers to their site of action in themembrane.

In conclusion, it has been somewhat surprising to find that all of the studied enhancershas yielded essentially the same intrinsic potency under the E¼ 10 conditions. As thepresent study is perhaps the first of its kind on this particular aspect, further work is needed.Despite the need of future testing, the data are so far consistent with the hypothesis that theenhancer polar head and alkyl groups act only to assist in the transfer of the enhancer fromthe aqueous phase to the SC intercellular lipid lamellae and make the enhancer availablefor its action in the transport rate-limiting domain. The essentially constant enhancerconcentration in the intercellular lipid lamellae (Figure 3.8) and the correlation between




in the SC lipid domain). The data in Figure 3.8, together with those in Figure 3.3, suggest

action and the macroscopic SC intercellular lipid ‘‘phase’’ can both be represented bywater-saturated liquid n-octanol. This is discussed further in the next section.

Microenvironment of the Site of Enhancer Action

the logarithm of the partition coefficient for enhancer partitioning between the aqueousphase and the intercellular lipid domain (log KSC lipid/PBS) vs. log Koctanol/PBS (He et al.,2004). Data from previous studies with branched alkanols (Chantasart et al., 2004) and 2-phenylethanol (unpublished data) are also included in the figure. The KSC lipid/PBS valueswerecalculated from enhancer intercellular lipids uptake data and isoenhancementaqueous concentration data (both under E¼ 10 conditions). Arguments are now presentedon the basis of Figure 3.3, Figure 3.8, and Figure 3.9 in support of the hypothesis:the microenvironment for the enhancer site of action is well mimicked by liquid n-octanol.It can be seen in Figure 3.9 that all enhancers fall essentially on the same line (with modestdata scattering). The data correlation is quite good with a regression slope closeto unity over a more than 1000-fold range of Koctanol/PBS values. Also, there is no significantdifference among the correlations of the enhancers with different polar headgroups and different alkyl chain length (slope¼ 1.00 to 1.30), suggesting that the micro-environment of the site of enhancer action in the SC is essentially the same for all theenhancers. A slope closer to unity in Figure 3.9 would indicate an even closer similarity

0

1

2

3

4

5

0 1 2 3 4 5 6

Log K octanol/PBS

Log

K S

C lip

id/P

BS

AP API AZ AL AD AM

AG TAL CAL bAL PE

y = 1.06x - 1.11R2 = 0.961

Figure 3.9 Correlation between the partition coefficient of the enhancer between PBS and theintercellular lipid domain (KSC lipid/PBS) vs. octanol/PBS partition coefficient (Koctanol/PBS). Eachdata point represents the average value without showing the standard deviation because theerror bar generally lies within the symbol in the log–log plot. Enhancer abbreviations are




the E¼ 10 isoenhancement concentrations and octanol/PBS partition coefficients (Figure3.3) also support the interpretation that the microenvironment of the enhancer site of

Figure 3.9 is a replot of the data in Figure 3.3 and Figure 3.8. It shows the relationship of

provided in Figure 3.2 except for branched alkanols (bAL) and 2-phenylethanol (PE).

between the microenvironment of the enhancer site of action and n-octanol. The smallpositive deviation of the slope from unity may be attributed to (a) somewhat strongerinteractions between the studied enhancers and the SC intercellular lipid domain thanthose of the n-octanol phase and (b) the SC intercellular lipid domain being slightly morelipophilic than the n-octanol phase. Despite this deviation, the essentially constant slopefor the enhancers in the log KSC lipid/PBS vs. log Koctanol/PBS plot suggests the same micro-environment of the enhancer site of action for the enhancers studied to date. It is important

inappropriate to conclude that the microenvironment for the enhancer site of action is wellmimicked by the n-octanol organic phase. The reason for this is that the SC lipid lamellae isnot expected to behave as a conventional, liquid homogeneous phase and there may wellbe a distribution of regions into which the enhancer may partition (Bouwstra et al.,2002a,b; Kuempel et al., 1998); it would not have been unreasonable to find regions inthe lipid lamellae into which the enhancer molecule may partition but may be relativelyineffective in contributing toward permeation enhancement. However, taken together,Figure 3.3 and Figure 3.8 would support the argument that a correlation likely existsbetween the n-octanol phase and both the transport rate-limiting domain and the micro-environment of the enhancer site of action in SC.

lipid ‘‘phase’’ involved in the E¼ 10 enhancer uptake experiments is well-mimicked byliquid (water-saturated) n-octanol. It is quite remarkable, considering the diversity ofmolecular types in this group of enhancers, that such a correlation (over a range of threeorders of magnitude) would exist, especially when one recognizes that the SC lipidlamellae would not generally be described as a conventional, homogeneous liquidphase from the standpoint of chemical composition or structural order. One might expectthat different regions (domains) in the SC lipid lamellae would favor partitioning ofdifferent enhancer molecules differently, especially when such a wide range of moleculartypes are considered. It may be possible, however, that there is such a distribution ofregions (domains) but this distribution can be relatively narrow, especially for amphi-philic molecules (such as our studied enhancers) and especially when the SC lipidlamellae has been fluidized to a significant extent (e.g., at E 4).

Transdermal Drug Delivery

To apply the quantitative structure–enhancement relationship and the microenvironmentdata in the development of transdermal drug delivery systems, certain limitations shouldbe noted. They will be discussed as follows.

First, transdermal delivery systems usually employ nonaqueous vehicles and cosol-vents. Some components of nonaqueous vehicles are able to partition into SC and alter thepolarity of the microenvironment of the SC intercellular lipid domain. They can also act aspermeation enhancers in the SC. These effects should be taken into consideration whenutilizing the structure–enhancement relationship to predict the effects of enhancers. In thetransdermal patch, the nonaqueous vehicles or cosolvents may also alter the thermo-dynamic activity of the enhancers and therefore the partitioning of the enhancers fromthe patch vehicle into the SC. This effect is less complicated when the SC is not altered bythe vehicle or cosolvents because it can then be predicted with thermodynamics.

The effects of permeation enhancers upon transdermal transport can also beaffected by the drug of interest when high drug concentration is used. In a high drug




to emphasize that based on either Figure 3.3 or Figure 3.8 alone, it would have been

A point of emphasis is that Figure 3.9 is evidence that the macroscopic SC intercellular

concentration transdermal delivery system, possible interactions between the drug andthe enhancer in the transdermal patch may affect enhancer partitioning into the SC. Thepresence of drug in high concentration in the SC may also affect the microenvironment ofthe SC transport pathway, again altering the amount of enhancer present at its site ofaction. These effects can be drug dependent.

One interesting point to note here is the amount of the enhancers in the SC under the

to 0.12 mmol of the enhancers are partitioned into the intercellular lipids of 1 mg of dry SC(SC dry weight measured before enhancer treatment) at E¼ 10. Assuming that the SCintercellular lipids are homogeneous, a simple calculation will indicate that high concen-trations of the enhancers are present in the SC intercellular lipids under these conditions(around 5 to 10%, w/w). Even at these moderately high concentrations (of enhancers ofdifferent molecular sizes and polarities), the chemical microenvironment of the enhancersite of action appears to remain the same and similar to liquid n-octanol; this is probably notunreasonable in light of data showing that the n-octanol/PBS partition coefficients them-selves are not significantly different at these concentrations in the n-octanol phase (Warneret al., 2003). In previous studies, the microenvironment of the rate-limiting domains forsteroidal permeant transport across SC in buffered saline (control without the presence ofenhancers) has been shown to be well mimicked by the n-octanol liquid phase (Andersonet al., 1988, 1989; Raykar et al., 1988). Together, this suggests that the microenvironment ofthe enhancer site of action for permeation enhancement in SC is not significantly altered bythe enhancers. This analysis leads one to speculate that (a) the results obtained in thepresent study will likely hold even in transdermal systems of high drug concentration andwith nonaqueous vehicles and (b) the alteration of thermodynamic activity of the enhancerin the transdermal patch system is likely to be the remaining factor.

An important issue has been ignored by intent in the experiments and discussion of thepresent chapter: that is the asymmetric enhancer situation in transdermal drug delivery(enhancer concentration gradient in SC). Avoiding such a situation is by design in thepresent study for the purpose of mechanistic interpretation of the results and identifying aquantitative structure–enhancement relationship without the complications arising fromenhancer concentration gradients across the membrane. However, the asymmetric situ-ation is generally encountered in in vivo transdermal delivery, and dealing with theresulting enhancer concentration gradients in the SC would be an important factor in theselection of effective permeation enhancers for transdermal drug delivery. Complexmodeling is required in this situation for data analysis. The relationship between thephysicochemical properties of the enhancers and enhancer concentration gradients in

Acknowledgment

The authors thank Kevin S. Warner, Ning He, and Doungdaw Chantasart for their contri-butions in the project and the financial support by NIH Grants GM 043181 and GM 063559.

ReferencesAnderson BD, Higuchi WI, Raykar PV, Heterogeneity effects on permeability–partition coefficient

relationships in human stratum corneum, Pharm Res, 5, 566–573, 1988.




E¼ 10 conditions in the present study. Thedata in Figure 3.8 shows that approximately 0.06

SC is discussed in Chapter 18, ‘‘Mechanistic studies of permeation enhancers,’’ this volume.

Anderson BD, Raykar PV, Solute structure–permeability relationships in human stratum corneum,J Invest Dermatol, 93, 280–286, 1989.

Aungst BJ, Structure/effect studies of fatty acid isomers as skin penetration enhancers and skinirritants, Pharm Res, 6, 244–247, 1989.

Aungst BJ, Rogers NJ, Shefter E, Enhancement of naloxone penetration through human skin in vitrousing fatty acids, fatty alcohols, surfactants, sulfoxides and amides, Int J Pharm, 33, 225–234,1986.

Barry BW, Mode of action of penetration enhancers in human skin, J Control Release, 6, 85–97,1987.

Brain KR, Walters KA, Molecular modeling of skin permeation enhancement by chemical agents,In: Pharmaceutical Skin Penetration Enhancement, Walters KA and Hadgraft J, Editors. 1993,New York: Marcel Dekker, p. 389–416.

Bouwstra JA, Pilgram GSK, Ponec M, Does the single gel phase exist in stratum corneum? J InvestDermatol, 118, 897–898, 2002a.

Bouwstra JA, Gooris GS, Dubbelaar FER, Ponec M, Phase behavior of stratum corneum lipidmixtures based on human ceramides: the role of natural and synthetic ceramide 1, J InvestDermatol, 118, 606–617, 2002b.

Chantasart D, Li SK, He N, Warner KS, Prakongpan S, Higuchi WI, Mechanistic studies of branched-chain alkanols as skin permeation enhancers, J Pharm Sci, 93, 762–779, 2004.

Cooper ER, Increased skin permeability for lipophilic molecules, J Pharm Sci, 73, 1153–1156, 1984.Goates CY, Knutson K, Enhanced permeation of polar compounds through human epidermis. I.

Permeability and membrane structural changes in the presence of short chain alcohols, BiochimBiophys Acta, 1195, 169–179, 1994.

He N, Li SK, Suhonen TM, Warner KS, Higuchi WI, Mechanistic study of alkyl azacycloheptanonesas skin permeation enhancers by permeation and partition experiments with hairless mouseskin, J Pharm Sci, 92, 297–310, 2003.

He N, Warner KS, Chantasart D, Shaker DS, Higuchi WI, Li SK, Mechanistic study of chemical skinpermeation enhancers with different polar and lipophilic functional groups, J Pharm Sci, 93,1415–1430, 2004.

Kim YH, Ghanem AH, Mahmoud H, Higuchi WI, Short chain alkanols as transport enhancers forlipophilic and polar/ionic permeants in hairless mouse skin: mechanism(s) of action, Int JPharm, 80, 17–31, 1992.

Kuempel D, Swartzendruber DC, Squier CA, Wertz P, In vitro reconstitution of stratum corneumlipid lamellae, Biochim Biophys Acta, 1372, 135–140, 1998.

Lee VHL, Yamamoto A, Kompelia UB, Mucosal penetration enhancers for facilitation of peptideand protein drug absorption, Crit Rev Ther Drug Carrier Syst, 8, 91–192, 1991.

Liu P, Higuchi WI, Ghanem A-H, Kurihara-Bergstrom T, Good WR, Assessing the influence ofethanol in simultaneous diffusion and metabolism of estradiol in hairless mouse skin for the‘asymmetric’ situation in vitro, Int J Pharm, 78, 123–136, 1992.

Norlen L, Skin barrier structure and function: the single gel phase model, J Invest Dermatol, 117,830–836, 2001.

Ogiso T, Iwaki M, Paku T, Effect of various enhancers on transdermal penetration of indomethacinand urea, and relationship between penetration parameters and enhancement factors, J PharmSci, 84, 482–488, 1995.

Potts RO, Guy RH, Mechanisms of Transdermal Drug Delivery. 1997, New York: Marcel Dekker.Raykar PV, Fung MC, Anderson BD, The role of protein and lipid domains in the uptake of solutes

by human stratum corneum, Pharm Res, 5, 140–150, 1988.Smith EW, Maibach HI, Percutaneous Penetration Enhancers. 1995, Boca Raton: CRC Press, Inc.Walters KA, Hadgraft J, Pharmaceutical Skin Penetration Enhancers. 1993, New York: Marcel

Dekker.Warner KS, Li SK, Higuchi WI, Influences of alkyl group chain length and polar head group on

chemical skin permeation enhancement, J Pharm Sci, 90, 1143–1153, 2001.




Warner KS, Li SK, He N, Suhonen TM, Chantasart D, Bolikal D, Higuchi WI, Structure–activityrelationship for chemical skin permeation enhancers: probing the chemical microenvironmentof the site of action, J Pharm Sci, 92, 1305–1322, 2003.

White SH, Mirejovsky D, King GI, Structure of lamellar lipid domains and corneocyte envelopes ofmurine stratum corneum. An x-ray diffraction study, Biochemistry 27, 3725–3732, 1988.

Williams AC, Barry BW, Skin absorption enhancers. Crit Rev Ther Drug Carrier Syst, 9, 305–353,1992.

Yoneto K, Ghanem AH, Higuchi WI, Peck KD, Li SK, Mechanistic studies of the 1-alkyl-2-pyrrolidones as skin permeation enhancers, J Pharm Sci, 84, 312–317, 1995.

Yoneto K, Li SK, Higuchi WI, Jiskoot W, Herron JN, Fluorescent probe studies of the interactions of1-alkyl-2-pyrrolidones with stratum corneum lipid liposomes, J Pharm Sci, 85, 511–517, 1996.




Chapter 4

The Role of Prodrugsin Penetration Enhancement

Kenneth B. Sloan and Scott C. Wasdo

CONTENTS

Push versus Pull Mechanisms for Penetration Enhancers .............................................................. 52Basis for Prodrugs as Penetration Enhancers.................................................................................. 52Acyl versus Soft Alkyl Promoieties .................................................................................................. 54Mechanisms for Penetration Enhancement ..................................................................................... 54

Decrease Crystal Lattice Energy by Masking Hydrogen BondDonor Functional Groups ........................................................................................................ 54

Incorporation of Water Solubility Enhancing Functional Groupsinto the Promoiety .................................................................................................................... 59

Conclusion ........................................................................................................................................ 62References......................................................................................................................................... 63

A penetration enhancer is by definition anything that is used to improve the delivery ofa chemical substance across some chemical barrier. For the present purposes, we areinterested in defining chemical, as opposed to physical-, electrical-, or mechanical-basedapproaches to improving the delivery of a drug into or through the skin: improvingdermal or transdermal delivery, respectively, or collectively improving the topical deliv-ery of a drug. These same chemical-based approaches could also be used to improve thedelivery of cosmeceuticals but there the task would be specifically only for improvingtheir dermal delivery.


51


Push versus Pull Mechanisms for Penetration Enhancers

From a mechanistic point of view, there are two general ways to accomplish the task ofimproving topical delivery using a chemical-based approach. The first approach toincrease the ‘‘push’’ of the vehicle components on the drug is to drive it into the skin.1

One way to increase the ‘‘push’’ of the vehicle is to use vehicle components in which thedrug is more soluble but which are volatile. Evaporation of the volatile components afterapplication of the drug–vehicle combination leaves a supersaturated solution of the drugin a state of heightened thermodynamic activity (av),

2 that is, av greater than one. Thesecond approach is to increase the ‘‘pull’’ on the drug into the skin by components ofthe vehicle that have permeated the skin and have decreased the resistance of the skinto permeation by the drug.1 One way to increase the ‘‘pull’’ on the drug is to usevehicle components that permeate the skin and increase the solubility of the drug inthe skin: those components that interact with the skin. Such components of the vehicledo not have to permeate the skin faster than the drug. However, another way to increasethe ‘‘pull’’ on the drug by components of the vehicle is to use components that dopermeate the skin faster than the drug and pull the drug along with them — a drageffect.3

The basis for the two chemical-based approaches to enhancing topical delivery(decreasing the solubility of the drug in the vehicle and increasing its solubility in theskin; ‘‘push’’ and ‘‘pull,’’ respectively) lies in the form of the equation that describes flux.The flux, J, of the drug through skin is directly related to the concentration of the drug inthe first layer of the skin, Cs, from Fick’s law: J¼ (Cs – Cx)Ds/Ls, where Cx is the concen-tration of the drug in the last layer of the skin (and is assumed to approach zero at steady-state), Ds is the diffusion coefficient of the drug in the skin and Ls is the thickness of themembrane. The concentration of the drug in the skin, Cs, is generated from its equilib-rium with the concentration of the drug in the vehicle, Cv, through the product of itspartition coefficient between the two phases, KCs:Cv, and Cv. The concentration of thedrug in the skin approaches its saturated solubility in the skin, Ss, and a thermodynamicactivity (as) of one when Cv approaches the saturated solubility of the drug in anoninteractive vehicle, Sv, that is, av also is one. Regardless of the value for Sv, thehighest concentration of drug in the skin that is possible from a drug applied in anoninteractive vehicle is Ss. As Sv increases KCs:Cv decreases and as Sv decreases KCs:Cv

increases: Ss remains as a constant. Ss can only be increased by using an interactivecomponent in the vehicle that changes the solubilizing capacity of the skin, the ‘‘pull,’’ orby using a volatile vehicle component that creates an environment at equilibrium wherethe thermodynamic activity of the drug in the vehicle, av, is greater than one, the ‘‘push,’’and hence the activity of the drug in the skin, as, is also greater than one.

Basis for Prodrugs as Penetration Enhancers

Although increasing the ‘‘push’’ can be easily accomplished by manipulating the com-ponents of the vehicle in which the drug is applied (its formulation), increasing the ‘‘pull’’can be more easily accomplished using a prodrug approach that changes the solubilityproperties of the drug. A prodrug is a chemically or enzymatically reversible derivative ofa parent drug that improves the physicochemical or biological properties of the parentdrug molecule to overcome some intrinsic problem associated with its therapeutic use: inthis case, poor solubility in the skin and low topical delivery.4 The particular combination




of functional groups that is added to the parent drug is called the promoiety and thereversible connection between the promoiety and the parent drug is called the enablingfunctional group. A prodrug approach, then, can be envisaged as a 1:1 molecularcombination of the drug and a promoiety that contains functional groups that willincrease its solubility in the skin.5 This prodrug approach stands in sharp contrast tomost formulation approaches where large molar excesses of penetration enhancers asvehicle components are routinely needed to increase Ss for the drug.

What are the properties of the functional groups in the promoiety which, when addedto the parent drug, could be reasonably expected to cause an increase in the solubility ofthe resulting prodrug in the skin and hence to cause an increase in flux? Since it is verydifficult to measure the solubility of a prodrug in the skin, it is infinitely more convenientto measure its flux through the skin in diffusion cell experiments and assume that, basedon Fick’s law, there is a direct relationship between increased flux and increased solu-bility in the skin. Using increases in flux as the criterion, it has been observed forhomologous series of more lipophilic prodrugs that the more water soluble membersof the series were the ones that gave the greatest increases in flux and not the more lipidsoluble ones.4,6,7 More recently a database (n¼ 42) for the delivery of parent drugs byprodrugs from an isopropyl myristate (IPM) vehicle through hairless mouse skin in vitrowas collected and analyzed to put the previous qualitative observations on a firmquantitative foundation. When data for solubilities in IPM (log SIPM) and water (logSAQ), molecular weight (MW) and flux (log J) were fit to a transformation of the Potts–Guy equation8 (the Roberts–Sloan equation)9 the coefficients for the parameters showedalmost as strong a dependence on SAQ as on SIPM: log J¼ xþ y log SIPMþ (1 y) logSAQ zMW where x¼0.211, y¼ 0.534, z¼ 0.00364, and r2¼ 0.937. A similar strongdependence on SAQ was obtained when data (n¼ 18) for delivery of parent drugs byprodrugs from a water vehicle through hairless mouse skin in vitro was fit to the Roberts–Sloan equation: x¼1.497, y¼ 0.660, z¼ 0.00469, and r2¼ 0.765.10 Finally, when datafor the delivery of ten nonsteroidal antiinflammatory drugs (NSAIDs) from a mineral oilvehicle through human skin in vivo11 were fit to the Roberts–Sloan equation, a significanteffect of SAQ of the drug on flux was observed: x¼1.459, y¼ 0.72, z¼ 0.00013, andr2¼ 0.934.12 Regardless of whether the vehicle was lipoidal or aqueous, whether it washairless mouse skin or human skin, and whether in vivo or in vitro, SAQ of the drug wasan important parameter from which to calculate flux. Thus the functionazl groups in thepromoiety should increase both the lipid (SIPM) and aqueous (SAQ) solubilities of theparent drug to increase its flux and hence by inference increase the solubility of the drugin the skin.

The reason that increasing both lipid and aqueous solubilities of the drug is importantto increasing its solubility in the skin, and hence its topical delivery, can be found in thestructure of the barrier to topical delivery — the intercellular compartment of the stratumcorneum (SC). The intercellular compartment consists of lamellar double bilayers com-prised of lipid components such as ceramides, cholesterol, and fatty acids which havepolar groups attached to them. These polar head groups have water associated with themso that for a permeant to cross these bilayers perpendicular to the axis of the bilayers, itmust alternately cross lipid and aqueous phases.5,6 Thus a balance of solubility in bothlipid and aqueous phases by the drug is necessary for its most efficient permeation of theintercellular compartment of the SC. The agreement between the experimentally meas-urable physicochemical parameters in the theoretically derived Roberts–Sloan equationand in the biochemically-based model for the barrier to permeation is encouraging.


Prodrugs in Penetration Enhancement & 53


Acyl versus Soft Alkyl Promoieties

The promoieties that have been used to increase lipid and aqueous solubilities can bedivided into two types based on whether they are attached directly to the functionalgroup in the parent drug, that is to be modified, or indirectly through a methylene orvinylogous methylene (aryl methylene) spacer.4,5,7 In each type, the enabling functionalgroup is usually a carbonyl type functional group because of its sensitivity to cleavageby chemical or enzymatic hydrolysis. Generally these types have been referred as acyland soft alkyl type promoieties, respectively. Cleavage of the acyl type promoietyregenerates the parent drug directly while cleavage of the acyl group in the soft alkylpromoiety generates an intermediate drug–X–CHR–X’H from drug–X–CHR–X’–(C¼X’’)–X’’’R. The intermediate is designed to be intrinsically unstable and undergoes rapid andcomplete chemical hydrolysis to the parent drug–X–H. The advantage of the soft alkylprodrug approach is that the stability of the prodrug (as well as its attendant physico-chemical properties) is not limited by the functional group in the parent drug to which itis attached. Generally, changing X will change the biochemical and/or pharmacologicalactivity of the drug, but changing X’ to obtain a more stable or more soluble prodrug willnot. Of course X’’ and X’’’ can be changed in the same ways that they could have been ifan acyl prodrug approach had been used.

Mechanisms for Penetration Enhancement

Decrease Crystal Lattice Energy by Masking HydrogenBond Donor Functional Groups

Regardless of whether the prodrug is derived from an acyl or soft alkyl type promoiety,there are two general mechanisms by which both types of promoieties can increase bothlipid and aqueous solubilities. The first mechanism has its basis in decreasing the crystallattice energy of the parent drug by modifying polar groups capable of forming intermo-lecular hydrogen bonds. In many if not most drug molecules the X in drug–X–H is aheteroatom which causes X–H to be polarized because of the difference in electronega-tivitites between X and H. This polarized drug–X–H bond is capable of forming intermo-lecular hydrogen bonds within the crystal lattice which leads to low solubilities especiallyin lipids but also frequently in water. The polarization is further attenuated if an electronwithdrawing carbonyl type functional group is attached to X–H to give drug–(O¼C)–X–H. Examples of this type of drug molecule, which can be measurably but not highlyionized at physiological pH, include heterocycles such as 5-fluorouracil (5-FU) (drug–(O¼C)–NH) and 6-mercaptopurine (6-MP) (drug–(S¼C)–NH) which are very highmelting and exhibit low solubilities in both water and lipids. In other examples such asparent drugs containing a carboxylic acid functional group (drug–(O¼C)–OH), thefunctional group is so highly polarized that it becomes highly ionized at physiologicalpH which does not allow it to readily cross the lipid phase of the alternating lipid:aqu-eous phases of the biological barrier. An important class of drugs that belong to thiscategory are the NSAIDs. Another example of this class are the nucleotide-based drugswhere the highly ionized functional group is a phosphate group. Simply masking thehydrogen bond donating abilities of the functional group by replacing the H in drug–X–Hwith either an acyl or a soft alkyl group decreases the melting point (mp) and increasesthe lipid solubility (SIPM) as well as frequently increasing the aqueous solubility (SAQ) of




the prodrug compared to the parent drug, especially for the shorter alkyl chain membersof a homologous series.

An illustration of the results that can be obtained from application of the principles ofthis first mechanism for increasing SIPM and SAQ to the design of prodrugs can be seen forseveral prodrugs of 5-FU. The mp, SAQ, SIPM, log partition coefficients between IPM andpH 4.0 buffer (log K) and rates of delivery of total 5-FU containing species throughhairless mouse skin from an IPM vehicle in vitro ( JIPM) for four different series of prodrugof 5-FU are given in Table 4.1: three acyl types and a one soft alkyl type. The first acyl

Table 4.1 Prodrugs of 5-Fluorouracil

Prodrugs, R¼ a mpb SIPMc SAQ

c,d log Ke JIPMf

1-AAC-5-FU1, C1NHC¼O 212 0.30 3.69 1.09 0.2082, C2NHC¼O 180 2.79 7.76 0.44 0.6003, C3NHC¼O 139 12.4 8.98 0.14 0.7464, C4NHC¼O 133 24.6 5.11 0.68 0.5155, C6NHC¼O 113 44.9 0.36 2.09 —6, C8NHC¼O 91 46.9 0.030 3.21 0.060

1-AOC-5-FU7, C1OC¼O 160 2.13 112 1.72 2.628, C2OC¼O 128 13.1 175 1.12 5.929, C3OC¼O 126 15.2 42.2 0.44 2.3110, C4OC¼O 98 33.8 24.1 0.15 2.2311, C6OC¼O 67 153 4.94 1.49 1.5412, C8OC¼O 98 36.2 0.13 2.45 0.29

1-AC-5-FU13, C1C¼O 130 22.1 120 0.73 9.314, C2C¼O 131 36.4 47.6 0.12 4.315, C3C¼O 146 17.4 6.50 0.43 1.316, C4C¼O 121 39.2 3.48 1.05 1.017, C5C¼O 102 112.7 2.94 1.58 1.118, C7C¼O 84 110.7 0.15 2.88 0.60

1-ACOM-5FU19, C1(C¼O)OCH2 124 3.29 183 1.74 2.8820, C2(C¼O)OCH2 102 9.83 167 1.23 3.8221, C3(C¼O)OCH2 91 14.4 42.4 0.47 2.5722, C4(C¼O)OCH2 88 14.8 12.3 0.08 1.2923, C5(C¼O)OCH2 91 14.7 2.23 0.82 0.5624, C7(C¼O)OCH2 108 9.99 0.17 1.77 0.12

5-FU,H 284 0.049 85.4g 3.24h 0.240

aC1,C2, etc. refer to the number of carbons in alkyl chain.bUnits of 8C.cSolubilities in units of mM.d Estimated from SIPM/K.ePartition coefficient between IPM and pH 4.0 buffer at 23 + 18C.fValues for the delivery of total species containing 5-FU through hairless mouse skin from IPM in vitro in units ofmmol cm2 h1.gSolubility in pH 4.0 buffer.hLog solubility ratio between pH 4.0 buffer and IPM.




type of prodrug of 5-FU that was evaluated for its ability to increase the delivery of 5-FUwas the alkylaminocarbonyl-5-FU (1-AAC-5-FU) prodrugs (Table 4.1 and Figure 4.1).Initially only the longer alkyl chain members of the series were evaluated (4 to 6),13

but subsequently the shorter alkyl chain members (1 to 3) were evaluated and one ofthem, 3, was found to give the greatest increase in the delivery of total 5-FU containing

Figure 4.1

naproxen; 41 to 45, naproxen; 46 to 51, ketoprofen; and 53 to 57, diclofenac.




Chemical structures of prodrugs for compounds 1 to 57 are listed in Table 4.1 toTable 4.4: 1 to 24, 5-fluorouracil; 26 to 29, levonogestrel; 31 to 34, indomethacin; 36 to 39,

species, JIPM.14 All of the 1-AAC-5-FU prodrugs exhibited lower mp than 5-FU and all ofthem were more soluble in IPM than 5-FU: from 6 times for 1 to almost 1000 times for 6.However, the most lipid soluble member evaluated, 6, gave only 0.25 times the flux of5-FU. None of the 1-AAC-5-FU prodrugs was even as soluble in water as 5-FU, and the C3member (3), not the shortest alkyl chain member of the series (1), gave the highest SAQ

value: only 0.11 times SAQ for 5-FU. The C3 member also gave the greatest increase in JIPM

value for the series; albeit only three times. Thus as predicted,4,6,7 for a more lipid solublehomologous series of prodrugs, the more water soluble member gave the highestJIPM value. The low increase in JIPM can be attributed to the low SAQ values exhibitedby the 1-AAC-5-FU prodrugs compared to subsequent series; and the low SAQ valuescan be attributed to the fact that one of the hydrogen bond donor functional groups,(O¼C)-NH, in 5-FU was merely replaced with another hydrogen bond donor group,N-(O¼C)-NH, in the promoiety. The potential for forming intermolecular hydrogenbonds was not decreased significantly and the added alkyl group in the promoiety furtherdepressed SAQ.

The second acyl type of prodrug of 5-FU that was evaluated was the alkyloxycarbo-15 In this series the hydrogen bond donat-

ing group in the parent drug has not been replaced with another hydrogen bond donatinggroup in the promoiety so the mp are somewhat lower than the corresponding membersin the 1-AAC-5-FU series except for the C8 member of the series. Consequently, themembers of the 1-AOC-5-FU series were all somewhat more soluble in IPM thanthe members of the 1-AAC-5-FU series except for the C8 member, 12; and the worstmember of the series in terms of increased SIPM was 43 times instead of 6 times moresoluble in IPM than 5-FU. However, the big difference between the two series was in theSAQ values. Not only were two members of the series more water soluble than 5-FU, 7and 8 (1.3 and 2 times, respectively), but they were all more water soluble than thecorresponding members of the 1-AAC-5-FU series (from 30 to 4.3 times). Thus, since the1-AOC-5-FU series was more soluble in lipids and in water, as predicted,4,6,7 theydelivered more total 5-FU species through hairless mouse skin than the 1-AAC-5-FUseries (from 3 to 12.5 times). Also, as predicted,4,6,7 the C2 member, 8, which was themost water soluble member of the series and not the most lipid soluble member ofthe series, 11, gave the greatest increase in JIPM compared to 5-FU (24.7 times), and thenext most water soluble member, 7, gave the next greatest increase in JIPM compared to5-FU (11 times).

Based on previous literature, the 1-AOC series was expected to be more stable thanthe 1-AAC series of prodrugs of 5-FU. Whereas the amount of intact prodrug deliveredby the 1-AAC series was in the 6 to 10% range, the amount delivered by the 1-AOC serieswas in the 40 to 70% range and was up to 90% for the best performing member of theseries, 8. If delivery through the skin and subsequent slower release of 5-FU systemicallywas the target of topical delivery, then the members of the 1-AOC-5-FU series performedwell. On the other hand, if delivery into the skin was the target, then a more rapidlyhydrolyzing type of prodrug of 5-FU was required.

The third acyl type of prodrug of 5-FU that was evaluated was the alkylcarbonyl-5-FU(1-AC-5-FU) prodrugs (Table 4.1).16 The members of this series were known to hydrolyzequite rapidly (t1/2¼ 3 to 5min); so it was expected that only 5-FU would be deliveredthrough the skin. This expectation was realized and only 5-FU and no intact prodrug wasobserved in the receptor phase after application of 1-AC-5-FU prodrugs in IPM indiffusion cell experiments. All of the members of the 1-AC series were much more solublein IPM than 5-FU (355 to 2300 times), and one member, C1 (13), was more soluble in




nyl-5-FU (1-AOC-5-FU) prodrugs (Table 4.1).

water than 5-FU: 1.4 times. However, direct comparisons between the 1-AC series andeither the 1-AOC or the 1-AAC series based only on the alkyl chain length in thepromoiety would be misleading without taking into account the added heteroatom inthe latter two series. For example, we will compare the OC1 member (7) of the 1-AOCseries with the C2 member of the 1-AC series (14), the OC2 with the C3, the OC3 with theC4, the OC4 with the C5 and the OC6 with the C7. Using these intererseries comparisons,the members of the 1-AC series were more soluble in IPM (1.3 to 17 times) than those ofthe 1-AOC series, except for 18 compared to 11. On the other hand, the members of the1-AOC series were more soluble in water (2.4 to 33 times) than those of the 1-AC series;and as predicted,4,6,7 they all gave higher JIPM values than the corresponding members ofthe 1-AC series, except for OC1, 7, versus C2, 14. 7 was only 2.4 times more solublein water than 14 while 14 was 17 times more soluble in IPM than 7. 14 exhibited asomewhat better balance of SAQ and SIPM than 7 and gave a higher JIPM value: 1.6 times.However, within the 1-AC series the C1 member, 13, which was the more water solublemember of the series and not one of the more lipid soluble members, gave the greatestenhancement in JIPM: 39 times that of 5-FU.4,6,7

In the 1-AC series the effect of mp on solubilities and ultimately on flux can be readilyillustrated. The C3 member of the series, 15, exhibited a higher mp than either theshorter, 14, or longer alkyl chain members, 16, and hence exhibited a lower SIPM valuethan those members. The SAQ value of 15 also dropped off more rapidly than expected asdid its JIPM value. On the other hand, the log K values appeared normally spaced and themethylene p values derived from the log K values only varied by 10%: p¼ 0.59 + 0.05.Log K values are no substitute for experimental solubilities for purposes of predictingtrends in J.

The soft alkyl example given here of the results that can be obtained from theapplication of the principles of the first mechanism for increasing SIPM and SAQ to thedesign of prodrugs is also a 5-FU prodrug: the 1-alkylcarbonyloxymethyl-5-FU (1-ACOM-

17 As expected each of the 1-ACOM-5-FU prodrugs exhibited alower mp than 5-FU since a hydrogen bond donor group had been masked in theprodrug. Also as expected each was much more soluble in IPM than 5-FU (67 to 302times) and there were two members, 19 and 20, that were more soluble in water than 5-FU(2.1 and 1.9 times, respectively). As predicted,4,6,7 19 and 20 were the members that gavethe greatest enhancement in JIPM (12 and 16 times, respectively) and not the more lipidsoluble, longer alkyl chain members of the series. However, to compare members of the 1-ACOM series with members of any one of the 1-acyl series, the added heteroatom andmethylene spacer in the 1-ACOM series needs to be taken into account. Thus comparisonsshould be made between the C1 member of the 1-ACOM series, 19, and the C3 member ofthe 1-AC, 15, or the C2 member of the 1-AOC series, 8; the C2 member of the 1-ACOMseries, 20, and the C4 member of the 1-AC series, 16, or the C3 member of the 1-AOCseries 9, etc. Using these interseries comparisons, the members of the 1-ACOM series wereless soluble in IPM but much more soluble in water (15.0 to 48.0 times) than the membersof the 1-AC series; and their JIPM values were greater except for the comparison between23 and 18 where the JIPM values were equivalent. On the other hand, althoughthe members of the 1-ACOM series were less soluble in IPM than the members ofthe 1-AOC series, in this comparison only two members of the 1-ACOM series, 20 and21, were substantially more soluble in water (4.0 and 1.8 times, respectively) and hencegave a greater JIPM value than the corresponding members of the 1-AOC series. Inthe comparison of 19 with 8, the SAQ values were very close and 8 was 4 times morelipid soluble, so 8 gave a 2 times greater increase in JIPM. Similarly, 11 was 2.2 times more




5-FU) prodrugs (Table 4.1).

water soluble and 10 times more IPM soluble than 23, so 11 gave a 3 times greater increasein JIPM.

Thus, the general mechanism for increasing lipid and aqueous solubilities of a drugby decreasing its ability to form intermolecular hydrogen bonds in the crystal lattice canbe very effective: 11 to 40 times enhancement of flux. But it is essential to evaluate theshorter alkyl chain members of any series to be considered because those are themembers that are most likely to be more water soluble as well as more lipid soluble. Inthe examples based on 5-FU, the increases in flux realized with these acyl and soft alkylprodrug approaches are more than sufficient to enlarge the indicated use of topical 5-FUfrom treating only actinic keratoses of the scalp18 to treating recalicitrant psoriasis on lesspermeable areas of the body.19

Incorporation of Water Solubility Enhancing FunctionalGroups into the Promoiety

The second general mechanism by which acyl and soft alkyl promoieties can be used toincrease the lipid and aqueous solubilities of prodrug compared to its parent drug is toincorporate polar, water solubilizing groups into the promoiety. In the examples illustrat-ing the previous mechanism, the primary effect of the prodrug modification was to increaselipid solubility because the promoiety contained only an enabling functional group anda simple alkyl group. Although large increases in SIPM were realized for all members ofhomologous series, increases in SAQ were usually modest (less than 2 times) and only forthe shorter alkyl chain members. In the examples illustrating the second general mechan-ism the promoiety contains an additional amine, amide, ether, or diol functional groupwhich in retrospect could have been designed specifically to increase SAQ. However, inmost examples SAQ values were not available from the original references.

The first example is the use of a diol functional group in the promoiety to increase theSAQ of the prodrug and hence J. Although the stated rationale was that more hydrophilicprodrugs could overcome the perceived rate limiting contribution of the aqueous viableepidermis part of the barrier to permeation of the skin by highly lipophilic drugs,20 thesuccess of such prodrugs would also support a model for permeation where alternatinglipid:aqueous barriers must be crossed in the intercellular compartment of the SC.4,6,7 InTable 4.2 the mp 8C, solubilities in mixtures of ethanol:water (Sv), log K between octanol

Table 4.2 Prodrugs of Levonorgestrel

Prodrugs, R ¼ mpa Svb log Kc Jd

25, levonorgestrel 240 19.2 (100) 3.70 0.0001926, C5H11 86 604 (95)

12.9 (62) 0.0005827, C4H9 170 28.3 (95) 0.0002628, OCH2CH(OH)CH2OH 148 30.2 (40) 3.22 0.006329, O(CH2)4CH(OH)CH2OH 53 396 (40) 3.75 0.0030

aUnits of 8C.bSolubilities in mixtures of ethanol:water in units of mM where the value in parenthesis ispercentage of ethanol in the mixture.cPartition coefficient between octanol and water at 248C.dValues for delivery of total species containing levonorgestrel from suspensions in mixturesof ethanol:water (given in the Sv column) through rat skin in vitro in units of mmol cm2 h1.




and pH 7.4 buffer and fluxes of total species delivered from suspensions in ethanol:waterthrough rat skin in vitro ( J ) are given for the evaluation of four acyl prodrugs oflevonorgestrel.

Neither were representative of the shorter alkyl chain members of the series which wouldhave had the greatest potential for increased aqueous as well as lipid solubility. Since 26and 27 were both more soluble in 95% ethanol than levonorgestrel, 25, was in 100%ethanol, it is reasonable to assume they would also be more soluble in octanol and hencebe defined as more lipohilic than 25. Since partition coefficients for 26 and 27 could notbe obtained because no 26 or 27 could be measured in the aqueous phase (while 25could), it is reasonable to assume that 26 and 27 were less hydrophilic than 25. Finally,since the flux of 25 from various ethanol:water (40 to 100%) mixtures did not varysignificantly, it can be assumed that delivery of total species containing 25 by theprodrugs from widely different ethanol:water mixtures can be compared to the averageflux generated by the application of 25 (0.00020 mmol cm2 h1) in ethanol:watermixtures. Thus 26 and 27, which were more soluble in lipids but less soluble in water,gave 3 and 1.3 times greater J values, respectively, than 25. Only 25 was observed in thereceptor phases.

By comparison, since the two prodrugs containing a diol functional group in thepromoiety, 28 and 29, were both more soluble in an ethanol:water mixture that wasprimarily aqueous in composition (40% ethanol) than 25 was in 100% ethanol, it can bereasonably assumed that 28 and 29 were more soluble in water than 25. In addition,since 28 and 29 exhibit log K that were comparable to that of 25 and were more solublein water than 25, it can be reasonably assumed that 28 and 29 were more soluble inoctanol than 25, that is, more lipophilic. Thus, since 28 and 29 were more soluble in alipid and in water than their parent drug, as predicted, they gave much larger increases inJ than the simple alkylcarbonyl prodrugs that were only more soluble in a lipid: 31 and 15times, respectively.4,6,7 However, because of their greater stabilities as carbonate estersthey delivered mostly intact prodrug through the skin: 80 and 96%, respectively.

The second example is the use of an amide functional group in the promoiety toincrease SAQ and hence J. The first report of the synthesis of a promoiety containing anamide functional group as part an effort to increase topical delivery was for theopylline:7-(N, N-diethylsuccinamoyloxymethyl) theophylline.21 However, the prodrug was nevercompletely evaluated. More recently 1-alkylazacycloalkan-2-one esters of indomethacin,30,22 and naproxen, 35,23

of SIPM, SAQ, and rates of delivery of total species containing 30 or 35 from water throughhuman skin in vitro (JAQ) are given. For the indomethacin series, the second member ofthe series, 32, was the only member of the series that exhibited a greater SAQ thanindomethacin, and although it was barely as soluble in IPM as indomethacin, it causedthe greatest enhancement of JAQ (4 times). The more lipid soluble but less water solublemembers gave lower enhancement of JAQ. For the naproxen series, the first member ofthe series, 36, was more soluble in water (8 times) than naproxen and was more solublein water than the other members of the series. 36 was also more soluble in IPM than theother members of the series but none were as soluble as naproxen. Thus, 36, which wasmore soluble in lipids and water than the other members of the series, gave the greatestenhancement in JAQ (2.7 times) as would be predicted.4,6,7

There are two additional observations that can be made about these two series ofprodrugs which have an amide functional incorporated into the promoiety. First, al-though the SIPM values for the two series are comparable, the SAQ values for the naproxen




Two of the prodrugs in Table 4.2 were simple alkylcarbonyl prodrugs: 26 and 27.

have been synthesized and evaluated. In Table 4.3 the values

series (36 to 39) are almost uniformly 10 times greater than those for the indomethacinseries (31 to 34) and consequently the JAQ values for the naproxen series are almostuniformly 10 times greater. Second, although more labile soft alkyl type prodrugs (n¼ 1)had been synthesized, they were never evaluated because they were considered to betoo labile. On the other hand, the n¼ 2 prodrugs were too stable, and only 10 to 12% ofeither parent drug was observed in the receptor phases of the diffusion cell experimentsin which they were evaluated. It would have been interesting to have evaluated the n¼ 1series of prodrugs using an IPM vehicle in which they would have been stable todetermine how effective they might have been at delivering the parent drug.

The third example is the use of an amine functional group in the promoietyto increase the SAQ of the prodrug and hence J. Again the first report of the synthesisof a promoiety containing an amine functional group as part of an effort to increasetopical delivery was for theophylline: 7-(N, N-dimethylaminoacetyloxymethyl) theophyl-line.21 However, again the prodrug was never completely evaluated. More recentlythe 17-(4’-dimethylaminobutyrate) ester prodrug of testosterone was evaluated using a10% solution of the prodrug in pH 7.4 buffer.24 Compared to delivery from a suspensionof testosterone in pH 7.4 buffer, the prodrug was 60 times more effective at deliveringtestosterone. Although no solubility data were reported, a 10% solution of the prodrugwas evaluated which suggests that it is substantially more soluble in water than testos-terone which was soluble to the extent of 0.004%. The 2-diethylaminoethyl ester prodrugof indomethacin was also evaluated by the same group.25 It was reported that theprodrug was 3.7 times more soluble in pH 7.4 buffer and its partition coefficient betweenoctanol and pH 7.4 buffer was 6.2 times greater than that of indomethacin so the prodrugwas also much more soluble in octanol: 23 times. Thus, it was entirely predictable that theprodrug gave a 4.3 times enhancement in the delivery of total indomethacin containingspecies through human skin in vitro.4,6,7

The fourth example is the use of an ether functional group in the promoiety toincrease the SAQ of the prodrug and hence J. There are numerous reports in the literaturewhere polyoxyethylene (POE) esters have been used as prodrugs to enhance oraldelivery26 but only a few where POE esters have been used to enhance topical delivery.One report that is typical is the use of POE esters to enhance the topical delivery of

Table 4.3 Prodrugs of Indomethacin and Naproxen

Prodrug SIPMa SAQ

a JAQb

30, Indomethacin 7.82 0.011 0.2331, n¼2, m¼3 6.00 0.0096 0.8032, n¼2, m¼4 7.34 0.016 0.9633, n¼2, m¼5 19.0 0.012 0.7734, n¼2, m¼6 27.5 0.0074 0.1935, naproxen 23.5 0.045 5.136, n¼2, m¼3 21.1 0.355 13.837, n¼2, m¼4 18.8 0.249 8.938, n¼2, m¼5 16.7 0.032 4.039, n¼2, m¼6 7.64 0.011 2.8

aSolubilities in units of mM.bValues for the delivery of total species containing parent drug from waterthrough human skin in vitro in nmol cm2 h1.




NSAIDs where naproxen, ketoprofen, and diclofenac were modified.27 The experimentallog partition coefficients between octanol and water (log K) and delivery of total speciescontaining the NSAIDs from an ethanol deposited film through human skin in vitro ( J )are reported in Table 4.4 along with the corresponding calculated SAQ values. For allthree series, as the number of oxyethylene units increased, log K values decreased andcalculated SAQ values increased. All of the POE prodrugs were more lipophilic than theirparent drugs based on log K values except for the n¼ 6 members of most series.However, it was the more water soluble, n¼ 6 members of two of the series that gavethe greatest enhancement in flux values: 2.3 times for the naproxen series and 2.4 timesfor the diclofenac series. It is not clear how far the trend could have been extended fornaproxen and diclofenac, but for ketoprofen n¼ 5 appeared to be as far as the trend inadding oxyethylene groups to the promoiety and obtaining higher J values went. Onlyintact prodrug was observed in the receptor phases. The authors suggested that lack ofreversion was an artifact of the in vitro diffusion cell experiment.

Conclusion

Recognizing that one of the mechanisms for topical penetration enhancement involvesincreasing the solubility of the drug in the skin and that prodrugs increase the delivery ofdrugs into and through the skin by achieving the same, then it is quite clear that prodrugs

Table 4.4 Prodrugs of Naproxen, Ketoprofen,and Diclofenac

Prodrug Log Ka SAQb Jc

40, Naproxen 3.0 2.85 2.2541, n¼ 2 3.9 0.31 1.4642, n¼ 3 3.7 0.51 2.1743, n¼ 4 3.5 0.83 1.8344, n¼ 5 3.2 1.74 4.4645, n¼ 6 2.9 3.65 5.1346, Ketoprofen 2.8 4.68 5.2947, n¼ 2 3.6 0.65 2.7148, n¼ 3 3.4 1.06 4.6249, n¼ 4 3.3 1.36 4.0450, n¼ 5 3.1 2.23 14.351, n¼ 6 2.6 7.67 7.9252, Diclofenac 3.2 1.74 3.3353, n¼ 2 4.1 0.19 2.5854, n¼ 3 3.9 0.31 4.0455, n¼ 4 3.5 0.83 4.8856, n¼ 5 3.2 1.74 5.8357, n¼ 6 2.6 7.67 8.04

aLog partition coefficient between octanol and water.bSolubilities in units of mM calculated from log SAQ¼1.072 log Kþ 0.672.cValues for the delivery of total species containing parent drug from ethanoldeposited compounds through human skin in vitro calculated from Q24 (inmmol cm2)/24 h to give nmol cm2 h1.




constitute one type of penetration enhancer separate from formulation approaches.An even more powerful approach to enhancing topical delivery would be to usecombinations of prodrugs with formulation approaches to enhancing topical delivery.So far there have been no reports of the use of such combinations except for simpleone component vehicles which have obviously not been optimized.28 However, thepossibilities with the use of such a combination approach would seem to be limitless.

References1. Kadir, R. et al. Delivery of theophylline into excised human skin from alkanoic acid solutions: a

‘‘push–pull’’ mechanism, J. Pharm. Sci., 76, 774, 1987.2. Coldman, M.F., Poulson, B.J., Higuchi, T. Enhancement of percutaneous absorption by use of

volatile: nonvolatile systems as vehicles, J. Pharm. Sci., 58, 1098, 1969.3. Friend, D.R., Smedley, S.I. Solvent drag in ethanol/ethyl acetate enhanced skin permeation of

d-norgestrel, Int. J. Pharm., 97, 39, 1993.4. Sloan, K.B. Functional group considerations in the development of prodrug approaches to

solving topical delivery problems, in Prodrugs: Topical and Ocular Drug Delivery, Sloan, K.B.,Ed., Marcel Dekker, Inc., New York, 1992, Chapter 2.

5. Sloan, K.B., Wasdo, S. Designing for topical delivery: prodrugs can make the difference, Med.Res. Rev., 23, 763, 2003.

6. Sloan, K.B., Koch, S.A.M., Siver, K.G. Mannich base derivatives of theophylline and5-fluorouracil: syntheses, properties and topical delivery characteristics, Int. J. Pharm., 21,251, 1984.

7. Sloan, K.B. Prodrugs for dermal delivery, Adv. Drug Delivery Rev., 3, 67, 1989.8. Potts, R.O., Guy, R.H. Predicting skin permeability, Pharm. Res., 9, 663, 1992.9. Robert, W.J., Sloan, K.B. Correlation of aqueous and lipid solubilities with flux of prodrugs of

5-fluorouracil, theophylline and 6-mercaptopurine: a Potts–Guy approach, J. Pharm. Sci., 88,515, 1999.

10. Sloan, K.B. et al. Topical delivery of 5-fluorouracil (5-FU) and 6-mercaptopurine (6-MP) bytheir alkylcarbonyloxymethyl (ACOM) prodrugs from water: vehicle effects on design ofprodrugs, Pharm. Res., 20, 639, 2003.

11. Wenkers, B.P., Lippold, B.C. Skin penetration of nonsteroidal antiflammatory drugs out oflipophilic vehicle: influence of the viable epidermis, J. Pharm. Sci., 88, 1326, 1999.

12. Roberts, W.J., Sloan, K.B. Application of the transformed Potts–Guy equation to in vivo humanskin data, J. Pharm. Sci., 90, 1318, 2001.

13. Sasaki, H. et al. Transdermal delivery of 5-fluorouracil and alkylcarbamoyl derivatives,Int. J. Pharm., 60, 1, 1990.

14. Sloan, K.B. et al. Transdermal delivery of 5-fluorouracil (5-FU) through hairless mouseskin by1-alkylaminocarbonyl-5-FU prodrugs: physicochemical characterization of prodrugs and cor-relation with transdermal delivery, Int. J. Pharm., 93, 27, 1993.

15. Beall, H., Prankerd, R., Sloan, K. Transdermal delivery of 5-fluorouracil (5-FU) through hairlessmouse skin by 1-alkyloxycarbonyl-5-FU prodrugs: physicochemical characterization of pro-drugs and correlations with transdermal delivery, Int. J. Pharm., 111, 223, 1994.

16. Beall, H.D., Sloan, K.B. Transdermal delivery of 5-fluorouracil (5-FU) by 1-alkylcarbonyl-5-FUprodrugs, Int. J. Pharm., 129, 203, 1996.

17. Taylor, H.E., Sloan, K.B. 1-Alkylcarbonyloxymethyl prodrugs of 5-fluorouracil (5-FU): synthe-ses, physicochemical properties and topical delivery of 5-FU, J. Pharm. Sci., 87, 15, 1998.

18. Dillaha, C.J. et al. Further studies with topical 5-fluorouracil, Arch. Dermatol., 92, 410, 1965.19. Tsuji, T., Sugai, T. Topical admistered fluorouracil in psoriasis, Arch. Dermatol., 105, 208, 1972.20. Friend, D. et al. Transdermal delivery of levonorgestrel II: effect of prodrug structure on skin

permeability in vitro, J. Control. Release, 7, 251, 1988.




21. Sloan, K.B., Bodor, N. Hydroxymethyl and acyloxymethyl prodrugs of theophylline: enhanceddelivery of polar drugs through skin, Int. J. Pharm., 12, 299, 1982.

22. Bonina, F.P. et al. 1-Alkylazacycloalkan-2-one esters as prodrugs of indomethacin forimproved delivery through human skin, Int. J. Pharm., 77, 21, 1991.

23. Bonina, F.P. Montenegro, L., Guerrera, F. Naproxen 1-alkylazacycloalkan-2-one esters asdermal prodrugs: in vitro evaluation, Int. J. Pharm., 100, 99, 1993.

24. Milosovich, S. et al. Testosteronyl-4-dimethylaminobutyrate HCl: a prodrug with improvedskin permeation rate, J. Pharm. Sci., 82, 227, 1993.

25. Jona, J.A. et al. Design of novel prodrugs for the transdermal penetration of indomethacin,Int. J. Pharm., 123, 127, 1995.

26. Greenwald, R.B. PEG drugs: an overview, J. Control. Release, 74, 159, 2001.27. Bonina, F.P. et al. In vitro and in vivo evaluation of polyoxyethyene esters as dermal prodrugs

of ketoprofen, naproxen and diclofenac, Eur. J. Pharm. Sci, 14, 123, 2001.28. Waranis, R.P., Sloan, K.B. The effects of vehicles and prodrug properties and their interactions

on the delivery of 6-mercaptopurine through skin: bisacyloxymethyl-6-mercaptopurine pro-drugs, J. Pharm. Sci., 76, 587, 1987.




VEHICLE EFFECTS

IN PENETRATION

ENHANCEMENT

II

Smith and Maibach / Percutaneous Penetration Enhancers 2nd edn TF2152_c005 Final Proof page 65 9.9.2005 5:07am


Chapter 5

Penetration Enhancementby Skin Hydration

Jin Zhang, Carryn H. Purdon, Eric W. Smith, Howard I. Maibach, andChristian Surber

CONTENTS

Introduction ...................................................................................................................................... 67Stratum Corneum Hydration by Occlusion ..................................................................................... 68Stratum Corneum Hydration by Exogenous Chemicals ................................................................. 69Conclusions ...................................................................................................................................... 70References......................................................................................................................................... 71

Introduction

Normal skin is a partially hydrated tissue that maintains a consistent water content ofapproximately 5 to 15%, regardless of how much the humidity of the environment varies.The superficial epidermal layer, the stratum corneum (SC), plays a dual barrier role byminimizing the transepidermal water loss (TEWL), and by preventing external ingressdue to its extremely high impermeability. The hydration state of the skin affects thepermeability of the SC, thus percutaneous absorption enhancement may be achieved bysimply increasing the water content in the SC. While water, an endogenous constituent ofskin, acts as the penetration enhancer, it causes minimal irritancy or toxicity to the skin,and any transient manifestations are reversible in a short time. Skin hydration can beachieved quite simply by applying an occlusive vehicle (e.g., an ointment) to the skin[1, 2], or more elegantly by incorporating specific moisturizing factors into the vehicle orby using polymer patch delivery systems.


67


Stratum Corneum Hydration by Occlusion

The normal physiological water content of the SC depends on the balance between waterretention and water loss factors of the tissue. A continuous passive diffusion of wateroccurs from within the body through the SC and into the environment (TEWL) because ofthe water concentration gradient in the SC. Any external factor that binds water in thetissue or retards the evaporation of water from the skin surface will result in a hyper-normal hydration of the SC. This hydration is associated with a swelling of the corneo-cytes and an increase in the water content of the inter-corneocyte lipid bilayers. Both ofthese phenomena appear to facilitate xenobiotic delivery across the tissue, in a totallyreversible manner.

The predominating effect of total or partial skin occlusion is increased hydration of theSC; swelling of the corneocytes, and uptake of water into intercellular lipid domains, oftenincreasing the water content of the SC by up to 50% [3]. Furthermore, occlusion of thedosed skin prevents loss of the surface-deposited chemical by friction or by exfoliation andthereby topical delivery may be increased. Hydration of the SC can be achieved by simpleocclusion of the skin surface, even by the use of matrix-type topical formulations. Lipid-rich vehicles are most effective for this occlusive purpose; emulsions, ointments, and gelsare traditional preparations with occlusive characteristics. Several novel formulations havealso been developed that appear to have excellent occlusive and hydrating effects. Wissingand Muller [4] used a validated corneometer to investigate the effects of solid lipidnanoparticles (SLN) on the viscoelastic properties and hydration of treated skin, whencompared with a conventional O/W cream. After 4 weeks of treatment there were signifi-cant changes in skin hydration for both formulations, however, the SLN-enriched creamwas significantly more effective at hydrating the skin than the conventional cream (þ24%versusþ31%). SLN formulations appear to be a promising new family of vehicles that haveseveral beneficial properties for both therapeutic and cosmetic applications.

Early in the refinement of dermatological vehicles, McKenzie and Stoughton [5]investigated the importance of delivery vehicle-induced, or occlusive-wrap-inducedhydration of the skin. They showed that the degree of vasoconstriction of corticoids(i.e., the extent of active drug delivery) could be increased substantially by occluding thedosed site. Magnus et al. [6] showed that the intensity of vasoconstriction over a pro-longed observation time was directly related to the time of occlusion, and that the drugdelivery appears to maximize between 12 and 16 h after initial dosing. Furthermore,Feldmann and Maibach [7] showed that percutaneous hydrocortisone penetration intosystemic circulation increased tenfold under occlusion. In dermatological practice,Griffiths et al. [8] performed a clinical and immunohistologic study to investigate theeffects of a prolonged occlusive dressing, fluocinonide alone, and occlusion plus fluoci-nonide in patients with psoriasis. It was reported that the combination of fluocinonideointment and prolonged occlusion produced significantly greater clinical improvementthan either treatment alone (p < 0.01). There are a multitude of other examples availablein the literature. Faergemann et al. [9] have shown that occlusion increases the transepi-dermal flux of chloride and carbon dioxide, increases the microbial counts on skin, andincreases the pH of skin from 5.6 to 6.7. Treffel et al. [10] compared the in vitropermeation profiles of two molecules with different physicochemical properties underoccluded versus unoccluded conditions. The data showed that occlusion increased thepermeation of citropten (lipophilic compound) 1.6 times whereas that of caffeine(amphiphilic compound) remained unchanged. Furthermore, Makki et al. [11] assessedtopical absorption under occlusion of three furanocoumarins: 5-methoxypsoralen




(5-MOP), 8-methoxypsoralen (8-MOP), and trimethylpsoralen (TMP), in relation to theirmolecular polarity. The results illustrated that the effect of occlusion depended onthe chemical polarity; absorption of the moderately lipophilic molecules (8-MOP and5-MOP) was increased under occlusion, but that of the highly lipophilic TMP wasnegligible. These results give some clue as to the effects that hydration may have onthe SC barrier properties. Though the precise mechanism of penetration enhancement isstill under investigation, several modes have been proposed. Research has shown thatrather than just simply increasing the pore size of the membrane, a more complexrelationship may exist in that hydration opens up the dense structure of the SC and resultin fluidity or disorder between the intercellular lipid bilayers, thus allowing drugs topermeate the skin more easily through the relative disorder of the bilayer barrier.

Understanding the mechanism of action of penetration enhancers at a molecular levelis important in their judicious inclusion in transdermal formulations to enhance perme-ation of drugs across skin so as to achieve therapeutically effective blood concentrations.Due to the complexity of SC lipid composition, Narishetty and Panchagnula [12] used asimple model SC lipid system to study the effect of 1,8-cineole and L-menthol on phasebehavior and conformational order of SC lipid alkyl chains. After incorporation of 1,8-cineole and L-menthol into the model SC lipid system, hydration levels in terpene-incorporated model SC lipid systems were significantly increased as seen approximatelyfrom the 3300 cm1 frequency band in the ATR-FTIR results. This breaking of intermo-lecular hydrogen bonding might have led to hydration of the ceramide head groups andformation of new polar pathways.

Nair and Panchagnula [13] investigated the influence of electrical parameters on theiontophoretic transport of a small peptide, arginine–vasopressin (AVP). In vitro studiesusing rat skin were conducted to assess the effect of different current densities (CDs),durations, duty cycles, and alternating polarity on vasopressin permeation. FTIR and TGAwere used to understand the biophysical changes caused in skin due to passage ofcurrent. FTIR spectroscopic studies were used as a semiquantitative tool to investigatethe changes in water content of skin after exposure to different CDs and duration ofapplication. Skin hydration was found to increase up to CD 1 mA/cm2, whereas furtherincrease in density did not cause a corresponding rise in hydration, probably due to thedamage of structures in skin that hold water. Also hydration levels were constant after 6 hof current duration probably due to polarization of skin as exemplified by our observa-tion that there was no significant difference in the amount of AVP permeated between 6and 8 h current application. TGA was used to further substantiate the results of FTIRinvestigations. The first derivative thermogram of epidermis treated with different CDsshowed an increase in percent water loss as a function of CD. There was no significantdifference in water loss between 6 and 8 h of current duration. The results correlated wellwith the observations in FTIR spectra. Using FTIR and TGA in tandem, Nair and Pan-chagnula [13] established definitive increase in skin hydration levels as a function of CDand duration of current application during iontophoresis. Thus, all events that take placein skin during iontophoresis may be explained by increased skin hydration as a functionof CD and duration because polarization depends on water content in skin.

Stratum Corneum Hydration by Exogenous Chemicals

In the SC, the major factor in maintaining normal hydration is the intracellular hydrophilicand hygroscopic substances called ‘‘natural moisturizing factors’’ (NMF) [14]. The NMF


Penetration Enhancement by Skin Hydration & 69


comprise approximately 10% of SC dry weight and are principally composed of aminoacids (40%), carboxylic pyrrolidone acid (12%), lactic acid (12%), urea (8%), salts (18%),and other unidentified compounds (10%). NMF are principally contained within thecorneocytes and are formed during epidermal differentiation. Keratinization plays animportant part in the formation of NMF, that exhibit strong osmotic potential and attractwater molecules to the cellular environment. The binding of water to NMF (in thecorneocytes) is proposed to be a static aspect of cutaneous hydration.

The dynamic aspect of NMF chemicals is the effect these substances may have on thebarrier properties of the intercellular lipid bilayers. The suggestion is made that if NMFcomponents may be delivered to the intercellular domains, their hygroscopic propertieswould enhance the hydration of these bilayers, disrupting their organization and therebyincreasing the permeability of the SC. Cosmetic formulations attempt to make use of thisprinciple in the delivery of ‘‘moisturizing chemicals’’ to the skin. In this situation cos-metics may supply hydrophilic substances to the SC, capable of attracting and retainingwater (thereby acting as a moisturizer), or capable of restoring the normal skin barrierconditions (thereby restoring normal water loss or protection).

Some chemical components of the NMF composite have already been evaluatedindividually as penetration enhancers to improve drug delivery to the skin. Thesechemicals have been evaluated in detail in the first edition of this volume. The followingare a representative sample of chemicals from this family. Urea is a colorless, odorless,slightly hygroscopic, and very water soluble crystalline powder that facilitates hydrationof the SC. It forms hydrophilic diffusion channels within the barrier, and also acts as amild keratolytic agent that could affect the adhesiveness of SC corneocytes [15]. Pyrroli-dinones have been researched extensively as topical penetration enhancers [16], and2-pyrrolidinone-5-carboxylic acid is a component of the NMF in the skin. This enhancermay increase the hydration of the lipid bilayer, and may also be influential in morecomplex lipid rearrangements of the intercellular structure. The hygroscopic nature ofglycerol may explain its hydration effect on the SC. Bettinger et al. [17] found that thepenetration of hexyl nicotinate was increased at glycerol-treated sites, and that there wasa rapid reconstitution of the protective skin barrier following glycerol use.

Conclusions

Skin hydration may be utilized beneficially in clinical practice for enhancing transdermaldrug delivery. Hydration by occlusive systems or topical vehicles is still the most facileway to obtain a reduction in barrier potential of the SC. To date, clinicians and pharma-ceutical scientists still regard occlusion as a convenient and safe method of enhancingtransdermal drug delivery. However, often a simple increase in the water content of theSC may not sufficiently improve the transdermal delivery of specific drugs to a levelwhere therapeutic concentrations are achieved. In these cases, other methods of percu-taneous penetration enhancement, often in addition to the hydration, are essential. It isinteresting that although we have known about, and clinically used, occlusion to increasepercutaneous drug delivery for several decades, we are only now really starting tounderstand the complex array of biochemical and histological events that take placewhen the surface of the skin is occluded. Undoubtedly this picture will continue toevolve as we develop more sophisticated systems for quantifying the kinetics of thewater retention process.




References1. Surber, C. and Smith, E.W. (2005) The mystical effects of dermatological vehicles. Dermatol-

ogy, 210, 157–68.2. Smith, E.W., Meyer, E., and Haigh, J.M. (1990) Blanching activities of betamethasone formu-

lations. The effect of dosage form on topical drug availability. Arzneimittelforschung, 40,618–21.

3. Mak, V.H., Potts, R.O., and Guy, R.H. (1991) Does hydration affect intercellular lipid organ-ization in the stratum corneum? Pharm Res, 8, 1064–5.

4. Wissing, S.A. and Muller, R.H. (2003) The influence of solid lipid nanoparticles on skinhydration and viscoelasticity — in vivo study. Eur J Pharm Biopharm, 56, 67–72.

5. McKenzie, A.W. and Stoughton, R.B. (1962) Method for comparing percutaneous absorption ofsteroids. Arch Derm, 86, 608–10.

6. Magnus, A.D., Haigh, J.M., and Kanfer, I. (1980) Assessment of some variables affecting theblanching activity of betamethasone 17-valerate cream. Dermatologica, 160, 321–7.

7. Feldmann, R.J. and Maibach, H.I. (1965) Penetration of 14c Hydrocortisone through normalskin: the effect of stripping and occlusion. Arch Dermatol, 91, 661–6.

8. Griffiths, C.E., Tranfaglia, M.G., and Kang, S. (1995) Prolonged occlusion in the treatment ofpsoriasis: a clinical and immunohistologic study. J Am Acad Dermatol, 32, 618–22.

9. Faergemann, J., Aly, R., Wilson, D.R., and Maibach, H.I. (1983) Skin occlusion: effect onPityrosporum orbiculare, skin PCO2, pH, transepidermal water loss, and water content. ArchDermatol Res, 275, 383–7.

10. Treffel, P., Muret, P., Muret-D’Aniello, P., Coumes-Marquet, S., and Agache, P. (1992) Effect ofocclusion on in vitro percutaneous absorption of two compounds with different physico-chemical properties. Skin Pharmacol, 5, 108–13.

11. Makki, S., Muret, P., Said, A.M., Bassignot, P., Humbert, P., Agache, P., and Millet, J. (1996)Percutaneous absorption of three psoralens commonly used in therapy: effect of skin occlu-sion (in vitro study). Int J Pharm, 133, 245–52.

12. Narishetty, S.T. and Panchagnula, R. (2005) Effect of L-menthol and 1,8-cineole on phasebehavior and molecular organization of SC lipids and skin permeation of zidovudine. J ControlRelease, 102, 59–70.

13. Nair, V.B. and Panchagnula, R. (2004) Influence of electrical parameters in the iontophoreticdelivery of a small peptide: in vitro studies using arginine–vasopressin as a model peptide.Farmaco, 59, 583–93.

14. Marty, J.P. (2002) NMF and cosmetology of cutaneous hydration. Ann Dermatol Venereol, 129,131–6.

15. Clarys, P., Gabard, B., and Barel, A.O. (1999) A qualitative estimate of the influence ofhalcinonide concentration and urea on the reservoir formation in the stratum corneum. SkinPharmacol Appl Skin Physiol, 12, 85–9.

16. Williams, A.C. and Barry, B.W. (2004) Penetration enhancers. Adv Drug Deliv Rev, 56, 603–18.17. Bettinger, J., Gloor, M., Peter, C., Kleesz, P., Fluhr, J., and Gehring, W. (1998) Opposing effects

of glycerol on the protective function of the horny layer against irritants and on the penetrationof hexyl nicotinate. Dermatology, 197, 18–24.


Penetration Enhancement by Skin Hydration & 71


Chapter 6

Enhancement of Deliverywith Transdermal Sprays

Barrie C. Finnin and Jonathan Hadgraft

CONTENTS

Introduction ...................................................................................................................................... 74Enhancement of Delivery of the Dose to the Skin ......................................................................... 75

Metering Valves............................................................................................................................. 75Spray Pattern and Droplet Size .................................................................................................... 75

Applicator Design ..................................................................................................................... 75Control of Area ............................................................................................................................. 75

Volume Delivered Per Unit Area.............................................................................................. 75Containment of Spray............................................................................................................... 76Minimizing Nonproductive Loss .............................................................................................. 76

Enhancement of Delivery into the Stratum Corneum..................................................................... 76Topology of Skin Surface ............................................................................................................. 76Sebum ........................................................................................................................................... 77Available Area ............................................................................................................................... 78Increase in Thermodynamic Activity of Drug ............................................................................. 78

Evaporation of the Volatile Component of the Vehicle .......................................................... 78Choice of Nonvolatile Solvent.................................................................................................. 78Increasing the Solubility of Drug in the Stratum Corneum .................................................... 78

Enhancement of Permeation through Skin ..................................................................................... 79Solvent Drag ................................................................................................................................. 79Effect on Lipid Arrangement ........................................................................................................ 79Extraction of Lipids....................................................................................................................... 79Selective Extraction of Cholesterol .............................................................................................. 79

Enhancement of Bioavailability ....................................................................................................... 80


73


Characteristics of Solvents and Enhancers Required for Metered Dose Spray Delivery ............... 80Volatile Solvent ............................................................................................................................. 80Nonvolatile Solvent....................................................................................................................... 80Penetration Enhancer ................................................................................................................... 81

References......................................................................................................................................... 81

Introduction

There has been much published regarding the enhancement of transdermal penetrationusing chemical penetration enhancers.1 However, most of the studies have focused onhow a particular chemical can change the permeation rate of drugs across the stratumcorneum from solutions of drugs where an ‘‘infinite’’ dose is used.2 The point of manyof these studies is to investigate the mechanisms of enhancement3,4 and structure–activity relationships.5–8 Much less has been published regarding the measurement ofenhanced permeation from commercially viable preparations in vivo. Studies usingskin blanching as a measure of penetration have been performed.9 Use has also beenmade of noninvasive techniques such as ATR FTIR to measure the enhancement ofpenetration into the stratum corneum in vivo.10 This review canvasses the strategiesthat have been employed with a particular delivery system namely a metered dosetransdermal spray.

Morgan and others11 showed that application of estradiol to the skin via a metereddose aerosol could be used to deliver systemically therapeutically relevant amountsof estradiol in postmenopausal women. Subsequently a number of clinical studieshave been performed to characterize the behavior of similar systems with a numberof different drugs including estradiol, testosterone, and buspirone (unpublished

metered dose pumps rather than by aerosol. The essential features of these preparationswere:

1. A means of applying a metered dose of drug evenly over a defined area of skin2. Presence of a nonvolatile solvent to ensure that the drug remained in solution until

it had penetrated into the stratum corneum3. Presence of a chemical enhancer to facilitate the partitioning of the drug into the

stratum corneum and decrease the resistance to diffusion through the stratumcorneum

4. Presence of a volatile solvent to enable uniform spreading of the dose of drugand enhancer across the required application area and subsequently on evapor-ation increasing the concentration of drug on the skin encouraging penetration intothe stratum corneum

If we give a broad definition to ‘‘enhancement’’ and include all factors that lead toimproved performance of the delivery system then we should consider not only howrapidly the drug is delivered but more importantly the amount of drug delivered perdose, the bioavailability and the uniformity of dose not only between patients but alsowithin patients. The design of the formulation might be expected to influence all of thesemeasures. The following is a consideration of the contribution of each of the elements ofthe formulation to these measures.




data — refer to www.acrux.com.au). The latter studies were performed with manual

http://www.acrux.com.au

Enhancement of Delivery of the Dose to the Skin

Metering Valves

To minimize between dose variation it is important to deliver an accurate amount of druguniformly to a defined area. The spray achieves this by using metering valves that havebeen developed for nasal delivery or inhalation. Compendial and regulatory standardsfor the accuracy tolerances of doses delivered using these pumps have been developed.12

The accuracy achievable with such pumps is significantly better than with other methodsof topical application of doses except for patches.

Spray Pattern and Droplet Size

Apart from the actual dose delivered from the pump the spray pattern and droplet size areimportant variables. There is an optimal range of droplet size to ensure efficient depos-ition of most of the dose delivered onto the skin. If the droplet size is too small then asignificant proportion of the dose will remain as an aerosol and will be lost. This may alsocause some concern regarding the potential for inhalation of part of the dose. If thedroplet size is too large then an uneven distribution of the drug across the administrationsite is likely.13 The spray pattern and the droplet size are mainly controlled by the use ofan appropriate actuator nozzle. The droplet size and the velocity of the droplets as ejectedfrom the nozzle will also have an effect on the amount of bounce-back that occurs. If thevelocity of a drop is too high when it reaches the skin then a significant proportion of thedose will bounce off the skin. Enhancement of the delivery of the dose to the skin thusinvolves an accurate metering device, an appropriate nozzle and an optimal distancebetween the nozzle, and the skin. The separation between the nozzle, and the skin can beaccurately controlled by enclosing the nozzle in a shroud that is placed against the skin

Applicator Design

Another factor that is important in achieving reproducible delivery of the dose to the skinrelates to the ergonomics of the delivery system. The force of actuation will have asignificant effect on the velocity and droplet size of the spray. Application of sub-optimalforces means that the spray is not uniform and this leads to an uneven distribution of thedose across the area of application. The pumps are chosen such that once an initialresistance is met the actuation resistance is low and thus similar speed of actuation willoccur provided the initial resistance is overcome. It has been shown that thumb strengthgenerally is greater than forefinger strength14 and so applicators designed to be activatedusing the thumb will give more reproducible results.

Control of Area

Volume Delivered Per Unit Area

The uniformity of dispersion of the dose across the skin will be influenced by the volumeper unit area applied. If the volume applied per unit area is too low then it will not coverthe area of application. If the volume applied per unit area is too high then it will spread


Enhancement of Delivery with Transdermal Sprays & 75


during application. Figure 6.1 shows how the spray is designed to be applied.

beyond the area of application and may run off the skin altogether. For alcoholicsolutions the optimal volume is 2.5 ml/cm2. The area over which the spray is deliveredwill be controlled by the distance of the nozzle from the skin and the angle of the spray.

Containment of Spray

In addition to setting the distance from the skin the use of a shroud helps to define thearea of application and to contain bounce back. A cross-sectional drawing of a typical

Minimizing Nonproductive Loss

The proportion of a dose that is lost after deposition on the skin will be controlled by theproportion that actually penetrates into the stratum corneum. Reddy and others15 havecalculated that, except for high molecular weigh drugs and highly lipophilic drugs, only asmall proportion of a dose that has penetrated into the stratum corneum will be lost bydesquamation.

Enhancement of Delivery into the Stratum Corneum

Topology of Skin Surface

An important factor that will decide the efficiency of drug penetration into the stratumcorneum is the topology of the skin surface. It is generally accepted that the predominant

Figure 6.1 Method of use of a metered dose transdermal spray. The applicator shroud is heldagainst the skin and the spray is actuated with the thumb.




metered dose transdermal spray device is shown in Figure 6.2.

route of delivery of drugs is through the lipid matrix between the corneocytes.16 Before adrug can penetrate via this route it must first gain access to the spaces between thecorneocytes. The viscosity of the solution applied and the interfacial tension betweenthe corneocytes and the solution will be important determinants. The low viscosity andinterfacial tension associated with ethanolic solutions may contribute to enhancingdelivery into the spaces between the corneocytes.

Sebum

The first barrier that will be encountered by drug solutions sprayed onto the skin is thelayer of sebum (that is generally accepted to form a discontinuous film across the surface)and the residues of other topically applied preparations such as moisturizing creams andlotions, etc. The importance of sebum to the barrier properties of the skin was dismissedby Kligman17 on the basis of the thickness of sebum layers required to decrease transe-pidermal water loss. Further Higuchi18 concluded that — on the basis of the thickness ofthe film, and the fact that the diffusion barrier would only be significant for polarcompounds — sebum was not important for topical preparations such as creams orointments. These arguments do not apply where the drug is applied in small amountssuch as with the metered dose transdermal spray. Even if it does not act as a physicalbarrier to the access of drug to the skin it is likely to form a lipophilic environment19 intowhich drug might partition becoming less available for penetration into the stratumcorneum. The relative importance of this effect should not be discounted lightly sincein oily skin, the content of sebum on the surface of the skin can be over 250 mg /cm,2,20

which is greater than the amount of lipid in the stratum corneum. It is feasible that aneffect of the alcohol in the spray solution may be to affect the sebum layer increasingaccess of the drug to the lipid between the corneocytes. The application rate of topicallyapplied lotions and creams is of the order of 2 mg /cm2. If about 50% of this is water thatevaporates there will still be a residual of approximately 1000 mg/cm2. Again as with the

Skin surface(Shroud is held againstskin surface duringactuation)

h

Actuator button

HDPE bottle

Manual metered-dosepump valve

Shroud

Spray plume geometry

Housing (RHS)

Actuator nozzle

α

Figure 6.2 A cross-sectional drawing of a metered dose transdermal spray illustrating thecritical dimensions: h the distance of the nozzle from the skin surface and a the angle of thespray.




sebum content this is significant in comparison to the amount of lipid actually in thestratum corneum and might present a significant reservoir into which the drug mayunproductively partition.

Available Area

The majority of the area that a drug encounters once it has passed the barrier of thesebum will be taken up by the corneocytes rather than the gaps between them. Therelative areas will depend on the spaces between the corneocytes and the average area ofand shape of the corneocytes. In the superficial layers of the stratum corneum the inter-corneocyte gaps will be a function of the state of desquamation. If the wider gapsassociated with desquamation are ignored the calculated proportion of the surfacerepresented by the gaps between the corneocytes will be of the order of 1%. (Averagediameter of corneocyte 50 mm, gap between corneocytes 0.1 mm.) Thus if a drug isprecipitated from solution, on evaporation most of the drug will be deposited on thecorneocyte surface and be unavailable for absorption. To enhance the amount deliveredin a spray solution, it is thus important to ensure that after evaporation of the volatilevehicle the drug remains in solution to allow diffusion across the surface of the corneo-cytes. This can be achieved by the inclusion of a nonvolatile solvent as part of thedelivery system.

Increase in Thermodynamic Activity of Drug

Evaporation of the Volatile Component of the Vehicle

The driving force for partitioning of the drug into the lipid of the stratum corneum willbe the thermodynamic activity of the drug in the solution in contact with the lipid. Theevaporation of the volatile solvent will lead to an increase in concentration of drug in thesolution remaining on the skin. This is likely to lead to an increase in the thermodynamicactivity of the drug provided that the solubility of the drug in the nonvolatile solventleft on the skin is not significantly higher than its solubility in the solution originallysprayed on.

Choice of Nonvolatile Solvent

Clearly there will be an optimum solubility in the nonvolatile solvent since if the solubilityis too high, the drug will be less likely to partition into the lipid while if the solubility istoo low, the drug will precipitate on evaporation of the ethanol and become unavailablefor penetration. Enhancement of delivery into the stratum corneum thus relies onappropriate choice of the nonvolatile solvent. The disposition of the nonvolatile solventitself will also be important. If the solvent is taken into the stratum corneum, this willresult in an increase in the concentration of drug in the remaining solvent on the skinsurface.

Increasing the Solubility of Drug in the Stratum Corneum

Some penetration enhancers are thought to achieve their effect by increasing the solu-bility of drugs in the stratum corneum.21 Because of the relatively low amounts of lipid in




the stratum corneum, it is likely that solvents penetrating will alter the solvent propertiesof the stratum corneum.

Enhancement of Permeation through Skin

Solvent Drag

Ethanol is well recognized to increase the rate of permeation of a variety of drugsacross the skin. This effect has been attributed to a number of different mechanisms,all of which require the continuing presence of ethanol in the skin22 because mostof the ethanol administered in sprays will evaporate23 before it permeates across theskin. The ethanol in typical formulations evaporates in less than 60 sec and thus it isunlikely to have a significant role in enhancing the permeation by altering the diffusionwithin the stratum corneum.

Effect on Lipid Arrangement

The low rate of diffusion of compounds through the skin is generally attributed to thestructured arrangement of the lipids between the corneocytes of the stratum corneum.24

The actual mechanism by which this structuring impedes permeation is a matter forconjecture. Much depends on the mechanism of diffusion. If the diffusion occurs assolute within a lipid solution then the energy required to displace molecules of the lipidwill be high in an ordered solution and thus create a barrier (free volume theory25). If thediffusion occurs between the lipid areas as in diffusion through a polymer gel wherethere may be considered to be two distinct phases26 then the change in diffusivity can berelated to the enthalpy of mixing.27 Chemical permeation enhancers have been shown bya variety of techniques to disrupt the ordered packing of the lipids.28–30 The mechanismsof this effect might include insertion of the enhancer into the structure decreasingpacking.31 The enhancer may exert its influence by unilateral hydrogen bonding withthe polar regions of the ceramides in the stratum corneum.7

Extraction of Lipids

Extraction of lipids from the stratum corneum with solvents has been shown to result in adecreased permeation barrier.32 It is likely that the alcoholic solutions applied with thespray will result in the extraction of some lipids from the outer layers of the stratumcorneum.

Selective Extraction of Cholesterol

The mechanism by which extraction of lipids results in a decreased barrier is not clear;however, it is of interest that the lipid most likely to be extracted by the alcoholicsolutions of lipophilic enhancers is cholesterol.33 It is also of interest that the return ofbarrier function after delipidization requires synthesis of cholesterol.34 Given the import-ance of cholesterol in controlling the permeability of drugs through the lipid layers ofliposomes,35 it is possible that selective extraction of cholesterol from the lipid mixture inthe stratum corneum may disrupt the ordered structure of the lipids.




Enhancement of Bioavailability

Where the objective of transdermal delivery is chronic administration of drug, the extentof delivery is more important than the rate of delivery. The bioavailability of drugsapplied to the skin in finite doses depends on the amount of drug applied per unitarea. Wester and Maibach36 showed that the bioavailability of testosterone from asolution in acetone applied to the skin was 8.8% when 30 mg/cm2 was administeredbut only 2.8% when 400 mg/cm2 was administered. Numerous other studies have alsoshown that the bioavailabilty of drugs applied to the skin in finite doses is dose related.Notwithstanding this, generally the total amount delivered (rather than the proportion ofdose delivered) usually increases with increasing dose. The key to achieving highbioavailability is to be able to deliver the maximum proportion of the dose to the stratumcorneum. It has been shown that the amount of drug in the stratum corneum half an hourafter application in a finite dose is a good indicator of the eventual systemic bioavail-ability.37 This indicates that the most important factor will be the partitioning of the drugfrom the site of application into the stratum corneum.

Characteristics of Solvents and Enhancers Requiredfor Metered Dose Spray Delivery

Volatile Solvent

The most important properties for the volatile solvent in such formulations are:

1. Volatility — for convenience the applied solution should dry within a reasonableperiod, say, 3min.

2. Solvency — it is important for the solubility of drug in the formulation doesnot limit the concentration that can be applied. The volatile solvent should alsobe able to dissolve the nonvolatile solvent and any combination of enhancersused.

3. Lack of toxicity and irritation — it needs to be borne in mind that only a very smallproportion of the administered dose is likely to absorbed. Concerns such asextraction of lipid and consequent causation of ‘‘dry’’ skin are likely to be reducedby the presence of the nonvolatile component.

Clincal studies with metered dose sprays have been reported for estradiol11 with ethanol90% v/v as the volatile solvent.

Nonvolatile Solvent

The function of the nonvolatile solvent is to keep the drug in solution after evaporationof the volatile component. The objective of balancing the solubility of drug betweenbeing too soluble and precipitating can be achieved by adjusting the relative proportionsof drug and nonvolatile solvent in the formulation. For maximum delivery it is preferablefor the nonvolatile solvent itself to have a high affinity for the lipid of the stratumcorneum. In the previously mentioned studies with estradiol11 and testosterone, octisa-late and padimate O have acted both as the nonvolatile solvent and as penetrationenhancers.




Penetration Enhancer

The important features of enhancers for use in metered dose transdermal sprays include:

1. Solubility in the combination of volatile and nonvolatile solvents2. Affinity for the stratum corneum3. Ability to increase partitioning into the skin

In summary, enhancement of delivery of drugs from a metered dose transdermal spraycan be achieved by using a device to deliver a uniform amount of drug over a definedarea of skin using a combination of solvents and enhancers that will lead to maximaldelivery of drug into the stratum corneum.

References1. Williams, A.C. and Barry, B.W., Penetration enhancers, Adv. Drug Deliv. Rev., 56, 603, 2004.2. Bach, M. and Lippold, B.C., Percutaneous penetration enhancement and its quantification,

Eur. J. Pharm. Biopharm., 46, 1, 1998.3. Suhonoe, T.M., Bouwstra, J.A., and Urtti, A., Chemical enhancement of percutaneous absorp-

tion in relation to stratum corneum structural alterations, J. Control. Release, 59, 149, 1999.4. He, N., Warner, K.S., Chantasart, D., Shaker, D.S., Higuchi, W.I., and Li, S.K., Mechanistic study

of chemical skin permeation enhancers with different polar and lipophilic functional groups,J. Pharm. Sci., 93, 1415, 2004.

5. Kanikkannan, N., Kandimalla, K., Lamba, S.S., and Singh, M., Structure–activity relationship ofchemical penetration enhancers in transdermal drug delivery, Curr. Med. Chem., 7, 593, 2000.

6. Kim, N., El-Kattan, A.F., Asbill, C.S., Kennette, R.J., Sowell Sr., J.W., Latour, R., and Michniak,B.B., Evaluation of derivatives of 3-(2-oxo-1-pyrrolidine)hexahydro-1H-azepine-2-one as der-mal penetration enhancers: side chain length variation and molecular modelling, J. Control.Release, 73, 183, 2001.

7. Hadgraft, J., Peck, J., Williams, D.G., Pugh, W.J., and Allan, G., Mechanisms of action of skinpenetration enhancers/retarders: Azone and analogues, Int. J. Pharm., 141, 17, 1996.

8. Warner, K.S., Li, S.K., He, N., Suhonen, T.M., Chantasart, D., Bolikal, D., and Higuchi, W.I.,Structure–activity relationship for chemical skin penetration permeation enhancers: probingthe chemical microenvironment of the site of action, J. Pharm. Sci., 92, 1305, 2003.

9. Smith, E.W. and Haigh, J.M. Assessing penetration enhancers for topical corticosteroids, inPercutaneous Penetration Enhancers, Smith, E.W. and Maibach, H.I., Eds, CRC Press, Inc.,Boca Raton, FL, 1995, Chapter 16.4.

10. Mak, V.H., Potts, R.O., and Guy, R.H., Percutaneous penetration enhancement in vivo meas-ured by attenuated total reflectance infrared spectroscopy, Pharm. Res., 7, 835, 1990.

11. Morgan, T.M., O’Sullivan, H.M.M., Reed, B.L., and Finnin, B.C., Transdermal delivery ofestradiol in postmenopausal women with a novel topical aerosol, J. Pharm. Sci., 87, 1226,1998.

12. Guidance for Industry — nasal spray and inhalation solution, suspension, and sprayproducts — chemistry, manufacturing and controls documentation, U.S. Department of Healthand Human Services, Food and Drug Administration, Center for Drug Evaluation andResearch, 2002.

13. Halbert, M.K., A perspective on particulate size analysis of consumer aerosol products,Environ. Res., 33, 189, 1984.

14. Hertzberg, H.T.E., Engineering anthropology, in Human Engineering Guide to EquipmentDesign, Van Cott, H.P. and Kinkade, R.G., Eds, U.S. Department of Defense, Washington, D.C.,1972, pp. 563–565.




15. Reddy, M.B., Guy, R.H., and Bunge, A.L., Does epidermal turnover reduce percutaneouspenetration? Pharm. Res., 17, 1414, 2000.

16. Walters, K.A., Stratum corneum: biological and biochemical considerations. Transdermal drugdelivery, in Development Issues and Research Initiatives, Hadgraft, J. and Guy, R.H., Eds,Marcel Dekker, Inc., New York, 1989, pp. 23–57.

17. Kligman, A.M., The uses of sebum, Br. J. Dermatol., 75, 307, 1963.18. Higuchi, T., Physical chemical analysis of percutaneous absorption process from creams and

ointments, J. Soc. Cosmet. Chem., 11, 85, 1960.19. Stewart, M.E., Sebaceous gland lipids, Semin. Dermatol., 11, 100, 1992.20. Rode, B., Ivens, U., and Serup, J., Degreasing method for the seborrheic areas with respect to

regaining sebum excretion rate to casual level, Skin Res. Technol., 6, 92, 2000.21. Harrison, J.E., Watkinson, A.C., Green, D.M., Hadgraft, J., and Brain, K., The relative effect of

Azone and Transcutol on permeant diffusivity and solubility in human stratum corneum,Pharm. Res., 13, 542, 1996.

22. Yum, S., Lee, E., Taskovich, L., and Theeuwes, F., Permeation enhancement with ethanol:mechanism of action through skin, in Drug Permeation Enhancement Theory and Applica-tions — Drugs and the Pharmaceutical Sciences, vol. 62, Hsieh, D.S., Ed., Marcel Dekker, Inc.,New York, 1994, pp. 143–170.

23. Pendlington, R.U., Whittle, E., Robinson, J.A., and Howes, D., Fate of ethanol topically appliedto skin, Food Chem. Toxicol., 39, 169, 2001.

24. Golden, G.M., McKie, J.E., and Potts, R.O., Role of stratum corneum lipid fluidity in transder-mal drug flux, J. Pharm. Sci., 76, 25, 1987.

25. Kasting, G.B., Smith, R.L., and Copper, E.R., Effect of lipid solubility and molecular size onpercutaneous absorption, in Skin Pharmacokinetics, vol. 1, Shroot, B. and Schaefer, H., Eds,Karger, Basel, 1987, pp. 138–153.

26. Clough, S.B., Read, H.E., Metzner, A.B., and Behn, V.C., Diffusion in slurries and in non-newtonian fluids, AIChE J., 8, 346, 1962.

27. Li, S.U. and Gainer, J.L., Diffusion in polymer solutions, Ind. Chem. Eng. Fund., 7, 433, 1968.28. Goodman, M. and Barry, B.W., Differential scanning calorimetry (DSC) of human stratum

corneum: effect of Azone, J. Pharm. Pharmacol., 37, 80P, 1985.29. Takeuchi, Y., Yasukawa, H., Yamaoka, Y., Morimoto, Y., Nakao, S., Fukumori, Y., and Fukuda,

T., Destabilization of whole skin lipid bio-liposomes induced by skin penetration enhancersand FT-IR/ATR (Fourier transform infrared/attenuated total reflection) analysis of stratumcorneum lipids, Chem. Pharm. Bull., 40, 484, 1992.

30. Anigbogu, A.N.C., Williams, A.C., Barry, B.W., and Edwards, H.G.M., Fourier transform ramanspectroscopy of interactions between the penetration enhancer dimethyl sulfoxide and humanstratum corneum, Int. J. Pharm., 125, 265, 1995.

31. Francoeur, M.L., Golden, G.M., and Potts, R.O., Oleic acid: its effects on SC in relation to(trans)dermal drug delivery, Pharm. Res., 7, 621, 1990.

32. Scheuplein, R. and Ross, L., Effects of surfactants and solvents on the permeability of epider-mis, J. Soc. Cosmet. Chem., 21, 853, 1970.

33. Monteiro-Riviere, N.A., Inman, A.O., Mak. V., Wertz, P., and Riviere, J.E., Effect of selectivelipid extraction from different body regions on epidermal barrier function, Pharm. Res., 18,992, 2001.

34. Feingold, KR., Man, M.Q., Menon, G.K., Cho, S.S., Brown, B.E., and Elias, P.M., Cholesterolsynthesis is required for cutaneous barrier function in mice, J. Clin. Invest., 86, 1738, 1990.

35. Lelievre, J. and Rich, G.T., Permeability of lipid membranes to nonelectrolytes, Biochim.Biophys. Acta, 298, 15, 1973.

36. Wester, R.C. and Maibach, H.I., Relationship of topical dose and percutaneous absorption ofrhesus monkey and man, J. Invest. Dermatol., 67, 518, 1976.

37. Rougier, A., Dupuis, D., Lotte, C., Roguet, R., and Schaefer, H., In vivo correlation betweenstratum corneum reservoir function and percutaneous absorption., J. Invest. Dermatol., 81,275, 1983.




Chapter 7

Hydrogel Vehicles forHydrophilic Compounds

Teresa Cerchiara and Barbara Luppi

CONTENTS

Hydrogels.......................................................................................................................................... 83Introduction .................................................................................................................................. 83Physical and Chemical Properties of Hydrogels ......................................................................... 84

Applications of Hydrogels and Patches in Transdermal Delivery.................................................. 87Hydrogels...................................................................................................................................... 89Transdermal Patches..................................................................................................................... 90

Conclusions ...................................................................................................................................... 91References......................................................................................................................................... 91

Hydrogels

Introduction

Hydrogels date back to 1960 when Wichterle and Lim first proposed the use of hydro-philic networks of poly(2-hydroxyethylmethacrylate) (PHEMA) in contact lenses [1].Since then, the use of hydrogels has extended to various biomedical [2] and pharmaceut-ical [3] applications. In particular, due to their physical properties similar to those ofhuman tissues (water content, soft, and pliable consistence) hydrogels have been usedfor different administration routes such as oral, rectal, ocular, epidermal, and subcutane-ous [3, 4].

Hydrogels are composed of hydrophilic macromolecules forming three-dimensionalinsoluble networks able to imbibe large amounts of water or biological fluids [5].Commonly, the polymers utilized to make hydrogels are insoluble due to the presenceof permanent or reversible crosslinks [6]. Permanent crosslinked hydrogels [1, 7, 8] arecharacterized by covalent bonds forming tie-points or junctions, whereas reversible


83


crosslinked hydrogels [9–11] present ionic, hydrophobic, or coiled-coil physical inter-actions. These kind, of crosslinks in the polymer structure yield insoluble materials ableto swell in aqueous environments retaining a significant fraction of water in theirstructure, up to thousands of times their dry weight in water.

Hydrogels can be divided into homopolymer or copolymers based on the preparativemethod, but they can also be natural polymers, synthetic polymers, or derivatives.In nature hydrogels can be found in plants (pectin, pullulan), various species ofbrown seaweed (alginic acid, agar, carrageenan), crustaceans (chitin) and animal tissue(hyaluronic acid, collagen, fibrin). Typical simple synthetic materials applied for general-purpose hydrogels are poly(ethylene oxide), poly(vinyl alcohol), poly(vinyl pyrroli-done), poly(hydroxyethylmethacrylate) and poly(n-isopropyl acrylamide). Moreover,the synthetic pathway offers more possibilities to create hydrogels with modified func-tional properties. In fact, several physiologically responsive hydrogels are obtained fromchemical or physical modifications of natural and synthetic polymers and tested for usein the so-called ‘‘intelligent biomaterials’’ [12–15] because they are capable of reactingto various environmental stimuli (temperature, pH, ionic strength, solute concentration,electric radiation, light, sound, etc.).

Hydrogels can be homogeneous, when the pores between polymer chains are theonly spaces available for mass transfer and the pore size is within the range of moleculardimensions (a few nanometers or less), or porous when the effective pore size is over 10nm. In homogeneous hydrogels, the transfer of water or other solutes is achieved by apure diffusional mechanism, which restricts the rate of absorption and to some extent thesize of species that are absorbed.

Porous hydrogels can be made by different polymerization methods in the presenceof dispersed porosigens (ice crystals, oil, sucrose crystals) which can be removed later toleave an interconnected meshwork, where the pore size depends on the size of theporosigens [16]. The introduction of a porosigen reduces mechanical strength signifi-cantly making porous hydrogels weaker than homogeneous hydrogels.

In medical, engineering, and pharmaceutical technology, hydrogel degradation isconsiderably important. In fact, investigators have focused on controlling the degradationbehavior of hydrogels to design polymers that can be cleared from the body once theycomplete their roles [17, 18]: for this reason labile bonds are frequently introduced in thegels. These bonds can be present either in the polymer backbone or in the crosslinksused to prepare the gel. Labile bonds can be broken under physiological conditionseither enzymatically or chemically, in most cases by hydrolysis [19–21].

Physical and Chemical Properties of Hydrogels

An important property of hydrogels is their swelling behavior: it depends upon thepolymer, extent of crosslinking, temperature, polymer–solvent interactions, and extentof ionization [22]. In particular, the extent of crosslinking can be changed to achieve arelatively strong and yet elastic hydrogel. Long chain crosslinkers and low crosslinkingratios (the ratio of moles of crosslinking agent to the moles of polymer repeating units)produce extremely weak hydrogels, while short chain crosslinkers and high crosslinkingratios produce extremely brittle hydrogels. Tightly crosslinked hydrogels will swell lessthan the same hydrogels with high crosslinking ratios or long crosslinkers chains.

The presence of hydrophilic or hydrophobic groups in the chemical structure of thepolymer affects the swelling behavior of hydrogels. When a dry hydrogel begins to




absorb water, the first water molecules entering the matrix will hydrate polar hydrophilicgroups. As the polar groups are hydrated, the network swells exposing hydrophobicgroups which also interact with water molecules, leading to hydrophobically boundwater. Finally an equilibrium swelling level is reached when the network imbibesadditional water (‘‘free water’’) which fills the space between the polymer chains. Ashydrophobic groups minimize their exposure to the water molecule, hydrogels contain-ing hydrophobic groups will swell much less than hydrogels containing hydrophilicgroups.

As stated earlier, the dissolution of polymer chains and consequently hydrogel swell-ing ability is prevented by the presence of crosslinking in the three-dimensional network.Different chemical and physical crosslinking methods have been employed to preparehydrogels [23]. In chemically and physically crosslinked gels, dissolution is prevented bycovalent bonds and physical interactions between different polymer chains, respectively.Chemically crosslinked gels can be obtained by radical polymerization of low molecularweight monomers in the presence of crosslinking agents, chemical reaction of comple-mentary groups, and high energy irradiation. Physically crosslinked gels can be obtainedby ionic interactions, hydrogen bonds, crystallization, and aggregation of the hydropho-bic segments of multiblock copolymers or graft copolymers. An example of crosslinkingby radical polymerization is the synthesis of hydrogels of Wichterle and Lim [1], acopolymerization of HEMA with the crosslinker EGDMA (ethylenglycol-dimethacrylate)in the presence of AIBN (2,2’-azo-bis-isobutyronitrile), the radical initiator (Figure 7.1).

linkages with functional groups of polymers, such as activated hydroxylic groups of

CH2 CH2

CH3 CH3

CO

O

OH

CO

O

CH2 CH2

CH3

CH3CO

O

CO

O

OH

m

CH CH

CH3 CH3

CO

O

CH2 CH2 CH2 CH2

CH2 CH2

CH2

CH2

CH2

CH3

CH2

OH

CO

O

CH

CO

O

+ nAIBN

90C

HEMA

EGDMA

P(HEMA-co-EGDMA)

Figure 7.1 Schematic representation of radical polymerization. Hydrogels are formed by thecopolymerization of HEMA with EGDMA using AIBN as the radical initiator.


Hydrogel Vehicles for Hydrophilic Compounds & 85


Chemical crosslinking agents, such as acyl dichlorides (Figure 7.2), can establish covalent

poly(vinyl alcohol) [24]. Polyaldehydes are utilized to crosslink proteins such as albumin[25] and gelatin [26], or natural polysaccharides such as hyaluronic acid [27]. However, asignificant disadvantage of chemical crosslinking agents is their toxicity. Among variousmethods applied for the production of hydrogels, the radiation technique [28] is a simple,efficient, clean, and environment-friendly process Hydrogels can beobtained by radiation technique in a few ways, including irradiation of solid polymer[29], monomer (in bulk or in solution) [30], or aqueous solution of polymer [31]. Forirradiation technologies, the main irradiating sources include gamma rays from radio-active isotopes such as cobalt 60, electron beams from electron accelerators, and x-raysconverted from electron beams.

Physical crosslinking of hydrogels also avoids the use of chemical crosslinking agents.Such agents can potentially inactivate the active principle and covalently link it to thehydrogel network. Examples of ionically crosslinked alginate hydrogels have beenreported [32]. Alginate is a family of linear polysaccharides composed of mannuronicacid (M) and guluronic acid (G). The chemical composition and sequence of M and Gresidues depend on the source from which the alginate has been extracted. The gelationof alginate is mainly achieved by the exchange of sodium ions with divalent cations suchas Ca2þ, Cu2þ, Zn2þ, or Mn2þ, which can form cation bridges between adjacent mol-ecules. The ‘‘egg-box’’ model of Grant et al. [32] is generally taken into consideration toexplain the formation of a rod-like crosslinked complex due to the bonding of thedivalent cations in the interchain cavities. Some polymeric complexes can be heldtogether by hydrogen bonds: poly(acrylic acid) and poly(methacrylic acid) providephysically crosslinked hydrogels with poly(ethylene glycol) (PEG) due to the formationof hydrogen bonds between the oxygens of PEG and the carboxylic groups of the acrylicpolymers [33]. Another physical method for producing physically crosslinked hydrogels isthe formation of crystalline regions in the polymer network, obtained by casting dilute,aqueous solutions of poly(vinyl alcohol), then cooling to –20 8C and thawing back toroom-temperature several times [34]. These frozen and thawed gels have demonstrated

Cl CO CH2 CO Cln

CH2 CH

OH

xCH2 CH

O

CO

x

CH2

CO

O

CH CH2

y

n

+

n = 2 SUCCINYLn = 4 ADIPOYLn = 8 SEBACOYL

Figure 7.2 Preparation of PVA hydrogels crosslinked by acyl dichlorides.




(Figure 7.3).

enhanced physical properties, such as high mechanical strength and high elasticity, thatmake them suitable for biomedical applications.

Finally, physically crosslinked hydrogels can be obtained by hydrophobic modifica-tion of polymers and in particular of polysaccharides such as chitosan, dextran, pullulan,and carboxymethyl curdlan [35–37]. Glycol chitosan substituted with palmitoyl chains isan example of a hydrophobized polysaccharide. The attachment of hydrophobic groups

[38]. Noncovalent crosslinking is achieved by the hydrophobic interactions of the palmi-toyl groups and a gel matrix is formed. Finally, our research group [37] reported physic-ally crosslinked chitosan hydrogels with lauric, myristic, palmitic, or stearic acid preparedby freeze-drying and studied for topical use These polymers producehydrogels with different functional properties related to the different acyl chains intro-duced in the polymer structure. In particular, the permeation of hydrophilic substancesthrough the skin can be modulated by increased or decreased drug solubility due to theinteraction of the different acyl chains with the stratum corneum.

Applications of Hydrogels and Patches in Transdermal Delivery

Transdermal drug delivery is an important route for delivering drugs that are destroyedby passing the first-pass metabolism and offers several other advantages [3] over con-ventional routes:

O OOCH2 CH2 CH2 CH2 CH2 CH2 CH2 CH2

CH O OO CH2CH2CH2CH2CH2CH2CH2

.

high radiation energy

crosslinking

Figure 7.3 Schematic representation of the radiation method to design hydrogels.




(Figure

to glycol chitosan yields an amphiphilic polymer capable of self-assembly into vesicles

7.4).

O

O

O

NH2

CH2OH CH2OH

OH OH O

NH

CO

CH2

CH3

n

CO

NH

O OO

H

OH

H

OH

H

OH

H

O

O

O

H

OH

H

OH

NH2

H

OH

H

CO

HN

OO

O

H

OH

H

OH

H

OH

H

O

O

O

H

OH

H

OH

H2NH

HO

H

n = 10 lauroyl chainn = 12 miristoyl chainn = 14 palmitoyl chainn = 16 stearoyl chain

Figure 7.4 Structural representation of physically crosslinked chitosan hydrogels.




& Drug delivery is noninvasive with no trauma or risk of infection [39].& Drugs are transported through the skin into the blood circulation.& Constant drug levels are reached and maintained.& Drug delivery can be interrupted by removing the system.

However, major disadvantage of transdermal drug delivery is that the drug itself or thematerials used to fabricate the vehicles may sometimes induce an irritation or reaction ofthe skin [40, 41].

In spite of the advantages, few molecules have been successfully delivered transder-mally, possibly due to the stratum corneum that forms a barrier to the permeation ofhydrophilic drugs. Consequently, recent research trends are focusing on approachessuch as microneedles [42], iontophoresis and electroosmosis [43], electroporation [44],and radiofrequency energy [45]. Particular attention has been paid to hydrogels asvehicles for transdermal delivery and therefore we reported the main characteristics ofhydrogels and patches.

Hydrogels

Recently, the topical use of hydrogels has increased and polymers such as chitosan,polyvinylalcohol (PVA), PEG, and many others used to design hydrogels are now widelyinvestigated as percutaneous penetration enhancers. In fact, the tight junctions present inthe epidermal barrier can be transiently opened by the interaction with these hydrogelsand may offer a pathway of absorption of hydrophilic and macromolecular drugs [46].

Gayet and Fortier [47] synthesized a new hydrogel obtained by the copolymerizationof bovine serum albumin (BSA) and PEG of various molecular masses. The release ofhydrophilic and hydrophobic drugs was a Fickian diffusion-controlled process due to thehigh water content of these BSA–PEG hydrogels. They suggested using the new family ofBSA–PEG hydrogels as drug-release devices in the field of wound dressing.

Another important polymer utilized for transdermal delivery is chitosan, a polysac-charide comprising copolymers of glucosamine and N-acetylglucosamine and derived bythe partial deacetylation of chitin. It is nontoxic and bioabsorbable [48] and has beenexplored for the release of many drugs [49, 50]. The research group of Bernkop–Schnurch[51] generated a novel bioadhesive polymer by covalent attachment of EDTA to chitosan.The polymer conjugate is more bioadhesive than unmodified chitosan and readilyhydratable, so it has been tested for possible topical use. In particular, the NaChito-EDTA gels are microbially stable and have excellent swelling properties [52].

Recently, our research group [37] described physically crosslinked chitosan hydrogelswith lauric, myristic, palmitic, or stearic acid able to enhance the skin permeation ofpropranolol hydrochloride selected as a hydrophilic model drug. The aim of the workwas to improve the permeation of drugs through biological membranes, as reported byNoble and co-workers [36] using hydrogels made of amphiphilic polymer. The concomi-tant presence of hydrophobic and hydrophilic groups in the polymer influenced theswelling properties. So, at pH 7.4 all hydrogels swelled slowly and their behaviorinfluenced the drug release. Among the different chitosan gels, chitosan laurate, andchitosan myristate enhanced drug permeation through the skin with respect to chitosanpalmitate and chitosan stearate hydrogels. This could be explained by the interaction ofthe hydrogels with the stratum corneum, increasing the solubility of the drug in the skin.

Hydrophilic and biocompatible polymers can be used to design hydrogels that areable to release hydrophilic drugs through the skin. Due to its properties, PVA crosslinked




with succinyl, adipoyl, or sebacoyl chloride [24] was employed as a supporting materialto release, generally hydrophilic drug. These hydrogels increased the solubility of thedrug which was attributed the polymer’s ability to be partially solubilized in the stratumcorneum by the presence of hydrophilic and hydrophobic groups.

Hydrosoluble drugs, as methotrexate (MTX), were incorporated into hydrogels pre-pared with the monomers acrylic acid and acrylamide and were used as vehicles for thelocal treatment of psoriasis [53, 54]. Studies of swelling have shown that these systems canbe utilized for the transdermal delivery of MTX preserving its stability.

Recently, Luppi and co-workers [55] used the crosslinked poly(methyl vinyl ether-co-maleic anhydride) as a topical vehicle for pyridoxine hydrochloride, selected as ahydrophilic model drug. In particular, poly(methyl vinyl ether-co-maleic anhydride)was crosslinked with ethylene glycol, butanediol, 1,6-exanediol, 1,8-octanediol, 1,10-decanediol or 1,12-dodecanediol. In vitro permeation studies were influenced bythe nature of the crosslinker: the decrease in crosslinker acyl chain length providesvehicles accelerating drug permeability through the skin.

Transdermal Patches

At present, only a few drugs are available in the form of transdermal drug deliv-ery patches. The patches consisted of membrane-controlling transdermal drug deliverysystems as reported, for example, by Tacharodi and Panduranga Rao [56]. In particular,they utilized chitosan membranes crosslinked with different concentrations of glutaral-dehyde and chitosan gel as drug reservoir. The membranes’ mechanical properties andin vitro drug release depended on the concentrations of crosslinker and on the area ofthe devices. In conclusion, chitosan membranes might be promising candidates astransdermal devices.

Another example of a matrix for transdermal delivery containing propranolol wasprepared using three different polymers (hydroxypropylmethylcellulose, polyisobutileneand Ucecryl1MC808) [57]. Among three different patches, promising results wereobtained with hydroxypropylmethylcellulose matrices coated with an aqueous disper-sion of an acrylic copolymer (Ucecryl MC808) and the presence of propylene glycolaccelerating the drug diffusion rate through matrices. This can be attributed to the drugcrystals dispersed in the matrix making the rate of drug release constant.

An adhesive hydrogel patch based on a hydrophilic matrix of poly(N-vinylpyrroly-done) (PVP) and oligomeric short-chain PEG was reported by Feldstein and co-workers[58]. They observed that the drug delivery rates from the hydrophilic transdermal systemswere higher than from the hydrophobic ones and depended on drug solubility in water.Kinetic studies were evaluated from this hydrogel to quantify the influence of the matrixand the membrane properties to drug delivery rate control [59].

Ethylene–vinyl acetate (EVA) matrix was tested as a membrane for transdermaldelivery of atenolol with the presence of plasticizers increasing the rate of drug release[60]. The effect of drug concentration, temperature, and plasticizers was investigated.Drug release from the EVA matrix follows a diffusion-controlled model and consequentlythe system could be used for transdermal delivery of hydrophilic drug.

Recently Padula and co-workers proposed a film not adhesive in the dry state, butbioadhesive when applied on wet skin [61]. The film studied was applied to the skin inthe presence of a certain amount of water. Water swelled the film on the surface incontact with the skin, transforming a dry polymeric matrix into a jellified polymer layer.




This particular film is flexible, mechanically resistant, and can avoid skin occlusionbecause of its permeability to water vapor. Compared to a typical patch, the bioadhesivefilm has a monolayer structure which includes backing, adhesive, and drug reservoirfunctions in one layer. Devising a simple delivery system composed of a smaller numberof layers could simplify the preparation procedure and innovate the field.

Conclusions

Transdermal delivery is a major administration route for drugs that are destroyed by theliver when taken orally [62]. Recently, much attention has been paid to hydrogels asvehicles for transdermal drug delivery and the success of hydrogels can be attributed tothe different methods of preparations. Many of the polymers used to design hydrogels arenontoxic and biocompatible and the incorporation of drugs into hydrogels permitsmodulation of their release kinetics [63]. Today, few drugs have been successfullydelivered transdermally utilizing hydrogels and patches as vehicles so that their futureuse in transdermal delivery will increase.

References1. Wichterle, O. and Lim, D., Hydrophilic gels for biological use, Nature, 185, 117, 1960.2. Hoffman, A.S., Hydrogels for biomedical applications, Adv. Drug Deliv. Rev., 54, 3, 2002.3. Peppas, N.A., Hydrogels in pharmaceutical formulations, Eur. J. Pharm. Biopharm., 50, 27,

2000.4. Guy, R.H., Current status and future prospects of transdermal drug delivery, Pharm. Res., 13,

1765, 1996.5. Peppas, N.A. and Mikos, A.G., Preparation methods and structure of hydrogels, in Hydrogels in

Medicine and Pharmacy, vol. 1, Peppas, N.A., Ed., CRC Press, Boca Raton, FL, 1986, 1.6. Berger, J. et al., Structure and interactions in covalently and ionically crosslinked chitosan

hydrogels for biomedical applications, Eur. J. Pharm. Biopharm., 57 (1), 19, 2004.7. Xiao, C. and Zhou, G., Synthesis and properties of degradable poly(vinyl alcohol) hydrogel,

Polym. Degrad. Stabil., 81 (2), 297, 2003.8. Brasch, U. and Burchard, W., Preparation and solution properties of microhydrogels from

poly(vinyl alcohol), Macromol. Chem. Phys., 197, 223, 1996.9. Watanabe, T. et al., NMR studies on water and polymer diffusion in dextran gels. Influence of

potassium ions on microstructure formation and gelation mechanism, Magn. Reson. Med., 35,697, 1996.

10. Wang, C., Steward, R.J., and Kopecek, J., Hybrid hydrogels assembled from synthetic polymersand coiled-coil protein domains, Nature, 397, 417, 1999.

11. Qu, X., Wirsen, A., and Albertson, A.-C., Synthesis and characterization of pH-sensitivehydrogels based on chitosan and D,L-lactic acid, J. Appl. Polym. Sci., 74, 3186, 1999.

12. Miyata, T., Uragami, T., and Nakamae, K., Biomolecule-sensitive hydrogels, Adv. Drug Del.Rev., 54 (1), 79, 2002.

13. Hoffman, A.S., Environmentally sensitive polymers and hydrogels — ‘‘smart’’ biomaterials,MRS Bull., 42, 1991.

14. Murdan, S., Electro-responsive drug delivery from hydrogels, J. Control. Rel., 92 (1–2), 1, 2003.15. Chen, L., Tian, Z., and Du, Y., Synthesis and pH sensitivity of carboxymethyl chitosan-based

polyampholyte hydrogels for protein carrier matrices, Biomaterials, 25 (17), 3725, 2004.16. Hickey, A.S. and Peppas, N.A., Mesh size and diffusive characteristics of semicrystalline

poly(vinyl alcohol) membranes prepared by freezing/thawing techniques, J. Membr. Sci.,107, 229, 1995.




17. Anderson, J.M. and Shive, M.S., Biodegradation and biocompatibility of PLA and PLGA micro-spheres, Adv. Drug Deliv. Rev., 28, 5, 1997.

18. Timmer, M.D. et al., Characterization of the cross-linked structure of fumarate-based degrad-able polymer networks, Macromolecules, 35, 4373, 2002.

19. Lee, K.Y., Bouhadir, K.H., and Mooney, D.J., Controlled degradation of hydrogels using multi-functional cross-linking molecules, Biomaterials, 25 (13), 2461, 2004.

20. Eliaz, R.E. and Kost, J., Characterization of a polymeric PLGA-injectable implant deliverysystem for the controlled release of proteins, J. Biomed. Mater. Res., 50, 388, 2001.

21. Damink, L.H.H.O. et al., In vitro degradation of dermal sheep collagen cross-linked using awater-soluble carbodiimide, Biomaterials, 17, 679, 1996.

22. Khare, A.R. et al., Measurement of the swelling force in ionic polymeric networks. I. Effect ofpH and ionic content, J. Control. Rel., 22, 239, 1992.

23. Hennink, W.E. and van Nostrum, C.F., Novel crosslinking methods to design hydrogels, Adv.Drug Del. Rev., 54 (1), 13, 2002.

24. Orienti, I. et al., Crosslinked polyvinylalcohol hydrogels as vehicles for hydrophilic drugs,Arch. Pharm. Pharm. Med. Chem., 333, 421, 2000.

25. Sahin, S. et al., Preparation, characterization and in vivo distribution of terbutaline sulfateloaded albumin microspheres, J. Control. Rel., 82 (2–3), 345, 2002.

26. Draye, J.P. et al., In vitro release characteristics of bioactive molecules from dextran dialdehydecross-linked gelatin hydrogel films, Biomaterials, 19, 99, 1998.

27. Luo, Y., Kirker, R.K., and Prestwich, G.D., Crosslinked hyaluronic acid hydrogels films: newbiomaterials for drug delivery, J. Control. Rel., 69, 169, 2000.

28. Safrany, A., Radiation processing: synthesis and modification of biomaterials for medical use,Nucl. Instr. Methods Phys. Res. Sect. B: Beam Interact. Mater. Atoms, 131 (1–4), 376, 1997.

29. Nedkov, E. and Tsvetkova, S., Effect of g-irradiation on the crystalline structure of ultra highmolecular weight poly(ethylene oxide), Radiat. Phys. Chem., 43, 397, 1994.

30. Rosiak, J.M., Hydrogel dressings, in Radiation Effects on Polymers, ACS Symposium Series 475,Clough, R.L. and Shalaby, S.W., Eds, ACS, Washington, D.C., 1991, 271.

31. Kabanov, V.Ya., Preparation of polymeric biomaterials with the aid of radiation-chemicalmethods, Russ. Chem. Rev., 67, 783, 1998.

32. Grant, G. et al., Biological interaction between polysaccharides and divalent cations: the egg-box model, FEBS Lett., 32 (1), 195, 1973.

33. Eagland, D., Crowther, N.J., and Butler, C.J., Complexation between polyoxyethylene andpolymethacrylic acid — The importance of the molar mass of polyethylene, Eur. Polym. J., 30,767, 1994.

34. Stauffer, S.R. and Peppas, N.A., Poly(vinyl alcohol) hydrogels prepared by freezing–thawingcyclic processing, Polymer, 33 (18), 3932, 1992.

35. Sludden, J., Uchegbu, I.F., and Schatzlein, A.G., The encapsulation of bleomycin withinchitosan based polymeric vesicles does not alter its biodistribution, J. Pharm. Pharmacol.,52, 377, 2000.

36. Noble, L. et al., A non-covalently cross-linked chitosan based hydrogel, Int. J. Pharm., 192 (2),173, 1999.

37. Cerchiara, T. et al., Physically cross-linked chitosan hydrogels as topical vehicles for hydro-philic drugs, J. Pharm. Pharmcol., 54, 1453, 2002.

38. Uchegbu, I.F. et al., Polymeric chitosan-based vesicles for drug deliver, J. Pharm. Pharmacol.,50, 453, 1998.

39. Denet, A.R., Vanbever, R., and Preat, V., Skin electroporation for transdermal and topicaldelivery, Adv. Drug Del. Rev., 56, 659, 2004.

40. Kurihara-Bergstrom, T. et al., Skin compatibility of transdermal drug delivery systems, J. Contr.Rel., 15, 271, 1991.

41. De Mey, C. et al., Transdermal delivery of mepindolol and propranolol in normal man,Arzneim-Forsch/Drug Res., 39, 1508, 1989.




42. Henry, S. et al., Microfabricated microneedles: a novel approach to transdermal drug, J. Pharm.Sci., 87, 922, 1998.

43. Pikal, M.J., The role of electroosmotic flow in transdermal iontophoresis, Adv. Drug Del. Rev.,46, 281, 2001.

44. Hu, Q. et al., Enhanced transdermal delivery of tetracaine by electroporation, Int. J. Pharm.,202, 121, 2000.

45. Sintov, A.C. et al., Radiofrequency-driven skin microchanneling as a new way for electricallyassisted transdermal delivery of hydrophilic drugs, J. Control. Rel., 89, 311, 2003.

46. Malminen, M. et al., Immunohistological distribution of the tight junction components ZO-1and occludin in regenerating human epidermis, Br. J. Dermatol., 149, 255–260, 2003.

47. Gayet, J.C. and Fortier, G., High water content BSA-PEG hydrogel for controlled releasedevice: evaluation of the drug release properties, J. Control. Rel., 38, 177, 1996.

48. Muzzarelli, R. et al., Biological activity of chitosan: Ultrastructural study, Biomaterials, 9, 247,1988.

49. Chandy, T. and Sharma, C.P., Chitosan beads and granules for oral sustained delivery ofnifedipine: in vitro studies, Biomaterials, 13, 949, 1992.

50. Chandy, T. and Sharma, C.P., Chitosan matrix for oral sustained delivery of ampicillin, Bio-materials, 14 (12), 939, 1993.

51. Bernkop-Schnurch, A., Paikl, Ch., and Valenta, C., Novel bioadhesive chitosan–EDTA conju-gate protects leucine enkephalin from degradation by aminopeptidase N, Pharm. Res., 14, 917,1997.

52. Valenta, C., Christen, B., and Bernkop-Schnurch, A., Chitosan–EDTA conjugate: a novelpolymer for topical gels, J. Pharm. Pharmacol., 50, 445, 1998.

53. Hwang, G.C. et al., Development and optimization of a methotrexate topical formulation,Drug Dev. Ind. Pharm., 21, 1941, 1995.

54. Alvarez-Figueroa, M.J. and Blanco-Mendez, J., Transdermal delivery of methotrexate: ionto-phoretic delivery from hydrogels and passive delivery from microemulsions, Int. J. Pharm.,215, 57, 2001.

55. Luppi, B. et al., Crosslinked poly(methyl vinyl ether-co-maleic anhydride) as topical vehiclesfor hydrophilic and lipophilic drugs, Drug Del., 10, 239, 2003.

56. Thacharodi, D. and Panduranga Rao, K., Development and in vitro evaluation of chitosan-based transdermal drug delivery systems for the controlled delivery of propranolol hydro-chloride, Biomaterials, 16, 145, 1995.

57. Guyot, M. and Fawaz, F., Design and in vitro evaluation of adhesive matrix for transdermaldelivery of propranolol, Int. J. Pharm., 204, 171, 2000.

58. Feldstein, M.M. et al., Hydrophilic polymeric matrices for enhanced transdermal drug delivery,Int. J. Pharm., 131, 229, 1996.

59. Iordanskii, A.L. et al., Modeling of the drug delivery from a hydrophilic transdermal thera-peutic system across polymer membrane, Eur. J. Pharm. Biopharm., 49, 287, 2000.

60. Kim, J. and Shin, S-C., Controlled release of atenolol from the ethylene–vinyl acetate matrix,Int. J. Pharm., 273, 23, 2004.

61. Padula, C. et al., Bioadhesive film for the transdermal delivery of lidocaine: in vitro and in vivobehavior, J. Control. Rel., 88(2), 277, 2003.

62. Langer, R., Transdermal drug delivery: past progress, current status and future prospects, Adv.Drug Del. Rev., 56, 557, 2004.

63. Kim, S.W., Bae, Y.H., and Okano, T., Hydrogels: swelling, drug loading and release, Pharm.Res., 9, 283, 1992.




Chapter 8

Enhanced Skin PermeationUsing Ethosomes

Elka Touitou and Biana Godin

CONTENTS

Introduction ...................................................................................................................................... 95Definition, Structure, and the Main Properties of Ethosomes ........................................................ 96Mechanism of Skin Permeation Enhancement by Means of Ethosomes ....................................... 97Various Aspects of Ethosomal Delivery System .............................................................................. 99

Efficiency to Enhance Drug Permeation into and through the Skin .......................................... 99Skin Permeation Performance of Ethosomes versus Liposomes and

Hydroethanolic Solution...................................................................................................... 99Enhanced In Vitro Delivery of Molecules with Diverse Chemical Properties ..................... 100Proof of Concept in Animals and in Clinical Trials ............................................................... 102

Safety Evaluation In Vitro, in Animals and in Human Studies ................................................. 104Stability and Manufacture........................................................................................................... 104

Summary ......................................................................................................................................... 105Acknowledgment ........................................................................................................................... 106References....................................................................................................................................... 106

Introduction

Liposomes were the first vesicular carrier studied for the delivery of drugs into the skin.Since the early works of Mezei using liposomes for topical drug delivery [1–2], numerousstudies have shown that classic liposomes are able to increase drug accumulation in theupper layer of the skin, the stratum corneum (SC) [1–7]. Drug delivery from liposomes ischaracterized by the formation of a drug reservoir in SC and by lack of penetration intothe deeper layers of the skin. Studies by Mezei and Gulusekharam [1, 2] demonstrated thatapplication of triamcinolone acetonide encapsulated within liposomes to depilated rabbitskin in vivo resulted in a fourfold increase in the amount of drug accumulated in


95


epidermis, as compared to application of the same drug concentration in an ointmentbase. Touitou et al. [3] found that the application of dyphylline incorporated in unila-mellar liposomes caused a high localization of the drug into the skin. In a further study,following the delivery of caffeine from small, unilamellar liposomes, high reservoir of thedrug was obtained in the skin [5]. These authors by using quantitative skin autoradio-graphy have examined the distribution of the drug in the various strata of the skin wherethe greatest concentration was found in the epidermis and the lowest in the dermis.Junginger’s group has shown by scanning and freeze-fracture electron micrographies(SEM and FFEM) that liposomes and nonionic surfactant vesicles, niosomes, adsorb onthe surface of SC [6, 7]. It was suggested that for these kinds of vesicles, it appears to bethermodynamically favorable upon application to the skin to aggregate, fuse, and adhereto the SC surface in stacks of lamellar sheets. Moreover, no ultrastructural changes in thedeeper layers of SC were reported with conventional liposomal systems tested.

A recent new approach for facilitating permeation of molecules through the skin is thedesign of vesicular carriers with appropriate characteristics to interfere with skin barrierfunction and to allow for enhanced delivery by passive transport to the beneath skinstrata and transdermally. Ethosomes, such vesicular carriers, by their structure andfunction are different from classic liposomes and niosomes [8, 9]. The main characteristicsof ethosomal vesicles are that they are soft, malleable, and penetrate the SC lipid bilayers.Our chapter reviews the physico-chemical and transmembrane delivery features of thisinnovative vesicular carrier.

Definition, Structure, and the Main Properties of Ethosomes

Ethosomes are specially tailored vesicular carriers for enhanced delivery of active agentsinto the deep layers of the skin and through the skin [8]. The delivery enhancementproperty is due to the composition and structure of ethosomes [8–10]. The ethosomalsystems contain unilamellar [11] or multilamellar [10, 12] soft lipid vesicles, with their sizesranging from 30 nm to mm. The main components of ethosomes are bilayer forming lipidssuch as phospholipids, important concentrations of volatile alcohols, such as ethanol andwater. Due to the interdigitation effect of ethanol on lipid bilayers, it was commonlybelieved that vesicles cannot coexist with high concentrations of ethanol [13–15].The vesicular structure of ethosomes was evidenced by a number of methodsincluding31 P-NMR, transmission electron micrography (TEM), and SEM.

Phosphorous NMR spectra of ethosomal systems showed a typical configuration ofthe phospholipid bilayer generally observed in phosphatidylcholine vesicles in water.Furthermore, the paramagnetic-ion NMR spectra indicated that the phospholipid inethosomes is in a more fluid state and the membrane is more permeable to cations, incomparison to liposomes [10]. These results suggested the existence of vesicles with a softmalleable structure, which could be due to the fluidizing effect of ethanol on thephospholipid bilayers. The existence of vesicles was further confirmed by electronmicroscopy visualization of the ethosomal systems. SE micrographs demonstrated athree-dimensional nature of ethosomal vesicles [10–12], while negatively stained TEmicrographs showed that multilamellar ethosomes are characterized by a bilayer struc-

Compounds of various hydrophilicities could be effectively entrapped in ethosomes.Ultracentrifugation studies demonstrated that the encapsulation efficiency of ethosomescould be as high as 90% in the case of a lipophilic drug testosterone [10]. Visualization of




ture throughout the vesicle (Figure 8.1).

entrapped fluorescent probes by means of confocal laser scanning micrography (CLSM)revealed that three molecules possessing different characteristics (lipophilic, amphy-philic, and hydrophilic) filled the entire volume of the ethosomes. In contrast, in classicliposomes, the lipophilic and amphiphilic probes were localized only in the bilayer, whilethe hydrophilic molecule was present mainly in the aqueous core of the vesicle [10, 12].These results could be explained by the structure of liposomes in which a small numberof bilayers surround an aqueous core. On the other hand, the unique structure ofethosomes is different from liposomes. The presence of ethanol together with vesiclelamellarity, that in the case of multilamellar ethosomes resembles a ‘‘fingerprint,’’ allowsfor efficient entrapment of lipophilic and amphiphilic molecules.

Mechanism of Skin Permeation Enhancement by Meansof Ethosomes

A proposed mechanism of action of the permeation enhancing ethosomal carrier ac-counts for a dual fluidizing effect of ethanol on the ethosomal lipid bilayers and on the SClipids. The soft ethosome penetrates the disturbed skin lipid bilayers creating a pathway

Figure 8.1 Visualization of multi-lamellar ethosomes containing 0.1% FITC-Bac by TEM. Bar100 nm. (From Godin, B. and Touitou, E., J. Control. Rel., 94, 365, 2004. With permission fromElsevier.)


Enhanced Skin Permeation Using Ethosomes & 97


through the skin and fuses with cell membranes in the deeper skin layers releasing theactive agent there [10]. Penetration through pilosebaceous pathways may also occur.

A scheme of the proposed model of penetration enhancement by the ethosomalsystem through the SC lipids is given in Figure 8.2. This sequence of synergistic processeswas suggested based on the results obtained in fluorescent anisotropy and differentialscanning calorimetry (DSC) experiments as well as in skin permeation studies.

Free energy measurements of the vesicle bilayers were assessed by DSC and fluores-cence anisotropy studies to gain insights into the characteristics of ethosomes that mightallow them to efficiently enhance drug delivery into the skin [10]. The transition tem-perature value of ethosomal phospholipids as shown by DSC thermograms was 20 to35 8C lower than in liposomes from the same components without ethanol, which canbe explained by the fluidity of phospholipids bilayers in ethosomes [10, 12, 16–18]. Thisbehavior was further confirmed by fluorescent anisotropy measurements of AVPC(9-Antrivinyl labeled analog of phosphatidylcholine) where a 20% lower value wasmeasured in comparison to liposomes [10].

Additional data which may shed further light on understanding the mechanism ofpermeation enhancement by ethosomes were obtained from skin permeation studies inwhich the vesicle components transported through the skin had been measured [10, 18].A significant amount (10.5% of initial) of phosphatidylcholine (PL) permeated the skinduring a 24-h experiment from a system composed of 2% PL, 30% ethanol, and water [10].These results suggest that the vesicles may have traversed the skin strata.

In further studies in which a fluorescently labeled polypeptide, bacitracin (FITC-Bac),was in vivo applied on rat abdomen with ethosomes and two controls, confocal laser

Figure 8.2 Proposed mechanism of enhanced permeation of molecules from ethosomal systemacross the lipid domain of SC. (From Touitou, E. et al., J. Control Rel., 65, 403, 2000. Withpermission from Elsevier.)




scanning(CLS) micrographs showed that the antibiotic delivered from ethosomes pene-trated the skin through the inter-corneocyte pathways, which typically exist along thelipid domain of the SC [12]. The high fluorescence intensity in the intercellular space dueto the penetration of FITC-Bac allowed for visualization of clear shadows of hexagonaland pentagonal corneocytes. Significantly lower fluorescence intensity through the inter-cellular penetration pathway was observed with FITC-Bac hydroethanolic control solu-tion, which points toward the influence of ethanol on the intercellular lipid domain of SC.On the other hand, with classic liposomes, CLS images appeared entirely different whereno inter- or intra-corneocyte fluorescence was observed [12].

The presence of significant ethanol concentrations (up to 50%) imparts to ethosomesincreased fluidity, a property that contributes to their high skin penetrating ability.A number of experiments revealed that from phospholipid vesicular carriers containing10% ethanol or less, skin permeation enhancement was negligible [8, 9, 19]. For example,an ethosomal system with 35% ethanol caused an eight times increased diclofenac skinpermeation than a preparation containing only 5% ethanol (309.2 vs. 37.7 mg/cm2) [8, 9].In an additional study, the lack of transdermal permeation of a-tocopherol incorporatedin phospholipid vesicular systems containing 10% ethanol, was explained by DSC resultsshowing that the vesicles were less fluid than ethosomes [19]. These results suggest thatlipid vesicles in the presence of low alcohol concentrations do not possess the skinpermeation enhancing characteristics of ethosomes.

Various Aspects of Ethosomal Delivery System

A new approach for administration of active agents to and through the skin could fail dueto the inefficient delivery of the molecule, local skin irritation associated with formula-tion, instability of the delivery system, and the complicated manufacturing process. Theseimportant points should be dealt with throughout the various stages of design of newtechnology to guarantee the move from a concept to clinical use. These issues wereaddressed in numerous studies with ethosomal delivery system which are further dis-cussed in this chapter.

Efficiency to Enhance Drug Permeation into and through the Skin

Skin Permeation Performance of Ethosomes versus Liposomesand Hydroethanolic Solution

Touitou et al. [10] evaluated the ability of ethosomes to deliver their contents deep intothe skin using a phospholipid fluorescent probe rhodamine red dihexadecanoyl glycer-

after 8-h application of RR from ethosomes, hydroalcoholic solution, and liposomes.These experiments showed that no deep penetration of RR from liposomal dispersionwas visualized. The highest intensity of fluorescence up to a depth of 150 mm was withethosomal systems. The application of the hydroalcoholic solution containing the sameconcentration of ethanol resulted in very low fluorescence intensity at the same skindepth as ethosomes. Because RR is used as an indicator of lipid fusion, which does notusually cross lipid bilayers, the results obtained in these experiments suggest that etho-somes traversed the skin strata to a high depth, in contrast to liposomes which remainedon the skin surface.




ophosphoethanolamine (RR). Figure 8.3 presents CLS micrographs of nude mice skin

In other experiments, the delivery of a lipophilic drug, minoxidil, into and across theskin was determined following a 24-h in vitro application of four different compositionsall containing 0.5% w/w of the drug [10, 20, 21]. The authors evaluated the quantities ofminoxidil that permeated the skin (Qr) and accumulated in the skin (Qs) at the end of theexperiment. The ethosomal systems resulted in 10, 45, and 35 times higher Qr and 2, 7,and 5 times higher Qs, compared to phosphatidylcholine ethanolic solution, hydroetha-nolic solution, or absolute ethanolic solution of the drug, respectively.

The results summarized above indicate that the ethosomal system is a much moreeffective permeation enhancer than absolute ethanol, aqueous ethanolic solutions, etha-nolic phospholipid solutions, or classic liposomes.

Enhanced In Vitro Delivery of Molecules with DiverseChemical Properties

It is commonly agreed that due to the inherent barrier properties of SC, only small,uncharged molecules with mild lipophilicity (log P 1–3) can penetrate it unassisted.

Several reports revealed that ethosomes are capable of enhancing permeability ofcompounds with a wide spectrum of physico-chemical features. An example of deliveryof lipophilic fluorescent probe into the deep skin layers was given in the previoussection. In further CLSM studies, delivery of two additional fluorescent probes possessingdistinct characteristics, a hydrophilic probe calceine and an amphiphilic cationic probeD-289 (4-[4-diethylamino] styryl-N-methylpyridinium iodide) from ethosomes and con-trol systems was visualized [11, 22]. The authors reported that for calceine a maximumfluorescence intensity (MaxFI) value of 150 arbitrary units (AU) was obtained whendelivered from ethosomes. This value was reached at the skin depth of 30 mm, remainedconstant throughout approximately 50 mm, and dropped to zero only at 160 mm. Incontrast, when calceine was applied from liposomes or from a hydroethanolic solution,lower MaxFI values were obtained followed by a sharp decrease of fluorescent intensityto zero at the depths of 60 and 80 mm, respectively [21, 22].

Similar results were acquired using the amphiphilic cationic probe D-289, in thepresence of cationic molecule trihexyphenidyl. Again, the ethosomes delivered the

Figure 8.3 Penetration of RR from ethosomes (a), hydroethanolic solution (b), or liposomes (c).The systems containing RR were applied nonocclusively to the back skin of 8-week male nudemice. At the end of the experiment, the skin was excised and analyzed by CSLM. The squaresections in each micrograph represent the optical slices of increasing skin depths. (From Touitou,E. et al. J. Control Rel., 65, 403, 2000. With permission from Elsevier.)




amphiphilic probe deeply into the skin (170 mm) and had much greater intensity thanthe two control systems [11].

Due to the lipophilic nature of the SC, highly lipophilic molecules (log P> 5) gener-ally show low transdermal absorption. These compounds accumulate within this layerand might encounter problems at the SC–viable epidermis interface where they mustpartition into a predominantly aqueous environment. Lodzki et al. examined a transder-mal transport of cannabidiol (CBD), a molecule with log P ~ 8, mediated by an ethosomalcarrier [23, 24]. The data from in vitro permeation experiments through nude mice skinindicated that following 24-h application of 100 mg ethosomal composition containing3% CBD, a significant amount of the drug permeated the skin (559 mg/cm2) and CBD skinreservoir (845 mg/cm2) was generated. This study demonstrated that ethosomes possessthe ability not only to enhance the partitioning into the lipophilic layers of the skin, butalso to enhance the clearance of the drug into the hydrophilic environment leading totransdermal delivery.

It is well documented in the literature that polypeptides, due to their size andhydrophilicity, generally do not penetrate through the intact skin. A recent work de-scribed the ethosomal delivery of fluorescently labeled bacitracin (FITC-Bac), a polypep-tide antibiotic (MW ~ 1.4 kDa), through human cadaver skin in vitro. Occlusive andnonocclusive application of FITC-Bac ethosomes resulted in flux values of 340+ 3 and290+ 2 mg/h cm2, respectively, showing that ethosomes delivered FITC-Bac through theskin and that an occlusion had almost no effect on skin permeation of this high MW drugfrom ethosomes [12].

Charged molecules were another group of challenging compounds tested with anethosomal carrier [11, 25], since there is strong evidence that SC is much more permeableto neutral molecules than the salts of weak acids or bases [26, 27]. The ability ofthe ethosomal system to deliver trihexyphenidyl hydrochloride (THP) to and throughthe skin was investigated using side-by-side diffusion cells. THP is an antimuscariniccationic drug used in the treatment of Parkinsonian syndrome. Figure 8.4 shows the

00

0.5

1.5

1

Qr

(mg)

2

5 10 15 20

Time (h)

Ethosomes Hydroethanolic soln Liposomes Phosphate buffer

Figure 8.4 Quantity of THP that permeated the skin (Qr) as a function of time from ethosomaland control systems. (From Dayan, N. and Touitou, E., Biomaterials, 21, 1879, 2000. Withpermission from Elsevier.)




kinetic profiles of THP skin permeation from ethosomes, classic liposomes, and phos-phate buffer, all containing 1% drug. Ethosomes allowed the highest THP skin transport[11, 25]. More detailed, the flux of THP from ethosomes (0.21 mg/cm2 h), was 87 timeshigher than from liposomes. The amount of drug accumulated within the skin at theend of an 18-h experiment was also significantly greater (p < 0.01) for ethosomes thanfor liposomes.

Proof of Concept in Animals and in Clinical Trials

A number of pharmacokinetic and pharmacodynamic studies on animals and in humanswith ethosomal systems were reported and are summarized below [10, 17, 20, 21, 23,24, 28, 29].

In an in vivo study on rabbits comparing two formulations, TestodermR patch (Alza)and an ethosomal patch, Testosome, the transdermal delivery of testosterone was esti-mated [10, 20]. Patches containing 12.25 mg testosterone were applied daily to the rabbitpinna skin for 5 consecutive days and blood samples were collected and analyzed byradio-immuno assay at the end of the experiment. AUC and Cmax values obtained withthe ethosomal system were 2.2 and 2.4 times higher, respectively. These results showedthat ethosomes allowed for enhanced in vivo transdermal absorption of the steroidhormone.

In another study the effect of the ethosomal insulin system on lowering blood glucoselevels (BGL) in vivo in SD1 rats was investigated. The results of this work show thatinsulin delivered from an ethosomal patch caused a significant reduction (up to 60%) inBGL in both normal and diabetic rats. On the other hand, insulin application from acontrol nonethosomal formulation was not able to decrease the blood levels of glucose[21, 28, 29]. The researchers outlined that with ethosomal insulin it was possible tomanage the hypoglycemic effect by adjusting the system composition to obtain variousglucose levels and duration of pharmacodynamic response. Moreover, the prolongedplateau effect that lasted for at least 8 h demonstrates the advantage of ethosomes fortransdermal delivery of insulin.

Delivery to pilosebaceous and hair follicular units could highly improve the treatmentefficiency for therapies directed at skin appendages related disorders such as seborrhea,hair loss, and acne. The current therapy with minoxidil, a lipid soluble drug administeredtopically on the scalp for alopecia treatment, is barely efficient [30]. To facilitate thetransport of the drug through the hair follicles, minoxidil was incorporated into etho-somes and the system was evaluated for localization of the drug into the pilosebaceousunits in vivo. Following the application of compositions containing 0.5% minoxidil and50 mCi tritiated drug to the dorsal region of hairless rats in vivo for up to 24 h, thelocalization of H3-minoxidil within the pilosebaceous units was observed [20, 21, 31]and measured by quantitative skin autoradiography [32, 33]. The study results demon-strated that the ethosomal system was superior in delivery of minoxidil to the pilosebac-eous elements of the skin than liposomes (22 vs. 4.5 nmol/g tissue, respectively, p< 0.005) [20, 21, 31]. It could be anticipated that targeting the hair follicles by usingethosomal minoxidil formulations will provide a feasible improved therapy for alopecia.

CBD is a potent agent in rheumatoid arthritis and other autoimmune diseases withan antiinflammatory effect several hundred times that of aspirin [34, 35]. In a study byLodzki et al. in vivo application of CBD ethosomes to the abdominal skin of CD1 nudemice resulted in significant accumulation of the drug in the skin and in the underlying




muscle [36]. In detail, following a 24-h application, a reservoir of CBD was detected in theabdominal skin (110.07+ 24.15 mg/cm2) and abdominal muscle (11.537 mg CBD/gmuscle) [36]. Upon application of the ethosomal system to the abdomen of ICR micefor 72 h, the kinetic profile of CBD’s plasma concentration shows that steady-state (SS)levels, stabilized at a value of 0.67 mg/ml, were reached at about 24 h and lasted at leastuntil the end of the experiment. Further, the antiinflammatory effect of CBD ethosomalsystems was evaluated using carrageenan-induced aseptic paw inflammation in ICR mice.Upon injection into mice paws, carrageenan provokes a local inflammatory reaction,which is a suitable method for evaluating antiinflammatory agents [37]. In these experi-ments for each mouse, the thickness of saline-injected paw was deduced from that of thecarrageenan-injected inflamed paw meaning that each mouse served as its own control[36]. Figure 8.5 summarizes results from these experiments. A significant difference in thepharmacodynamic profiles can be observed between CBD treated and untreated animalsat all times during the experiment, indicating that the inflammation was prevented bytrandermal delivery of ethosomal CBD.

Poor penetration of drugs into the skin (and partially, the permeation across the SC)often limits the efficacy of topical formulations. The potential of ethosomes to treat theviral herpetic infection was evaluated and confirmed in the study with topical acyclovir(ACV) in humans. Horowitz and co-wokers [38] reported the efficiency of ethosomal5%-acyclovir system (EA) compared to a 5%-acyclovir cream (Zovirax R, ZC) for treatmentof herpetic infection in a two-armed, double-blind, randomized clinical study. The meas-ured parameters included the proportion of lesions not progressive beyond the papularstage (abortive lesions), time to crust development, and time to loss of crust. Significantimprovement in all evaluated clinical parameters in both parallel and cross-over arms wasdemonstrated in this trial when the disorder was treated with EA. For instance, in theparallel arm on the third day from the beginning of herpetic episode 80% of lesionscrusted after treatment with EA comparing to only 10% in the ZC group. In the cross-overarm, the number of days to loss of crust (the healing time) was 4.2 for EA vs. 5.9 for ZC.

0.0 0.5 1.0 2.0 3.0 4.03.5 4.52.51.5−0.5−0.1

0

0.1

0.2

0.3

0.4

0.5

Time post injection (h)

Del

ta p

aw thi

ckne

s (m

m)

CBD pretreatment

No treatment** **

**

Figure 8.5 Antiinflammatory effect of CBD transdermal ethosomal patch, applied 19 h prior tothe injection, is compared to no pretreatment: D: (mean+SEM) between the thickness of carra-geenan-injected and saline-injected paws of the same mouse at different time points postinjec-tion. **p < 0.01; *p < 0.05. (From Lodzki et al., J. Control Rel., 93, 377, 2003. With permissionfrom Elsevier.)




In the previous studies a nonsufficient penetration of ACV, a hydrophilic molecule, intothe basal epidermis where the replication of the virus occurs, was proposed as a mainreason for a nonefficient topical therapy [39, 40]. The results of this clinical pilot trial showthat treatment with ethosomal acyclovir formulation significantly improved all evaluatedclinical parameters.

Another study in humans evaluated the antierythemal efficacy of ammonium glycyr-rhizinate (AG) delivered from ethosomes vs. ethanolic and aqueous solutions [41]. AG,obtained from liquorice root extract, is a natural antiinflammatory agent effective intreating acute and chronic dermatitis. The research was carried out on 12 healthyvolunteers pretreated with AG systems 1, 3, and 5 h prior to application of methylnicotinate (MN), a vasodilating agent, and tested the induced erythema at six sites onthe ventral surface of each forearm. The erythema index (DEI) was monitored for 8 h byusing a reflectance visible spectrophotometer. The authors reported that pretreatmentwith ethosomal AG significantly reduced both the intensity and duration of MN inducederythema as compared to pretreatment with AG solutions in water and ethanol. Forexample, the maximal DEI following 5 h pretreatment with the ethosomal system was29.6% while for the ethanolic and aqueous controls the measured maximal DEI valueswere 62.7 and 60.7%, respectively [41].

The reported pharmacokinetic and pharmacodynamic data suggest that the enhance-ment of skin permeation of various molecules achieved with ethosomes could lead toefficient delivery of active agents for which the skin barrier function should be overcome.

Safety Evaluation In Vitro, in Animals and in Human Studies

A good penetration enhancer will interfere with the barrier function of the skin withoutdamaging its cells or causing unwanted effects such as irritation.

An in vitro live /dead viability/ cytotoxicity test carried out in cultured fibroblasts forvarious vesicular systems and controls indicated that ethosomal carriers were not toxic to3T3 fibroblasts and that cultured cells kept their viability [22].

Animal studies were performed to assess safety of ethosomes prior to human studies.No acute skin irritation in rabbits was observed following a single-dose 48-h occlusiveapplication of patches containing the ethosomal systems. Furthermore, cumulative14-day repeated ethosomal patch application also did not generate any significanterythema [10].

Recently, Paolino et al. [41] performed human tolerability experiments with etho-somes and control systems on healthy volunteers utilizing a noninvasive technique ofreflectance spectrophotometry. The authors reported no signs of erythema following 12-,24-, or 48-h application of ethosomal carrier containing 2% PL and 45% ethanol. More-over, no significant difference in DEI was measured between skin areas treated with

Stability and Manufacture

In contrast to liposomes, ethosomes are prepared by methods which do not requirespecial equipment. The entire process could be run at temperatures between 20 and408C [8–10].

Stability of vesicular systems could be evaluated by measuring changes in vesiclesize distribution and visualization of the vesicles. The stability of ethosomal systems




ethosomes and saline (Figure 8.6).

incorporating various drugs was assessed in a number of studies by comparing theaverage diameter and the structure of the vesicles during the 2-year period in roomtemperature [10, 11]. No changes in the mean size of empty and cationic trihexyphenidyl

visualization by negative stain TEM confirmed that the vesicular structure of the etho-somes persisted after 2 years of storage and no significant structural changes occurredover that time in both systems. In another study with ethosomes containing a lipophilicantibiotic erythromycin (1%), 2% phosphatidylcholine, and 30% ethanol, no significantvariations in the dimensions of ethosomes throughout the storage in room temperaturewere measured. The initial mean size of the vesicles was 123+ 15 nm, while the diameterof ethosomes following a 1-year interval was 117+ 18 nm. TEM micrographs confirmedthat erythromycin unilamellar ethosomes kept their configuration during the stabilityevaluation experiments.

Data on SupraVir cream (Trima, Israel), a marketed ethosomal formulation of acyclo-vir, indicate that the formulation and the drug had long shelf-lives with no stabilityproblems. Acyclovir in SupraVir cream has been shown by HPLC assay to be stable forat least 3 years at 258C. Furthermore, skin permeation experiments showed that thecream after 3 years retains its initial penetration enhancing capacity [42].

Summary

Ethosomal carriers are systems containing phospholipid fluid vesicles in the presence ofhigh concentrations of ethanol. The mechanism of permeation enhancement suggeststhat ethanol has a fluidizing effect on both the intercellular SC lipids and on the ethosomephospholipid bilayers. The fluidized (soft) vesicle penetrates through the disorganized SC

6 24 480

20

40

60

80

100

Ethosomes45% EtOH soln.0.9% NaCl

Time (h)

∆EI

(%)

Figure 8.6 In vivo human skin tolerability of ethosomes vs. hydroethanolic solution and salinefollowing 6-, 24-, or 48-h application. Results are expressed as a mean value of variation in DEI(+SD), n 5 6 [41].




loaded vesicles were observed during the storage interval as seen in Figure 8.7. Moreover,

lipids into the deep skin strata where it releases the incorporated molecules by fusionwith cell membranes. Recent studies suggest that intracellular permeation may also occur.In vitro, in vivo and clinical studies summarized in this chapter show that this passivedelivery system possesses a penetration-enhancing action on the molecules with a widerange of physico-chemical characteristics and structures. In terms of safety, no localirritation was detected following skin application of ethosomes. Moreover, since thecarrier consists of materials approved for pharmaceutical and cosmetic use, no systemictoxicity is anticipated. The enhanced penetration of drugs deep into and across the skinby means of ethosomal carrier could be valuable in a variety of existing and newemerging therapies.

Acknowledgment

Prof. Elka Touitou is also affiliated with The David R. Bloom Center for Pharmacy, TheHebrew University of Jerusalem, Jerusalem, Israel.

References1. Mezei, M. and Gulusekharam, V., Liposomes, a selective drug delivery system for the topical

route of administration, Life Sci., 26, 1473, 1980.2. Mezei, M. and Gulusekharam, V., Liposomes, a selective drug delivery system for the topical

route of administration: gel dosage form, J. Pharm. Pharmacol., 34, 473, 1982.3. Touitou, E. et al., Diphylline liposomes for delivery to the skin, J. Pharm. Sci., 81, 131, 1992.4. Egbaria, K. and Weiner, N., Liposomes as a topical drug delivery system, Adv. Drug Deliv. Rev.,

5, 287, 1990.5. Touitou, E. et al., Modulation of caffeine skin delivery by carrier design: liposomes versus

permeation enhancers, Int. J. Pharm., 103, 131, 1994.

-

50

100

150

200

Mea

n size

(nm

)

1 14 39 75 105 730

Storage period (days)Empty ethosomes

1% THP ethosomes

Figure 8.7 Stability of ethosome vesicles determined by assessing the size of the vesicles overtime. The measurements were conducted on three batches of ethosomes that were kept at roomtemperature. Mean size was measured by DLS. Empty ethosomes composition: 2% soybeanphosphatidylcholine (PL), 30% ethanol and water. Trihexphenidyl (THP) ethosomes compos-ition: 1% THP, 2% PL, 30% ethanol and water.




6. Touitou, E. et al., Liposomes as carriers for topical and transdermal delivery, J. Pharm. Sci., 83,1189, 1994.

7. Hofland, H.E. et al., Interactions between liposomes and human stratum corneum in vitro:freeze fracture electron microscical visualization and small angle diffraction scattering studies,Br. J. Dermatol., 132, 853, 1995.

8. Touitou, E., Compositions for applying active substances to or through the skin, U.S. Patent5,540,934, 1996.

9. Touitou, E., Composition for applying active substances to or through the skin, U.S. Patent5,716,638, 1998.

10. Touitou, E. et al., Ethosomes — novel vesicular carriers for enhanced delivery: characterizationand skin penetration properties, J. Control. Rel., 65, 403, 2000.

11. Dayan, N. and Touitou, E., Carriers for skin delivery of trihexyphenidyl HCl: ethosomes vs.liposomes, Biomaterials, 21, 1879, 2000.

12. Godin, B. and Touitou, E., Mechanism of bacitracin permeation enhancement through the skinand cellular membranes from an ethosomal carrier, J. Control. Rel., 94, 365, 2004.

13. Chin, J.M. and Goldstein, D.B., Membrane disordering action of ethanol: variation with mem-brane cholesterol content and depth of the spin label probe, Mol. Pharmacol., 13, 435, 1997.

14. Harris, R.A. et al., Effect of ethanol on membrane order: fluorescence studies, Ann. N.Y. Acad.Sci., 492, 125, 1987.

15. Pang, K.Y.Y. et al., The perturbation of lipid bilayers by general anesthetics: a quantitative testof the disordered lipid hypothesis, Mol. Pharmacol., 18, 84, 1980.

16. Godin, B. and Touitou, E., Intracellular and dermal delivery of polypeptide antibiotic baci-tracin, Drug research between information and life sciences, ICCF, 3rd Symposium Abstracts,Bucharest, 2002.

17. Godin, B., Rubinstein, E., and Touitou, E., A new approach to interfere with microorganisms’resistance to antibiotics, 31st Annual Meeting and Exposition of the Controlled Release Society,Honolulu, Hawaii, 2004, 354.

18. Touitou, E. et al., Ethosomes: novel vesicular carriers for enhanced skin delivery, Pharm. Res.,14, S305, 1997.

19. Lavy, S., Ethosomes for Enhancement of Skin Penetration of Alpha-tocopherol, M.Sc. thesis,The Hebrew University of Jerusalem, Jerusalem, Israel, 2002.

20. Touitou, E., Godin, B., and Weiss, C., Enhanced delivery of drugs into and across the skin byethosomal carriers, Drug Dev. Res., 50, 406, 2000.

21. Godin, B. and Touitou E., Ethosomes: new prospects in transdermal delivery, Crit. Rev. Ther.Drug Carrier Syst., 20, 63, 2003.

22. Touitou, E. et al., Intracellular delivery mediated by an ethosomal carrier, Biomaterials, 22,3053, 2001.

23. Lodzki, M., Transdermal Delivery of Cannabidiol by Ethosomal Carrier, M.Sc. thesis, TheHebrew University of Jerusalem, Jerusalem, Israel, 2002.

24. Touitou, E. et al., Transdermal delivery of cannabinoids by ethosomal carriers, 4th WorldMeeting ADRITELF/APV/APGI Abstracts, Florence, 2002.

25. Dayan, N., Enhancement of Skin Permeation of Trihexyphenidyl HCl, Ph.D. thesis, TheHebrew University of Jerusalem, Jerusalem, Israel, 2000.

26. Swarbrick, J. et al., Drug permeation through human skin. II. Permeability of ionizablecompounds, J. Pharm. Sci., 73, 1352, 1984.

27. Green, P.G., Hadgraft, J., and Ridout, G., Enhanced in vitro permeation of cationic drugs,Pharm. Res., 6, 628, 1989.

28. Dkeidek, I. and Touitou, E., Transdermal absorption of polypeptides, AAPS Pharm. Sci., 1,S202, 1999.

29. Dkeidek, I., Transdermal Transport of Macromolecules, M.Sc. thesis, The Hebrew University ofJerusalem, Jerusalem, Israel, 1999.

30. Meidan, V. and Touitou, E., Treatments for androgenetic alopecia and alopecia areata: currentoptions and future prospects, Drugs, 61, 53, 2001.




31. Touitou, E., Meidan, V., and Horwitz, E., Methods for quantitative determination of druglocalized in the skin, J. Control. Rel., 56, 7, 1998.

32. Fabin, B. and Touitou, E., Localization of lipophilic molecules penetrating rat skin in vivo byquantitative autoradiography, Int. J. Pharm., 74, 59, 1991.

33. Godin, B., Alcabez, M., and Touitou, E., Minoxidil and Erythromycin targeted to pilosebaceousunits by ethosomal delivery systems, Acta Technologiae et Legis Medicament, 10, 107, 1999.

34. Evans, T., Formukong, E.A., and Evans, F.J., Actions of cannabis constituents on enzymes ofarachidonate metabolism: anti-inflammatory potential, Biochem. Pharmacol., 36, 2035, 1987.

35. Malfait, A.M. et al., The nonpsychoactive cannabis constituent — cannabidiol is an oral anti-arthritic therapeutic in murine collagen-induced arthritis, Proc. Natl Acad. Sci., 97, 9561, 2000.

36. Lodzki, M. et al., Cannabidiol — transdermal delivery and anti-inflammatory effect in a murinemodel, J. Control. Rel., 93, 377, 2003.

37. Sammons, M. et al., Parsons, a method for determining thermal hyperalgesia and inflammationin the mouse hind paw, Br. J. Pharmacol., 122, 334, 1997.

38. Horwitz, E. et al., A clinical evaluation of a novel liposomal carrier for acyclovir in the topicaltreatment of recurrent herpes labialis, Oral Surg. Oral Med. Oral Pathol. Oral Radiol. Endod.,88, 700, 1999.

39. Shaw, M. et al., Failure of acyclovir cream in the treatment of recurrent herpes labialis, Br. Med.J., 291, 7, 1985.

40. Worral, G., Topical acyclovir for recurrent herpes labialis in primary care, Can. Fam. Phys., 37,92, 1991.

41. Paolino, D. et al., An in vivo evaluation of ethosomes for dermal administration of a naturalanti inflammatory agent as ammonium glycyrrhizinate. European Conference of Drug Deliveryand Pharmaceutical Technology, Seville, Spain, 2004, O41.

42. Trima Israel Pharmaceutical Products Maabarot Ltd, data on SupraVir cream file.




Chapter 9

Microemulsions in TopicalDrug Delivery

Sari Pappinen and Arto Urtti

CONTENTS

Introduction .................................................................................................................................... 109Microemulsions in Topical Drug Delivery..................................................................................... 111

Introduction ................................................................................................................................ 111Solubilization of Drugs by Microemulsions............................................................................... 111Microemulsions and Drug Permeability .................................................................................... 112In Vivo Results ............................................................................................................................ 114Irritation ...................................................................................................................................... 114

Conclusions .................................................................................................................................... 114References....................................................................................................................................... 115

Introduction

Microemulsion was first introduced when conventional emulsion was transformed totransparent solution after the addition of a co-surfactant [1–3]. Ever since, microemulsionshave been used in several fields of industry. The first articles about the potential use ofmicroemulsions as topical formulations were published in the 1980s [4, 5]. During thepast few decades microemulsions have attracted a lot of interest in cutaneous drugdelivery since the importance of their drug delivery potential, compared to conventionaltopical formulations, was realized.

Microemulsions typically consist of an aqueous phase, an organic phase, and asurfactant or co-surfactant The combination of these three

Interfacial tension of these structures is low leading to the spontaneousformation of the microemulsion without the use of externally applied energy (unlike


109


components can be plotted as a percentage on a pseudo-ternary phase diagram (Figure

in the case of coarse emulsions). The low interfacial tension is achieved by adding a

9.2).

component (Figure

very

9.1).

Bicontinuous

o/w

w/o

OIL

Drug

Surfactant

Co-surfactant

WATER

Figure 9.1 The basic microemulsion structures formed by oil phase, water phase, and surfactantand co-surfactant interfacial film.

80 60 40 20

20

40

60

80

80

60

40

20

S + CW

IPM

3aφ

3aφ3φ

1φ

2φ

2φ2aφ2bφ2dφ2uφ

1φ1aφ1bφ1cφ1dφ

3φ3aφ3bφ3cφ3dφ3dφ3uφ3fφ3dφ

2dφ

3dφ3fφ

1aφ

1bφ1cφ2bφ2aφ

2aφ3bφ

WIPMS+C

: Double distilled water: Isopropyl myristate: Tween 80/Span 80/

1,2-octanediol(3:1:1.2 w/w)

: L2: L1

: LC ()

: Lβ: Lα

: L2/Lα

: O/E/W: O/Lα/W: O/L2/Lα

: O/Me/W: L1/Lα/W: O/L2/W: Lα/Lβ/W

: O/L1/Lα (^)

: E/L1

: L2/W

: Lβ/L1: O/L1

Figure 9.2 Pseudo-ternary diagram of the water:isopropylmyristate:surfactant and co-surfactantsystem at 258C. The diagram is characterized by the presence of regions of one, two, or threephases (1f, 2f, 3f). L2 and L1 phases delimit the regions of W/O and O/W microemulsions,respectively. La and Lb correspond to lamellar-liquid crystals and lamellar-gel phases. LC phase isan anisotrophic region of liquid crystals. In the case of the regions with several phases, Me standsfor a bicontinuous microemulsion and E for emulsion. (From Baroli et al., J. Control. Release, 69,209 [2000]. With permission.)




second surfactant or co-surfactant, like short-chain alcohol, to the system. Microemulsioncan appear over a wide range of oil–water–surfactant compositions, but in the majorityof cases they exist only over a narrow range of concentrations. There are three basictypes of microemulsions: water-in-oil (W/O), oil-in-water (O/W), and bicontinuous

An interfacial film that consists of both surfactant and co-surfactant,surrounded by continuous external phase, encloses the dispersed phase. Typical trans-parency and low viscosity of microemulsions arises from their small droplet diameter(10 to 100 mm). Microemulsions are thermodynamically stable systems due to theirspontaneous formation, but there are also dynamic systems in which the interface

surfactant film is a very important factor that enables the existence of the bicontinuousstructures.

Existence of microemulsion structure, droplet size, and effect of interparticle inter-actions in phase system have been studied using different methods [7–9]. An interfacialfilm that consists of both surfactant and co-surfactant molecules encloses the dispersedphase of a microemulsion (Figure 9.1). Microemulsions are fluctuating systems in whichthe droplets are aggregated due to the very flexible interfacial film and then againseparated to smaller droplets [6, 7, 10]. Obviously this kind of dynamic system favorsdrug movement in the vehicle.

Microemulsions have been shown to facilitate the cutaneous delivery of wide range ofboth lipophilic and hydrophilic drugs [10–16]. Microemulsions are an attractive formula-tion alternative for the dermal formulations.

Microemulsions in Topical Drug Delivery

Introduction

Dermal formulations should not prevent drug diffusion to the skin surface but oftenthey should solubilize the drug and improve its permeability in the skin. This is acomplex system where properties of drug, vehicle, and skin barrier play a role and alsointeract [17].

Solubilization of Drugs by Microemulsions

It has been observed in several studies that a large amount of drugs can be incorporatedto microemulsions due to their solubilization capacity. In microemulsions, the drug is inequilibrium between the dispersed and continuous phase (Figure 9.1). Microemulsioncomponents, like oil phase, surfactant or co-surfactant, have different solubilizing cap-acities. Solubility of the drug in the formulation depends on its solubility in the neatvehicle, and also on the microemulsion structure. Mostly drugs are associated with theinterfacial surfactant film between inner and outer phases [18]. The structure of micro-emulsion may contain even equal amounts of oil and water phases and a high surfactantcontent (20 to 80%) enabling incorporation of large fractions of both lipophilic andhydrophilic drugs to the formulation. Also various oils are incorporated into the micro-emulsion in different ways and have different effects on drug permeation [13, 19, 20].It has been suggested that small molecular volume oils were located in the interfacialsurfactant monolayer in the same way as a co-surfactant, while the larger oil moleculestend to locate in the center of the droplet. It has been shown that the rate of diffusion of


Microemulsions in Topical Drug Delivery & 111


(Figure

film is continuously and spontaneously fluctuating [6]. This kind of flexibility of the

9.1).

the drug from dispersed systems decreases with increasing surfactant concentration[16, 21]. Even simple changes in the relative ratios of the water, oil, and surfactant or

According to Fick’s first law, high concentration gradient (or more exactly, thegradient of thermodynamic activity) across the skin is the main driving force behinddrug permeation. Accordingly, the high solubilization potential of microemulsions allowsincreasing the concentration gradient across the skin and thereby improving the pene-tration rate of drugs compared to many other topical formulations. It should be, however,noted that the thermodynamic activity of the drug (i.e., ‘‘escaping tendency’’ from thevehicle) is the real driving force in drug release from the vehicle into the skin. Maximalthermodynamic activity is achieved at saturated solution and the thermodynamic activityat certain concentration can be described as the ratio between the said concentrationand saturated concentration. Since the solubility in the microemulsion is also increased,the thermodynamic activity cannot be directly concluded from the concentration of thesolubilized drug.

Microemulsions and Drug Permeability

In addition to the high solubilization potential, it is not obvious if high delivery potentialof microemulsions is related to a special microstructure, or not. However, transform-ation of microemulsion to other colloidal structures, like micelles and lamellar vesicles,has been seen to significantly decrease drug delivery through the skin [22]. Microemul-sions have also been demonstrated to improve transdermal delivery of several drugs overthe conventional topical preparations like emulsions, gels and liposomes, or aqueoussolutions, and neat oil phases [7, 12, 16]. The greater extent of drug delivery fromthe microemulsions may be partly related to high drug diffusivity in the vehicle.This promotes faster drug diffusion to the skin surface, a prerequisite of transdermalflux. High drug mobility in the microemulsion structure is probably due to the smalldroplet size and high density of droplets in microemulsion, which cause a large surfacearea of droplets. Therefore, droplets settle down in close contact with the skin provid-ing high local concentration gradient on the skin surface. This should improve drugpermeation.

Drugs will also partition between aqueous and hydrophobic phases depending on

coefficient of the drug between the oil and water phases of the microemulsion. Thereis always some drug in the continuous phase. This drug may partition into the skin andsubsequently drug release from the droplets to the continuous phase compensates for the

disappearing, and newly forming droplets, the droplets may also interact with the stratumcorneum surface after colliding with it. A mechanistic study of Peltola et al. [23] demon-strated that in addition to drug absorption from the continuous phase the droplets mayinteract with the stratum corneum as depicted in Figure 9.3. This interaction may result inincreased drug absorption due to the high concentration locally at the very surface of thestratum corneum.

Another possible mechanism is the fusion of the droplet in the skin resulting inpotential penetration enhancing effects (Figure 9.3). In fact, many components of micro-emulsions are known to be drug permeation enhancers, for example, some oils, surfact-ants, and many co-surfactants. Although in many experiments increased drug penetration




co-surfactant phases can affect the phase structures (Figure 9.2) and drug permeation.

their lipophilicity (Figure 9.1). Relative drug concentrations depend on the partition

absorbed drug (Figure 9.3). Since microemulsions are dynamic systems of colliding,

from the microemulsions were suggested to be due to enhancer effects of individualcomponents, the flux enhancement was primarily due to an increase in drug concentra-tion [11]. Permeation enhancement results are contradictory: addition of some penetra-tion enhancers to microemulsions increased the skin permeation [14, 24–26] but notalways [11, 15, 23, 27].

It appears that in general the microemulsion formulations improve the solubility in the

on the permeability in the skin seem to depend on the specific formulations.

Figure 9.3 Mechanisms of drug release from microemulsion droplets. Drug is always distrib-uted between inner and outer phase according to its solubility to these components. In the firstmechanism drug penetrates to skin only from outer phase. In the second mechanism dropletsbreak down upon contact with the skin and then release the contents of droplet into the skin. Themicroemulsion components may mix with skin lipids. In the third mechanism high density ofdroplets in microemulsion enables the direct drug diffusion to the skin from a microemulsiondroplet.




vehicle and augment drug access to the stratum corneum (Figure 9.4), while the effects

In Vivo Results

A few cutaneous drug delivery studies with microemulsions in vivo, have been per-formed in animal [18, 25, 28, 29] and human skin [27, 30–32]. Bioavailability, drug con-centration profiles in the systemic circulation, topical effects, and toxicity have beenreported. Improved drug delivery across the skin was also seen in vivo, suggesting thatthe concentration gradient influences the rate of drug delivery from microemulsions. Thein vitro results do not always match with in vivo studies: cyclodextrin containing micro-emulsion improved piroxicam permeation sixfold in vitro, but the therapeutic effect inthe skin was similar with and without cyclodextrin [28]. Although, microemulsionsfacilitate the percutaneous penetration in vivo, the therapeutic relevance depends alsoon the drug potency. One positive feature of microemulsions is the shortening of the lagtime of drug permeation compared to conventional topical preparations [18, 27].

Irritation

Skin toxicity is a common disadvantage of microemulsions in topical drug delivery. Thisis typically due to a co-surfactant in the vehicle, but surfactant and oil components canalso be irritating. Typically co-surfactants are short-chain alcohols, but alternate co-surfactants with better tolerability can be used. Nonionic surfactants are in general lesstoxic and have been used in microemulsions instead of ionic surfactants. Microemulsionswithout co-surfactants have also been described [33] as having permeation enhancementprofiles almost as good as co-surfactants containing microemulsions, but their structurecan be more easily destabilized by changes in composition and temperature.

Conclusions

Unfortunately the structure–property relationships of microemulsions in topical drugdelivery have not been investigated in adequate detail so far. Therefore, the correlations

2.6

10.8

13.6 13.2

0.20

4

8

12

16

OA Tween 20 EtOH Total Me D

Am

ount

of e

stra

diol

(m

g)

Figure 9.4 Solubility of estradiol in microemulsion D (Oleic acid 41.2%, Tween 20 8.3%,Ethanol 34.7%, and phosphate buffered saline 15.7%), and in each individual microemulsioncomponents. The histogram shows that high solubility potential is dependent on microemulsioncomponents rather than the physical structure of the microemulsion. (From Peltola, S. et al., Int.J. Pharm., 254, 99 [2003]. With permission.)




between the microemulsion structure and its performance in improving drug delivery arestill unclear. Furthermore, the toxicity of the microemulsion components still imposeslimitations on the use of microemulsions.

References1. Hoar, T.P. and Schulman, J.H., Transparent water in oil dispersions: oleopathic hydromicelle,

Nature, 152, 102, 1943.2. Shulman, J.H., Matalon, R., and Cohen, M., Discussions Faraday Soc., 11, 117, 1951.3. Shulman, J.H., Stoeckenius, W., and Prince, L.M., J. Phys. Chem., 63, 1677, 1959.4. Martini, M.C. et al., Role of microemulsions in the percutaneous absorption of alpha-tocoferol,

J. Pharm. Belg., 39, 348, 1984.5. Wang, J.C. et al., The release and percutaneous permeation of antralin products, using

clinically involved and uninvolved psoriatic skin, J. Am. Acad. Dermatol., 16, 812, 1987.6. Lam, A.C. and Schechter, R.S., The theory of diffusion in microemulsion, J. Colloid Interface

Sci., 120, 56, 1987.7. Kreilgaard, M., Pedersen, E.J., and Jaroszewski J.W., NMR characterization and transdermal

drug delivery potential of microemulsion systems, J. Control. Reease, 69, 421, 2000.8. Shukla, A. et al., Microemulsions for dermal drug delivery studied by dynamic light scattering:

effect of interparticle interaction in oil-in-water microemulsions, J. Pham. Sci., 92, 730, 2003.9. Mrestani, Y., Neubert R.H.H., and Krause, A., Partition behavior of drugs in microemulsions

measured by electrokinetic chromatography, Pharma. Res., 15, 799, 1998.10. Lee P.J., Langer R., and Shastri V.P., Novel microemulsion enhancer formulation for simultan-

eous transdermal delivery of hydrophilic and hydrophopic drugs, Pharm. Res., 20, 264, 2003.11. Baroli et al., Microemulsions for topical delivery of 8-methoxsalen, J. Control. Release, 69, 209,

2000.12. Kriwet, K. and Muller-Goymann C.C., Diclofenac release from phospholipid drug systems and

permeation through excised human stratum corneum, Int. J. Pharm., 125, 231, 1995.13. Trotta, M., Morel, S., and Gasco, M.R., Effect of oil phase composition on the skin permeation

of felodipine from o/w microemulsions, Pharmazie, 52, 50, 1997.14. Rhee, Y-S. et al., Transdermal delivery of ketoprofen using microemulsions, Int. J. Pharm.,

228, 161, 2001.15. Alvarez-Figueroa, M.J. and Blanco-Mendez, J., Transdermal delivery of methotrexate: ionto-

phoretic delivery from hydrogels and passive delivery from microemulsions, Int. J. Pharm.,215, 57, 2001.

16. Ktistis, G. and Niopas I., A study on the in vitro percutaneous absorption of propranolol fromdisperse systems, J. Pharm. Pharmacol., 50, 413, 1998.

17. Nishihata, T. et al., Percutaneous absorption of diclofenac in rats and humans: aqueous gelformulation, Int. J. Pharm., 46, 1, 1988.

18. Sintov, A. and Shapiro, L., New microemulsion vehicle facilitates percutaneous penetrationin vitro and cutaneous drug bioavailability in vivo. J. Control. Release, 95, 173, 2004.

19. Alany, R.G. et al., Effects of alcohols and diols on the phase behavior of quaternary systems,Int. J. Pharm., 196, 141, 2000.

20. Malcolmson C. et al., Effect of oil on the level of solubilization of testosterone propionate intonon-ionic oil-in-water microemulsions, J. Pharma. Sci., 87, 109, 1998.

21. Malcolmson, C. and Lawrence M.J., A comparison of the incorporation of model steroids intonon-ionic micellar and microemulsion systems, J. Pharm. Pharmacol., 45, 141, 1993.

22. Trotta, M., Influence of phase transformation on indomethacin release from microemulsions,J. Control. Release, 60, 399, 1999.

23. Peltola, S. et al., Microemulsions for topical delivery of estradiol, Int. J. Pharm., 254, 99, 2003.24. Mei, Z. et al., Solid lipid nanoparticle and microemulsion for topical delivery of triptolide,

Eur. J. Pharm. Biopharm., 56, 189, 2003.




25. Escribano E. et al., Assessment of diclofenac permeation with different formulations: anti-inflammatory study of a selected formula, Eur. J. Pharm. Sci., 19, 203, 2003.

26. Lee, P.J., Langer R., and Shastri V.P., Novel microemulsion enhancer formulation for simultan-eous transdermal delivery of hydrophilic and hydrophobic drugs, Pharma. Res., 20, 264, 2003.

27. Paolino, D. et al., Lecitin microemulsions for the topical administration of ketoprofen: percu-taneous adsorption through human skin and in vivo human skin tolerability, Int. J. Pharm.,244, 21, 2002.

28. Dalmora, M.E., Dalmora S.L., and Oliveira A.G., Inclusion complex of piroxicam with cyclo-dextrin and incorporation in cationic microemulsion. In vitro drug release and in vivo topicalanti-inflammatory effect, Int. J. Pharm., 222, 45, 2001.

29. Zabka, M. and Benkova M., Microemulsions containing local anaesthetics. Part 6: influence ofmicroemulsion vehicle on in vivo effect of pentacaine, Pharmazie, 50, 703, 1995.

30. Lehmann L., Keipert S., and Gloor M., Effects of microemulsion on the stratum corneum andhydrocortisone penetration. Eur. J. Pharm. Biopharm., 52, 129, 2001.

31. Kreilgaard, M. et al., Influence of microemulsion vehicle on cutaneous bioequivalence of alipophilic model drug assessed by microdialysis and pharmacodynamics, Pharm. Res., 18, 593,2001.

32. Bonina, F.P. et al., Effect of phospholipid based formulations on in vitro and in vivo percu-taneous absorption of methyl nicotinate, J. Control. Release, 34, 53, 1995.

33. Garti, N. et al., Water solubilization in nonionic microemulsions stabilized by grafted siliconicemulsifiers, J. Colloid Interface Sci., 233, 286, 2001.




Chapter 10

Nanoparticles as Carriersfor Enhanced SkinPenetration

Shozo Miyazaki

CONTENTS

Introduction .................................................................................................................................... 117Microparticles for Enhancing Skin Penetration ............................................................................. 118Nanoparticles for Enhancing Skin Penetration ............................................................................. 118PNBCA Nanocapsules as a Carrier for Enhanced Skin Penetration of Indomethacin................. 119Conclusions .................................................................................................................................... 122References....................................................................................................................................... 123

Introduction

Recently, much research has focused on the discovery of methods for improving thepercutaneous absorption of drugs. Many reports have described efforts to change skinpermeability using chemical enhancers because the stratum corneum is recognized as abarrier for transdermal drug delivery. In addition to the use of chemical enhancers, it ispossible to increase the percutaneous absorption by use of physical methods such asiontophoresis (electric fields) or by phonophoresis (ultrasound). We demonstrated thattherapeutic continuous1,2 or pulsed output3 ultrasound at 1 MHz could enhance thepercutaneous absorption of nonsteroidal antiinflammatory drugs (NSAIDs) from anointment in rats.

Colloidal drug delivery systems such as micro and nanoparticles have been extensivelystudied as one of the most promising strategies to achieve site-specific drug delivery.4–6

These systems have been utilized for oral and parenteral administration of drugs, andthey could be useful in delivering several drugs into the skin. However, there are a few


117


investigations supporting the idea that the colloidal drug delivery systems may improvethe transdermal delivery of drugs.5 The objective of this chapter is to consider the potentialuse of micro and nanoparticles for enhancing percutaneous absorption of drugs.

Microparticles for Enhancing Skin Penetration

Polymeric microparticles have become one of the most popular controlled releasedosage forms. Natural polymers such as gelatin and synthetic polymers such as polylacticacid have been investigated for use in injectable delivery systems.7

Previous papers have explored the use of microparticles for topical application. Inorder to deliver drug to the skin over a prolonged period of time, and in order to reduceits systemic absorption, a sustained drug release would be necessary. Several authorshave used microparticle drug delivery systems for topical delivery of drugs such asretinol8 and 5-fluorouracil9.

Rolland et al.10 reported that adapalene-loaded microspheres (5 mm mean diameter)were specially targetted to the follicular ducts and did not penetrate via the stratumcorneum. In order to improve the therapeutic index of adapalene, a drug for the treatmentof acne, site-specific delivery to the hair follicles using 50:50 poly(DL-lactic-co-glycolicacid) (PLGA) microspheres as particulate carriers was investigated in vitro and in vivo.The percutaneous penetration pathway of the microspheres was shown to be dependenton their mean diameter. The 1-mm microspheres randomly distributed into the stratumcorneum and hair follicles. However, the main penetration pathway of these drug-loadedmicrospheres was the transepidermal route since the outer surface of follicular orificerepresents only 0.1% of the total skin surface. The largest microparticles (20 mm) did notpenetrate the skin and remained on the stratum corneum surface.

Recently, de Jaron et al.11 showed that the PLGA microparticles (1 to 10 mm) caneffectively enter porcine skin through the stratum corneum and reach the epidermis,although the largest particles remained on the skin surface. Distribution of PLGA micro-particles in porcine skin, after its topical application, was studied in vitro using micro-particles containing rhodamine as a fluorescent probe. PLGA microparticles loaded withrhodamine were prepared using a solvent evaporation technique. Skin distribution offluorescent microparticles was performed, by horizontal and vertical slicing of frozenskin. Fluorescent photomicrographs revealed that PLGA microparticles could penetratethrough the stratum corneum and reach the epidermis. However, permeation experi-ments showed that these microparticles were not able to reach the receptor compartmentof the diffusion cells even over a period of 24 h. These carriers could be used as vehiclesfor topical drug delivery in order to obtain a sustained drug release into the skin,improving the time intervals between doses.

Nanoparticles for Enhancing Skin Penetration

Polymeric nanoparticles are solid or semisolid colloidal particles ranging in size from 10to 1000 nm.4–6 They consist of macromolecular materials and are extensively employedas drug carriers. Nanoparticles can be prepared by polymerization techniques or bydispersion of preformed polymers. They have several advantages over conventionaldrug carriers: small particle size, ease of administration, drug targeting to the specificbody site, solubilization of hydrophobic drug.




Recently, nanocapsules formed from lipophilic droplets, a core surrounded by a thinwall of polymeric material prepared by anionic polymerization of alkylcyanoacrylatemonomer, have been proposed as vesicular colloidal polymeric drug carriers.12–21 Poly-alkylcycanoacrylate-based nanoparticles have been extensively investigated for oral22–24

administration. The oral use of colloidal carrier systems for peptide delivery and espe-cially as adjuvant for oral vaccination seems to be a promising application for such carriersystems.25 Colloidal carrier systems of polyalkylcycanoacrylate nanoparticles are one ofmany drug delivery systems that have been proposed for improving the poor ocularbioavailability of ophthalmic medications.26

Furthermore, colloidal carriers could be useful for delivering several drugs into theskin and the stratum corneum. This might set off a drug supply to the skin over aprolonged time period. A sustained drug release might reduce systemic drug absorptionand a local treatment of inflammation might reduce systemic side effects. Because of theirultrafine particle size and their oily vesicular nature, alkylcyanoacrylate nanocapsules cansustain drug release, and, as a result, this colloidal carrier system has great potential forenhancing skin penetration. As nanoparticles can cross the eye corneal epithelium, itwould be interesting to investigate the ability of nanoparticles to pass through the skin.However, there are a few articles supporting the idea that nanoparticles may improve thetransdermal delivery of drugs.

For example, Kohli and Alpar27 have recently investigated the effect of size andcharge on the permeation of nanoparticles through the skin as the first step in designinga transdermal vaccine delivery system. The particles tested were 50, 200, and 500 nmlatex particles that were positively charged, negatively charged, and neutral. Fluorescentparticles ranging in size and charge were applied to the surface of full thickness pig skinin a diffusion chamber and the receptor fluid was assayed to determine permeation. Theresults showed that only 50 and 500 nm particles that were negatively charged were ableto permeate the skin. This provides evidences of the potential of nanoparticles as deliveryvectors for antigens and DNA for the purpose of transdermal vaccination protocols.

Poly(n-butylcycanoacrylate) (PNBCA) nanoparticles as a drug carrier for 5-fluorour-acil (5-FU) intended for topical treatment of skin lesions were investigated.28 Bioadhesivepoly(butylcycanoacrylate) nanoparticles used as a sustained drug delivery system offerthe possibility of improvement of the therapeutic index and frequency of topical 5-FUtreatment. The nanoparticle suspension loaded with 5-FU by adsorption of the drug wasconsidered for further biological tests.

PNBCA Nanocapsules as a Carrier for Enhanced Skin Penetrationof Indomethacin

In our previous publications,29,30 thermally reversible gels of Pluronic F-127 were evalu-ated as vehicles for the percutaneous administration of NSAIDs (including indometha-cin). Therefore, drug delivery to the skin could possibly be realized if a nanoparticlecarrier system of indomethacin is incorporated into a Pluronic F-127 gel formulation. Weprepared PNBCA nanocapsules of indomethacin incorporated into a Pluronic F-127based gel delivery system, and evaluated the ability to deliver the drug systemicallyafter topical application.31

PNBCA nanocapsules of indomethacin were prepared by interfacial polymerization.The physicochemical characterization of the PNBCA nanocapsules was performed bymeasuring the drug content by HPLC and analyzing the particle size using scanning


Nanoparticles as Carriers for Enhanced Skin Penetration & 119


electron microscopy (Figure 10.1). The drug loading results indicated that approximately76.6% of indomethacin was loaded into the PNBCA nanocapsules, the average particlesize was 188 nm.

The in vitro permeation of indomethacin through excised rat skin was determined forPNBCA nanocapsules in pH 7.4 phosphate buffer (I), and in Pluronic F-127 gel (II) andwere compared against indomethacin incorporated into 25% w/w Pluronic F-127 gel

three formulations (I, II, and III) in both the flux at steady state and the cumulativeamounts permeated at 8 h. The nanocapsules of indomethacin, when dispersed in 25%w/w Pluronic F-127 gel (formulation II), showed smaller flux and cumulative amounts,due to the viscous environment provided by the Pluronic F-127 gel. These resultssuggested that PNBCA nanocapsules are able to permeate through rat skin in a periodof 8 h. The skin distribution of the rhodamine-loaded nanoparticles was determined byconfocal laser scanning microscopy (CLSM). The nanoparticles were clearly visualized inthe stratum corneum, epidermis, and dermis. When a rhodamine solution was applied tothe skin, no fluorescent nanoparticles were observed but only a red background corre-sponding to the free probe. These results revealed that the nanoparticles can penetratethrough the stratum corneum and reach the epidermis.

In vivo percutaneous absorption of indomethacin following the application of theintact PNBCA nanocapsules and the 25% w/w Pluronic F-127 gel containing drug to theabdominal rat skin was monitored by the determination of plasma drug levels. As shown in

of the intact PNBCA nanocapsules are in agreement with the in vitro permeation results

Figure 10.1 Scanning electron micrograph (SEM) of indomethacin-loaded PNBCA nanocap-sules.




alone (III). The in vitro results (Figure 10.2 and Table 10.1) indicated a rank order for the

Figure 10.3, the higher indomethacin plasma levels over 6 h following the application

(Figure 10.2), which show a higher permeation rate for this drug through excised rat skincompared to that of a 25% w/w Pluronic F-127 gel (formulation III). The higher plasmaconcentrations using the intact nanoparticles result in an increase in the AUC values byfactor 3.3 compared to the AUC values of the Pluronic F-127 gel formulation. These resultsrevealed that the PNBCA nanocapsules could penetrate through the stratum corneum andepidermis and reach the blood circulation because of their ultrafine particle size and oilyvesicular nature. The topical application of the PNBCA nanocapsules on the skin surfacewas observed over the duration of the experiment and compared with the Pluronic F-127gel. When the Pluronic F-127 gel was applied to the skin, it formed a thin, smooth film onthe skin surface. This is due to the evaporation of water from the gel. The lower drugplasma concentrations observed with the Pluronic F-127 gel may be due to this evapor-ation effect, but might be also attributed to a high affinity of the drug to the hydrophobicdomains of the Pluronic F-127. However, an interesting observation was that the intactnanocapsule formulation gradually disappeared from the skin surface over time.

These results suggested that the PNBCA nanocapsules could penetrate through thestratum corneum and the epidermis and reach the blood circulation. This could be due to

00

5

10

20

15

2 4 6 8Time (hr)

PNBCA Nanocapsules inpH 7.4 Phosphate Buffer (I)

PNBCA Nanocapsulesin PLF-127 Gel (II)

25% w/w PLF-127 Gel (III)C

umul

ativ

e am

ount

of I

ndom

etha

cin

perm

eate

d (µ

g/cm

2 )

Figure 10.2 Cumulative amount of drug permeating through excised rat skin when releasedfrom: PNBCA nanocapsules dispersion in pH 7.4 buffer, PNBCA nanocapsules dispersion inPluronic F127 gel, and 25% w/w Pluronic F-127 gel. Each experiment was repeated four timesand the error bars represent the standard error.

Table 10.1 Flux and Cumulated Amount of Indomethacin through Rat Skin fromDifferent Formulations In Vitro Permeation Studies

Dosage formFlux (6 to 8 h)

(mg/cm3h)Cumulative amount

at 8 h (mg/cm3)

PNBCA nanocapsules in pH 7.4 phosphate buffer (I) 3.29+0.82** 13.43+3.30**PNBCA nanocapsules in PLF-127 Sol(II) 0.74+0.19 4.46+0.64*25% w/w PLF-127 Sol(III) 0.24+0.08 1.44+0.40

Each value is the mean+ SE of four determinations. p * 0.01, **0.05 compared with Phuronic F-127.




their ultra fine particle size and their hydrophilic and hydrophobic surface characteristics.The increased skin permeation of indomethacin may be explained by a modification ofthe lipid organization in the skin due to the presence of nanocapsules. Indeed some of thenanocapsule components (benzyl benzoate) are known to act as absorption enhancers.However, the penetration mechanism of the nanocapsules through the skin is not knownand requires more research.

Conclusions

From the small amount of the published research reviewed in this chapter, it may bestated that polymeric nanoparticles hold promise as a carrier for enhanced skin penetra-tion. If nanoparticulate carriers can cross the stratum corneum then they can act asmicroreservoirs of a drug in the skin and provide a sustained drug delivery. However,enough specific information on the uptake of nanoparticles by stratum corneum has notyet been presented. The findings of our work described here indicate that PNBCAnanocapsules can be used as a carrier for topical drug delivery, in order to improve theskin permeation of drugs such as indomethacin, and presumably other more hydropho-bic drugs. It is suggested that a rapid onset of a pharmacological effect is sufficientlyinduced by free indomethacin in the skin and plasma followed by the absorption of theintact PNBCA nanocapsules. However, it is difficult to provide any conclusions as to themechanism by which this occurs. Despite this, since PNBCA nanocapsules showedenhanced permeation in the study, we believe that this provides evidence for thepotential of nanoparticulate delivery of drugs through the skin.

Consequently nanoparticle carriers seem to be promising systems for topical drugadministration. However, more work will have to be done to assess the efficacy of thisnew approach as carriers for enhanced skin penetration.

00.0

0.5

1.5

1.0

2.0

2.5

2 4 6Time (hr)

Pla

sma

conc

n. (

µg/m

L)

PNBCA Nanocapsules

25% w/w PLF - 127 Gel

Figure 10.3 Percutaneous absorption of indomethacin in rats from PNBCA nanocapsules and25% w/w Pluronic F-127 gel. Each experiment was repeated six times and the error barsrepresent the standard error.




References1. Miyazaki, S., Mizuoka, H., Oda, M., and Takada, M., External control of drug release and

penetration: enhancement of the transdermal absorption of indomethacin by ultrasoundirradiation, J. Pharm. Pharmacol., 43, 115, 1991.

2. Miyazaki, S., Mizuoka, H., Kohata, Y., and Takada, M., External control of drug release andpenetration. VI. Enhancement effect of ultrasound on the transdermal absorption of indo-methacin from an ointment in rats, Chem. Pharm. Bull., 40, 2826, 1992.

3. Asano, J., Suisha, F., Takada, M., Kawasaki, N., and Miyazaki, S., Effect of pulse outputultrasound on the transdermal absorption of indomethacin from an ointment in rat, Biol.Pharm. Bull., 20, 288, 1997.

4. Kreuter, J., Nanoparticles. In Colloidal Drug Delivery Systems, Kreuter, J., Ed., Marcel Dekker,New York, 1994, 219.

5. Alonso M.J., Nanoparticulate drug carrier technology. In Microparticulate Systems for theDelivery of Proteins and Vaccines, Cohen, S. and Bernstein, H., Eds., Marcel Dekker, NewYork, 1996, Chapter 7.

6. Allemann, E., Gurny, R., and Leroux, J.C., Biodegradable nanoparticles of poly(lactic acid) andpoly(lactic-co-glycolic acid) for parenteral administraton, in Pharmaceutical Dosage Forms:Disperse Systems, Vol. 3, 2nd ed., Lieberman, H.A., Rieger, M.M., Banker, G.S., Eds., MarcelDekker, New York, 1986, Chapter 5.

7. Kisse, T. and Koneberg, R., Injectable biodegradable mirospheres for vaccine delivery, inMicroparticulate systems for the delivery of proteins and vaccines, Cohen, S. and Bernstein,H., Eds., Marvel Dekker, New York, 1996, Chapter 2.

8. Rosseler, B., Kreuter, J., and Ross, G., Effect of collagen microparticles on the stability of retinoland its absorption into hairless mouse skin in vitro, Pharmazie, 49, 157, 1994.

9. Ghorab, F., Ertl, B., Wirth, M., and Mallinger, R., Ketoprofen-poly(D,L-lactic-co-glycolic acid)microspheres: influence of manufacturing parameters and type of polymer on the releasecharacteristics, J. Microencapsul., 16, 1, 1990.

10. Rolland, A., Wagner, N., Chatelus, A., Shroot, B., and Schaefer, H., Site-specific drug delivery topilosebaceous structures using polymeric microspheres, Pharm. Res., 10, 1738, 1993.

11. de Jaron, E.G., Blanco-Prieto, M.J., Ygartua, P., and Santoyo, S., PLGA microparticles: possiblevehicles for topical drug delivery, Int. J. Pharm., 226, 181, 2001.

12. Couvreur, P., Kante, B., Roland, M., Guiot, P., Baudin, P., and Speiser, P., Polycyanoacrylatenanocapsules as potential lysosomotropic carriers : preparation, morphological and sorptiveproperties, J. Pharm. Pharmacol., 31, 331, 1979.

13. Couvreur, P., Kante, B., Roland, M., and Speiser, P., Adsorption of antineoplastic drugs topolyalkylcyanoacrylate nanoparticles and their release in calf serum, J. Pharm. Sci., 68, 1521,1979

14. Couvreur, P., Kante, B., Lenaerts, V., Scailteur, V., Roland, M., and Speiser, P., Tissue distribu-tion of anticancer drugs associated to polyalkylcyanoacrylate nanoparticles, J. Pharm. Sci., 69,199, 1980.

15. Kante, B., Couvreur, P., Lenaerts, V., Guiot, P., Roland, M., Baudhuin P., and Speiser, P., Tissuedistribution of 3H-actinomycin D absorbed on polybutyl-cyanoacrylate nanoparticles,Int. J. Pharm.,7, 45, 1980.

16. Couvreur, P., Kante, B., Grislain, L., Roland, M., and Speiser, P., Toxicity of polyalkylcyanoa-crylate nanoparticles. II. Doxorubicin loaded nanoparticles, J. Pharm. Sci., 71, 790, 1982.

17. Kreuter, J., Mills, S.N., Davis, S.S., and Wilson, C.G., Polybutylcyanoacrylate nanoparticles forthe delivery of [75Se] norcholesterol, Int. J. Pharm., 16, 105, 1983.

18. El-Samaligy, M.S., Rohdevald, P., and Mahmoud, H.A., Polyalkyl cyanoacrylate nanocap-sules, J. Pharm. Pharmacol., 31, 216, 1986.

19. Damge, C., Michel, C., Aprahamian, M., and Couvreur, P., New approach for oral administra-tion of insulin with polyalkylcyanoacrylate nanocapsules as drug carrier, Diabetes, 37, 246,1988.




20. Allemann, E., Gurny, R., and Doelker, E., Drug-loaded nanoparticles preparation methods anddrug targeting issues, Eur. J. Pharm. Biopharm., 39, 173, 1993.

21. Das, S.K., Tucker, I.G., Hill, D.J., and Ganguly, N., Evaluation of poly(isobutylcyanoacrylate)nanoparticles for mucoadhesive ocular drug delivery. I. Effect of formulation variables onphysicochemical characteristics of nanoparticles, Pharm. Res., 12, 534, 1995.

22. Andrieu, V., Fessi, H., Dubrasquet, M., Devissaguet, J-Ph., Puisieux, F., and Benita, S., Phar-macokinetic evaluation of indomethacin nanocapsules, Drug Des. Del., 4, 295, 1989.

23. Araujo, L., Sheppard, M., Lobenberg, R., and Kreuter, J., Uptake of PMMA nanoparticles fromthe gastrointestinal tract after oral administration to rats: modification of the body distributionafter suspension in surfactant solutions and in oil vehicles, Int. J. Pharm., 176, 209, 1999.

24. Damge, C., Aprahamian, M., Balboni, G., Hoeltzel, A., Andrieu, V., and Devissaguet, J-Ph.,Polyalkylcyanoacrylate nanocapsules increase the intestinal absorption of a lipophilic drug,Int. J. Pharm., 36, 121, 1987.

25. Jung, T., Kamm, W., Breitenbach, A., Kaiserling, E., Xiao, J. X., and Kissel, T., Biodegradablenanoparticles for oral delivery of peptides: is there a role for polymers to affect mucosaluptake? Eur. J. Pharm. Biopharm., 50, 147, 2000.

26. Harima, T., Kreuter, J., Speiser, P., Boye, T., Gurny, R., and Kubis, A., Enhancement of themyotic response of rabbits with pilocarpine-loaded polybutyl-cyanoacrylate nanoparticles, Int.J. Pharm., 33, 187, 1986.

27. Kohli, A.K. and Alpar, H.O., Potential use of nanoparticles for transcutaneous vaccine delivery:effect of particle size and charge, Int. J. Pharm., 275, 13, 2004.

28. Simeonova, M., Velichkova, R., Ivanova, G., Enchev, V., and Abrahams, I., Poly(n-butylcycanoacrylate) nanoparticles for topical delivery of 5-fluorouracil, Int. J. Pharm.,263, 133, 2003.

29. Miyazaki, S., Tobiyama, T., Takada, M., and Attwood, D., Percutaneous absorption of indo-methacin from pluronic F127 gels in rats, J. Pharm. Pharmacol., 47, 455, 1995.

30. Takahashi, A., Suzuki, S., Kawasaki, N., Kubo, K., Miyazaki, S., Loebenberg, R., Bachynsky, J.,and Attwood, D., Percutaneous absorption of non-steroidal anti-inflammatory drugs fromin situ gelling xyloglucan formulations in rats, Int. J. Pharm., 246, 179, 2000.

31. Miyazaki, S., Takahashi, A., Kubo, K., Loebenberg, R., and Bachynsky, J., Polyn-butylcyanoacrylate (PNBCA) nanocapsules as a carrier for NSAIDs: in vitro release andin vivo skin penetration, J. Pharm. Pharmaceut. Sci., 6, 238, 2003.




Chapter 11

Solid Lipid Nanoparticlesfor Topical Delivery

Zhinan Mei

CONTENTS

Introduction .................................................................................................................................... 125Preparation of Solid Lipid Nanoparticles....................................................................................... 126

High Shear Homogenization and Ultrasonic Dispersion .......................................................... 126High Pressure Homogenization ................................................................................................. 126

Hot Homogenization .............................................................................................................. 127Cold Homogenization............................................................................................................. 127

Emulsification or Vaporization Method ..................................................................................... 127Microemulsion ............................................................................................................................ 127

Characterization of SLN.................................................................................................................. 127SLN Incorporation and Release of Drug........................................................................................ 128SLN for Topical Delivery ................................................................................................................ 128SLN as a Novel Carrier System for Sunscreens.............................................................................. 131

Occlusive Effect of SLN .............................................................................................................. 132In Vitro Occlusion of SLN .......................................................................................................... 132SLN In Vivo Occlusion................................................................................................................ 133


Introduction

Solid lipid nanoparticles (SLN) were developed at the beginning of the 1990s andattracted increasing attention during recent years as an alternative carrier systemto emulsions, liposomes, and polymeric nanoparticles for controlled drug delivery[1, 2]. Drugs are typically incorporated into biodegradable lipids, which are solid atroom temperature. There are three general components in SLN:lipids (such as solidtriglycerides), emulsifiers, or coemulsifiers (such as soybean lecithin, egg lecithin,


125


phosphatidylcholine, tyloxapol, sodium cholate, sodium glycocholate, taurocholic acidsodium salt, taurodeoxycholic acid sodium salt, butanol, butyric acid, dioctyl or sodiumsulfosuccinate), and the incorporated drug(s).

SLN play an important role as drug delivery systems for intravenous, peroral, paren-teral, pulmonary, or ocular administration, and for topical delivery. When SLN technol-ogy is used in lipid emulsion, the greatest best advantage is the decrease in the diffusioncoefficient of the drug, thereby slowing the release of the drug. For example, thepreocular retention of SLN in rabbit eyes was tested using drug-free, fluorescent SLN(F-SLN). These structures were retained for longer times on the corneal surface and in theconjunctival sac when compared with an aqueous fluorescent solution [3]. There are alsosome notable potential problems in SLN as drug delivery systems; for example, forintravenous administration the particle may jam the capillary vessel. SLN my also havepotential new application in vaccine administration.

SLN are well tolerated in living systems because they are made from physiologicalcompounds and, therefore, are easily metabolized. However, the toxicity of the emulsi-fiers has to be considered, but their potential toxicity is also relevant for other carriersystems. No problems should be observed for peroral or transdermal administration andi.m. or s.c. injection if appropriate surfactants are used. In contrast to polymeric nano-particles, SLN preserved at low concentrations do cause direct or indirect cytotoxic effectsin peritoneal macrophages [4].

Preparation of Solid Lipid Nanoparticles

SLN are particles made from solid lipids with a mean photon correlation spectroscopy(PCS) approximately 50 and 1000 nm. Several methods of preparing SLN have beenreported. One can derive them from parenteral nutrition emulsions by replacing the oil ofthe emulsion droplets with a solid lipid. In contrast to emulsions for parenteral nutritionwhich are normally stabilized by lecithin, SLN can be stabilized by other surfactants orpolymer mixtures.

High Shear Homogenization and Ultrasonic Dispersion

High shear homogenization and ultrasonic dispersion are the conventional methods usedfor dispersing emulsions, liposome, and solid lipid nanodispersions. The advantages ofthe two methods are ease of control, the disadvantages are the ease of forming particlesin the micron range, and the ease of metal contamination if the ultrasonic dispersion timeis too long (15 min).

High Pressure Homogenization

High pressure homogenization (HPH) has been demonstrated as a more effectivemethod for the production of submicron sized dispersions of solid lipids compared tohigh shear homogenization or ultrasonic dispersion [5–7]. This is a technique that hasbeen well established on the large scale for many years and is already available in thepharmaceutical industry. The working principle of HPH is pumping liquid at highpressure (10 to 200 MPa) through a narrow gap (in the range of a few microns), whichgreatly accelerates the liquid droplets (over 1000 km/m) over a short distance. The highshear stress and cavitational forces disrupt the droplets down to the submicron range [8].




The two basic production methods for SLN are the hot homogenization technique andthe cold homogenization technique.

Hot Homogenization

In hot homogenization the drug is dissolved or solubilized in the lipid at temperaturesabove the melting point of the lipid [2]. Higher temperatures result in smaller particlesizes but may also increase the degradation rate of the drug and the carrier [9]. Thehomogenization step can be repeated several times. We should be aware that HPHincreases the temperature of the sample (approximately 108C for 500 bar [10]), in mostcases, 3 to 5 homogenization cycles at 500 to 1500 bar are sufficient.

Cold Homogenization

In contrast, cold homogenization is carried out with the solid lipid and represents,therefore, a high pressure milling of a suspension. Effective temperature control andregulation is needed in order to ensure the solid state of the lipid due to the increase intemperature caused by homogenization [10].

Emulsification or Vaporization Method

The preparation of SLN can also use emulsification or vaporization methodology, whichis always used in polymer nanoparticle (PNP) manufacture. The lipophilic material isdissolved in a water-immiscible organic solvent (e.g., cyclohexane) that is subsequentlyemulsified in an aqueous phase. Upon evaporation of the solvent, a nanoparticle disper-sion is formed by precipitation of the lipid in the aqueous medium. However, a cleardisadvantage of the method is the use of organic solvents.

Microemulsion

Microemulsions are clear or slightly translucent solutions composed of a lipophilic phase(e.g., lipid), a surfactant (and in most cases also a co-surfactant), and water. This effect isexploited in the preparation method for SLN developed by Gasco [2, 11].

To form a microemulsion with a lipid being solid at room temperature, the micro-emulsion needs to be produced at a temperature above the melting point of the lipid. Thelipid (fatty acids, glycerides, or both) are melted, a mixture of water, co-surfactant(s) andthe surfactant is heated to the same temperature as the lipid and added under mild stirringto the lipid melt. A transparent, thermodynamically stable system is formed when thecompounds are mixed in the correct ratio for microemulsion formation. This microemul-sion is then dispersed in a cold aqueous medium (2 to 388C) under mild mechanicalmixing, thus ensuring that the small size of the particles is due to the precipitation and notmechanically induced by the stirring process [12, 13].

Characterization of SLN

An adequate characterization of the SLN dispersion is a necessity for the quality control ofthe product. However, the characterization of SLN is a challenge due to the small size


Solid Lipid Nanoparticles for Topical Delivery & 127


of the particles and the complexity of the system, which also includes dynamic phenom-ena. Several parameters have to be considered: particle size, zeta potential; degree ofcrystallinity, lipid modification, coexistence of additional colloidal structures (micelles,liposomes, supercooled melts, drug-nanoparticles) and the time scale of distributionprocesses. The size of the particle has a direct impact on the stability and release kineticsof the lipid. PCS and laser diffraction (LD) are the most powerful techniques for routinemeasurements of particle size [2]. Surfactants markedly affect the polymorphic statetransition, the diameter, and the stability of the nanopaticle. Investigation indicates thatthe surfactant, but not the size of SLN, influences viability and cytokine productionmacrophages [14]. Gelation phenomena describe the transformation of a low-viscositySLN dispersion into a viscous gel. This process may occur very rapidly and unpredictably.In most cases, gel formation is an irreversible processes that involves the loss ofthe colloidal particle size and can be stimulated by intense contact and shear forces [2].Lipid crystallization may not occur although the sample is stored at a temperaturebelow the melting point of the lipid. It is difficult to describe the physical state of thelipid as crystallized or noncrystallized, for the crystallized lipid may be present in severalmodifications of crystal lattice form. The main reason for the formation of super-cooled melts is the size dependence of the crystallization processes, crystallizationrequiring a critical number of crystallization nuclei to initiate [15]. The presence ofother colloidal species (micelles, liposomes, supercooled melts, drug-nanoparticles) isan important point to consider, but this aspect has been ignored in the majority of theSLN literature.

SLN Incorporation and Release of Drug

As discussed previously, drug loading might result in dramatic changes in the SLNcharacteristics (particle size distribution, zeta potential, lipid modification, etc.). Drugincorporation implies the localization of the drug in the solid lipid matrix. The crystallinestructure, physico-chemical nature of the lipid, and the polymorphic form are key factorsin determining whether a drug will be expelled or firmly incorporated in the long-term[11]. The degree of crystallinity and lipid modification are strongly correlated with drugincorporation and release rates. Thermodynamic stability and lipid packing densityincrease, and drug incorporation rates decrease in the following order: supercooledmelt < a-modification < b-modification < b’-modification < solid state.

However, several alternative incorporation sites (micelles, mixed micelles, liposomes,drug-nanosuspensions) exist in addition to the complex physico-chemical status of thelipid (supercooled melt and several modifications). Most of the data about in vitro drugrelease mechanisms were generated by Mehnert et al., studying the model drugs tetra-caine, etomidate, and prednisolone [16–18]. Burst release as well as sustained release hasbeen reported for SLN suspensions, however, these aspects have also been neglected inthe design and discussion of most SLN experiments.

SLN for Topical Delivery

As topical delivery system, SLN possess many obvious advantages that have been men-tioned in several articles, such as the possibility for systemic drug therapy, and theavoidance of first pass metabolism. It is assumed that SLN dispersion, when administered




directly onto skin in a small but sufficient quantity, would cause fewer side effects, if any,than the currently available formulations. SLN seem to be well suited for use on damagedor inflamed skin because they are formed from nonirritative and nontoxic lipids. They areconsidered as being the next generation of delivery system after liposomes [19, 20].Similar to liposomes they are composed of well-tolerated excipients and due to theirsmall particle size they possess similar adhesive properties leading to film formation onthe skin. Distinct advantages of SLN are the solid state of the particle matrix, the ability toprotect labile ingredients against chemical decomposition, and the possibility to modu-late drug release. Furthermore, the small particle size of SLN ensures that nanoparticlesare in close contact with the stratum corneum, thus increasing the amount of encapsu-lated agents penetrating into the viable skin. The penetration of active compounds intohuman skin was studied using the Tesa stripping test — investigated compoundsincluded coenzyme Q10 [19, 21] and retinol [22–25].

Burst release as well as sustained release has been reported for SLN dispersions,both features are of interest for transdermal application. Burst release can be usefulto improve the penetration of drugs, while sustained release becomes important foractive ingredients with irritating effects at high concentrations or to supply drug to theskin over a prolonged period of time. Fick’s law of diffusion does not seem tobe applicable in this case but may be applied for different parts of the release. SLNtechnology has been reported for topical delivery of triptolide [26], a purified componentof a traditional Chinese medicine isolated from a shrub-like vine named Tripterygiumwilfordii Hook F. Triptolide is effective in the treatment of a variety of inflammatoryand autoimmune diseases, especially rheumatoid arthritis (RA), and has been shownto have other functions, such as antifertility and antineoplastic activity. However, theclinical use of triptolide has some practical disadvantages, mainly due to low watersolubility and toxic effects. In vitro cutaneous permeation studies

from a microemulsions within the first 6 h. Furthermore, the flux was not constant butincreased over the next 6 h. A possible explanation for the release profile from SLNwas that within the first 6 h the carrier remained essentially unchanged and burst drugrelease occurred because of the solid matrix of the particles. Thereafter, due to theexperimental settings, water evaporated from the SLN dispersions during the experiment.Within 12 h the fluid SLN dispersion slowly turned into a semisolid gel. Gel formation ofSLN could be correlated with polymorphic transitions of the lipid matrix [27, 28]. Sincedifferent polymorphic forms differ in their ability to include the drug molecules [29] intheir lattice, drug expulsion as a consequence of this transition was likely. The expelledagent was poorly soluble in water and hence increased thermodynamic activity, explain-ing the higher diffusion rate of triptolide as compared to the first 6 h. The highestcumulative amounts of drug were obtained from the smallest particle size of SLNdispersion. The steady-state flux ( Js) and permeability coefficient (Kp) of triptolidein the first 6 h for the smallest particle size formulation of SLN dispersion were3.1+ 0.4 mg/cm2/h and 0.0124+ 0.001 cm/h. These values were 3.45 and 7.02 timeshigher than those of triptolide solution, respectively. The antiinflammatory activity of SLNdispersion was stronger than that of a microemulsion vehicle in carrageenan-induced ratpaw edema. However, the results were the reversed in complete Frenud’s adjuvant-induced paw edema.

Volkhard Jenning and co-workers [30] use glyceryl behenate SLN loaded with vitaminA (retinol and retinyl palmitate) and incorporated in a hydrogel and o/w-cream to testtheir influence on drug penetration into porcine skin. Vitamin A concentrations in the




(TableFigure 11.1) showed that the permeation from an SLN dispersion was higher than that

11.1,

Table 11.1 The Compositions of the Tested SLN Dispersion Formulations (From Zhinan, M. et al., Eur. J. Pharm. Biopharm. 56,189, 2003. With permission.)

SLNformulation TSG SA P188 SL PEG400MS Particle size PI z (mV) Jx (mg/cm2 per h) Kp (cm/h)

A 5.00 — 1.20 — 3.60 147+1.5 0.27 42 2.8+ 0.3 0.0112+ 0.001B 5.00 — — 1.20 3.60 123+0.9 0.19 45 3.1+ 0.4 0.0124+ 0.002C — 5.00 — 1.20 3.60 157+1.2 0.29 40 2.3+ 0.8 0.0092+ 0.004D — 5.00 1.20 — 3.60 173+2.3 0.24 39 1.9+ 0.4 0.0076+ 0.002

TSG, tristearin glyceride; SA, stearic acid; P188, poloxamer 188; SL, soybean lecithin; PI, polydispersity index; z zeta potential; Jx, flux for the first 6 h; Kp, permeability coefficientfor the first 6 h.

Smith

and

Maib

ach/

Percu

taneous

Penetratio

nEnhan

cers

2nd

edn

TF2152_c0

11

Fin

alPro

of

pag

e130

8.9

.2005

10:2

7pm

130

&P

ercu

tane

ou

sP

en

etratio

nEn

han

cers


skin tissue suggested a certain drug localizing effect. High retinol concentrations werefound in the upper skin layers following SLN preparations, whereas the deeper regionsshowed only very low vitamin A levels. However, the drug localizing action appears to belimited for 6 to 24 h following the application to the skin, after which a polymorphictransition of the lipid carrier occurs with subsequent drug expulsion. Optimal resultswere obtained with retinol SLN incorporated in the o/w cream with respect to drugexpulsion. The penetration of the occlusion sensitive drug retinyl palmitate was evenmore influenced by SLN incorporation.

SLN can also be used as drug carriers for topical glucocorticoids; Santos Maia and co-workers [31] investigate both in vitro and in vivo delivery of Prednicarbate (PC). The riskto benefit ratio of PC was reported to exceed those of halogenated topical glucocorticoidsby about twofold. To obtain a further highly desirable increase by drug targeting to theviable epidermis, PC was incorporated into SLN. Keratinocyte and fibroblast monolayercultures, reconstructed epidermis and excised human skin served to evaluate SLN toxicityand PC absorption. Well-tolerated preparations (e.g., cellular viability 94.5% following18 h incubation of reconstructed epidermis) were obtained. PC penetration into humanskin increased by 30% as compared to PC cream, permeation of reconstructed epidermisincreased even threefold. The present study shows the great potential of SLN to improvedrug absorption by the skin.

SLN as a Novel Carrier System for Sunscreens

Protection against UV radiation has played an important role in preventing skin cancer [5].Two basic UV protection systems include molecular UV absorbers (sunscreens) andphysical, particulate compounds such as titanium dioxide. Molecular blockers oftenhave photoallergic and phototoxic effects; particulate blockers such as small titaniumdioxide (e.g., 5 to 20 nm) penetrate into the skin and can interact with the immune system[32]. It has been found in vitro that SLN have UV reflecting properties [33]. The UVreflectance is related to the solid state of the lipid, and was not evident in nanoemulsionsof comparable composition [34].

3 6 9 12 1500

5

10

20

30

40

35

25

15

Time (h)

Cum

ulat

ive

amou

nt o

ftr

ipto

lide

(µg/

cm2 )

formulation Aformulation Bformulation Cformulation Dformulation H

Figure 11.1 Permeation profiles of triptolide through excised rat skins from SLN formulations(A–G), H was a triptolide solution. (From Zhinan, M. et al., Eur. J. Pharm. Biopharm., 56, 189,2003. With permission.)




Occlusive Effect of SLN

An adhesive effect is claimed for small particles, such as liposomes that form a film on theskin after application, the adhesion increasing with decreasing particle size. The sameadhesive effect was postulated for SLN some years ago [7]. Occlusion can enhance thepenetration of drugs through the stratum corneum by increased hydration. Apart fromnonspecific occlusion effects, drug penetration might also be affected by the SLN carrieritself as the large surface area of the nanometer-sized SLN facilitates contact of theencapsulated drugs with the stratum corneum [35]. Furthermore, stabilization of chem-ically unstable drugs by incorporating them into a lipid matrix might be possible. On theother hand, soybean lecithin is also known to improve the safety of co-applied agents[36]. Moreover, the co-applied lipids are likely to minimize the danger of allergic contactdermatitis that may be induced by the drug [31].

In Vitro Occlusion of SLN

Intensive in vitro studies were performed to quantify the occlusivity of SLN in terms ofthe so-called occlusion factor [11]. The first systematic occlusion study was performed byWissing et al. [37], investigating the chemical nature of the lipid, crystallinity of the lipidmatrix, and particle size [11]. A first model for the film formation by SLN on the skin wasdeveloped by Muller and Dingler [20] — a hexagonal packaging in a monolayer wasassumed. Figure 11.2 shows the difference for 2 mm lipid microparticles compared to200 mm (note the figure shows correct size relations). In hexagonal packing, about 76% of

Section:

2 µm 200 nm

Top view:

Fusion:

skin

small“capillary pores”

application andcapillary forces

H2O evaporation

large pores

Figure 11.2 Model of film formation on the skin for lipid 2-mm particles and lipid 200 nmparticles shown as section (upper) and from the top (middle) perspectives. A new model of fusionof the nanoparticles into a pore-less film (lower). (From Muller, R.H., Radtke, M., and Wissing,S.A., Adv. Drug. Deliv. Rev., 54 (Suppl. 1), S131, 2002. With permission.)




the surface is covered, 24% is uncovered, meaning the uncovered surface is identical forboth the micro- and the nanoparticles. However, the ‘holes’ in between the microparticlesare relatively large and this favors the evaporation of water. In contrast, only nanosizedpores exist in the monolayer of SLN and evaporation of water is unfavorable from thesepore dimensions. These pore-water dynamics are similar to the occurrence of capillarycondensation in silica gel; water condensates in the pores due to their small size andreduced vapor pressure (La Place equation). Thus the pores in the SLN film would ratherattract than lose water. Recent investigations by electron microscopy showed that afterevaporation of the water from SLN dispersions, a continuous, pore-less film was formed

Since sunscreens are intended to act on the surface of the skin, they should penetrateas little as possible into the viable epidermis, dermis, and into the systemic circulation[34]. Also, the presence of sunscreens on top of the horny layer prevents phototoxic andphotoallergic reactions which have been observed for various molecular UV-blockers[39, 40]. Incorporation of molecular sunscreens into SLN has a synergistic effect on theirprotective characteristics. Studies using a membrane-free model by Wissing and Muller[38, 16] clearly showed that incorporation of the molecular sunscreen oxybenzone inSLNs decreased the rate of release by up to 50% compared to equally sized emulsions.Using SLN as the carrier system offers further advantages in that the SLN act as sunscreensthemselves (due to the solid state of its lipids), so the concentration of potentiallyhazardous molecular sunscreen (UV blockers) or TiO2 can be decreased. SLN are ableto decrease the rate of release and penetration of the sunscreen chemical, thereforethe sunscreen remains on the surface of the skin for longer periods. However, if thesunscreen formulation is highly occlusive already, addition of SLN will have little orno effect [7].

SLN In Vivo Occlusion

To detect the adhesiveness of SLN on human skin, a Tesa strip test has been performed[41, 42]. The strips were analyzed using electron microscopy at different magnificationsand showed the presence of SLN even at the largest magnification. It appears, from theseobservations, that SLN adhere to the skin surface, explaining the film formation andincreased skin hydration. The increase in the water content within the skin generated by

Figure 11.3 Electron micrograph of an air-dried SLN dispersion. (From Wissing, S.A. andMuller, R.H., Pharmazie, 56, 783, 2001. With permission.)




(Figure 11.2 lower and Figure 11.3) [1].

SLN formulations can reduce the symptoms of atopic eczema and improve the appear-ance of the skin.

Conclusions

SLN were developed at the beginning of the 1990s as an alternative carrier system toemulsions, liposomes, and polymeric nanoparticles. They are attractive carriers fortopical cosmetic and pharmaceutical products. However, more human studies need tobe conducted with SLN delivery vehicles to produce ‘‘real life’’ data. Especially, a betterunderstanding is needed of how lipid nanoparticles modify drug penetration into theskin, and how lipid particles interact with the lipids of the stratum corneum, and howthey then affect drug penetration.

References1. Muller, R.H., Radtke, M., and Wissing, S.A., Solid lipid nanoparticles (SLN) and nanostructured

lipid carriers (NLC) in cosmetic and dermatological preparations, Adv. Drug. Deliv. Rev., 54(Suppl.1), S131, 2002.

2. Mehnert, W. and Mader, K., Solid lipid nanoparticles: production, characterization and appli-cations, Adv. Drug Deliv. Rev., 47, 165, 2001.

3. Roberta Cavalli, M. et al., Solid lipid nanoparticles (SLN) as ocular delivery system for tobra-mycin, Int. J. Pharm., 238, 241, 2002.

4. Scholer, N. et al., Preserved solid lipid nanoparticles (SLN) at low concentrations do causeneither direct nor indirect cytotoxic effects in peritoneal macrophages, Int. J. Pharm., 196, 235,2000.

5. Muller, R.H. et al., Solid lipid nanoparticles (SLN) — an alternative colloidal carrier system forcontrolled drug delivery, Eur. J. Pharm. Biopharm., 41, 62, 1995.

6. Siekmann, B. and Westesen, K., Submicron-sized parenteral carrier systems based on solidlipids, Pharm. Pharmacol. Lett., 1, 123, 1992.

7. Muller, R.H and Lucks, J.S. Arzneistofftrager aus festen Lipidteilchen, Feste Lipidnanospharen(SLN), Eur. Pat. No. 0605497, 1996.

8. Lippacher, A., Muller, R.H., and Mader, K., Investigation on the viscoelastic properties of lipidbased colloidal drug carriers, Int. J. Pharm., 196, 227, 2000.

9. Lander, R. et al., Homogenization: a mechanistic study, Biotechnol. Prog., 16, 80, 2000.10. Jahnke, S., The theory of high pressure homogenization, in Emulsions and Nanosuspen-

sions for the Formulation of Poorly Soluble Drugs, Muller, R.H., Benita, S., and Bohm B.,Eds, Medpharm Scientific Publishers, Stuttgart, 1998, 177.

11. Muller, R.H., Mader, K., and Gohla, S., Solid lipid nanoparticles (SLN) for controlled drugdelivery — a review of the state of the art, Eur. J. Pharm. Biopharm., 50, 161, 2000.

12. Gasco, M.R., Solid lipid nanospheres from warm micro-emulsions, Pharm. Technol. Eur., 9, 52,1997.

13. Boltri, L. et al. Lipid nanoparticles: evaluation of some critical formulation parameters, Proc.Int. Symp. Contr. Rel. Bioact. Mater., 20, 346, 1993.

14. Scholer, N. et al. Surfactant, but not the size of solid lipid nanoparticles (SLN) influencesviability and cytokine production macrophages, Int. J. Pharm., 221, 57, 2001.

15. Boistelle, R., Fundamentals of nucleation and crystal growth, in Crystallization and Poly-morphism of Fats and Fatty Acids, Garti, N. and Sato, K. Eds, Marcel Dekker, Inc., NewYork, Basel, 189, 1988.

16. Wissing, S.A. and Muller, R.H., Solid lipid nanoparticles as carrier for sunscreens, in vitrorelease and in vivo skin penetration, J. Control. Rel., 81, 225, 2002.




17. Muhlen, A.Z and Mehnert, W., Drug release and release mechanism of prednisolone loadedsolid lipid nanoparticles, Pharmazie, 53, 552, 1998.

18. Mehnert, W. et al, Solid lipid nanoparticle ein neuartiger Wirkstoff-Carrierfur Kosmetika undPharmzeutika, Wirkstoff-Inlorporation, Freiserzungund Sterilizierbarkeit, Pharm. Ind., 4, 511,1997.

19. Muller, R.H. and Dingler, A., Feste Lipid-Nanopartikel (Lipopearls) als neuartiger Carrier furkosmetische und dermatologische Wirkstoffe, Pharmazeutische Zeitung Dermopharmazie,49, 11, 1998.

20. Muller, R.H. and Dingler, A., The next generation after the liposomes: solid lipid nanoparticles(SLNe

ˆ, Lipopearlse

ˆ) as dermal carrier in cosmetics, Eurocosmetics, 7/8, 19, 1998.

21. Dingler, A., Feste Lipid-Nanopartikel als kolloidale Wirk stofftragersysteme zur dermalenApplikation, PhD thesis, Free University of Berlin, 1998.

22. Jenning, V., Schafer-Korting, M., and Gohla, S., Vitamin A-loaded solid lipid nanoparticles fortopical use: drug release properties, J. Control. Rel., 66, 115, 2000.

23. Jenning, V. et al. Vitamin parenteral administration of lipid nanoparticles. A loaded solidlipid nanoparticles for topical use: occlusive properties and drug targeting to the upper skin,Eur. J. Pharm. Biopharm., 49, 211, 2000.

24. Jenning, V. and Gohla, S., Encapsulation of retinoids in solid lipid nanoparticles (SLN),J. Microencapsul., 18, 149, 2001.

25. Jenning, V., Feste Lipid-Nanopartikel (SLN) als Tragersystem fur die dermale Applikation vonRetinol: Wirkstoffinkor-poration,-freisetzung und Struktur, PhD thesis, Free University ofBerlin, 1999.

26. Zhinan, M. et al. Solid lipid nanoparticle and microemulsion for topical delivery of triptolide,Eur. J. Pharm.Biopharm., 56, 189, 2003.

27. Gasco, M.R., Morel, S., and Carpigno, R., Optimization of the incorporation of desoxycortisoneacetate in lipospheres, Eur. J. Pharm. Biopharm., 38, 7, 1992.

28. Muhlen, Z.A., Schwarz, C., and Mehnert, W., Solid lipid nanoparticles (SLN) for controlled drugdelivery — Drug release and release mechanism, Eur. J. Pharm. Biopharm., 45, 149, 1998.

29. Bunjes, H., Westesen, K., and Koch, M.H.J., Crystallization tendency and polymorphic transi-tions in triglyceride nanoparticles, Int. J. Pharm., 129, 159, 1996.

30. Jenning, V. et al. Vitamin A loaded solid lipid nanoparticles for topical use: occlusive propertiesand drug targeting to the upper skin, Eur. J. Pharm. Biopharm., 49, 211, 2000.

31. Maia, C.S., Mehnert, W., and Korting, M.S., Solid lipid nanoparticles as drug carriers for topicalglucocorticoids, Int. J. Pharm., 196, 165, 2000.

32. Hagedorn-Leweke, U. and Lippold, B.C., Accumulation of sunscreens and other compounds inkeratinous substrates, Eur. J. Pharm. Biopharm., 46, 215, 1998.

33. Freitas, C. and Muller, R.H., Correlation between long-term stability of solid lipid nanoparticles(SLN) and crystallinity of the lipid phase, Eur. J. Pharm. Biopharm., 47, 125, 1999.

34. Potard, G. et al. The stripping technique: in vitro absorption and penetration of five UV filterson excised fresh human skin, Skin Pharmacol. Appl. Skin Physiol., 13, 336, 2000.

35. Jenning, V. et al., Vitamin A loaded solid lipid nanoparticles for topical use: occlusive prop-erties and drug targeting to the upper skin, Eur. J. Pharm. Biopharm., 49, 211, 2000.

36. Cevc, G. and Blume, G., New, highly efficient formulation of diclofenac for the topicaltransdermal administration in ultradeformable drug carriers, Transfersomes, BBA 1514, 191,2001.

37. Wissing, S.A., Lippacher, A., and Muller, R.H., Investigations on the occlusive properties ofsolid lipid nanoparticles (SLNe

ˆ), J. Cosmet. Sci., 52, 313, 2001.

38. Wissing, S.A. and Muller, R.H., Solid lipid nanoparticles (SLNeˆ) — a novel carrier for UV

blockers, Pharmazie, 56, 783, 2001.39. Goossens, A. et al., Adverse cutaneous reactions to cosmetic allergens, Contact Dermat., 40,

112, 1999.40. Ricci, C., Pazzaglia, M., and Tosti, A., Photocontact dermatitis from UV filters, Contact Dermat.,

38, 343, 1998.




41. Dingler, A. Feste Lipid-Nanopartikel als kolloidale Wirk-stofftragersysteme zur dermalenApplikation, PhD thesis, Free University of Berlin, 1998.

42. Dingler, A. et al., Solid lipid nanoparticles (SLNeˆ/Lipiopearlse

ˆ) — a pharmaceutical and

cosmetic carrier for the application of vitamin E in dermal products, J. Microencapsul., 16,751, 1999.




Chapter 12

Fatty Alcohols and Fatty Acids

R. J. Babu, Mandip Singh, and Narayanasamy Kanikkannan

CONTENTS

Introduction .................................................................................................................................... 137General Overview of Fatty Alcohols and Fatty Acids on the Skin Permeation ........................... 138Fatty Alcohols ................................................................................................................................. 139Fatty Acids....................................................................................................................................... 144

Effect of Carbon Chain Length of Fatty Acids ........................................................................... 145Saturated versus Unsaturated Fatty Acids .................................................................................. 145The Number and Position of Double Bonds in Unsaturated Fatty Acids ................................ 146Fatty Acid Esters.......................................................................................................................... 146Chemical Modification of Drugs with Fatty Acids..................................................................... 147

Mechanism of Penetration Enhancement...................................................................................... 148Lipid Disruption at the Intercellular Level: The Enhancer Disrupts Stratum

Corneum Lipid Organization, Making It Permeable to Drugs............................................. 148Fatty Acids Form Solvated Complexes or Molar Addition Compounds and Permeate

through Skin Simultaneously with Drugs ............................................................................. 149Fatty Acids Increase the Diffusivity and Partitioning of Drugs and Vehicles

through Stratum Corneum..................................................................................................... 150Formulation Considerations........................................................................................................... 150

Enhancer Configuration in TDS ................................................................................................. 150Fatty Alcohols and Acids in Transdermal Patches..................................................................... 151

Skin Irritation Potential................................................................................................................... 152Summary ......................................................................................................................................... 153References....................................................................................................................................... 154

Introduction

Fatty alcohols, fatty acids, and their derivatives are used in a variety of skin and generalhealthcare products, to name a few: moisturizing creams, shampoos, hair products,shaving products, bath oils, lipsticks, and perfumed products. The wide usage of these


137


as topical ingredients implies that they are nontoxic and considered safe for topical use.Pimecrolimus 1% cream (Elidel1 cream), Tretinon cream (Renova1 cream), and Fluor-ouracil 0.5% cream are some examples of topical formulations employing fatty alcohols,or fatty acids, or both as dermatological ingredients. Alpha-lipoic acid, a ring containingfatty acid, is an active medicament for treating photo-damaged skin and is deemed safeas a topical agent. Fatty alcohols and fatty acids have been extensively investigated tocharacterize their interactions with stratum corneum (SC) and to describe their effects inmodulating the skin barrier function.1 This chapter deals with the utility of fatty alcoholsand fatty acids as promising skin penetration enhancers for topical and transdermaldelivery of drugs. The role of vehicle or other ingredients on the enhancement effectsof the topical or transdermal formulation and their skin irritation potential has also beendiscussed.

General Overview of Fatty Alcohols and Fatty Acidson the Skin Permeation

Both saturated and unsaturated fatty acids and alcohols have been established as skinpermeation enhancers. Most fatty acids are straight-chain compounds with carbon chain-lengths between 2 and 24. Medium chain (C6–C10) and long-chain (C12–C24) fatty acidsare frequently reported as skin penetration enhancers. In addition, fatty acids withdifferent structural configurations, unsaturated carbon atoms and branched carbonchains, and also fatty acid esters have been reported as skin penetration enhancers.

tion enhancers. Fatty acids increase the skin permeation by disordering the highlyordered structure of skin lipid barrier and they are generally believed to increase thediffusivity and partitioning of drugs across SC. Fatty acids have been studied as skinpenetration enhancers mainly for lipophilic drugs and in some instances these also have

some studies in which the fatty acids were used as skin penetration enhancers for drugswith varied physico-chemical properties. In many cases the vehicle has strong influenceon the efficacy of fatty acids and some vehicles like propylene glycol or ethanol havesynergistic effect along with fatty acids in enhancing the skin permeation of drugs. A largenumber of patents describing the utility of fatty acids or alcohols as enhancers in

These patents claim topical formulations, which are nonirritating to skin and comprise avariety of therapeutic agents.

Fatty alcohols have generally lower melting points and higher solubilities than corre-

penetration enhancers. It is often considered that the chemical enhancement is a conse-quence of interactions in the polar head group region, which results in the increasedfluidity of the alkyl chains.2 It is difficult to classify fatty alcohols and fatty acids in terms oftheir efficacy as penetration enhancers as their enhancement effect is dependent on thephysico-chemical properties of both the drug molecule and the enhancer under study.Earlier investigations3–5 on the influence of n-alkanols, alkyl pyrrolidones, alkyl diols, andalkyl dimethylamides as skin permeation enhancers for steroid molecules as permeantsdemonstrated that the enhancer potency of these four homologous series was the samewhen compared at the same alkyl chain length; that is, the contribution of the hydroxyl,pyrrolidone, diol, and dimethylamide groups to enhancer potency was the same. This




Table 12.1 presents the list of saturated and unsaturated fatty acids used as skin penetra-

been reported as enhancers for peptides and hydrophilic permeants. Table 12.2 lists

transdermal formulations have been reported; some examples are given in Table 12.3.

sponding fatty acids. Table 12.4 presents some frequently reported fatty alcohols as skin

implies that the enhancer potency of fatty acids and fatty alcohols depends on their alkylchain and the contribution of polar head group may not be very significant. In contrast, thedata from our laboratory6,7 on the permeation of melatonin through porcine skin indicated

Studies on the skin-permeation enhancement abilities of lauric acid and lauryl alcoholindicated substantial difference between these two compounds; lauric acid showed 30-fold higher flux of naloxone through human skin as compared with lauryl alcohol.8

Fatty Alcohols

Ethyl alcohol has been widely employed as a penetration enhancer in several marketedtopical and transdermal formulations (e.g., Estraderm1 patches, Duragesic1 patches,

Table 12.1 Fatty Acids Reported in the Literature as Skin Penetration Enhancers

Chemical name Common name Molecular weight

Saturated Fatty Acids with Linear Hydrocarbon ChainPentanoic Valeric 102.1Hexanoic Caproic 116.1Octanoic Caprylic 144.2Nonanoic Pelargonic 158.2Decanoic Capric 172.3Dodecanoic Lauric 200.3Tetradecanoic Myristic 228.4Hexadecanoic Palmitic 256.4Octadecanoic Stearic 284.4Eicosanoic Arachidic 412.5Docosanoic Behenic 340.5Tetracosanoic Lignoceric 368.6

Saturated Fatty Acids with Branched Hydrocarbon ChainPentan-2-oic Isovaleric acid 102.12,2-Dimethyl pentanoic Neoheptanoic 130.22,2-Dimethyl heptanoic Neononanoic 158.22,2-Dimethyl octanoic Neodecanoic 172.32-Heptyl undecanoic Isostearic acid 284.4

Unsaturated Fatty Acidscis-9-Tetradecenoic acid Myristoleic acid 226.4cis-9-Hexadecenoic acid Palmitoleic acid 254.4All cis-9,12,15-octadecadienoic acid a-Linolenic acid 278.4All cis-6, 9 and 12-octadecatrienoic acid g-Linolenic acid 278.4All cis-9, 12-octadecadienoic acid a-Linolenic acid 280.4cis-11-octadecenoic acid Asclepic acid 282.4cis-6-octadecenoic acid Petroselinic acid 282.4trans-9-octadecenoic acid Elaidic acid 282.4cis-9-octadecenoic acid Oleic acid 282.4All cis-5, 8, 11 and 14-eicosatetranoic acid Arachidonic acid 304.5cis-11-eicosenoic acid Gondoic acid 310.5cis-13-docosenoic acid Erucic acid 338.6


Fatty Alcohols and Fatty Acids & 139


clear differences between different fatty acids and corresponding alcohols (Table 12.5).

and Nimulid1 gel) and is often the solvent of choice for use in patches.1 The penetrationenhancement effect increases with the increase in ethanol volume fraction, and at veryhigher ethanol volume fractions a reduction in the permeation rate was observed fortestosterone9 and estradiol.10 However, ethanol pretreatment enhanced the permeationof insulin through rat skin better than with ethanol/water skin pretreatment.11

The effect of a series of straight chain alkanols as penetration enhancers on thetransdermal delivery of levonorgestrel through excised rat and human cadaver skin wasstudied.12 The steady-state flux of levonorgestrel increased as the alkyl chain increasedfrom C2 to C4, but decreased as the chain length increased above C4 alcohol. Goosenet al.13 reported that short-chain alkanols (C1 to C6) were better in enhancing thepermeation of methyl thalidomide than medium chain alkanols (C7 to C12). Thesteady-state flux was highest with C1 (methanol) and the flux decreased linearly withthe increase in the carbon chain length of alkanol up to C6 (hexanol). The medium chainalkanols showed a parabolic relationship between the flux of methyl thalidomide and thecarbon chain length of the alkanol with peak permeation rate at C10 (decanol), as shown

Sloan et al.14 measured the fluxes of theophylline through hairless mouseskin from formulations containing straight chain alkanols. The flux of theophylline was

Table 12.2 Fatty Acids and Vehicles as Penetration Enhancers for Different DrugMolecules

Drug Enhancer Vehicle Skin speciesEnhancement

factor

Flurbiprofen94 5% Oleic acid CMC Hydrogel Rat skin 7.65% Linolic acid 8.05% Linolenic acid 9.0

Levosimendan36 5% Oleic acid 40% ethanol Human epidermis 22.7Arginine

vasopressin955% Oleic acid Ethanol:water Rat 19.5

5% Linolic acid 19.75% Linolenic acid 19.5

Diclofenac Na96 20% Oleic acid Transcutol:water (59:20)

Rat skin 16.7

20% Lauric acid 18.915% Oleic acidþ 5% limonene

121.0

Ondansetron97 Oleic acid(pretreatment)

60% ethanol Shed snake skin 45.5

Lauryl alcohol 28.0Tenoxicam33 3% Oleic acid Propylene glycol Hairless mouse skin 72.8

3% Linolenic acid 237.53% Lauric acid 9.83% Capric acid 6.03% Caprylic acid 4.4

Flurbiprofen28 5% Palmitoleic acid Propylene glycol Rat skin 7.05% Pleic acid 5.85% Linoleic acid 5.25% Linolenic acid 17.55% Arachdonic acid 2.8




in Table 12.6.

Table 12.3 Fatty Alcohols or Acids Reported as Skin Penetration Enhancers in the Patent Literature

Drug Fatty acid Vehicle SkinEnhancement

factor Important claims

Chlorpheneraminebase97

11% Undecylenic acid24% Isostearic acid

Isopropyl myristateIsopropyl myristate

Human skin 1.41.6

Reduction of irritancy of theamine base drug was claimed

Nicotine base98

Interferon a2b99

(IFNa2b)Palmitic acid(IFNa2b was conjugated)

0.1%methyl cellulose gel Human skin 1.84 8.2-fold Higher cutaneous depositionin viable skin layers by palmitoylIFNa2b was claimed

Oxymorphone78 5% Linolenic acid Propylene glycol: triacetin Hairless mouse 13.33 Linolenic acid in aqueoussystem with no skinirritation was claimed

10% Linolenic acid 0.3% carbopol gel 17.1720% Linolenic acid (with 2.5% Tween 20) 22.225% Linolenic acid 27.5610% Linolenic acid 34.0920% Linolenic acid 32.24

Norethindroneacetate67

6% Oleic acidþ12%linolenic acid

DuroTek — ethylcellulose matrix

Pig skin 1.5 Good adhesiveness and low skinirritation was claimed. The systemacts by push–pull mechanism

Molsidomine100 10% Lauric acid Propylene glycol Human cadaverskin

15.0 —

Smith

and

Maib

ach/

Percu

taneous

Penetratio

nEnhan

cers

2nd

edn

TF2152_c0

12

Fin

alPro

of

pag

e141

8.9

.2005

10:3

1pm

FattyA

lcoh

ols

and

FattyA

cids

&141


the least from methanol (C1) and increased almost 100-fold by alkanols with carbonatoms between 5 and 9 (C5 to C9) and then decreased to 10-fold for the flux oftheophylline by undecanol (C11).

The enhancement effect of saturated (C8 to C14) and unsaturated fatty alcohols (C18with one, two, or three double bonds) at 5% concentration in 60% ethanol was studied formelatonin.7 A parabolic relationship between the carbon chain length of fatty alcohol andthe flux of melatonin with peak permeation rate by decanol was established. Amongunsaturated fatty alcohols, as the level of unsaturation increased from one to two doublebonds, there was an increase in the permeation of melatonin, both in porcine and humanskin. However, a decrease in the permeation was observed with fatty alcohol with threedouble bonds.7 In another study, we reported that the fatty alcohols (decanol, undecanol,and lauryl alcohol), which showed greater permeation of melatonin, also producedgreater transepidermal water loss (TEWL), skin blood flow, and erythema in hairlessrats. Octanol and nonanol may be the most useful enhancers for the transdermal deliveryof melatonin considering their lower skin irritation and a reasonably good permeationenhancement effect.15 The influence of 10% saturated alcohols (C8 to C18) in propylene

Table 12.4 Fatty Alcohols Reported as Skin Penetration Enhancers

Chemical name Common name Formula

n-Alcohols1-Decanol Capric alcohol C10H21OH1-Dodecanol Lauryl alcohol C12H25OH1-Tetradecanol Myristyl alcohol C14H29OH1-Hexadecanol Cetyl alcohol C16H33OH1-Octadecanol Stearyl alcohol C18H37OH

Iso-alcohols10-Methyl-1-hendecanol Isolauryl alcohol C12H25OH12-Methyl-1-tridecanol Isomyristyl alcohol C14H29OH14-Methyl-1-pentadecanol Isopalmityl alcohol C16H33OH16-Methyl-1-heptadecanol Isostearyl alcohol C18H37OH

Table 12.5 Steady-State Flux Values of Melatonin across PorcineSkin Using Saturated Fatty Alcohols and Fatty Acids

Fatty alcohols (flux mg/cm2/h) Fatty acids (flux mg/cm2/h)

Control 4.5+ 0.8 Control 5.34+0.59Nonanol 28.9+ 2.5 Nonanoic acid 7.77+0.55*Decanol 30.7+ 2.6 Decanoic acid 18.79+1.59*Undecanol 23.9+ 1.3 Undecanoic acid 23.70+2.64Lauryl alcohol 17.2+ 2.3 Lauric acid 24.98+1.45*Myristyl alcohol 10.7+ 1.5 Myristic acid 17.29+1.31*Oleyl alcohol 12.3+ 1.3 Oleic acid 11.52+1.29Linoleyl alcohol 18.6+ 1.5 Linoleic acid 9.14+0.65*Linolenyl alcohol 12.5+ 1.2 Linolenic acid 10.23+0.52

*Significantly different from that of corresponding fatty alcohol (P < 0.05). (Modified fromAndega, S., Kanikkannan, N., and Singh, M., J. Control. Release, 77, 17, 2001 and Kandimalla,K.K. et al. J. Pharm. Pharmacol., 51, 783, 1999.)




glycol on the flux of naloxone through the human skin was studied.8 A parabolic effect ofalkyl chain length with C10 and C12 being most effective was demonstrated. It was alsoshown that with an increase in the number of double bonds in the C18 fatty alcohol, thepermeation of naloxone is increased.

The effect of a series of fatty alcohols in ethanol/Panasate 800 and ethanol/water onthe permeation of Tegafur across hairless mouse skin was studied.16 With ethanol/Panasate 800 system, all fatty alcohols, except stearic acid, increased the skin permeabil-ity of Tegafur, but the degree of penetration enhancement was similar among differentfatty acids. In contrast, with ethanol/water (6:4) binary vehicle, all fatty alcohols signifi-cantly enhanced the flux of Tegafur and the flux increased with an increase in alkyl chainlength, reached peak permeation by C12-alcohol, and then decreased as further increasein the alkyl chain length. These results suggest that vehicle plays an important role in thepermeation enhancement effect of fatty alcohols.

The effect of various alkanols in isopropyl myristate on the permeation of benztropinethrough hairless mouse and human cadaver skin was studied.17 Interestingly, the binarycosolvents consisting of IPM and short-chain alkanols such as ethanol, isopropanol, andtertiary butanol, in particular a 2:8 combination produced a marked enhancementof benztropine flux from the mesylate salt, whereas a retarding effect was noticed forthe permeation of benztropine base. The enhancement potency for the benztropinemesylate permeation increased linearly with the carbon number of the alcohol presentin the binary mixtures.

The effect of n-alkanols on the permeation of a polar, nonelectrolyte penetrant,nicotinamide through hairless mouse skin in vitro was studied. The enhancement versusalkanol chain length profile was parabolic with a peak permeation rate at C6. This studydemonstrated the utility of fatty alcohols as enhancers for polar penetrants.18 The authorssuggested that the principal mechanism by which alkanols enhance percutaneousabsorption of polar penetrants is by extraction of SC intercellular lipids. Seki andMorimoto19 demonstrated that medium chain aliphatic alcohols (C8–C12) enhanced thepermeation of both hydrophilic and lipophilic model drugs (6-carboxyfluorescein and

Table 12.6 Steady-State Flux (mg/cm2/h+ SD) Values of Thalidomide and ItsN-Alkyl Analogs from a Series of n-Alcohols

Steady-State Flux (mg/cm2/h+ SD)

Vehicle Thalidomide N-Methylthalidomide N-Propylthalidomide

Methanol (C1) 0.147+ 0.005 4.498+ 0.220 1.730+0.768Ethanol (C2) 0.066+ 0.053 2.819+ 0.391 0.378+0.226Propanol (C3) 0.039+ 0.002 0.822+ 0.096 0.621+0.058Butanol (C4) 0.037+ 0.016 0.813+ 0.196 0.123+0.03Pentanol (C5) 0.039+ 0.014 0.722+ 0.034 0.291+0.134Hexanol (C6) 0.028+ 0.007 0.468+ 0.037 0.347+0.111Heptanol (C7) 0.029+ 0.002 0.250+ 0.016 0.129+0.023Octanol (C8) 0.035+ 0.010 0.185+ 0.003 0.164+0.022Nonanol (C9) 0.036+ 0.018 0.174+ 0.008 0.225+0.062Decanol (C10) 0.018+ 0.001 0.351+ 0.220 0.201+0.081Undecanol (C11) 0.010+ 0.003 0.223+ 0.040 0.101+0.047Dodecanol (C12) 0.025+ 0.004 0.172+ 0.030 0.063+0.016

Modified from Goosen, C. et al. Pharm. Res., 19, 434, 2002.




indomethacin) through excised rat skin. The enhancing effects of the aliphatic alcoholsfor 6-carboxyfluorescein and indomethacin decreased with the increase in carbon chainlength. Although the relationships between the structure and skin permeation-enhancingeffect of the aliphatic alcohols used in this study are not yet fully understood, theyare possible candidates as permeation enhancers for both hydrophilic and lipophilicdrugs.

The influence of hydrocarbon chain branching or positioning of polar head group(–OH) in the alkyl chain on the permeation enhancement effect was examined byChantasart et al.20 The effects of x-heptanol, x-octanol, and x-nonanol, (where x is theposition of the hydroxyl group ranging from 1 up to 5) on the transport of a probepermeant, corticosterone across hairless mouse skin were investigated. To compareenhancer potencies among different enhancers, the concept of isoenhancement concen-tration was introduced; which is defined as the aqueous solution concentration of anenhancer in equilibrium with the SC yielding the same transport enhancement factor E asthat of a reference enhancer solution. For example, E¼ 10 is where the transdermal fluxof a permeant is increased 10-fold by the enhancer over the control (no enhancer) for thesame thermodynamic gradient across skin. The isoenhancement concentrations of2-alkanol, 3-alkanol, 4-alkanol, and 5-alkanol to induce E¼ 10 were approximately1.9-, 2.6-, 3.1-, and 3.9-fold higher, respectively, than those of 1-alkanols of the samemolecular formula. This suggests that the branched chain-alkanols have lower enhancerpotency than 1-alkanols of the same molecular formula; the potency decreases as thehydroxyl group moves from the end of the chain towards the center of the enhancer alkylchain. Branching of the alkyl chain reduces the ability of the enhancer to effect lipidfluidization in the SC lipid lamellae at the target site.20

It may be significant that the most effective chain lengths (C10–C12) correspond to thelength of the steroid nucleus of cholesterol, suggesting that these may act by disruptingceramide–cholesterol or cholesterol–cholesterol interaction.2 Ackermann et al.21 studiedthe permeation of a series of alkanols (C1–C8) across the nude mouse skin. The per-meability coefficients of alkanols increased linearly as the chain length increased. Further,the permeability coefficients of n-alkanols correlated well with their ether–water partitioncoefficients. These results could be used to explain the permeation enhancement effectof different alkanols. The increase in the enhancement effect of lower alkanols withincrease in the alkyl chain length may be attributed to the increased permeation ofalkanols through the skin. While short-chain alkanols (polar) traverse the skin, thelong-chain alkanols (nonpolar) are largely retained in the SC and this appears to makesuch combinations superior enhancer systems.13

Fatty Acids

Fatty acids have been extensively studied as skin penetration enhancers for the devel-opment of successful topical and transdermal delivery systems of different classes ofdrugs. Structurally, fatty acids consist of an aliphatic hydrocarbon chain and a terminalcarboxyl group. These fatty acids differ in their hydrocarbon chain length, in the number,position, and configuration of double bonds, and have branching and other substituents.It is generally suggested that the C12 and C14 hydrophobic groups have an optimalbalance of partition coefficient and affinity for the skin.22,23 However, oleic acid — a C18unsaturated fatty acid — is one of the most extensively studied penetration enhancersamong the fatty acids.24,25




Effect of Carbon Chain Length of Fatty Acids

There are several reports on the effect of carbon chain length of fatty acids on the skinpermeation enhancement of drugs. Hsu et al.26 studied the effect of saturated fatty acids(C12–C18) on the percutaneous absorption of piroxicam through rat skin and theenhancement effect was decreased linearly with increasing carbon number of saturatedfatty acid from 12 to 18. The permeation-enhancing effects of saturated fatty acids formelatonin through excised hairless mouse skin was reported to be in the followingdecreasing order27: C10>C12>C14>C16>C18. Meanwhile, oleic acid (C18) dramatic-ally enhanced the skin permeability coefficient of melatonin more than 950-fold overthe effect of propylene glycol alone.27 We reported earlier that the saturated fatty acids(C9–C14) enhanced the permeation of melatonin across excised rat skin. A sharp increasein the permeation of melatonin was observed, when the fatty acid chain length increasedup to 11 carbons, and then the permeation rate decreased.6 Among the series of saturatedfatty acids investigated on the permeation of propranolol through rabbit skin, lauric acidand myristic acids were the most potent agents in increasing the permeation of propra-nolol from the gel formulations.22 Among capric, lauric, and myristic acids, lauric acidwas found to be optimum in enhancing the permeation of flurbiprofen across rat skin.28

The other two fatty acids showed no significant increase in the permeation of flurbipro-fen as compared with control vehicle (propylene glycol). In another investigation, theskin permeation of albuterol across hairless mouse skin was studied using Klucel 0.5% gelcontaining capric, lauric, and myristic acid as skin penetration enhancers. The resultssuggested that lauric acid preferentially enhanced albuterol diffusion compared to otherfatty acids.29

It has been proposed that acids with a certain chain length, that is, around 12 carbons,possess an optimal balance between partition coefficient or solubility parameter andaffinity to skin.22 Shorter chain fatty acids would have insufficient lipophilicity for skinpermeation, whereas longer chain fatty acids would have much higher affinity to lipids inSC and thereby retarding their own permeation and that of other permeants. The paralleleffect with the permeation enhancement suggests that the mode of action of saturatedfatty acids as enhancers is dependent on their own permeation across the SC or skin.30,31

Saturated versus Unsaturated Fatty Acids

It has been well established that unsaturated fatty acids are more potent permeationenhancers than the saturated species.27 Chi et al.28 compared the permeation enhance-ment effects of saturated and unsaturated fatty acids for flurbiprofen. Among saturatedfatty acids (capric, lauric, and myristic acids), only lauric acid was effective as penetrationenhancer. The permeation rate of flurbiprofen was increased 5.8 to 17.5 times with theaddition of unsaturated fatty acids. Linolenic acid showed the most potent enhancingeffect, followed by oleic, palmitoleic, linoleic, and arachidonic acid. Fang et al.25 evalu-ated the efficacy of unsaturated fatty acids for the enhancement of flurbiprofen perme-ation through mouse skin. Unsaturated fatty acids showed greatest enhancement offlurbiprofen permeation than other classes of enhancers like terpenes or Azone. Theflux of flurbiprofen increased with the increase in the number of double bonds of theunsaturated fatty acid (linolenic acid> linoleic acid>oleic acid). However, oleic acidproduced higher skin retention of flurbiprofen than the other unsaturated fatty acids.

Gwak et al.32 investigated the effects of different penetration enhancers on the in vitropermeation of ondansetron hydrochloride across hairless mouse skin. The greatest flux




was attained by unsaturated fatty acids; the enhancement factors with the addition ofoleic acid or linoleic acid (3% w/w) to propylene glycol were about 1250 and 450,respectively. Saturated fatty acids failed to show a significant enhancing effect. In anotherstudy,33 utilizing the same study design, the enhancement factors for tenoxicam with theaddition of oleic acid or linoleic acid to propylene glycol were 348 and 238, respectively,whereas the saturated fatty acids (lauric, capric, and caprylic acids) had no effect on theskin permeation of tenoxicam. Oleic acid has been shown in numerous studies to beeffective skin permeation enhancer,34–37 while stearic acid usually has not had skinpermeation-enhancing effects.27

The Number and Position of Double Bondsin Unsaturated Fatty Acids

As the number of double bonds increased from one (oleic acid) to two (linoleic acid),there was a substantial increase in the flux of naloxone. However, an increase in thenumber of double bonds to three (linolenic acid) did not increase the flux further.38

Golden et al.39 evaluated the effects of position and configuration of unsaturated (18:1)fatty acids using porcine SC and vehicle containing 0.15 M fatty acid in ethanol. The cisisomers were effective permeation enhancers, whereas the corresponding trans isomershad less or no enhancing effect.

The effect of number of double bonds (in cis-conformation) in straight chain poly-unsaturated acids on the permeation of para amino benzoic acid in human SC wasstudied.30 Compared to monounsaturated fatty acid (with only one double bond),polyunsaturated fatty acids — linoleic, linolenic, and arachidonic acid with, respectively,two, three, and four double bonds produced a significantly higher increase in thepermeation of para amino benzoic acid. However, there was no significant differencein effects among the polyunsaturated fatty acids. Carelli et al.40 also reported that theenhancement of flux of alprazolam by linoleic acid (two double bonds) was greater thanthat of oleic acid (one double bond) through hairless mouse skin. In contrast, Morimotoet al.41 reported that the flux of indomethacin was unaffected by the number of doublebonds.

An earlier investigation showed that among unsaturated fatty acids, oleic acid is aneffective skin penetration enhancer for polar and nonpolar drugs.42 Cis-unsaturated fattyacids (viz., oleic acid, linoleic acid, and linolenic acid) have been reported to formseparate domains within SC lipids, which effectively decrease either diffusional pathlength or the resistance.43 The presence of double bonds in the structure has beenproposed to cause the formation of kinks in the lipid structure to allow water permeationacross the skin.42

Fatty Acid Esters

Medium chain aliphatic alcohols (C8–C12) and methyl or propyl esters of medium chainfatty acids (C8–C12) enhanced the permeation of 6-carboxyfluorescein (hydrophilic), andindomethacin (lipophilic) through rat skin. The enhancing effects of the aliphatic alco-hols for both drugs decreased with the increase in carbon chain length. In the case of fattyacid esters, the enhancing effects were lower than those of aliphatic alcohols and fattyacids. Although the relationships between the structure and skin permeation-enhancing




effect of the aliphatic alcohols and fatty acid esters used in this study are not yet fullyunderstood, they are possible candidates as permeation enhancers for hydrophilic andlipophilic drugs.19 Song et al.44 investigated the effects of oleic acid and of a group ofchemically related cis-(ricinoleic acid) and trans-(ricinelaidic acid) 12-monohydroxylatedderivatives and their corresponding ethyl and methyl esters on the skin permeation ofmodel hydrophobic (hydrocortisone) and hydrophilic (5-fluorouracil) drugs using ex-cised hairless mouse skin. Whereas the addition of oleic acid markedly enhanced thetransdermal flux of both drugs relative to propylene glycol alone (hydrocortisoneapproximately 1800-fold; 5-fluorouracil approximately 330-fold), that of a cis- or trans-12-monohydroxylated analog of oleic acid resulted in only a small increase (1.4 to2.7-fold for hydrocortisone; 4.4 to 6.6-fold for 5-fluorouracil). On the other hand,the methyl and ethyl esters of cis- and trans-12-hydroxy-9-octadecenoic acid exerteda much greater enhancing effect (327 to 720-fold for hydrocortisone, 42 to 74-fold for5-fluorouracil) than the corresponding parent fatty acids. Additionally, the esters oftrans-12-hydroxy-9-octadecenoic acid promoted permeation to an extent comparableto that achieved with their cis-counterparts.

The effects of a series of polyol fatty acid esters (sefsols) on diclofenac permeationthrough rat skin were investigated.45,46 Among four monoesters and one diester ofsefsol, all monoesters except the glyceryl monoester enhanced the percutaneous perme-ation of diclofenac. The highest enhancement was observed in propylene glycol mono-caprylate.

Chemical Modification of Drugs with Fatty Acids

Yamamoto et al.47 synthesized three novel lipophilic derivatives of phenylalanyl-glycine(Phe-Gly), C4-Phe-Gly, C6-Phe-Gly, and C8-Phe-Gly by chemical modification withbutyric acid (C4), caproic acid (C6), and octanoic acid (C8). The effect of the acylationon the stability, permeability, and accumulation of Phe-Gly in the rat skin was investi-gated. The stability (in skin homogenates) and permeability of Phe-Gly were improvedby chemical modification with fatty acids and this enhanced permeability of Phe-Gly bythe acylation may be attributed to the protection of Phe-Gly from the enzymatic degrad-ation in the skin and the increase in the partition of Phe-Gly to the SC. Of all the acyl-Phe-Gly derivatives, C6-Phe-Gly was the most permeable compounds across the intact skin.

Setoh et al.48 studied the in vitro permeability of chemically modified tetragastrin withfatty acids through the rat skin. The permeation of tetragastrin across the intact skin wasimproved by chemical modification with acetic acid and butyric acid. However, tetra-gastrin and caproyl-tetragastrin did not permeate across the intact skin. The stability oftetragastrin in skin homogenate was also significantly improved by chemical modificationwith fatty acids.

Various fatty acid ester derivatives of cycloserine were synthesized to improve skinpermeation of cycloserine.49 The skin permeation of cycloserine across the hairlessmouse skin was increased up to 20-fold by the fatty acid esters indicating their potentialuse in treatment of various skin infections.

Yahalom et al.50 prepared transdermally deliverable analogs of gonadotropin releas-ing hormone (GnRH) with various aliphatic acids (acetic, caproic, lauric, or stearic acid)to the amino side chain of [D-Lys]6GnRH, a superactive GnRH agonist. Nevertheless,analogs with 12-carbon or shorter aliphatic acids were shown to be GnRH superagonists,




with in vitro and in vivo potencies similar to that of [D-Lys]6GnRH. The transdermalpenetration of the peptides as evaluated by in vivo functional experiments in ratsis gradually lowered in increasingly hydrophobic analogs. The skin permeation ofthe peptide dramatically decreased by the attachment of a fatty acid, either due tosignificantly increased molecular weight or due to substantial conformational changes.These factors should be considered in the design of transdermally deliverable analogs ofsmall peptides.

Mechanism of Penetration Enhancement

The protection of the skin is provided primarily by the SC, which is only 10 to 20 mm thickand this provides the primary barrier to the percutaneous absorption of drugs as well as totransepidermal water loss. The viable epidermis is a stratified epithelium consistingof basal, spinous, and granular cell layers. The keratinocytes synthesize and expressnumerous different structural proteins and lipids during their maturation and transforminto chemically and physically resistant corneocytes. The corneocytes surrounded byextracellular nonpolar lipids constitute the SC barrier. Fatty alcohols and acids act aspermeation enhancers by interaction with the skin barrier by one or more of thefollowing mechanisms.

Lipid Disruption at the Intercellular Level: The EnhancerDisrupts Stratum Corneum Lipid Organization, MakingIt Permeable to Drugs

The major lipid classes in SC are ceramides, cholesterol, and free fatty acids. The acylchain length of ceramides and free fatty acids is between C22 and C26. Cholesterol ispresent in small amounts (typically 2 to 5% w/w of total lipids). The intercellular lipids arearranged in an ordered structure of multiple bilayers of polar and nonpolar regions.These ordered structures of lipid bilayers can be monitored by Differential scanningcalorimetry and Fourier transform infrared spectroscopy (FT-IR). Several studies havedemonstrated that fatty acids increase the fluidity of SC lipids by disrupting the packingorder of alkyl chains of lipids in SC.20,31,51

The effect of saturated and unsaturated fatty acids on the permeation of imipraminehydrochloride and amitriptyline hydrochloride was investigated using rat skin. Unsatur-ated fatty acids at 5% w/v in ethanol (oleic acid, linoleic acid, linolenic acid) inducedperturbation or increased bilayer fluidity of SC as observed by FT-IR.52 Oleic acid isreadily absorbed into SC and increases rotational movement of the hydrocarbon chainsand decreases the order of bilayer structure.53,54 Furthermore, oleic acid induced incom-plete lipid bilayer structures, prominent dilatation of lacunar domains, and the loss offollicular epidermal calcium gradient in the skin of rabbits.55 It was also proposed thatoleic acid disrupts the packed structure of the intercellular lipids because of the incorp-oration of its cis double bond.

However, the corresponding C18 saturated compound, stearic acid, even though is amajor component of SC is not efficient enough to increase the fluidity of ceramides.56

This explains why oleic acid is a much more effective permeation enhancer thanstearic acid. Further, the shorter chain saturated hydrocarbon fatty acids and branchedchain hydrocarbons disrupted lipid structures containing long-chain fatty acids. This is




consistent with the observations that C10 to C12 saturated fatty acids are more effective inenhancing the permeability than long-chain saturated fatty acids.57

Fatty Acids Form Solvated Complexes or Molar AdditionCompounds and Permeate through Skin Simultaneouslywith Drugs

The transdermal delivery of tamoxifen and linolenic acid from borage oil containing 25%linolenic acid was monitored using full thickness human skin including dermis.58 Initialuptake into SC would be a consequence of the miscibility of the vehicle and theintercellular lipids of the SC. The dermis is polar in nature relative to the lipids of theSC and is generally perceived as a barrier to the ingress of highly lipophilic molecules.From the relatively high amounts of both tamoxifen and linolenic acid that permeatedthe skin, it appears that the driving force of the vehicle (combination of liquid oil,infinite dose) was sufficient to overcome this barrier. From this, it appears that therate of permeation of linolenic acid and tamoxifen from the oil was linked to that ofsolute via a fixed solvation cage, rather than discrete molecules. When nonsteroidal anti-inflammatory drugs (NSAIDs, ibuprofen or ketoprofen) were applied to pig ear skin in afish oil vehicle, there were similar large fluxes of ibuprofen or ketoprofen concomitantwith eicosapentaenoic (EPA) and docosahexaenoic (DHA) acids.59 In the same study, itwas found that the rate of permeation of NSAID was linked to the rate of permeation ofEPA and DHA in the fish oil, suggesting permeation involved fixed NSAID and triacyl-glycerol complexes.59 The overall data from this study suggests that when a solutemolecule traverses the skin from a liquid (and perhaps semisolid) vehicle it does sostill whilst remaining at least partially associated with its solvation cage from the applieddose. Consequently, the permeation rates observed pertains not simply to the permeantof interest, but of the overall solvated complex.

The mechanism of the fatty acid enhancing effect on metaproterenol sulfate wasexamined by measuring 1H NMR spectra and the apparent partition coefficient.23 Thepossible involvement of carbonyl group of lauric acid and NH group of metaproterenolsulfate in the formation of ion-pair complex was demonstrated by FT-IR spectrometry. Inaddition, the partition coefficient of metaproterenol sulfate between n-octanol anddistilled water for the lauric acid 1:1 and 2:1 molar ratios to metaproterenol sulfate wassignificantly higher than that of metaproterenol alone. This result indicates higher lipo-philicity of the complex formed between metaproterenol sulfate and the fatty acid.A possible mechanism of drug-diffusion enhancement may be due to improved partition-ing of the drug through the SC when a complex is formed with the fatty acid. Theprobable mechanism of skin penetration enhancement of metaproterenol sulfate byfatty acids is the capability of the formation of a metaproterenol sulfate–fatty acidcomplex and partitioning with the lipophilic route of the SC.

Recently Stott et al.60 demonstrated the formation of 1:1 molar addition compounds ofpropranolol with lauric or capric acid from their binary mixtures (as determined by FTIR).The additional compounds are formed by the interaction between the carbonyl groupof the fatty acid and the amino group of the b-blocker, to form a salt. The oppositelycharged species of the salt have been shown to permeate the human epidermal mem-brane by an ion-pair mechanism. Green and Hadgraft61 reported similar findings suggest-ing the formation of ion-pairs between different beta-adrenoceptor blocking agents andlauric and oleic acids.




Fatty Acids Increase the Diffusivity and Partitioning of Drugsand Vehicles through Stratum Corneum

The permeation of drug into skin is influenced by (a) permeation of vehicle into SC and(b) affinity of drug to the vehicle. If the vehicle permeates skin readily and the drug hashigh affinity to the solvent, the permeation through skin can be increased by ‘‘solventdrag mechanism’’ (drug–vehicle permeating together).31 Propylene glycol is able topermeate better than mineral oil and the long-chain fatty acids further increased thepermeation of propylene glycol in porcine SC.62 Further, under the influence of propyl-ene glycol, while palmitic acid localized in the SC lipids, myristic acid was able topenetrate the deeper epidermal layers of the skin.63 This indicates a mutual increase inthe permeation of fatty acids and propylene glycol under the influence of each other andthis explains ‘‘the solvent drag mechanism’’ by a combination of fatty acid and propyleneglycol. The permeation rates of drugs and propylene glycol were correlated together forindomethacin64 and Molsidomine65,66 using vehicles containing fatty acids.

A ‘‘push–pull mechanism’’ of enhancement by fatty acids (combination of lauric acidand oleic acid) was proposed for steroidal hormones like estradiol and testosterone andnorethindrone.67 It is possible that fatty acids are mainly distributed to the SC because oftheir lipophilicity, and interact with the SC lipids causing a ‘‘pull effect’’ for the drug, andthe fatty acids that remain within the formulation increase the thermodynamic activity ofthe active agent within the formulation causing a ‘‘push effect.’’ Use of combination ofpenetration enhancers of the same chemical family resulted in sustained and controlledpercutaneous absorption of the drugs from an adhesive matrix formulation.

Formulation Considerations

Enhancer Configuration in TDS

Many studies in the past have assessed the permeation-enhancing activity of compoundsas a result of placing the pure enhancer or its solutions onto the skin surface. This maynot be relevant to the incorporation of an enhancer into a transdermal system and thatmust be designed such that both the active drug and enhancer are released into the skin.During the preformulation stage of transdermal product development, the concentrationof the enhancer, its chemical constitution, its compatibility with polymer, and othermatrix or gel components must be evaluated by the development scientist. Theenhancers to be incorporated and the final transdermal devices should possess theseattributes: (a) compatible with formulation ingredients and system components,(b) chemically and shelf stable in the system, (c) promote drug release from systemand should get released to act on skin to reduce its barrier function, (d) nonirritating,nonsensitizing, nonphototoxic, and pharmacologically inert, (e) produce rapid onset ofaction with high degree of potency.

Of the 35 transdermal patch products spanning 13 molecules, at least half of theproducts are formulated with cosolvents that possess enhancer activity. Optimizing thedesign of a TDS is more complex when the enhancer is formulated into the device,because of the possibilities of adverse interactions among enhancer, drug, and othersystem components.

The major attraction of fatty acids, fatty alcohols, or fatty acid esters is that some ofthese materials were classified as Generally Recognized As Safe (GRAS) by Food and




Drug Administration (FDA). Few examples of GRAS status compounds include oleic acid,oleyl alcohol, cetyl alcohol, stearyl alcohol, etc. Theratech, Inc. uses a combination ofglyceryl monooleate and lauryl lactate to enhance the diffusion of testosterone acrossnonscrotal skin in hypogonadal males in the Androderm patch. Discouragingly, many ofthe fatty acids and alcohols are potential skin irritants upon direct contact to the skin.Therefore, there is a critical need to develop a transdermal formulation that could controlthe delivery of the fatty acid or alcohol to the skin surface.

Fatty Alcohols and Acids in Transdermal Patches

Despite the knowledge that fatty acids are generally irritant, there are several patents

caused irritation when they are applied to human skin in undiluted form or as concen-trated solution as, for example oleic acid should not be condemned, a priori, becauseirritation of the skin and other toxic effects can be suppressed by controlled release.68

One of the requirements of transdermal patch is constant release and penetration afterlong storage points. Transdermal systems with a multiple layer design may containnonhomogeneous content of the enhancer. In some cases, the enhancer will migrate todifferent layers into entire matrix until equilibrium is established.

The technological aspects of fatty acids and alcohols in the transdermal formulationshave not been systematically investigated. There are not many papers dealing withincorporation of fatty acids and alcohols in the transdermal patches. A polyacrylate orpolyisbutylene adhesive patch containing saturated or unsaturated fatty acids (C6–C18)and estradiol or estradiol and progestin mixture as active drugs was described.69 Thispatent claims a nonirritating patch for 3-day delivery of hormones in hormone replace-ment therapy. A novel formulation of a monolithic transdermal device comprising acombination of fatty acids, or fatty alcohols, or both as penetration enhancers wasdescribed.67 This patent claims that a combination of oleic and lauric acid acts as themost adequate composition for many active agents. It was also claimed that a combin-ation of fatty acid(s) or fatty alcohol(s) or both with different chain lengths as penetrationenhancer provides controlled drug permeation rates at all application times.

Fatty alcohols were incorporated into monolithic adhesive matrix type patches con-taining captopril for enhanced transdermal delivery.70 Oleic acid and propionic acid wereincorporated into a transdermal patch formulation containing physostigmine.71,72 It wasshown that inclusion of oleic acid allowed the amount of physostigmine and the size ofthe transdermal patch to be substantially reduced, whilst maintaining effective deliveryrates. The formulation containing oleic acid was not irritant to guinea-pigs when appliedto the skin for 48 h.

Different fatty alcohols and fatty acids were incorporated at 5% w/w in monolithicdrug-in adhesive (Eudragit E 100) transdermal patches of melatonin.73 Decanol, myristylalcohol, and undecanoic acid showed significantly higher flux values through hairless ratskin (enhancement ratios 1.7, 1.5, and 1.6, respectively). Ketotifen transdermal deliverysystems were prepared using polyisobutylene, liquid paraffin, and lauric acid. Thepharmacokinetics of Ketotifen patch was determined by applying the skin patch to thedorsal skin of rabbits. The therapeutic plasma levels were maintained at a constant levelup to 30 h.74

The effects of coating thickness, type of adhesive (Duro-Tak 87–2196 and Duro-Tak87–2097), and type and concentration of enhancer (0 or 10% of either caprylic acid or




proposed for their use in transdermal drug delivery (Table 12.3). The compounds that

methyl laurate) on the mechanical properties of two acrylic pressure-sensitive adhesiveswere investigated. Coating thickness, concentration of enhancer, and type of adhesivewere inter-related and these parameters affected the adhesive peel strength and releaseliner peel strength of the patch.75

Fatty acids, fatty alcohols, and their esters have been frequently reported in the patent,literature of their potential use as penetration enhancers in topical gels,76 creams,77 andadhesive tape or patches78; and these formulations were claimed to be effective andnonirritating to skin of rats or humans.

Skin Irritation Potential

Fatty acids and alcohols are well known to cause skin irritation.79 Prior studies on a seriesof saturated and unsaturated fatty acids of different chain lengths under occlusive patchtest revealed that the saturated fatty acids of carbon chain length C8 to C12 and a C18dienoic unsaturated fatty acid (linoleic) were most irritant to human skin.80 In general,unsaturated fatty acids cause more skin irritation than saturated fatty acids.81,82 A recentstudy demonstrated that unsaturated fatty acids at an extremely low concentration(0.0015% in propylene glycol) induced the production of prostaglandin-E2.

25 Further-more, unsaturated fatty acids increased IL-1a and IL-8 mRNA levels in cultured epidermis(human skin equivalent) whereas saturated fatty acids were not effective.82 It was alsoshown that a low level of oleic acid (0.01 to 0.03%) is capable of elevating IL-la mRNAlevels in the living cell layers.83

Application of a 5% oleic acid or propylene glycol vehicle to the skin of six humansubjects for 6 h resulted in a minor irritation; severe irritation occurred with a 20% oleicacid or propylene glycol vehicle.84 A 5% oleic acid in 66.6% ethanol (gelled with HPMC)induced significant histopathological changes (collagen fiber swelling, inflammatory cellinfiltration, and sub-epidermal edema) in rat skin.85 An aqueous vehicle containing 10%oleic acid was applied to the skin of nude mice for 24 h under occlusion, and resulted inulcerative eruptions, hyperplasia and edema of the epidermis, and inflammation of thedermis.86 While 10% oleic acid was severely irritating to nude mice skin, 10% oleylalcohol induced no discernible change in the histological appearance of the skin.86

Occlusive topical application of oleic acid in mice led to creation of pores on the surfaceof corneocytes and also prominent effects on epidermal Longerhan cells leading toquestions on skin immunosuppression upon chronic exposures.87

Skin irritation and barrier disruption effects of fatty acids and alcohols werecharacterized by: (a) bioengineering methods such as TEWL measurements6,88 andlaser doppler velocimetry (LDV) and imaging,15,81,89 (b) biomarker expression measure-ments,79,82,83 (c) histopathological changes in the skin,85 and (d) visual scoring by Draizemethod.

It appears that both TEWL and skin permeation enhancement effect of fatty acids areassociated with the involvement of SC lipids. Recently, we compared the effect of fattyacids and alcohols on the enhancement of skin permeation of melatonin across rat skinin vitro and the enhancement of TEWL in rats, in vivo.6,15 Interestingly, there was acorrelation between the TEWL and the enhancement effect of different penetrationenhancers across rat skin. In the case of saturated fatty acids, as the carbon chain lengthincreased from C9 (nonanoic acid) to C11 (undecanoic acid), there was a substantialincrease in the TEWL. However, the TEWL decreased below the level of control (vehicle)as the chain length was increased to C14 (myristic acid). Similarly, in case of unsaturated




fatty acids, there was a positive correlation between TEWL and permeation enhancementacross rat skin. In contrast, a study on series of fatty alcohols showed that tridecanol andmyristyl alcohol showed lower permeation enhancement effect than other fatty alcohols(decanol, undecanol, and lauryl alcohol) but caused greater TEWL. Therefore, we cannotgenerally assume TEWL measurement to be a prediction tool for skin permeation of apenetrant.

Tanojo et al.90 reported that the increase in the TEWL by unsaturated fatty acids wasgreater than that of straight chain fatty acids having 6 to 12 carbon atoms in humanvolunteers. In another study, Tanojo et al.81 reported the effects of saturated fatty acids(having 6 to 12 carbon atoms) and unsaturated fatty acids (oleic, linoleic, linolenic, andarachidonic acids) on skin barrier function (as assessed by measuring TEWL) and irritantskin response using LDV in combination with visual scoring. Saturated fatty acids onlycaused a slight irritation and increase in TEWL, whereas unsaturated fatty acids caused asignificant increase in TEWL and LDV (irritation) responses. Similar results were observedby Boelsma et al.82 on the irritancy potential of a series of saturated and unsaturated fattyacids under short-term exposure conditions using TEWL, LDV as parameters. The unsat-urated fatty acids increased both TEWL and LDV whereas the saturated fatty acids werenot very effective.

Tanojo et al.89 observed a good correlation between the permeation rate of hexylnicotinate (a model permeant) and increase in the LDV values by oleic acid, indicatingthat LDV can be used to elucidate the effects of enhancers on the skin. Conversely,Aramaki et al.91 reported that TEWL measurement is a more suitable test, which coulddifferentiate mild skin irritation, whereas LDV measurement is more appropriate toevaluate pronounced skin reaction of the enhancers.

We reported recently that magnetic resonance imaging (MRI) could be used toinvestigate the alterations in the skin ultrasturcture after topical exposure to jet fuels.Jet fuels (JP-8 and JP-8þ100) were applied by both occlusive and unocclusive methods.The skin of treated and control (untreated) sites was excised and analyzed by MRI.Exposure to JP-8 showed the largest difference from the control with regard to visualobservations of the SC and hair follicles, while JP-8þ100 appeared to affect the hairfollicle region.92 Therefore, we believe MRI can be used as an effective tool to investigatethe alterations in the skin morphology after exposure to skin penetration enhancers likefatty acids and alcohols.

Aungst93 stated three approaches to separate the skin irritation of fatty acids from theirpermeation-enhancing effects. (a) To control the concentration and delivery of fattyacids; (b) selection of less irritating fatty acids, for example, myristic acid; (c) inclusionof other ingredients in the vehicle to overcome the skin irritation induced by fatty acids,for example, glycerin, vitamin E, and squalene. As described in the previous section,there are several patents utilizing fatty acids as skin penetration enhancers in the trans-dermal formulations and these were claimed to be effective and nonirritating to skin.

Summary

Fatty acids and alcohols are broadly effective in enhancing the skin permeation of severalclasses of drugs including peptides. Saturated fatty acids and alcohols of medium chainlength (in particular, capric, lauric, myristic) and unsaturated fatty acids or alcohols (oleic,linoleic, and linolenic) are frequently reported to be more effective. Fatty acid esters andfatty acid ester prodrugs also have been frequently reported for enhancing transdermal




delivery of drugs. These compounds mainly act by disrupting packed structures ofintercellular lipids of SC. Formation of solvated complexes and ion-pairing of penetrantswith drugs also have been proposed as other mechanisms of permeation enhancement.Discouragingly, unsaturated fatty acids are more skin irritants and induce inflammatorymediators in the skin, in addition to their permeation enhancement on drugs. The effectsof these enhancers are highly dependent on the vehicle, with propylene glycol generallyproviding maximum permeation enhancement. A major challenge is to successfullyincorporate these compounds in the transdermal formulations for permeation enhance-ment of drugs and avoiding skin irritation.

References1. Williams, A.C. and Barry, B.W. Penetration enhancers, Adv. Drug Deliv. Rev., 56, 603, 2004.2. Brain, K. and Walters, K. Molecular modeling of skin permeation enhancement by chemical

agents, in Pharmaceutical Skin Penetration Enhancement, Walters, K. and Hadgraft, J., Eds,vol. 59, Drugs and Pharmaceutical Sciences, Marcel Dekker, Inc., New York, 1993, Chapter18, p. 389.

3. Warner, K.S., Li, S.K., and Higuchi, W.I. Influences of alkyl group chain length and polar headgroup on chemical skin permeation enhancement, J. Pharm. Sci., 90, 1143, 2001.

4. Kim, Y. et al. Short chain alkanols as transport enhancers for lipophilic and polar/ionicpermeants in hairless mouse skin: mechanism(s) of action, Int. J. Pharm., 80, 17, 1992.

5. Yoneto, K. et al. Mechanistic studies of the 1-alkyl-2-pyrrolidones as skin permeation en-hancers, J. Pharm. Sci., 84, 312, 1995.

6. Kandimalla, K.K., Kanikkannan, N., and Andega, S. Effect of fatty acids on the permeationof melatonin across rat and pig skin in-vitro and on the transepidermal water loss in ratsin-vivo, J. Pharm. Pharmacol., 51, 783, 1999.

7. Andega, S., Kanikkannan, N., and Singh, M. Comparison of the effect of fatty alcohols on thepermeation of melatonin between porcine and human skin, J. Control. Release, 77, 17, 2001.

8. Aungst, B.J., Rogers, N.J., and Shefter, E. Enhancement of naloxone penetration throughhuman skin in vitro using fatty acids, fatty alcohols, surfactants, sulfoxides and amides,Int. J. Pharm., 33, 225, 1986.

9. Kim, M.K., Lee, C.H., and Kim, D.D. Skin permeation of testosterone and its ester derivativesin rats, J. Pharm. Pharmacol., 52, 369, 2000.

10. Altenburger, R., Rohr, U.D., and Kissel, T. Rate control in transdermal beta-estradiol reservoirmembrane systems: the role of membrane and adhesive layer, Pharm. Res., 8, 1238, 1998.

11. Pillai, O., Nair, V., and Panchagnula, R. Transdermal iontophoresis of insulin: IV. Influence ofchemical enhancers, Int. J. Pharm., 269, 109, 2004.

12. Friend, D. et al. Transdermal delivery of levonorgestrel II: effect of prodrug structure on skinpermeability in vitro, J. Control. Release, 7, 243, 1988.

13. Goosen, C. et al. Percutaneous delivery of thalidomide and its N-alkyl analogs, Pharm. Res.19, 434, 2002.

14. Sloan, K.B. et al. Transdermal delivery of theophylline from alcohol vehicles, Int. J. Pharm.,171, 185. 1998.

15. Kanikkannan, N. and Singh, M. Skin permeation enhancement effect and skin irritation ofsaturated fatty alcohols, Int. J. Pharm., 248, 219, 2002.

16. Lee, C.K. et al. Effect of hydrophilic and lipophilic vehicles on skin permeation of tegafur,alclofenac and ibuprofen with or without permeation enhancers, J. Pharm. Sci., 82, 1155,1993.

17. Gorukanti, S.R, Li, L., and Kim, K.H. Transdermal delivery of antiparkinsonian agent, benz-tropine. I. Effect of vehicles on skin permeation, Int. J. Pharm., 192, 159, 1999.




18. Kai, T. et al. Mechanism of percutaneous penetration enhancement: effect of n-alkanols onthe permeability barrier of hairless mouse skin, J. Control. Release, 12, 103, 1990.

19. Seki, T. and Morimoto, K. Enhancing effects of medium chain aliphatic alcohols and esters onthe permeation of 6-carboxyfluorescein and indomethacin through rat skin, Drug Deliv., 10,289, 2003.

20. Chantasart, D. et al. Mechanistic studies of branched-chain alkanols as skin permeationenhancers, J. Pharm. Sci., 93, 762, 2004.

21. Ackermann, C. et al. Ether–water partitioning and permeability through nude mouseskin in vitro. II. Hydrocortisone 21-n-alkyl esters, alkanols and hydrophilic compounds,Int. J. Pharm., 36, 67, 1987.

22. Ogiso, T. and Shintani, M. Mechanism for the enhancement effect of fatty acids on thepercutaneous absorption of propranolol, J. Pharm. Sci., 79, 1065, 1990.

23. Elyan, B.M., Sidhom, M.B., and Plakogiannis, F.M. Evaluation of the effect of different fattyacids on the percutaneous absorption of metaproterenol sulphate, J. Pharm. Sci., 85, 101,1996.

24. Yu, B. et al. In vitro visualization and quantification of oleic acid induced changes intransdermal transport using two-photon fluorescence microscopy, J. Invest. Dermatol., 117,16, 2001.

25. Fang, J.Y., Hwang, T.L., and Leu Y.L. Effect of enhancers and retarders on percutaneousabsorption of flurbiprofen from hydrogels, Int. J. Pharm., 50, 313, 2003.

26. Hsu, L.R. et al. Percutaneous absorption of piroxicam from FAPG base through rat skin:Effects of oleic acid and saturated fatty acid added to FAPG base, Drug Dev. Ind. Pharm., 20,142–59, 1994.

27. Oh, H.J. et al. Effects of vehicles and enhancers on transdermal delivery of melatonin,Int. J. Pharm., 212, 63, 2001.

28. Chi, S., Park., E., and Kim, H. Effect of penetration enhancers on flurbiprofen permeationthrough rat skin, Int. J. Pharm., 126, 267, 1996.

29. Chisty, M.N. et al. In vitro evaluation of the release of albuterol sulfate from polymer gels:effect of fatty acids on drug transport across biological membranes, Drug Dev. Ind. Pharm.,28, 1221, 2002.

30. Tanojo, H. et al. In vitro human skin barrier modulation by fatty acids: skin permeation andthermal analysis studies, Pharm. Res., 14, 42, 1997.

31. Aungst, B.J. et al. Transdermal oxymophone formulaton development and methods forevaluating flux and lag times for two skin permeation-enhancing vehicles, J. Pharm. Sci.,79, 1072, 1990.

32. Gwak, H.S., Oh, I.S., and Chun, I.K. Transdermal delivery of ondansetron hydrochloride:effects of vehicles and penetration enhancers, Drug Dev. Ind. Pharm., 30, 187, 2004.

33. Gwak, H.S., and Chun, I. K., Effect of vehicles and penetration enhancers on the in vitropercutaneous absorption of tenoxicam through hairless mouse skin, Int. J. Pharm., 236, 57,2002.

34. Dimas, D.A., Dallas, P.P., and Rekkas, D.M. Use of an 8(1)3(2) asymmetrical factorial designfor the in vitro evaluation of ondansetron permeation through human epidermis, Pharm.Dev. Technol., 9, 39, 2004.

35. Thomas, N.S. and Panchagnula, R. Combination strategies to enhance transdermal perme-ation of zidovudine (AZT), Pharmazie, 58, 895, 2003.

36. Valjakka-Koskela, R. et al. Transdermal delivery of levosimendan, Eur. J. Pharm. Sci., 11, 343,2000.

37. Giannakou, S.A. et al. Development and in vitro evaluation of nimodipine transdermalformulations using factorial design, Pharm. Dev. Technol., 3, 517, 1998.

38. Aungst, B.J., Rogers, N.J., and Shefter, E. Enhancement of naloxone penetration throughhuman skin in vitro using fatty acids, fatty alcohols, surfactants, sulfoxides, and amides,Int. J. Pharm., 33, 225, 1986.




39. Golden, G.M., McKie, J.E., and Potts, R.O. Role of stratum corneum lipid fluidity in transder-mal drug flux, J. Pharm. Sci., 76, 25, 1987.

40. Carelli, V. et al. Enhancement effects in the permeation of Alprazolam through hairless mouseskin, Int. J. Pharm., 88, 89, 1992.

41. Morimoto, K. et al. Enhancing effects of unsaturated fatty acids with various structures on thepermeation of indomethacin through rat skin, J. Pharm. Pharmacol., 48, 1133, 1996.

42. Barry, B.W. Mode of action of penetration enhancers in human skin, J. Control. Release, 6, 85,1987.

43. Ongpipattanakul, B. et al. Evidence that oleic acid exists in a separate phasewithin SC lipids,Pharm. Res., 8, 350, 1991.

44. Song, J.F., Lau-Cam, C.A., and Kim, K.H. Monohydroxylation and esterification as determin-ants of the effects of cis- and trans-9-octadecenoic acids on the permeation of hydrocortisoneand 5-fluorouracil across hairless mouse skin in vitro, Int. J. Pharm., 212, 153, 2001.

45. Takahashi, K. et al. Effect of polyol fatty acid esters on diclofenac permeation through ratskin, Biol. Pharm. Bull., 19, 893, 1996.

46. Takahashi, K. et al. Characterization of the influence of polyol fatty acid esters on thepermeation of diclofenac through rat skin, J. Control. Release, 73, 351, 2001.

47. Yamamoto, A. et al. Enhanced transdermal delivery of phenylalanyl-glycine by chemicalmodification with various fatty acids, Int. J. Pharm., 250, 119, 2003.

48. Setoh, K. et al. Improvement of transdermal delivery of tetragastrin by lipophilic modificationwith fatty acids, J. Pharm. Pharmacol., 47, 808, 1995.

49. Thorsteinsson, T. et al. Cycloserine fatty acid derivatives as prodrugs: synthesis, degradationand in vitro skin permeability, Chem. Pharm. Bull., 50, 554, 2002.

50. Yahalom, D. et al. Synthesis and bioactivity of fatty acid-conjugated GnRH derivatives, LifeSci., 64, 1543, 1999.

51. Jasti, B.R. and Abraham, W. Fluorescence spectroscopic investigation of effect of excipientson epidermal barrier and transdermal systems, J. Invest. Dermatol. Symp. Proc., 3, 128, 1998.

52. Jain, A.K. and Panchagnula, R. Transdermal drug delivery of tricyclic antidepressants: effect offatty acids, Methods Find. Exp. Clin. Pharmacol., 25, 413, 2003.

53. Francocur, M.L., Golden, G.M., and Potts, R.O. Oleic acid: its effects on SC in relation to(trans)dermal drug delivery, Pharm. Res., 7, 621, 1990.

54. Gay, C.L. et al. An electron spin resonance study of skin penetration enhancers, Int. J. Pharm.,49, 39, 1989.

55. Choi, E.H., Ahn, S.K., and Lee S.H. The changes of stratum corneum interstices and calciumdistribution of follicular epithelium of experimentally induced comedones (EIC) by oleicacid. Exp. Dermatol., 6, 29, 1997.

56. Neubert, R. et al. Structure of stratum corneum lipids characterized by FT-Raman spectros-copy and DSC. II. Mixtures of ceramides and saturated fatty acids, Chem. Phys. Lipids, 89, 3,1997.

57. Aungst, B.J. Structure/effect studies of fatty acid isomers as skin penetration enhancers andskin irritants, Pharm. Res., 6, 244, 1989.

58. Karia, C. et al. Simultaneous permeation of tamoxifen and gamma linolenic acid acrossexcised human skin. Further evidence of the permeation of solvated complexes,Int. J. Pharm., 271, 305, 2004.

59. Heard, C.M. et al. The in vitro delivery of NSAIDs across skin was in proportion to thedelivery of essential fatty acids in the vehicle — evidence that solutes permeate skin associ-ated with their solvation cages? Int. J. Pharm., 261, 165, 2003.

60. Stott, P.W., Williams, A.C., and Barry, B.W. Mechanistic study into the enhanced transdermalpermeation of a model beta-blocker, propranolol, by fatty acids: a melting point depressioneffect, Int. J. Pharm., 219, 161, 2001.

61. Green, P.G. and Hadgraft, J. Facilitated transfer of cationic drugs across a lipoidal membraneby oleic acid and lauric acid. Int. J. Pharm., 37, 251, 1987.




62. Wang, M.Y., Yang, Y.Y., and Heng, P.W. Role of solvent in interactions between fatty acids-based formulations and lipids in porcine stratum corneum, J. Control. Release, 94, 207, 2004.

63. Cotte, M. et al. Synchrotron FT-IR microscopic study of chemical enhancers in transdermaldrug delivery: example of fatty acids, J. Control. Release, 97, 269, 2004.

64. Nomura, H. et al. Percutaneous absorption of indomethacin from mixtures of fatty alcoholand propylene glycol (FAPG bases) through rat skin: effects of oleic acid added to FAPG base,Chem. Pharm. Bull., 38, 1421, 1990.

65. Yamada, M. and Uda, Y. Enhancement of percutaneous absorption of molsidomine, Chem.Pharm. Bull., 35, 3390, 1987.

66. Yamada, M., Uda, Y., and Tanigawar, Y. Mechanism of enhancement of percutaneousabsorption of malsidomine by oleic acid, Chem. Pharm. Bull., 35, 3399, 1987.

67. Carrara, D. Composition for Controlled and Sustained Transdermal Administration, U.S.Patent, 6,231,885 B1, May 15, 2001.

68. Hille, T. Technological aspects of penetration enhancers in transdermal systems, in Pharma-ceutical Skin Penetration Enhancement, Walters, K. and Hadgraft, J., Eds, Vol. 59, Drugs andPharmaceutical Sciences, Marcel Dekker, Inc., New York, 1993, Chapter 15.

69. Gonella, J. Administration System for Estradiol, U.S. Patent, 5,665,377, September 9, 1997.70. Park, E.S. et al. Effects of adhesives and permeation enhancers on the skin permeation of

captopril, Drug Dev. Ind. Pharm., 27, 975, 2001.71. Jenner, J., Saleem, A., and Swanston, D. Transdermal delivery of physostigmine: a pretreat-

ment against organophosphate poisoning, J. Pharm. Pharmacol., 47, 206, 1995.72. Kochinke, F. and Baker, R.W. Device and Method for Enhanced Administration of Physos-

tigmine, U.S. Patent, 5,364,629, November 15, 1994.73. Kanikkannan, N. et al. Formulation and in vitro evaluation of transdermal patches of mela-

tonin, Drug Dev. Ind. Pharm., 30, 205, 2004.74. Chiang, C.H. et al. Therapeutic effect and pharmacokinetics of ketotifen transdermal delivery

system, Drug Dev. Ind. Pharm., 24, 213, 1998.75. Dimas, D.A. et al. Effect of several factors on the mechanical properties of pressure-sensitive

adhesives used in transdermal therapeutic systems. AAPS PharmSciTech, 1, E16, 2000.76. Betlach, C.J. Composition and Method for Transdermal Delivery of Diclofenac, Int. Patent,

PCT WO93/00873, January 21, 1993.77. Gruber, K. et al. Transdermal 2-(4-isobutylphenyl)-Propionic Acid Medication and Methods,

U.S. Patent, 4,555,524, November 26, 1985.78. Mehjour, M., Mauser, B., and Fawzi, F. Fatty Acids and their Small Chain Esters as Penetration

Enhancers in Aqueous Systems, U.S. Patent, 4,879,297, November 7, 1989.79. Kozema, H. et al. Evaluation of skin irritation in a reconstituted human dermal model (3-D

model) using water insoluble fatty acids, fatty alcohols and hydrocarbons, AATEX, 5, 201,1998.

80. Stillman, M.A., Maibach, H.I., and Shalita, A.R. Relative irritancy of free fatty acids of differentchain length, Contact Dermatitis, 1, 65, 1975.

81. Tanojo, H. et al. In vivo human skin barrier modulation by topical application of fatty acids,Skin Pharmacol. Appl. Skin Physiol., 11, 87, 1998.

82. Boelsma, E. et al. An in vivo–in vitro study of the use of a human skin equivalent for irritancyscreening of fatty acids, Toxicol. In Vitro, 11, 365, 1997.

83. Boelsma, E. et al. Assessment of the potential irritancy of oleic acid on human skin: evaluationin vitro and in vivo, Toxicol. In Vitro, 10, 729, 1996.

84. Loftsson, T., Gildersleeve, N., and Bodor, N. The effect of vehicle additives on the transdermaldelivery of nitroglycerin, Pharm. Res., 4, 436, 1987.

85. Narishetty, S.T. and Panchagnula, R. Transdermal delivery of zidovudine: effect of terpenesand their mechanism of action, J. Control. Release, 95, 367, 2004.

86. Lashmar, U.T., Hadgraft, J., and Thomas, N. Topical application of penetration enhancers tothe skin of nude mice: a histopathological study, J. Pharm. Pharmacol., 41, 118, 1989.




87. Touitou, E. et al. Oleic acid, a skin penetration enhancer, affects Langerhans cells andcorneocytes, J. Control. Release, 80, 1, 2002.

88. Jiang, S.J. et al. Structural and functional effects of oleic acid and iontophoresis on hairlessmouse stratum corneum, J. Invest. Dermatol., 114, 64, 2000.

89. Tanojo, H. et al. In vivo human skin permeability enhancement by oleic acid: a laser Dopplervelocimetry study, J. Control. Release, 58, 97, 1999.

90. Tanojo, H., Junginger, H.E., and Bodde, H.A. In vivo human skin permeability enhance-ment by oleic acid: transepidermal water loss and fourier-transform infrared spectroscopystudies, J. Control. Release, 47, 31, 1997.

91. Aramaki, J. et al. Which bioengineering assay is appropriate for irritant patch testing withsodium lauryl sulfate? Contact Dermatitis, 45, 286, 2001.

92. Kanikkannan, N., Locke, B.R., and Singh, M. Effect of jet fuels on the skin morphology andirritation in hairless rats, Toxicology, 175, 35, 2002.

93. Aungst, B.J. Fatty acids as skin permeation enhancers, in Percutaneous Penetration En-hancers, Smith, E. and Maibach, H.I., Eds, CRC Press, New York, 1995, Chapter 9.1, p. 277.

94. Fang, J.Y. et al. In vitro and in vivo evaluations of the efficacy and safety of skin permeationenhancers using flurbiprofen as a model drug. Int. J. Pharm., 255, 153, 2003.

95. Nair, V.B. and Panchagnula, R. Effect of iontophoresis and fatty acids on permeation ofArginine Vasopressin through rat skin, Pharmacol. Res., 47, 563, 2003.

96. Escribano, E. et al. Assessment of diclofenac permeation with different formulations: anti-inflammatory study of a selected formula, Eur. J. Pharm. Sci., 19, 203, 2003.

97. Takahashi, K. and Rytting, J.H. Novel approach to improve permeation of ondansetron acrossshed snake skin as a model membrane, J. Pharm. Pharmacol., 53, 789, 2001.

98. Brown, L.R., Cine, J.F., and Davidson, J. Transdermal Administration of Amines with MinimalIrritation and High Transdermal Flux Rate, U.S. Patent, 5,422,118, June 6, 1995.

99. Foldwari, M., Kwadwo, S., and Poku, A. Composition and Method for Dermal and Transder-mal Administration of a Cytokine, U.S. Patent, 6,444,200 B2, September 3, 2002.

100. Wong, O. and Nguyen, T.N. Enhancement of Transdermal Drug Delivery Using MonoalkylPhosphates and other Absorption Promoters, U.S. Patent, 5,358,715, October 25, 1994.




Chapter 13

Essential Oils and Terpenes

Rashmi A. Thakur, Yiping Wang, and Bozena B. Michniak

CONTENTS

Introduction .................................................................................................................................... 159Structural Classification .................................................................................................................. 160Structure–Activity Relationships..................................................................................................... 162Improving Permeation Enhancement Ability of Terpenes ........................................................... 163Mechanism of Action of Terpenes ................................................................................................. 165Terpene Derivatives as Enhancers................................................................................................. 168Conclusions .................................................................................................................................... 171References....................................................................................................................................... 171

Introduction

One of the most common approaches to drug penetration enhancement is the useof chemical agents which modify the skin barrier properties. For such an enhancerto be acceptable, the following properties are desired1: It should be nontoxic, nonirritat-ing, exhibit no pharmacological property of its own, exert reversible effects on theskin, and be cosmetically acceptable. While a number of chemicals such as sulfoxides,alcohols, pyrrolidones, fatty acids, Azone have been studied, none of them have beenproven to be outstanding often due to safety concerns. Terpenes may offer advantagesover such enhancers because of their natural origin as well as Generally Regarded AsSafe (GRAS) status. They are carbon, hydrogen, and oxygen containing nonaromaticcompounds found in essential oils, extracted from flowers, fruits, and other naturalproducts. These compounds have been used for a long time as fragrances and flavoring

examples of essential oils which have various terpenes or terpenoids as their mainconstituents.


159

agents in commercial preparations (sweets, toothpastes, cigarettes). Table 13.1 provides


Structural Classification

Structurally, terpenes contain isoprene units (C5H8 units derived from methylbuta-1,3-diene) and exist as hydrocarbons, alcohols, ketones, and oxides. They can be classifiedas shown in Table 13.2.

Monoterpenes and some sesquiterpenes are the chief constituents of the essential oils

Oxygen containing terpenes are termed as terpenoids. All the terpenes can be subdivided

Acyclic monoterpenes can be considered derivatives of 2,6-dimethyloctane.Monocyclic monoterpenes are derivatives of cyclohexane with isopropyl substituents.Bicyclic monoterpenes contain the same number of carbon atoms but these are

arranged in more than one aromatic ring.Sesquiterpene structures found mainly in higher plants are present as several acyclic,

mono, bi, tri, and tetracyclic systems.

Table 13.1 Essential Oils and Terpenes

Terpenes Source

D-Limonene Orange or lemon oilTerpineol and acetyl terpineol Cardamom seeds1,8-Cineole EucalyptusAscaridole ChenopodiumAnethole AniseGeraniol and linalool esters Ylang ylangMenthone Peppermint oilLimonene oxide Lemon grassCarvone Spearmint oilNerol Neroli oilEugenol Clove oilOctahydro-1,8-dimethyl-7-(1-methylethenyl)-naphthalene,

b-Bisabolene, a-PanasinsenLow polarity fraction of

Alpinia oxyphyllaOctahydro-1,8-dimethyl-7-(1-methylethenyl)-naphthalene,

cis-a-Copaene-8-ol, NootkatoneHigh polarity fraction of

Alpinia oxyphylla

Table 13.2 Classification of Terpenes

Terpenes Isoprene units Carbon atoms

1 Monoterpenes 2 102 Sesquiterpenes 3 153 Diterpenes 4 204 Sesterpenes 5 255 Triterpenes 6 306 Carotenoids 8 40



and have been extensively studied as transdermal penetration enhancers (Figure 13.1).

into acyclic, monocyclic, and bicyclic (Figure 13.2).


D-limonene

1,8-cineole

pulegone menthol limonene oxide

menthone carvone

a-terpineol carveol

OH

OH

OO

O OH

O

O

Figure 13.1 Structures of commonly studied terpenes as chemical enhancers.

geraniol

camphor

nerol g -terpinene a -phellandrene

a -pinene

CH2OH

O OH

CH2OH

Acyclic monoterpenes

Bicyclic monoterpenes Sesquiterpenes

Monocyclic monoterpenes

Figure 13.2 Classification of terpenes.


Essential Oils and Terpenes & 161


Structure–Activity Relationships

Terpenes are highly lipophilic compounds with large partition coefficients betweenoctanol and water. The physicochemical properties of terpenes differ widely and this inturn affects their skin penetration enhancing ability. A series of terpenes as skin pene-tration enhancers have been assessed by Williams and Barry2 for transdermal delivery of5-flourouracil (5-FU), a model polar penetrant. Cyclic terpenes representing chemicalclasses of alcohols, ketones, oxides, and hydrocarbons were used as neat liquids onexcised human epidermal membranes. It was observed that the hydrocarbons such asa-pinene only doubled the permeability coefficient for the hydrophilic 5-FU whilealcohols and ketones resulted in a much higher increase (1,8-cineole caused a 95-foldincrease in permeation of the drug). Also the enhancement ratios of 1,2-oxygen bridgedterpenes (epoxides) were lower than longer oxygen bridged terpenes (cyclic ethers), forexample, 1,8-cineole. Ring conformation seemed to play a role in the enhancing activityof these compounds. A five-membered cyclopentene oxide (which is under considerabletorsional strain when planar and hence takes a slightly ‘‘puckered’’ form) exhibitedgreater enhancing activity if compared to six-membered cyclohexene oxide (which isrelatively flat in the chair conformation and free of angular and torsional strain). A bulkyside chain was reported to play an important role in the enhancing activity. Hydrocarbonchain-substituted terpenoids such as limonene oxide were better accelerants than non-substituted terpenoids.

On the other hand, the same terpenes when studied for delivery of lipophilic drugs,for example, estradiol3 and indomethacin4 showed that hydrocarbon terpenes (D-limo-nene) were the most effective enhancers. In the study with estradiol, cyclic ethers were aseffective as hydrocarbons providing approximately fourfold increase in the permeabilitycoefficient of aqueous estradiol. Alcohols, ketones, and epoxides proved to be ineffect-ive. In a separate study conducted by Hori et al.5 to study the effect of enhancerlipophilicity on transdermal absorption of hydrophilic propranolol hydrochloride andlipophilic diazepam, interesting results were obtained. A number of hydrocarbon andoxygen containing terpenes were used as enhancers. It was observed that while bothtypes of terpenes enhanced the delivery of propranolol hydrochloride equally, theanalogs with hydrogen bonding capacity (cyclic ethers) did not enhance the deliveryof lipophilic diazepam. In a separate study,6 similar results were seen when theenhancing activities of four terpene enhancers (fenchone logP 2.13+ 0.30, thymollogP 3.28+ 0.20, D-limonene logP 4.58+ 0.23, and nerolidol logP 5.36+ 0.38)formulated in hydroxypropyl cellulose gel formulations were studied for permeationof four compounds nicardipine hydrochloride (logP 0.99+ 0.1), hydrocortisone (logP1.43+ 0.47), carbamazepine (logP 2.67+ 0.38), and tamoxifen (log P 7.87+ 0.75). Itwas observed that hydrophilic terpenes like fenchone and thymol were less effectivethan lipophilic terpenes like limonene. The authors attributed the higher enhancementactivity of limonene relative to thymol and fenchone to its higher thermodynamic activityin the gel since limonene was not completely soluble in the gel at 2% concentration ascompared to thymol and fenchone (completely soluble in the gel at the evaluatedconcentration). Generally, hydrocarbon terpenes are effective for lipophilic drugs andoxygen containing terpenes are effective for hydrophilic drugs.

Lipophilicityof the terpenesmayalsoplay an important role in thepermeationenhance-ment process. El-Kattan et al.7 observed a linear relationship between the flux for polarsteroid hydrocortisone and 12 terpene enhancers (logP 1.06 – 5.36) studied. As logP ofterpenes increased, a linear decrease in lag time was observed. Nerolidol




(logP¼ 5.36+ 0.38) provided the greatest enhancement for HC flux (35.3-fold over con-trol) while fenchone (logP¼ 2.13+ 0.30) exhibited the lowest enhancement of HC flux.

Sesquiterpenes, which are relatively large molecules, have also been evaluated invarious studies as penetration enhancers. One of the initial studies in this respect wasperformed by Cornwell and Barry8 wherein 12 different sesquiterpenes were investigatedwith 5-FU as the model drug. It was reported that in accordance with results for mono-terpenes, enhancers with polar functional groups were more potent than pure hydrocar-bons. The best enhancer was nerolidol (15 carbons) which increased the flux of 5-FU 20-fold. Also an additional observation was that enhancers which were more branched werepoor penetration enhancers. Sesquiterpenes had a long duration of action where theeffects did not diminish markedly even after 36 h of skin treatment. This also illustratedpoor reversibility of the effects caused by this group of compounds which could possiblybe due to their size and high lipophilicity which decreases their clearance from thestratum corneum (SC).

Improving Permeation Enhancement Ability of Terpenes

A. Using isolated terpenes: Many studies have been conducted using essential oils aspermeation enhancers. Camphor and eucalyptus oil in 50% ethanol as vehicle wasreported to increase the flux of nicotine (lipophilic) through hairless mouse skin.9 Inanother study using prednisolone (logP value of 1.62), an acetone extract of cardamomseed which had active constituents a-terpineol and acetyl terpineol was more effectivein promoting drug permeation than laurocapram (azone).10 Using isolated terpenesinstead of mixed agents in essential oils is the most common example of increasingthe effectiveness of terpenes. A comparative evaluation of the enhancing activity ofpure 1,8-cineole and ascaridole vis-a-vis oil of eucalyptus (contains approximately 75%1,8-cineole) and oil of chenopodium (contains approximately 70% ascaridole) was con-ducted by Williams and Barry2 for 5-FU. It was observed that while oil of eucalyptusexhibited an enhancement ratio of 34.2, the pure 1,8-cineole had an enhancement ratioof 94.5 (enhancement ratio¼permeability coefficient after enhancer treatment/permea-bility coefficient before enhancer treatment). Similarly it was observed that oil of cheno-podium was less effective than the corresponding pure ascaridole. The authorsconcluded that these results were probably due to the fact that the active constituentsare not at maximum thermodynamic activities when present in oils.

B. Using different solvents: Use of solvents with enhancing abilities themselves inconjunction with terpenes has been widely investigated for improved permeation en-hancement. Yamane et al.11 studied the effects of propylene glycol (PG)/water co-solventsystems and terpene permeation enhancers on absorption of 5-FU. Co-application of allthe terpenes in the study (1,8-cineole, (þ)-limonene, menthone, and nerolidol), in PG co-solvent systems increased the drug flux. Terpene activity was highest in 80% PG systems.When compared to flux from pure vehicles, the enhancement ratios were as high as 24,21, 4, and 18 with 1,8-cineole, menthone, (þ)-limonene and nerolidol, respectively.A comparison of enhancement ratios showed fourfold increases in permeation for thedrug when saturated formulations of terpenes in PG were applied as compared toenhancement with neat terpene application. The authors concluded that PG acts byincreasing the partitioning of the terpene enhancers into the SC.

Another study was conducted by Vaddi et al.12 in vitro with two terpenes having thesame functional group: limonene oxide and pinene oxide and these were used at




concentrations of 5% w/v in 50% ethanol and 100% PG to enhance permeation ofhaloperidol through human skin. Both terpenes in 50% ethanol were able to providerequired therapeutic plasma concentrations of the drug implying it probably was a goodvehicle for maximum enhancement effect.

Godwin and Michniak13 investigated 11 monoterpenes (limonene, menthone, terpi-nen-4-ol, a-terpineol, 1,8-cineole, carvone, verbenone, fenchone, cymene, neomenthol,and geraniol) applied with PG on hairless mouse skin, which were investigated usingthree different model drugs (caffeine, hydrocortisone, triamcinolone acetonide) withvarying lipophilicities. The enhancement for the drugs was seen in the following order:caffeine enhanced maximum by neomenthol and geraniol>hydrocortisone enhancedmost by terpineols> triamcinolone acetonide, most active terpene for this test compoundbeing terpineol. It was concluded that combination of terpenes with PG can significantlyincrease the transdermal penetration of the hydrophilic drug caffeine and the polarsteroid hydrocortisone.

C. Varying the degree of saturation: Morimoto et al. studied the effects of vehiclescontaining different compositions of water, ethanol, L-menthol on in vitro permeation ofmorphine hydrochloride.14 L-Menthol was either present in the vehicle below its limit ofsolubility in the system (i.e., unsaturated) or the system was saturated with L-menthol. Itwas seen that despite similar pseudo steady-state fluxes (maximum fluxes observed), thelag time for permeation of the drug from saturated systems was shorter than that fromunsaturated systems. It was also seen that saturation of L-menthol was important inobtaining a significant enhancing effect. An excessive amount of L-menthol in theaqueous solvent formed an o/w emulsion and showed decreased enhancing effectimplying the degree of saturation of the terpene in the vehicle was an importantparameter for the enhancement effect.

D. Using melting point depression: Another important point to consider is the effect ofmelting point depression of a permeant on transdermal delivery. Pure enantiomers ofchiral compounds differ in melting points as compared to their racemic mixtures. Thelower the melting point of a substance, the greater its solubility in the given solventincluding the vehicle used in permeation experiments and the skin lipids (ideal solubilitytheory). Mackay et al.15 studied the effect of melting point of chiral penetration enhancerson their SC uptake. Two terpenes, menthone and neomenthone, were applied to the SC,saturated in PG/water. Racemic (+)menthol melts at 9 to 108C below melting point of itspure enantiomers. It was observed that (+)menthol had higher solubility in the vehiclethan its enantiomers. On the other hand, (+)neomenthol has a melting point 268C higherthan its enantiomers and has lower solubility than the enantiomers. Consequently, inboth the cases, the lower melting point form exhibited higher uptake into the SC.

Another novel approach which uses the melting point depression concept to increasetransdermal flux is formation of eutectic systems between the drug and enhancers. Stottet al.16 used eutectic systems comprising of ibuprofen as the drug and seven terpene skinenhancers including alcohols, hydrocarbons, and ketones to study the melting pointdepression of the delivery system. A range of ibuprofen:terpene binary mixtures weremelted together, cooled, and recrystallized. DSC and FT-IR analysis of these mixturesshowed that alcohols and 1,8-cineole formed eutectic mixtures while menthone,p-cymene, and limonene failed to do so. Ibuprofen in the condensed state exists in itsdimeric form. In the eutectic series with the presence of certain terpenes, the hydrogenbonding state of ibuprofen changes from the dimeric form to carbonyl (C¼O–HO)hydrogen bonded form. Terpenes capable of providing the hydrogen for this bonding(example: alcohols with –OH) can form eutectic mixtures. The resultant melting point




depression of the eutectic mixture then correlated with increases in transdermal flux. Itwas also observed that the maximum flux was obtained when the eutectic mixtureexisted as two-phase system (some excess solid ibuprofen in equilibrium with liquidibuprofen and terpene mixture) at the temperature of the permeation experiments. Thiswas shown to be due to the fact that in such eutectic mixtures solid ibuprofen achievesmaximum thermodynamic activity and maximum driving force for permeation. Forma-tion of eutectic mixtures not only affects the flux of ibuprofen but also the flux ofterpenes. As the proportion of terpene increases, the driving force for this componentrises and hence increased penetration enhancing ability is obtained as compared toenhancement achieved by pretreatment procedure alone. The melting point depressionalso affects the saturation of the drug and enhancer. The ibuprofen:thymol system had amelting point depression of 328C, same as the temperature of the experiment and henceboth the drug and enhancer were saturated in the system. Apart from melting pointdepression, eutectic systems also maintain the terpene delivery throughout the experi-ment as compared to pretreatment where terpenes can be washed out of the membraneover the course of the experiment. This strategy of using eutectic mixtures to enhancetransdermal delivery of experiments can also be used for other therapeutic categoriessuch as b-blockers (example: propranolol with fatty acid as enhancer) and ACE inhibitors(example: captopril with fatty acids).

Mechanism of Action of Terpenes

The lipid partitioning theory by Barry2 attempted to explain the modes of action ofpenetration enhancers. This proposed three main mechanisms of enhancer activity:(1) disruption of the highly ordered structure of SC lipids, (2) interaction with intracellularprotein, and (3) improvement in partitioning of a drug, co-enhancer or co-solvent intothe SC. In the same report which evaluated monoterpenes for permeation of 5-FU, acorrelation was seen between increases in the diffusion coefficient and penetrationenhancement ratios. This implied that terpenes act in part by at least modifying intercel-lular lipids and disrupting their highly ordered structure. Differential Scanning Calorim-etry (DSC), small-angle x-ray diffraction (SAXD), thermogravimetric analysis (TGA),infrared spectroscopy have been used to elucidate the mechanism of action of enhancers,DSC being the most popular of the methods listed.

DSC of a 20 to 40% hydrated sample of human SC gives rise to four main peaks

(1008C) associated with intracellular protein denaturation.17 While endotherm T1 is oflower importance, modifications in T2 and T3 by any enhancer imply a change in packingof the intercellular lipid domain. An enhancer which modifies T4 most likely interfereswith the structure and folding of the skin proteins. While initial studies18,19 conductedto understand the mechanism of terpenes failed to show any correlation between DSCresults and permeation enhancement, a detailed study by Cornwell et al.20 usingD-limonene, 1,8-cineole, and nerolidol with or without PG was able to elicit differencesin the mode of action of different terpenes. DSC, SAXD studies were conducted onenhancer treated and untreated skin. The enhancers were either applied as neat solutionsor as solutions in PG. D-Limonene treatment shifted T2 and T3 without affecting theenthalpies of the peaks. 1,8-Cineole, however, not only shifted T2 and T3 but also reducedthe enthalpy of T3 by 50%. Consequent calculations revealed that the combined entropyassociated with T2 and T3 was reduced by 1,8-cineole treatment. This is indicative of the



(Figure 13.3): T1 (368C), T2 (728C), T3 (838C) associated with lipid melting and T4


fact that 1,8-cineole is lipid disruptive at normal physiological temperatures whilelimonene is probably not. This is in accordance with the in vitro studies conductedfor 5-FU with the same enhancers2 wherein D-limonene (hydrocarbon) failed to en-hance drug permeation while 1,8-cineole (oxygen containing) showed a significantincrease in drug permeation. A similar study was also conducted by Yamane et al.21

to study permeation of the 5-FU in the presence of oleic acid in PG, D-limonene, 1,8-cineole, menthone, and nerolidol. The enhancement effects of D-limonene and oleicacid were saturable within 6 h, reaching a limiting value of about 3.6- and 24-foldincrease in drug flux, respectively, whereas 1,8-cineole, menthone, and nerolidolshowed increasing effects with time leading to maximum enhancements of about 95-,42- and 25-fold increase, respectively, after 12 h while DSC results were similar to thestudy by Cornwell et al.20

Further, Yamane et al.11 in their studies of PG/water co-solvent systems and terpenepermeation enhancers on absorption of 5-FU concluded that terpenes caused a greatershift in melting transition of T2 endotherm which corresponded to short alkyl chain lipidsthan T3 endotherm that corresponded to lipids bound to keratin.

To prove synergy between PG and terpenes, Cornwell et al. conducted DSC experi-ments20 on skin treated with PG alone, neat terpenes, and PG with terpenes. It was seenthat PG alone reduced the transition temperatures of T2 and T3 by 2 and 68C but had nosignificant effect on their enthalpies. Terpenes applied in combination with PG causedlipid endotherm shifts which were usually additive rather than synergistic. However, PGdid promote the reduction in enthalpies produced by 1,8-cineole implying PG synergywith 1,8-cineole may occur through enhanced lipid disruption at normal skin temperat-ures. The reasons for this synergy could not be explained. Enhancer uptake studies haveshown that PG does not influence the uptake of terpenes in anyway.

Fourier Transform InfraRed spectroscopy (FTIR) and transepidermal water loss(TEWL) was employed by Zhao and Singh22 to investigate the biophysical changes inthe SC lipids in presence of enhancers and PG. SC lipid extraction is reflected by a

Control

25.0 50.0 100.0 125.075.0Temperature (C)

Heatflow

(mW)

T1

T2 T3 T4

N

M

OA

D-L

1,8-C

Figure 13.3 DSC thermograms of 20–40% hydrated SC (control) and SC treated for 12 h with1,8-cineole (1,8-C), D-limonene (D-L), oleic acid 5% w/w in propylene glycol (OA), menthone(M), and nerolidol (N). (From Yamane, M.A., Williams, A.C., and Barry, B.W., Int J Pharm, 116,1995, 237–51. With permission.)




decrease in C–H stretching absorbance intensity. Measurement of TEWL is widely used tocharacterize the macroscopic changes in the barrier properties of the skin and is con-sidered to be a relevant parameter for the prediction of percutaneous absorption ofsubstances. While in vitro studies showed that 5% limonene in 50% PG was the bestenhancer for the drug tamoxifen followed by 5% eugenol in 50% PG, in general 5%terpenes/50% PG produced greater decreases in peak heights and areas for C–H stretch-ing absorbances in comparison with 50% PG and untreated SC, indicating additionaleffects of terpenes beyond the effects of 50% PG alone. The maximum percent decreasein C–H stretching absorbances was achieved by 5% limonene/50% PG in accordance within vitro studies. The decrease in peak heights and areas of C–H stretching absorbances isrelated to SC lipid extraction (i.e., physical removal of the SC lipids), Significant increasein the in vitro TEWL was observed for treatments of the epidermis with 5% eugenol/50%PG and 5% limonene/50% PG more so than treatment with 50% PG. The 5% limonene/50% PG treatment produced greater in vitro TEWL values than treatment with menthoneor eugenol/50% PG. The authors concluded that an increase in SC lipid extraction andmacroscopic barrier perturbation was the main mechanism of action of terpenes. Thesame study also concluded that lower terpene concentrations (1, 2, and 3%) induced skindamage that was reversible and did not cause major skin damage unlike the higherconcentration (e.g., 5%).

Cornwell and Barry23 had previously suggested formation of new polar pathwaysbased on increased electrical conductivity of human epidermis following terpene treat-ment. These conclusions have been supported by a recent study conducted by Narishettyand Panchagnula.24 The effect of various oxygen-containing monoterpenes such ascineole, menthol, a-terpineol, menthone, pulegone, and carvone was investigated onex vivo permeation studies of zidovudine across rat coupled with saturation solubility,partition coefficient, and molecular modeling approaches. All the terpenes studiedsignificantly increased transdermal flux of the drug in comparison to vehicle. Saturationsolubility and SC/vehicle partition coefficient of the drug were not significantly alteredby terpenes. Interactions between terpenes and SC lipids were studied with molecularmodeling and found that terpenes form hydrogen bonds (bond lengths < 2 A) with lipidhead groups. The SC lipids are organized as highly ordered to less ordered lamellarsheets25 held together by van der Waals, electrostatic, hydrophobic, and hydrogenbonding interactions.26 Lateral and transverse hydrogen bonding between head groupsof SC lipids within and opposite lamellae, respectively, is crucial for bilayer structure andintegrity. It was seen that cineole, menthone, pulegone, and carvone formed hydrogenbonds with amide group while alcohol terpenes formed hydrogen bonds with either acidcarbonyl or hydroxyl group of sphingosine molecule (which plays a major role in theskin lipid barrier function). Based on these results combined with the solubility andpartition coefficient studies, it is suggested that hydrogen bonds between terpenes andceramide head groups break interlamellar hydrogen bonding network of lipid bilayerand furthermore, as interlamellar hydrogen bonding network breaks, the distance be-tween two opposite lamellae increases and therefore new polar pathways or channels are

Cornwell and Barry8 studied mechanism of action of larger terpenes such as sesqui-terpenes. SC and water drug partitioning studies have suggested that this class ofcompounds acts by increasing the drug diffusivity into the SC. Recently, Fang et al.27

isolated sesquiterpenes from Alpinia oxyphylla and tested them as skin penetrationenhancers Oxygenated sesquiterpenes (high polarity fraction termedAO-2) enhanced the delivery of indomethacin significantly as compared to the low



formed thus disrupting the SC barrier (Figure 13.4).

13.1).(Table


polarity fraction AO-1 comprising of hydrocarbons. In the same study, release of inflam-matory mediator, prostaglandin E2, from human skin fibroblasts which occurs uponenhancer treatment was investigated to screen skin toxicity. Contrary to the resultspertaining to enhancing capabilities of both the compounds, AO-2 inhibited release ofPGE2 while AO-1 slightly increased the release. This indicated that skin inflammationcould not be used as a measure to understand enhancing activities.of these compounds.

Terpene Derivatives as Enhancers

A few terpene derivatives have been synthesized and evaluated as transdermal penetra-tion enhancers. Relationship of their chemical structures and enhancing effects as well astheir irritancy have been investigated in an attempt to design a potent enhancer ofminimum toxicity.

Compounds containing azacycloalkanone rings similar to Azone and acyclic terpenechains were evaluated by Okamoto et al.28 The structure–activity relationship in theenhancement of these 1-alkyl- or 1-alkenylazacycloalkanone derivatives was examinedusing 6-mercaptopurine as a model drug in a guinea pig skin pretreated with theenhancers. With respect to tail chain length, derivatives with monoterpene (C10) orsesquiterpene (C15) chains show greater enhancement effects than those with a longerchains (C20). The saturation of the alkenyl chain has no significant effect. Increasing thesize of azacycloalkanone ring from five- to a seven-member ring has little effect on theenhancing activity, while the compound with two carbonyl groups on the polar head is

Figure 13.4 Effect of terpenes on SC lipid bilayer arrangement. In the absence of terpenes,lipids are held together in lamellae by lateral and transverse hydrogen bonding. Terpenes breaktransverse hydrogen bonding leading to widening of aqueous region near head groups therebyincreasing diffusivity of polar molecules. Note: Bond lengths and bond angles are not to bescaled. (From Narishetty, S.T. and Panchagnula, R., J Control Release, 95 (3), 2004, 367–79.With permission.)




less effective than the other enhancers with one carbonyl group. This study also revealedthat enhancers with trans double bonds in the tail chain have low irritancy in spite oftheir high enhancing activities. Okamoto et al.29 further investigated the effect of 1-alkyl-or 1-alkenylazacycloalkanone derivatives on the penetration of drugs with differentlipophilicities. Large penetration enhancement was observed for the drugs, such as5-FU and 6-mercaptopurine with octanol–water partition coefficient of approximatelyunity, in both aqueous and ethanolic vehicles. In another effort to enhance the deliveryof drugs with the aid of terpenes, 1-alkylazacycloalkan-2-one esters of indomethacinhave been synthesized as prodrugs and their flux across excised human skin wasstudied.30 While 1-methylazacycloalkan-2-one esters of indomethacin proved unstablein aqueous media, 1-ethylazacycloalkan-2-one esters were more stable. These werereadily hydrolyzed in vitro by porcine esterase and penetrated excised human skin betterthan the parent drug.

Thiomenthol derivatives were synthesized and their enhancing activity on the percu-taneous absorption of ketoprofen from hydrogels was evaluated in rats by Takanashiet al.31 Thiomenthol itself is a derivative of menthol containing a sulfur atom. Structures

containing ketoprofen and the enhancers were prepared for in vitro permeation studies.The strongest enhancing effect was observed with compound 13, which has a thioethylgroup at C3 position and methyl groups at the C1 and C4 positions, whereas compound3, 7, and 8 with the thioethyl group at the C3 position were less effective on the drugabsorption than compound 13. Comparing compounds 9 and 13, compound 9 does nothave the methyl group at the C1 position and its enhancing activity was much lower thancompound 13, indicating the methyl group at C1 position is important to the enhancingactivity. From a comparison of compounds 4 and 9, it was found that a bulky group at theC4 position (compound 4) led to a decrease in the enhancing activity. Mathematicmodeling of the enhancing activity showed that the steric energy values negativelyaffected the promoting activity of enhancers, that is, the more stable and smaller thesteric structure was, the stronger the enhancement action achieved. Irritancy of thesynthesized thiomenthol derivatives after 8-h application of ketoprofen hydrogels wasinvestigated in the same study. Unfortunately, irritancy of the enhancers was almostlinearly related to the enhancing activity.

L-Menthoxypropane-1,2-diol (MPD), also known as cooling agent 10, is a derivative ofL-menthol. It has a cooling effect without volatility or odor characteristics, and thus, iswidely used in cosmetics, toothpastes, and chewing gums. Enhancing activity of MPDwas compared with L-menthol using Yucatan micropig skin.32 Indomethacin, a lipophilicdrug, and antipyrine were used as model drugs. MPD (3%) increased indomethacinpermeation through full-thickness skin about three times over control, while 3%L-menthol increased the permeation about ten times. It was almost the same case withantipyrine. MPD (3%) increased antipyrine permeation three times, while 3% L-mentholincreased the permeation 11 times over control values. Effects of MPD and L-menthol onSC were studied by FTIR spectra and x-ray diffraction patterns. Spectra suggested thatL-menthol, but not MPD, disrupts the intercellular lipid structure of SC. From eitherenhancing activity or irritancy point of view, MPD is a moderate skin permeationenhancer compared to L-menthol.

O-Ethylmenthol (MET) is another derivative of L-menthol (Figure 13.5). Nakamuraet al.33 evaluated percutaneous absorption of ketoprofen from carboxyvinyl polymerhydrogels in rats in vitro and in vivo. The antiinflammatory action of ketoprofen hydro-gels was also evaluated with a rat paw edema test. All the studies showed high enhancing



of the thiomenthol derivatives are show in Figure 13.5. Carboxyvinyl polymer hydrogels


potency of MET compared to L-menthol. In the in vitro permeation study, at least 2%L-menthol was required to obtain the same activity as 0.25% MET. In order to obtainsignificant inhibitory action of ketoprofen on the rat paw edema, at least 1% L-mentholwas required, while only 0.25 to 0.5% MET could produce the same effect. Extension ofthe spaces between the SC cells was microscopically observed with 0.5 to 2% MET,whereas the change caused by L-menthol was relatively weak.

After achieving remarkable enhancing activity with MET, 1-O-ethyl-3-n-buthylcyclohexanol (OEBC) (Figure 13.5) was synthesized by the same research group.34 An in vivostudy in rats showed that the enhancing activity of OEBC was approximately two timeshigher than MET; however, the skin irritancy was almost same as MET. Morphologicalchanges of the SC surface were microscopically observed with 0 to 2% OEBC. An electronspin resonance study was performed to investigate the effect of OEBC on the intercellular

CH3

SCH2CH3

CH(CH3)2

CH3

SH

CH(CH3)2

CH3

OH

CH3H3C

CH3

OCH2CH3

CH3

CH3

OCH2CH3H3C

SCH2CH3

CH3

SCH2CH3

CH3

CH3

CH3

SCH3

CH(CH3)2

SCH2CH3

CH2CH2CH2CH3

a b c

Compound 8 Compound 9 Compound 13

Compound 7Compound 4Compound 3

Figure 13.5 The chemical structures of terpene derivatives compound 3,4,7,8,9,13-thio-menthol derivatives (a) L-menthol (b)O-Ethylmenthol (MET) (c) 1-O-ethyl-3-buthylcyclohexanol(OEBC).




lipid bilayer fluidity of the SC. As expected, the fluidity of the SC lipid increased as theaddition of OEBC.

In summary, no synthetic terpene derivative has shown both high enhancing potencyand low toxicity. Structure–activity and structure–irritancy studies with aid from math-ematical modeling have given some direction to reaching the final goal. Collection ofmore data is required to design a better functional database.

Conclusions

Terpenes represent one of the most widely studied classes of chemical skin penetrationenhancers. The different classes and the varying physicochemical properties among eachof the groupings make them a promising set of enhancers for drugs from a wide range oflipophilicities and other properties. Each class also differs in their mechanism of action.These attributes make it difficult to rationally select one particular terpene for a givendrug. In additon, those terpene derivatives which satisfy the criteria of good penetrationenhancing ability fall short in the field of safety. Probably a renewed classification of thisclass of enhancers depending on their mechanism of action as well as structures ratherthan structures only should be attempted. Nevertheless, they are promising candidatesbecause of their generally lower adverse side effects.

References1. Williams, A.C. and Barry, B.W., Penetration enhancers, Adv Drug Deliv Rev, 2004, pp. 603–18.2. Williams, A.C. and Barry, B.W., Terpenes and the Lipid–Protein-Partitioning Theory of Skin

Penetration Enhancement, 1991.3. Williams, A.C. and Barry, B.W., The enhancement index concept applied to terpene penetra-

tion enhancers for human skin and model lipophilic (oestradiol) and hydrophilic (5-fluorour-acil) drugs, Int J Pharm, 1991, pp. 157–68.

4. Okabe, H., Takayama, K., Ogura, A., and Nagai, T., Effect of limonene and related compoundson the percutaneous absorption of indomethacin, Drug Des Deliv, 1989, pp. 313–21.

5. Hori,M., Satoh,S.,Maibach,H. I., andGuy,R.H.,Enhancementofpropranololhydrochlorideanddiazepam skin absorption in vitro: effect of enhancer lipophilicity. J Pharm Sci 1991, pp. 32–35.

6. El-Kattan, A.F., Asbill, C.S., Kim, N., and Michniak, B. B., The effects of terpene enhancerson the percutaneous permeation of drugs with different lipophilicities, Int J Pharm, 2001,pp. 229–40.

7. El-Kattan, A.F., Asbill, C.S., and Michniak, B.B., The effect of terpene enhancer lipophilicity onthe percutaneous permeation of hydrocortisone formulated in HPMC gel systems, Int J Pharm,2000, pp. 179–89.

8. Cornwell, P.A. and Barry, B.W., Sesquiterpene components of volatile oils as skin penetrationenhancers for the hydrophilic permeant 5-fluorouracil, J Pharm Pharmacol, 1994, pp. 261–9.

9. Nuwayser, E. S., Gay, M.H., De Roo, D.J., and Blaskovich, P.D., Transdermal nicotine — an aidto smoking cessation, in Proceedings of 15th International Symposium on the ControlledRelease of Bioactive Materials, 1988, pp. 213–4.

10. Yamahara, J., Kashiwa, H., Kishi, K., and Fujimura, H., Dermal penetration enhancement bycrude drugs: in vitro skin permeation of prednisolone enhanced by active constituents incardamon seed. March 1989.

11. Yamane, M.A., Williams, A.C., and Barry, B.W., Terpene penetration enhancers in propyleneglycol/water co-solvent systems: effectiveness and mechanism of action, J Pharm Pharmacol,1995, pp. 978–89.




12. Vaddi, H.K., Ho, P. C., Chan, Y.W., and Chan, S.Y., Oxide terpenes as human skin penetrationenhancers of haloperidol from ethanol and propylene glycol and their modes of action onstratum corneum, Biol Pharm Bull, 2003, pp. 220–8.

13. Godwin, D.A. and Michniak, B.B., Influence of drug lipophilicity on terpenes as transdermalpenetration enhancers, Drug Dev Ind Pharm, 1999, pp. 905–15.

14. Morimoto, Y., Wada, Y., Seki, T., and Sugibayashi, K., In vitro skin permeation of morphinehydrochloride during the finite application of penetration-enhancing system containing water,ethanol and L-menthol, Biol Pharm Bull, 2002, pp. 134–6.

15. Mackay, K.M., Williams, A.C., and Barry, B.W., Effect of melting point of chiral terpenes onhuman stratum coreneum uptake, Int J Pharm, 2001, pp. 889–897.

16. Stott, P.W., Williams, A.C., and Barry, B.W., Transdermal delivery from eutectic systems:enhanced permeation of a model drug, ibuprofen, J Control Release, 1998, pp. 297–308.

17. Barry, B.W. and Williams, A.C., Terpenes as skin penetration enhancers, Pharmaceutical SkinPenetration Enhancement, Hadgraft, J. Marcel Dekker, New York, 1993, pp. 95–112.

18. Williams, A.C. and Barry, B.W., Permeation, FTIR and DSC investigations of terpene penetra-tion enhancers in human skin, J Pharm Pharmacol, 1989, p. 12P.

19. Williams, A.C. and Barry, B.W., Differential scanning calorimetry does not predict the activityof terpene penetration enhancers in human skin, J Pharm Pharmacol, 1990, p. 156P.

20. Cornwell, P.A., Barry, B.W., Bouwstra, J. A., and Gooris, G. S., Modes of action of terpenepenetration enhancers in human skin; differential scanning calorimetry, small-angle x-raydiffraction and enhancer uptake studies, Int J Pharm, 1996, pp. 9–26.

21. Yamane, M.A., Williams, A.C., and Barry, B.W., Effects of terpenes and oleic acid as skinpenetration enhancers towards 5-fluorouracil as assessed with time; permeation, partitioningand differential scanning calorimetry, Int J Pharm, 1995, pp. 237–51.

22. Zhao, K. and Singh, J., Mechanisms of percutaneous absorption of tamoxifen by terpenes:eugenol, D-limonene and menthone, J Control Release, 1998, pp. 253–60.

23. Cornwell, P.A. and Barry, B.W., The routes of penetration of ions and 5-fluorouracil acrosshuman skin and the mechanisms of action of terpene skin penetration enhancers, Int J Pharm,1993, pp. 189–94.

24. Narishetty, S.T. and Panchagnula, R.,Transdermal delivery of zidovudine: effect of terpenesand their mechanism of action, J Control Release, 2004, pp. 367–79.

25. Bouwstra, J.A., Thewalt, J., Gooris, G.S., and Kitson, N.,A model membrane approach to theepidermal permeability barrier: an x-ray diffraction study, Biochemistry, 1997, pp. 7717–25.

26. Moore, D.J. and Rerek, M.E., Insights into the molecular organization of lipids in the skinbarrier from infrared spectroscopy studies of stratum corneum lipid models, Acta DermVenereol Suppl (Stockh), 2000, pp. 16–22.

27. Fang, J.Y., Leu, Y.L., Hwang, T.L., Cheng, H.C., and Hung, C.F., Development of sesquiter-penes from Alpinia oxyphylla as novel skin permeation enhancers, Eur J Pharm Sci, 2003,pp. 253–62.

28. Okamoto, H., Hashida, M., and Sezaki, H., Structure–activity relationship of 1-alkyl- or1-alkenylazacycloalkanone derivatives as percutaneous penetration enhancers, J Pharm Sci,1988, pp. 418–24.

29. Okamoto, H., Hashida, M., and Sezaki, H., Effect of 1-alkyl- or 1-alkenylazacycloalkanonederivatives on the penetration of drugs with different lipophilicities through guinea pig skin,J Pharm Sci, 1991, pp. 39–45.

30. Bonina, F.P., Montenegro, L., De Capraris, P., Bousquet, E., and Tirendi, S., 1-Alkylazacycloal-kan-2-one esters as prodrugs of indomethacin for improved delivery through human skin, Int JPharm, 1991, pp. 21–9.

31. Takanashi, Y., Higashiyama, K., Komiya, H., Takayama, K., and Nagai, T., Thiomentholderivatives as novel percutaneous absorption enhancers, Drug Dev Ind Pharm, 1999,pp. 89–94.




32. Fujii, M., Takeda, Y., Yoshida, M., Utoguchi, N., Matsumoto, M., and Watanabe, Y., Comparisonof skin permeation enhancement by 3-l-menthoxypropane-1,2-diol and l-menthol: the perme-ation of indomethacin and antipyrine through Yucatan micropig skin and changes in infraredspectra and x-ray diffraction patterns of stratum corneum, Int J Pharm, 2003, pp. 217–23.

33. Nakamura, Y., Takayama, K., Higashiyama, K., Suzuki, T., and Nagai, T., Promoting effect ofO-ethylmenthol on the percutaneous absorption of ketoprofen, Int J Pharm, 1996, pp. 29–36.

34. Li, C.J., Higashiyama, K., Yoshimura, Y., Nagai, T., Takayama, K., and Obata, Y., Promotingmechanism of menthol derivative, 1-O-ethyl-3-buthylcyclohexanol, on the percutaneousabsorption of ketoprofen, Biol Pharm Bull, 2001, pp. 1044–8.




PHYSICAL METHODS

OF PENETRATION

ENHANCEMENT

III



Chapter 14

Iontophoresis: ClinicalApplications and FutureChallenges

Nada Abla, Aarti Naik, Richard H. Guy, and Yogeshvar N. Kalia

CONTENTS

Introduction .................................................................................................................................... 178Principles of Iontophoresis ............................................................................................................ 179

Mechanisms................................................................................................................................. 179Electromigration...................................................................................................................... 180Electroosmosis ........................................................................................................................ 181

Practical Considerations ............................................................................................................. 181Electrode Choice..................................................................................................................... 181Current .................................................................................................................................... 182Drug Concentration in the Donor.......................................................................................... 182pH............................................................................................................................................ 183

Exisiting Therapeutic Applications of Iontophoresis .................................................................... 184FDA Approved Applications ...................................................................................................... 184

Pilocarpine Delivery for the Diagnosis of Cystic Fibrosis..................................................... 184Tap Water Delivery for the Treatment of Hyperhidrosis ...................................................... 184Lidocaine Delivery .................................................................................................................. 185The GlucoWatch1 Biographer for Noninvasive Glucose Monitoring.................................. 187E-TRANS1 Fentanyl HCl (Ionsyse

ˆ) — Pending Approval from the FDA ........................... 189

Other Applications of Iontophoresis In Vivo ............................................................................ 190Physical Medicine ................................................................................................................... 191Dentistry and Other Oral Pathologies.................................................................................... 191Ophthalmology....................................................................................................................... 192Otorhinolaryngology .............................................................................................................. 192


177


Potential Candidates for Iontophoretic Delivery .......................................................................... 193Animal Models ............................................................................................................................ 193

Cardiovascular Agents ............................................................................................................ 193Dermal Applications ............................................................................................................... 193Opioids.................................................................................................................................... 200Antiinflammatory Agents ........................................................................................................ 201Miscellaneous Nonpeptidic Drugs ......................................................................................... 201Protein and Peptide Drugs ..................................................................................................... 201

Human Studies............................................................................................................................ 204Dermal Applications ............................................................................................................... 204Opioids.................................................................................................................................... 206Antiinflammatory Agents ........................................................................................................ 207Miscellaneous Nonpeptidic Drugs ......................................................................................... 207Protein and Peptide Drugs ..................................................................................................... 208

Conclusions .................................................................................................................................... 208Successes and Opportunities for the Future ............................................................................. 208Limitations................................................................................................................................... 208Remaining Challenges ................................................................................................................ 209

References....................................................................................................................................... 209

Introduction

The first detailed descriptions of the use of transdermal iontophoresis to deliver moleculesacross the skin date back to the early 20th century. Leduc showed, to dramatic effect, that apotential difference could be used to deliver strychnine and cyanide to rabbits (withobvious results) from the anodal and cathodal electrode compartments, respectively.1,2

In the 1930s and 1940s, iontophoresis was frequently used to deliver molecules across theskin, but the technique did not gain scientific prominence until several decades later.2 Theacceptance of transdermal delivery as a viable alternative administration route, togetherwith the need to extend the range of drugs amenable to this approach, spurred a revival ofscientific interest in iontophoresis during the 1980s. The constraints imposed by the skin’sbarrier function meant that passive transdermal delivery was limited to a few select (highlypotent) drugs with the appropriate physicochemical properties. This is illustrated by theconcise list of available passive systems: clonidine, estradiol, fentanyl, nicotine, nitrogly-cerin, scopolamine, testosterone, oxybutinin, and the combination products norelgestro-min or ethinyl estradiol and estradiol or norethindrone acetate.3 Numerous strategies havebeen developed in order to expand the range of drugs available for transdermal adminis-tration including the use of current application.

In addition to the significant advantages offered by passive transdermal delivery, forexample, avoidance of chemical and enzymatic degradation during gastrointestinal andhepatic first-pass transit, presence of a large and readily accessible surface area (1 to 2 m2)for delivery, as well as being an easy-to-use and noninvasive alternative to parenteraldelivery,3,4 iontophoresis has the additional benefits of allowing precise control over thedelivery rate and the input kinetics. Thus, it can be used to deliver drugs in very tightlycontrolled and individualized input regimens and is a relatively straightforward means ofenabling pulsatile delivery of therapeutic agents.3 Passive delivery is limited to neutraland essentially lipophilic molecules; in contrast, iontophoresis permits the passageof charged and polar molecules across the uppermost layer of the skin, the stratumcorneum, which is otherwise a very efficient barrier against the penetration of such




molecules. Furthermore, because it is a symmetrical process, that is, the electric field iscapable not only of driving molecules into the skin but also of extracting endogenoussubstances, it is suitable for therapeutic monitoring.

According to Tyle, many of the applications of iontophoresis prior to 1960 were basedon clinical impressions rather than on scientific data about the permeation rate of themolecules, and controlled clinical studies were rare.1 Furthermore, iontophoresis proto-cols were seldom optimized: inappropriate choice of electrodes and current densityfrequently led to avoidable adverse effects such as burns, which obviously (but unfairly)discredited the technique. Our knowledge of iontophoretic mechanisms and experimen-tal ‘‘know-how’’ have evolved over time, and generally more controlled studies areperformed today.

Despite the numerous promising results published, describing iontophoresis in vitroon different skin models, and in vivo either in animals or man, commercialized applica-tions are only now beginning to appear, and these recent developments are one of theprincipal subjects of this review. One reason for the late arrival of commercializedproducts is that progress in microelectronics and engineering processes was necessarybefore miniaturized and cost-effective delivery systems could be designed and manufac-tured. Therefore, the iontophoretic device can no longer be considered as a cumbersomemachine reserved for treatment when other techniques have failed (e.g., treatment ofhyperhidrosis), but as a portable, user-friendly device which has the potential to replacemore painful techniques (e.g., the GlucoWatch1 Biographer [Cygnus, Inc., RedwoodCity, CA] and the LidoSite1 lidocaine system [Vyteris, Inc., Fair Lawn, NJ]) and to deliverthe next generation of therapeutic drugs (e.g., peptides and proteins).

The purpose of this review is (i) to present clinical areas in which iontophoresis hasalready found its niche, and (ii) to take stock of interesting preliminary in vivo studies thatdescribe promising, though as yet unexploited, applications of iontophoresis.

Principles of Iontophoresis

Mechanisms

Iontophoresis consists of the application of a potential difference, generated by a powersource, between two electrodes, anode and cathode, which separately contact the skin

contained in the electrode compartment of the same charge. Iontophoresis is a symmet-rical process; hence, it also allows the extraction of analytes from the skin, which is thebasis for iontophoretic drug monitoring. Although the GlucoWatch1 Biographer (Cyg-nus, Inc., Redwood City, CA) described in the next section illustrates this concept, thisreview will essentially focus on iontophoretic drug delivery. During iontophoresis, themovement of a molecule across the skin can be attributed to three components: (en-hanced) passive diffusion, electromigration, and convective solvent flow, also calledelectroosmosis. Each phenomenon is independent. Therefore, the total flux of a mol-ecule during iontophoresis is usually described by the summation of fluxes due to thesethree processes (Nernst–Planck theory):5

JTOT þ JP þ JEM þ JEO (14:1)

where JTOT is the total flux, JP is the passive flux, JEM and JEO are the fluxes resultingfrom electromigration and electroosmosis, respectively. The role of passive diffusion in


Iontophoresis & 179


surface via an adequate formulation (Figure 14.1). The molecule to be delivered is usually

iontophoretic delivery is usually minor compared to the two other mechanisms, espe-cially in the case of charged and polar molecules. Passive transdermal delivery has beenextensively described elsewhere.4,6–13

Electromigration

When a potential difference is applied across a membrane, such as skin, dissolved ionsmigrate according to their charge: cations are driven from anode to cathode, while anionsmove in the opposite direction. This process (electromigration) may be describedby Faraday’s law: 3,5,14–17

JEM;D ¼ (ItD)=(AFzD) (14:2)

where JEM,D, tD, zD represent, respectively, the electromigration flux, transport number,and valence of the drug (D); I is the applied current; F is the Faraday constant; and A isthe contact area.

The inference from this equation is that the application of a constant current can beused to control the drug delivery rate, the other terms in the expression (tD,zD, A, F) beingfixed. A change in I is directly translated into a change in JEM,D. On the other hand, if aconstant voltage (V) is applied, then the current flow across the skin is given by Ohm’s law:

I ¼ V=R (14:3)

where R is the skin’s resistance. Since R is known to decrease upon the application of anelectric field across the skin, it follows that I must increase to maintain this proportion-ality. This results in a corresponding increase in JEM,D, but now in an uncontrolled fashionthat is subject to inter-skin variability (and thereby losing the principal advantage ofiontophoresis).

EM + EO EO

D+ X–D

EO

X

V

+ −

Systemiccirculation

Skin

D+ or D D−

EM – EO EM + EOE M – EO

X+ D–

Figure 14.1 Electromigration (EM) and electroosmosis (EO) phenomena and their respectivecontributions to the transport of charged and neutral molecules during iontophoresis underphysiological conditions. X/X1/X2 represents either an exogenous or an endogenous moleculeto be extracted, and D/D1/D2 is a therapeutic agent. Because the EO flow is directed from theanode to the cathode, it contributes to the transport of cations and opposes that of anions.




The transport number is a parameter that describes the fraction of the total chargecarried by each species:3,17,18

tD ¼ (cDzDuD).X

(ci zi ui) (14:4)

where cD, zD, and uD represent the concentration, valence, and mobility of the drug,respectively. The sum of the transport numbers of all ions (cations and anions) presentin the system equals unity. This equation is a useful predictive tool for estimating thetransport efficiency of a given species, but assumes that cD and uD within the skin matrixare equivalent to those in the aqueous formulation.17

According to Equation (14.4), highly mobile ions such as Naþ or Cl, which are oftenpresent at a high concentration in the anodal compartment (as will be discussed later)and are ubiquitous in vivo, will carry a large fraction of the current, thus competingstrongly with less mobile ions (such as higher molecular weight [MW] peptides), therebydecreasing their delivery. Theoretically, one alternative to increase the delivery efficiencyof such high MW drugs is to increase their concentration in the donor; however, inpractice, this is frequently not a viable option either due to poor solubility or on thegrounds of cost.3

Electroosmosis

Rein was among the first to report the presence of electroosmosis during transdermaliontophoresis,19 and this phenomenon was subsequently shown to be responsible for theanodal transport of water, as well as that of uncharged and ionic solutes across theskin.20–22

This highly complex process, often interpreted using nonequilibrium thermodynam-ics,23–26 can be conceptualized as follows. At physiological pH, the skin (having anisoelectric point of approximately 4 to 4.527) possesses a net negative charge. Therefore,it acts as an ion-exchange membrane, which is permselective to cations. As a conse-quence, under the influence of an electric field, a convective solvent flow is generated inthe anode to cathode direction. This electroosmotic flow contributes to the permeationof cations but opposes the movement of anions. Furthermore, neutral molecules (withreasonable MW, such as sugars) can also be transported from the anode into the body,and from the body to the cathode. The observation that uncharged molecules could alsobe delivered by iontophoresis has naturally extended its scope of application.23–26,28–31 Itis generally accepted that the relative contribution of electroosmosis to the permeation ofcations becomes more important with increasing MW. That is, while small molecules willbe mostly delivered by electromigration (e.g., for lidocaine, 90% of its flux from a 10 mMsolution in HEPES-buffered saline is due to electromigration32), larger cations, with lowermobilities, are primarily transported by electroosmosis.

Practical Considerations

Electrode Choice

Iontophoresis has often been performed with inert electrodes such as platinum. How-ever, such electrodes provoke the electrolysis of water, generating hydroxide ions at thecathode and protons at the anode.3,17,33 This is undesirable because it induces pH shifts


Iontophoresis & 181


that can impact upon delivery efficiency by affecting drug stability, changing the ioniza-tion state of the molecule to be delivered, and even reversing the electroosmotic flow.3 Inaddition to reducing delivery efficiency, the protons generated at the anode (and hy-droxide ions released at the cathode) can also cause chemical burns. Therefore, silver–silver chloride (Ag–AgCl) electrodes, which are reversible at low potential, chemicallystable and preclude pH changes, are usually the electrodes of choice.33

The reactions that occur at each electrode are:33

AgCl(s) þ e ! Ag(M) þ Cl(aq) (cathode)

Ag(M) þ Cl(aq) ! AgCl(s) þ e (anode)

The electronic current generated by the battery is converted to an ion flow at theelectrode–solution interface. The fraction of this ionic current carried by the drugmolecule will be determined by its mobility and concentration with respect to thoseof the other charge carriers in the system, which cannot always be excluded fromthe system. For instance, Ag–AgCl electrodes require chloride ions for the anodal elec-trochemistry.3,17,34 The most convenient source of chloride is the hydrochloride salt of thedrug; however, when this is not available at the requisite levels, an external source ofchloride ions must be provided at the anode. This is usually supplied in the formof sodium chloride (NaCl) to the formulation. As discussed above, sodium ions competevery effectively with other cations, including the drug to be delivered, and theirpresence in the formulation can significantly reduce delivery. Strategies to reducethis competition can involve physical (but not electrical) separation of the drug andelectrode compartment, such as that based on the well-known ‘‘salt-bridge’’ concept inelectrochemistry. With respect to patch design, this can be deployed in the form of acharge or size-selective membrane that separates the drug formulation from the electrodecompartment.

Current

According to Faraday’s law (Equation [14.2]), iontophoretic drug transport across the skinis dependent on the total current supplied. Therefore, in principle, delivery can beenhanced by increasing the magnitude of the applied current. Although the applicationof higher currents normally results in a proportional increase in electromigration, limitingcurrent densities above which there is no further increase in transport have been reportedin some studies.3 In terms of patient compliance and current tolerability, the upper limitfor the current density applied in vivo is considered to be approximately 0.5 mA/cm2.35

Although tingling and itching sensations as well as erythema (which resolves withoutsequelae) are frequent and well-tolerated side-effects, higher current densities can pro-voke pain and discomfort.35

Drug Concentration in the Donor

Equations (14.2) and (14.4), a priori, suggest that an increase in donor drug concentrationwill result in an increased transdermal flux, via an increase in tD. For example,the relationship between lidocaine concentration (1 to 40 mM) and flux is linear, in the

32 However, for certain molecules,




presence of background electrolyte (Figure 14.2).

the flux–concentration profiles reach a plateau above which further increases in concen-tration have only a limited or negligible effect on flux. For example, the iontophoreticfluxes of quinine and propranolol (which are more lipophilic than lidocaine) increasedonly by a factor of approximately 14 when donor concentration was raised by a factor of40 (1 to 40 mM), in the presence of background electrolyte.32 Similar behavior has beenobserved with peptides, such as nafarelin, the transport of which exhibits a nonlineardependence on concentration.36 These molecules are lipophilic cations, which arethought to interact with the negative charges of the skin; this neutralization of skin charge(and hence skin permselectivity) can result in a decrease, and even a reversal, of theelectroosmotic flow. The impact on drug delivery depends on the relative contributionof electroosmosis to the iontophoretic transport of the molecule in question.32,37 Otherfactors contributing to the nonlinearity of this relationship could arise in situations wherean increase in drug concentration in the formulation (i) is not mirrored by a proportionateincrease in membrane drug concentration, and (ii) provokes aggregation of drug withinthe skin. It is evident that the concentrations and mobilities of competing ions play animportant role in determining the flux–concentration profile.3 For instance, in the ab-sence of competing ions in the donor, iontophoretic fluxes of lidocaine, quinine, pro-pranolol,32 and hydromorphone hydrochloride38 were shown to be independent of thedrug concentration, suggesting that transport was dependent on mole fraction, and notthe absolute concentration, of drug in the formulation. In contrast, in vitro results withropinirole hydrochloride appeared to suggest that there was no relation between trans-port and mole fraction of drug.39 Clearly, further work is required in this area. Insummary, unless the presence of a buffer in the drug formulation is essential to ensurestability (and/or optimise pH), it should be employed sparingly to minimize the presenceof competing ions.

pH

As the two main mechanisms of iontophoretic drug transport are electromigration andelectroosmosis, drug permeation can be optimized by controlling the ionization stateof the drug and skin, through manipulation of the formulation pH. As we have seen,extreme pH values are to be avoided because they can cause chemical burns, provoke

0

500

1000

1500

2000

2500

0 50 100

Donor concentration (mM)

Flu

x at

ste

ady-

stat

e(n

mo

l.cm

-2.h

-1)

Figure 14.2 Relationship between donor concentration and steady-state flux of lidocaine.(Adapted from Marro, D., Kalia, Y.N., Delgado-Charro, M.B., and Guy, R.H., Pharm. Res., 18,1701, 2001.)


Iontophoresis & 183


drug instability, and change both drug and skin ionization state. Moreover, high concen-trations of Hþ or OH will also compete with the drug, thus reducing the deliveryefficiency. With respect to the skin, its charge should be such that it favors electroosmosisin the direction of drug movement. In the case of a cationic drug, it will usually be moreappropriate to keep the skin negatively charged, while in the case of anions, it maybe useful to use pH 5 or 6 to reduce the effect of electroosmosis. Thus, the choice offormulation pH should be based not only on the drug but also on the dominant transportmechanism (EM or EO) thereby avoiding, for example, enhancement of the degree ofdrug ionization at the expense of skin permselectivity for a molecule that is predomin-antly transported by EO. As an illustration, the transport of thyrotropin releasing hor-mone (TRH; pKa 6.2), at pH 8 with 98% uncharged peptide was twice that at pH 4 whereTRH is approximately 99% protonated.40 The optimization of electroosmosis is expectedto be more critical as the MW of the solute increases, where electromigration plays alesser role.

Exisiting Therapeutic Applications of Iontophoresis

FDA Approved Applications

Pilocarpine Delivery for the Diagnosis of Cystic Fibrosis

The diagnosis of cystic fibrosis by iontophoresis of pilocarpine was one of the earliestdescribed therapeutic applications of this technology.41 Cystic fibrosis is a hereditarysystemic disorder of the mucus-producing exocrine glands, which affects the pancreas,the bronchi, the intestine, and the liver. The perspiration contains an abnormal concen-tration of sodium and chloride ions,42 and the assay of the latter is the basis for thediagnostic test.43 The collection of sweat for diagnostic purposes is facilitated by thedelivery of pilocarpine, a cholinergic agent, which induces sweating. As a small (208 Da)positively charged molecule, pilocarpine is ideally suited to iontophoresis. This diagnos-tic test was first introduced by Gibson and Cooke,43 and received FDA approval in 1983.It has since become a standard screening test44 commonly used by pediatricians,45

although cases of false-positive and false-negative results have been documented. Asthe gene for the cystic fibrosis transmembrane conductance regulator (CFTR) has beenrecently identified, genotyping of CFTR mutations is now used to confirm this diagnosisand to perform antenatal and newborn diagnoses, populations for which the sweat test isnot recommended.46

Tap Water Delivery for the Treatment of Hyperhidrosis

Iontophoresis of tap water is an effective therapy for palmoplantar idiopathic hyperhi-drosis. This application was first described in 1936 and has been used by physiotherapistsand dermatologists since the 1950s.47,48 It consists of placing the region to be treated intap water and applying, during approximately 30 min (depending on the protocol), acurrent of up to 15 to 20 mA per palm or sole49 and chosen such that there is nodiscomfort45. This treatment procedure is repeated (for example, once a day,48 orseven times over 28 days,50 or once or twice a week49) until sweating is significantlyreduced. An anticholinergic substance can also be added to the water to accelerate thetreatment and to extend the period of relief 51,52 although this can be associated with




anticholinergic side-effects.45 The efficacy of tap water iontophoresis has been demon-strated in several clinical studies,48,50,53–55 some with home-use iontophoretic devices48,53

(Figure 14.3). Undesirable effects such as erythema, local burning, and vesicular forma-tion were minimal.50 Although no long-term side-effects have been described,48 onedisadvantage is the need for long-term maintenance treatment to avoid relapse,49,56 areason why this technique is sometimes perceived by the patient to be time-consumingand inefficient.57

The mechanism underlying this therapy remains unclear. It has been hypothesizedthat iontophoresis induces the blockage of neuroglandular transmission or the inhibitionof the secretory mechanism at the cellular level, rather than a mechanical obstruction ofthe eccrine ducts or a structural degeneration of acini in sweat glands.48 According to Satoet al.,58 the mechanism may be dependent on the acidification of sweat glands. There-fore, this is one of the few applications of iontophoresis where platinum electrodes arepreferred to Ag–AgCl electrodes, because of their capacity to electrolyse water, and hencedecrease the anodal pH.58 This is also consistent with the observation that treatment ismore efficacious at the anodal electrode. Interestingly, in a clinical study involving 112patients, in 65 of the 91 patients responding to the treatment of palmar hyperhidrosis,plantar hyperhidrosis also resolved simultaneously, suggesting that a biofeedback mech-anism could be involved.50

Thus, tap water iontophoresis remains a method of choice for the treatment ofpalmoplantar hyperhidrosis.51 In the case of axillary hyperhidrosis, other methods areusually preferred, such as the application of aluminium chloride salts.48,51

Lidocaine Delivery

Lidocaine is a local anesthetic, primarily delivered by injection, which is fast, effective,long-lasting, but painful. Therefore, topical lidocaine delivery for the noninvasive induc-tion of local anesthesia has been investigated, leading to the formulation of the widelyused EMLA cream (eutectic mixture of local anesthetics: lidocaine and prilocaine).However, this topical formulation is slow to take effect (1 to 2 h) and achieves amaximum depth of anesthesia of only 3 to 5 mm.59 These obvious drawbacks haveundoubtedly influenced the development of iontophoretic systems for the delivery of

Figure 14.3 Drionic1 device for tap-water iontophoresis. (From General Medical Co., LosAngeles, CA. With permission.)


Iontophoresis & 185


lidocaine. Among the first applications of lidocaine iontophoresis were local anesthesiaprior to tooth extraction or root canal surgery60 and anesthesia of the external ear canalfor myringotomy61 for which iontophoresis remains the preferred method41. Otherstudies focusing on the transdermal iontophoresis of lidocaine followed.32,62–66 Thesestudies established the potential of such systems to provide (i) sufficient anesthesia forbrief periods62 and (ii) a rapid onset of action compared to available passive formula-tions.65 Furthermore, the addition of vasoconstrictors such as epinephrine was reportedto promote a depot effect by decreasing the clearance (which is increased by theanesthetic itself 67).63,64 In ophthalmology, an in vivo study on human volunteers has

anesthesia of eyelid skin.68

In addition to the clear therapeutic rationale for lidocaine iontophoresis, the moleculealso possesses suitable physicochemical properties: a low MW (234 Da) and a pKa of7.9, which means that it is a cation primarily delivered by electromigration undertypical experimental conditions. Therefore, it is not surprising that lidocaine was the

iontophoresis (Iontocaine1, Iomed, Inc., Salt Lake City, UT). This device consists ofa microprocessor-controlled battery-powered DC current generator and electrodes;

Recently, a second-generation lidocaine iontophoreticdevicewasapproved (May2004)by the FDA (LidoSite1, Vyteris, Inc., Fair Lawn, NJ and entered the market early 2005). It isthe first prefilled iontophoretic product, designed to achieve local anesthesia before med-ical interventions such as insertion of intravenous (IV) catheters, needlesticks for blooddraws, and other diagnostic as well as dermatological surgical procedures. It combines fastonset of action with an easy-to-use, preprogrammed design.3 The device consists of amonolithic patch containing Ag–AgCl electrodes with 10% lidocaine and 0.1% epinephrinedispersed throughout a hydrogel matrix formulation at the anode (Figure 14.4).

Phase I clinical studies have shown that 10 min of iontophoresis were sufficientto anesthetize the skin to a depth of at least 6 mm (often 10 mm or more), whichis sufficient for needlesticks and dermatological procedures. Neither lidocaine norepinephrine were detected in the systemic circulation.3 Phase II studies demonstratedthe suitability of the device for pediatric patients (n¼ 48) requiring venipuncture, whilePhase III clinical studies established that patients (adults and children) receivinglidocaine from the LidoSite1 system reported significantly less pain upon venipunctureor IV cannulation than the corresponding control subjects.3,69 Furthermore, adult and

Figure 14.4 LidoSite1 device. (With permission.)




first drug approved by the FDA (in 1995 [http://www.fda.gov]) for administration by

the anode chamber is filled with a lidocaine or epinephrine solution before use (http://

shown that iontophoresis of lidocaine is effective for achieving short-term, superficial

www. iomed.com).

http://www.fda.gov

http://www.iomed.com

http://www.iomed.com

pedriatic patients treated with LidoSite1 experienced little or no pain during surgicalprocedures, such as incisional or excisional treatment of superficial skin lesions.3 Inlight of these results, then, the commercialization of such a device is expected tobring significant improvements to the quality of care associated with invasive medicalinterventions.

The GlucoWatch1 Biographer for Noninvasive Glucose Monitoring

Diabetes mellitus requires frequent monitoring of glycemia to reduce the long-termcomplications caused by high serum concentrations of glucose,70 and to avoid dangerousepisodes of hypoglycemia. This obliges the patient to resort to the ‘‘finger-stick’’ tech-nique, which is painful. Moreover, for tight glucose control, this procedure needs to beperformed several times a day. Consequently, patient compliance can be an issue, withobvious risks to health and, in the long term, to life expectancy. The feasibility of reverseiontophoretic glucose extraction was demonstrated first in vitro,71 and then in humanvolunteers.70,72 Because glucose is a small (180 Da) and neutral molecule, electroosmosisis the mechanism responsible for its extraction from the extracellular fluid. Glucose lendsitself to ‘‘reverse’’ iontophoresis because it is present at sufficiently high concentrations inthe interstitial fluid so as to still be detectable after extraction into the sampling compart-ment. Nevertheless, the concentrations detected are significantly lower than the milli-molar levels in the blood and analytical chemistry was therefore the greatest challenge toovercome in the development of the device.

The GlucoWatch1 Biographer (Cygnus, Inc., Redwood City, CA) received FDA ap-proval in 2001 for glucose monitoring. It consists of a wrist-worn device which continu-ously extracts (by iontophoresis) and measures glucose (via an electrochemical–enzymatic sensor73) over a period of 13 h (Figure 14.5). The correlation between glucoseconcentrations measured with this device and blood concentrations has been demon-strated in both adult and pediatric patients, making it suitable for home-use.73–76 Acomprehensive clinical study in Types I and II diabetics (92 patients; age: 42.1+ 15.1years) demonstrated a high correlation between blood glucose measurements made

Figure 14.5 GlucoWatch1 G2eˆ

Biographer. (Copyright 2004 Cygnus, Inc. With permission.)


Iontophoresis & 187


with the GlucoWatch1 Biographer and the comparator fingerstick sampling method(Figure 14.6).73 A detailed analysis of the paired data points using the Clarke error gridapproach showed that 96.8% of the data points lay in the combined A and B regions thatcorrespond to diagnoses that are either clinically accurate or where an error would eitherresult in a benign effect. Hence, only 3.2% of the data fell in regions C, D, and E wheremistreatment leads to progressively more damaging sequelae for the patient. Thesestatistics compared favorably with existing invasive blood glucose sampling methods.Subsequent studies in younger patients (28 patients, age: 30.9+ 6.9 years) with Type Idiabetes, confirmed the accuracy of the device, as compared to the comparator finger-stick method, in both the clinical and home environments.74 Again, more than 96% of thepaired data lay in regions A and B of the Clarke error grid.

To date, its main inconvenience is the necessity of daily calibration with a fingerstickblood glucose sample, in order to correlate the extracted glucose amounts with subder-mal levels. A second-generation product would ideally remove this invasive calibrationstep. Recently, an in vitro study has demonstrated the feasibility of using an endogenous‘‘internal standard’’ simultaneously extracted during iontophoresis, the serum concentra-tions of which (unlike that of glucose) are relatively constant — thus circumventing theneed for a fingerstick calibration.77 However, glucose extraction, in contrast to sodiumextraction, appeared to be subject to seasonal variations.78

There have been reports of irritation at the application site due to the electric current.Preapplication of corticosteroid preparations reduced this side-effect without affectingthe efficacy of the GlucoWatch1 Biographer.79

22 44 66 88 1010 1212 141400

22

44

66

88

1010

1212

1414

1616

1818

Elapsed Time, h

Blo

od G

luco

se, m

mol

/L

BiographerBGCal Pt

Figure 14.6 Glucose concentration vs elapsed time for 1 Subject as measured by the Gluco-watch Biographer and Reference Blood Glucose (BG) Methods. Cal Pt indicates calibrationpoint. To convert millimoles per liter to milligrams per deciliter, multiply by 18. (Reproducedwith permission from Tamada, J.A., Garg, S., Jovanovic, L., Pitzer, K.R., Fermi, S., and Potts,R.O., Cygnus Research Team, JAMA, 282, 1839, 1999. Copyright 1999. American MedicalAssociation. All rights reserved.)




In addition to invasive glucose monitoring, Type I diabetes patients need repeateddaily subcutaneous injections of insulin to regulate their glycemia. Therefore, boththe diagnosis and the therapy of this disease require long term, invasive, and henceinconvenient procedures. One highly sought after (and challenging) solution to thispredicament would be to develop a noninvasive closed-loop device combining con-trolled insulin delivery in response to the monitored glucose levels — all withoutrecourse to needles.80 Although iontophoresis would appear to be an attractive optionboth for extracting glucose and delivering insulin; the latter poses a considerable chal-lenge (as discussed below).

E-TRANS1 Fentanyl HCl (Ionsyseˆ

) — Pending Approval from the FDA

One of the major advantages of iontophoresis is the ability to control the rate of drugdelivery. Therefore, pain management is a field in which this technique could have agreat impact, potentially offering the relief of either acute (e.g., postoperative) or chronic(e.g., cancer-associated) pain.3 Opioid analgesics are good candidates for iontophoreticdelivery, because they are small (300 to 500 Da), and positively charged at physiologicalpH. Furthermore, they are potent compounds and low doses (and plasma concentrationsin the ng/ml range) are sufficient to obtain a therapeutic effect.81

Fentanyl has a modest MW of 336.5 Da and moderate lipophilicity (log Ko/w¼ 3.89);it is also a very potent opioid (100 to 500 times more potent than morphine).3 Thesephysicochemical and pharmacological properties, together with its pharmacokinetics —short half-life and significant hepatic first-pass effect — make it an ideal candidate fortransdermal delivery. In many respects, fentanyl has already ‘‘conquered’’ the passivetransdermal market: the Duragesic1 patch (ALZA Corporation, Mountain View, CA),which received FDA approval in 1990 for the management of chronic pain, has become

ability to provide constant plasma concentrations and hence continuous pain relief.However, the passive patch cannot provide a rapid ‘‘bolus’’ drug input to relieve acutepain. More to the point, Duragesic is specifically contraindicated for the management of

because of its slow onset of action,82 nor appropriate because it does not allow patient-controlled analgesia.17 Iontophoresis is clearly able to respond to this unmet need sinceelectrically-assisted delivery of the fentanyl cation can provide both controlled deliveryand rapid onset of analgesia.

The feasibility of iontophoretic fentanyl delivery has been demonstrated in vitro andin vivo83–85 and has led to the development of the Ionsys system (using E-TRANSelectrotransport technology; Alza Corporation, Mountain View, CA), which has recentlyreceived an ‘‘approvable letter’’ from the U.S. FDA (July 2004). It consists of a prepro-grammed, self-contained, on-demand drug delivery system that is activated by the patient

81 (cf. Duragesic input rates of25 to 100 mg/h for patch sizes of 10 to 40 cm2) and allows either a 24 h continuous or anon-demand pulsed delivery to be achieved.86

In clinical studies, steady-state fentanyl concentrations obtained with constant currentiontophoresis were proportional to the current applied. Pharmacokinetically, ionto-

86–88

Recent clinical studies have confirmed the efficiency of the Ionsys system and haveshown that the iontophoretic delivery of fentanyl was well tolerated when used to treat


Iontophoresis & 189


(Figure 14.7). It can deliver 40 mg of fentanyl during 10 min

phoretic delivery was shown to be able to match IV infusion kinetics (Figure 14.8).

a ‘‘billion dollar’’ product (http://www.alza.com). The key to its success has been its

acute or postoperative pain (http://www.fda.gov/cder), for which it is not very effective

http://www.alza.com

http://www.fda.gov

postoperative pain.81 Furthermore, the technology provides postsurgical pain controlequivalent to that of an intraveneous morphine pump, illustrating the potential impact ofthis device in medical care.89

Other Applications of Iontophoresis In Vivo

Iontophoretic devices have been used in several therapeutic areas, often dating backmany years, sometimes with mixed success, but without the benefit of clinical trials, inthe conventional sense, or FDA submission and approval. The Phoresor1 (Iomed, Inc.,Salt Lake City, UT), a hand-held device for applying a small electrical current, was first

Importantly, the Phoresor1 was approved exclusively as a device for use inhumans, and not approved for use with a specific drug, for example lidocaine ordexamethasone (see below). As such, the user is required to fill the electrode with adrug solution ‘‘on-site’’ immediately prior to use; even so the device has been frequentlyused to administer therapeutic agents for clinical applications.

On-demand button

System controller

Electronics and battery

Electrode

Drug reservoir

Adhesive

Figure 14.7 E-TRANS1 system for the iontophoretic delivery of fentanyl HCl (Ionsyseˆ). (With

permission.)

0

0.5

1

1.5

2

2.5

I = 0.15 mA I = 0.2 mA l = 0,25 mA 50 mg/20 minIV delivery

Cm

ax (

ng

ml-

1 )

Figure 14.8 Comparison of mean peak plasma concentrations (Cmax) obtained after iontophore-tic and IV delivery of Fentanyl in human subjects. (Adapted from Gupta, S.K., Southam, M.,Sathyan, G., and Klausner, M., J. Pharm. Sci., 87, 976, 1998.)




approved in 1987 (Motion Control, Inc., Salt Lake City, UT) (http://www.accessdata.fda.gov).

http://www.fda.gov

http://www.fda.gov

Physical Medicine

Iontophoresis has a well-established place in physical therapy. It enables the deliveryof ionic drugs for local effects, while minimizing their systemic levels. However, many ofthe early applications were empirical and implemented without a detailed understandingof the processes (and sometimes, safety issues) involved.90 Poor choice of electrodes orthe use of too high a current meant that iontophoretic treatment was sometimes accom-panied by chemical burns of the skin because of changes in solution pH.67 Clinicalstudies reported in the literature predominantly concern musculo-skeletal pathologies,such as plantar fasciitis,91 rheumatoid arthritis of the knee,92 and other rheumaticdiseases.93

In 1995, Costello and Jeske reported a resurgence in the use of iontophoresis,particularly for the delivery of lidocaine and antiinflammatory agents, which remain themain applications of iontophoresis in physical medicine.67 Corticosteroids, in particulardexamethasone sodium phosphate and methylprednisolone sodium succinate, havebeen extensively employed as topical antiinflammatory agents for the treatment ofmusculo-skeletal conditions such as tendinitis — where they can be combined withlidocaine. Although these applications date back to the 1960s, studies examining thetissue levels and distribution of the iontophoresed molecules followed much later.94 Theanodal iontophoresis of dexamethasone in Rhesus monkeys resulted in higher tissueconcentrations compared to those obtained by systemic administration, although thelevels were inferior to those measured after local injection. The drug was able to reachthe muscle and some of the underlying structures, such as the cartilage, at concentrationswhich were considered to be clinically adequate.94 In a double-blind, randomizedclinical study, iontophoretic treatment with dexamethasone led to an improvement inacute Achilles tendon pain.95 Similarly, a pilot study in five patients with rheumatoidarthritis of the knee reported a significant improvement with iontophoretic dexametha-sone compared to placebo.92

Acetic acid has also found a place in physical therapy, notably for the treatmentof calcifying tendinitis of the shoulder, although in the absence of supporting clinicalstudies.96,97 A recent double-blind randomized controlled trial showed that remissionwas not better when iontophoresis of acetic acid was coupled to physiotherapy, thanwhen patients were treated with physiotherapy alone.98 However, it was suggested that alarger group of patients was required to confirm the results.

Dentistry and Other Oral Pathologies

Iontophoresis has also been widely used in dentistry and, as mentioned earlier, one of thefirst applications of lidocaine iontophoresis was local anesthesia prior to tooth extractionor root canal surgery.60 Three basic applications in dentistry have been described:treatment of hypersensitive dentine using fluoride, therapy of oral ulcers and herpesorolabialis lesions using corticosteroids and antiviral drugs, respectively, and localanesthesia.67

Iontophoresis of a 2% lidocaine solution was shown to provide adequate anesthesiaduring extraction of loose deciduous teeth.45,99 Fluoride iontophoresis, in double-blindcontrolled, clinical studies, was found to be effective for the desensitization of hypersen-sitive dentine.41,45,99 In a separate study, the fluoride content of extremely thin layers ofdentine was reported to be higher after iontophoresis compared to the passive deliveryof fluoride.100 The nature of dentine may also impact on the efficacy of iontophoresis; the


Iontophoresis & 191


delivery of charged drugs through caries-affected dentine was significantly less than thatthrough intact dentine.101

Ophthalmology

The applications of iontophoresis in ophthalmology have been the subject of severalstudies, though they are often restricted to animal models.45,102,103 In humans, iontophor-esis of lidocaine has been shown to be effective for superficial anesthesia of eyelid skin.68

In this double-blind study, nine patients undergoing bilateral upper eyelid surgeryreceived lidocaine iontophoretically in only one eyelid, prior to the routine anestheticinjections. These patients reported less pain in the iontophoresed eyelid compared tothe control eyelid.68 Recently, the transscleral iontophoresis of methylprednisolonesodium succinate — for suppressing active corneal graft rejection — was clinicallyevaluated in 17 patients. The ‘‘Eyegate’’ iontophoresis applicator (Optis, France) madeof soft silicone rubber designed to closely fit the eye contour and palpebral opening(Figure 14.9) consists of a tungsten electrode immersed in the drug solution that flowsthrough two thin silicone tubes. Vision is not impaired during the treatment becausethe corneal surface is not covered by the device (which has an annular shape) or bythe drug solution. The treatment was successful for 15 of the 17 eyes treated (completereversal of the rejection processes) and did not induce any structural alterations inthe cornea at low current densities.104 The potential risk associated with ocular ionto-phoresis should, however, be taken into account: Monti et al.105 reported increasedcorneal hydration associated with the iontophoresis of two beta-blockers (timololmaleate and betaxolol hydrochloride), a result indicative of damage to the cornealepithelium.

Otorhinolaryngology

Iontophoresis is the method of choice for anesthesia of the tympanic membrane prior tosurgery.61,67,106 Other otological applications with iontophoresis such as the treatment oftinnitus with local anesthetics,45 and that of burned ear chondritis107 with antibiotics(gentamicin and penicillin) have been attempted with mixed success. There are isolatedreports of zinc iontophoresis for the treatment of allergic rhinitis, before the advent of theantihistamine drugs.67,108

Figure 14.9 Eyegate device. (Optis, France. With permission.)




Potential Candidates for Iontophoretic Delivery

To date, relatively few applications of iontophoresis have found a place in routine clinicalcare, despite the extensive studies conducted in this field, and the numerous drugsinvestigated. A detailed discussion of the myriad iontophoretic investigations conductedin vitro is outside the scope of this review. However, for the sake of completeness and to

in different therapeutic areas and which, to the best of our knowledge, have not beenfollowed up in vivo. In this section, we describe the results of some promising, but morepreliminary, iontophoretic studies in vivo. These range from early stage animal studies tomore advanced investigations in humans that have not (yet) culminated in the develop-

Animal Models

Cardiovascular Agents

The iontophoretic delivery of beta-blockers has been investigated in vivo, both within thecontext of mechanistic studies109,110 and for therapeutic investigations.111,112 The ionto-phoretic delivery of atenolol, pindolol, metoprolol, acebutolol, oxprenolol, and propra-nolol was investigated in vivo in male Sprague–Dawley rats to determine the relationshipbetween iontophoretic transport and drug lipophilicity.109 Drug concentrations in theskin, together with those in the cutaneous and systemic veins were measured. Skinconcentrations of beta-blockers generally increased as a function of lipophilicity, whiledrug transfer from the skin to the cutaneous vein was inversely proportional to log P.However, pindolol, a relatively hydrophilic molecule, presented the highest skin absorp-tion and transfer to cutaneous blood flow.109

The physiologic effects of beta-blockers, delivered by iontophoresis, have also beenevaluated by directly analysing the cardiac responses in rabbits.111,112 Pulsed-modeconstant current (0.5 mA) iontophoretic delivery of metoprolol, a beta-blocker thatundergoes significant hepatic first-pass elimination, to rabbits made hypertensive bymethoxamine IV infusion, induced a decrease in systolic and diastolic pressures within

111 Iontophoretic delivery of timolol (1 mg/ml) induced an inhibition ofisoprenaline induced tachycardia in rabbits. The effect was increased by pretreatment ofthe skin with a chemical enhancer (Azone1), and the dose delivered after pretreatmentfrom a 0.1 mg/ml donor solution was comparable to that with intravenous delivery oftimolol (30 mg/kg).112

Iontophoretic delivery of the angiotensin-converting enzyme (ACE) inhibitor, capto-pril, both in direct current and pulsed direct current modes, was also shown to reduce themean arterial pressure in induced hypertensive rabbits by 20% within 1 h.113

Dermal Applications

Triamcinolone acetonide, an inhibitor of protein synthesis used in the treatment ofhypertrophic scars and keloids, is usually injected as a suspension into the scar by ahigh-pressure device.114 The iontophoretic delivery of triamcinolone acetonide (from anaqueous N,N-dimethylacetamide solution) in hairless rats resulted in pharmacologicallyeffective concentrations in the skin tissues beneath the drug electrode.114,115 However,human studies were not conducted because of concerns that proliferation of the normal


Iontophoresis & 193


2 h (Figure 14.10).

better orient the reader, Table 14.1 and Table 14.2 provide a summary of in vitro studies

ment of a commercial product (Table 14.3 and Table 14.4).

Table 14.1 In Vitro Studies Using Animal Skin

Drugs Animals Observations References

Beta-blockers. Series (propranolol, timolol,

metoprolol, nadolol, atenolol)

. Propranolol

. Betaxololþ timolol

Full-thicknesshairless mouse skin

Full-thicknesshairless mouse skin

PigRabbit cornea

Decrease of EO inthe case of lipophilic and cationic molecules

Decrease of EODecrease of EO

Increased permeation but damage to cornea

37

3732

105

Verapamil Hairless mouseþ hairless rat Increased permeation rate and reduced lag-timecompared to passive delivery

Drug reservoir in the skin

191192193

TRH Nude mouse

Hairless ratRabbit inner pinna skin

Enhanced transport compared to passive delivery;greater contribution of EO than EM

Enhanced transport compared to passive deliveryMechanistic studies using pulsed current

40

194195

(D-Trp6, Pro9-NHEt)LHRH Hairless mouse 50-fold increase in donor concentration induced only5-fold increase in delivery rate

196

Sermorelin analogue Ro 23–7861 Hairless guinea pig Flux increased with increasing current density; Pt electrodes 197DHEA prodrugs Rabbit Up to 7-fold increase in flux compared to DHEA 198Domperidone Hairless rat Enhanced delivery compared to passive delivery,

but too low for therapeutic application199

Azidothymidine (AZT) Hairless ratHairless mouse

Enhanced delivery compared to passive deliverySynergistic effect of chemical enhancers

200201

Smith

and

Maib

ach/

Percu

taneous

Penetratio

nEnhan

cers

2nd

edn

TF2152_c0

14

Fin

alPro

of

pag

e194

12.9

.2005

12:0

9pm

194

&P

ercu

tane

ou

sP

en

etratio

nEn

han

cers


Table 14.2 In Vitro Studies with Human Skin

Drugs Type of skin Observations References

Beta-blockers. Sotalol

. Sotalol, Timolol, Propranolol

. Atenolol, Timolol

Cadaver abdominal skin (epidermis)

Cadaver abdominal skin (epidermis)

Abdominal skin from plastic surgery(stratum corneum)

Iontophoretic delivery not equivalent tosnake skin, and not increased by enhancers

Iontophoresis enhanced the flux of themost hydrophilic drugs

Decrease of EO by timolol, but not by atenolol

202

203

204

Nafarelin Post-surgical or cadaver skin (epidermis) Transport increased with decreasing donorconcentration (decrease of EO)

36

Melanotropin (6–9) Breast skin (epidermis) Permeation rate increased by a factor of 30compared to passive diffusion

Iontophoresis more effective than chemicalenhancement

205

9-desglycinamide, 8-arginine-vasopressin (DGAVP)

Abdominal skin from plasticsurgery (dermatomed skin)

Transported mainly by EO 206

Khellin Abdominal skin from abdominoplastysurgery (full-thickness skin)

Therapeutic amount deliveredDrug reached upper dermis

207

Nalbuphineþprodrugs Skin from breast reduction operations(full-thickness)

Modest enhancement compared topassive delivery

208

Rotigotine Abdominal or breast skin from surgery(full-thickness skin and stratum corneum)

(stratum corneum)

Flux linearly correlated to flux concentrationand current density

Therapeutic levels possibleInfluence of pH and NaCl concentrationEM is the main transport mechanism

209

210

Smith

and

Maib

ach/

Percu

taneous

Penetratio

nEnhan

cers

2nd

edn

TF2152_c0

14

Fin

alPro

of

pag

e195

12.9

.2005

12:0

9pm

Ion

top

ho

resis

&195


Table 14.3 In Vivo Animal Studies

Drugs Animals Observations References

Beta-blockers. Series (atenolol, pindolol, metoprolol,

acebutolol, oxprenolol, propranolol). Series (propranolol, oxprenolol,

timolol, metoprolol, sotalol). Metoprolol

. Timolol

Male Sprague–Dawley rats

Hairless rats

Hypertensive New Zealandwhite rabbits

Albino rabbits

Therapeutic amounts delivered. Correlationbetween lipophilicity and permeation

Therapeutic amounts delivered

Therapeutic effect (reduction of blood pressure)

Therapeutic effect (inhibition of isoprenalineinduced tachycardia)

109

110

111

112

Captopril Hypertensive rabbits Decrease in arterial pressure 113Triamcinolone Rats May be interesting for the treatment of keloid

and hypertrophic scars114115

Chlorhexidine digluconate Guinea pigs Pronounced reduction of resident stratum corneumbacteria relative to control treatments

211

Hydromorphone Pigs Iontophoretic delivery comparable to IV infusion 38Buprenorphine Weanling Yorkshire swine Therapeutic amounts delivered 117NSAIDS. Salicylic acid. Ketoprofen. Naproxen. Indomethacin

Sprague–Dawley rrats Correlation between lipophilicity and permeation 118

Ropinirole Hairless rats Possibility of delivering therapeutic amounts 119

Smith

and

Maib

ach/

Percu

taneous

Penetratio

nEnhan

cers

2nd

edn

TF2152_c0

14

Fin

alPro

of

pag

e196

12.9

.2005

12:0

9pm

196

&P

ercu

tane

ou

sP

en

etratio

nEn

han

cers


Antisense oligonucleotides Rats, mice

Rats

Physiological responses

Penetration into all corneal layers

121122123

Insulin Alloxan-diabetic rabbitsDiabetic rats

Decrease of blood glucose levels after removal of SCMonomeric human analogue (intact skin) and bovine

insulin (impaired barrier) induced decrease inblood glucose level

133130

Arg-vasopressinþ analog Rats Minor antidiuretic effect 136137

Desmopressin Rats with diabetes insipidus Pharmacological effect 139Calcitonin. Human. Salmon

Hairless ratsRabbits

Rats

Hypocalcemia comparable to IVTherapeutic effectPossible to match IV infusionPulse depolarization-iontophoresis induced hypocalcemia

142145144147

hPTH Ovariectomized ratsSprague–Dawley rats,

hairless rats, beagle dogs

Similar results to SC injectionAbsorption via hair follicles

149150

LHRH Yorkshire pigs Delivery of pharmacologically active LHRH 151GHRH Hairless guinea pigs Steady-state levels comparable to IV and SC 153GHRP Rats Therapeutic levels may be achieved 154Octreotide Rabbits Increased flux as a function of current and concentration 155

Smith

and

Maib

ach/

Percu

taneous

Penetratio

nEnhan

cers

2nd

edn

TF2152_c0

14

Fin

alPro

of

pag

e197

12.9

.2005

12:0

9pm

Ion

top

ho

resis

&197


Table 14.4 In Vivo Human Studies

Drugs Patient population Observations References

Idoxuridine 6 patients with 14 recurrent HSV lesions Clinically effective against HSV orolabialis 15699

Ara-AMPþ acyclovir 9 patients (for each treatment) with HSVorolabialis lesions

Only Ara-AMP effective 15741

Lidocaineþ epinephrinefollowed bymethylprednisolone

1250 patients with PHN Significant relief in 60 to 70% of patients 9941

Cisplatin 12 patients with 15 cancer lesions1 patient

Partial or complete response in 11 lesionsSuccessful treatment

160162

Vinblastine 4 healthy volunteers;5 HIV-1 infected patients

Partial to complete clearing of all lesions 163

5-Fluorouracil 26 patients with Bowen’s disease Effective 164ALAProdrugs of ALA

13 healthy volunteers10 healthy volunteers20 healthy volunteers

Treatment of skin cancers by photodynamic therapy

Sufficient amounts delivered for PDT anddose-dependent response

165166167

Methotrexate 1 patient with palmar psoriasis Marked improvement 168Tranilast 7 healthy volunteers; 8 eight patients Effective for the treatment of keloid and

hypertrophic scars116

Morphine 17 post-surgical patients

4 healthy volunteers

Minimal effective concentrations achieved,but ‘‘wheal-and-flare’’

Therapeutic amounts delivered

169

212Aspirin 80 patients with rheumatic disease

Male (18 years)Good results; no side-effectsSystemic adverse reactions

170172

Pirprofen 80 patients with rheumatic disease Good results; no side-effects 170Ketoprofen 3 healthy volunteers Pain relief 174

Smith

and

Maib

ach/

Percu

taneous

Penetratio

nEnhan

cers

2nd

edn

TF2152_c0

14

Fin

alPro

of

pag

e198

12.9

.2005

12:0

9pm

198

&P

ercu

tane

ou

sP

en

etratio

nEn

han

cers


Ketorolac Double-blind study on 60 patients withpain from rheumatic disease

Pain relief (superior to placebo) 175

Diclofenac Woman (aged 76) with painfulinflammation of sciatic nerve

Woman (aged 36) with lumbar pain

Systemic adverse reactions

Allergic contact dermatitis

173

177Piroxicam 4 healthy volunteers Tape-stripping after treatment showed

enhancement of delivery by iontophoresis178

Prednisolone Volunteers (skinþnail) Reservoir in epidermis 179

Dexamethasone5 patients with rheumatoid arthritis of the knee14 patients with acute Achilles tendon pain

Significant improvement compared to placeboSignificant improvement compared to placebo;

one-year follow-up

9295

Alniditan 8 human volunteers Cmax similar to SC injection, but minimaltherapeutic levels

180

R-apomorphine 10 Parkinson’s patients10 Parkinson’s patients

SubtherapeuticConcentrations at the lower end of therapeutic

concentration range

184182

Tacrine 10 healthy volunteers Plasma concentration similar to oral delivery 185Metoclopramide 7 human volunteers Therapy rates achievable 186Arbutamine 32 healthy male volunteers Good correlation of plasma concentrations with the

isolated perfused porcine skin flap model (IPPSF)213

Leuprolide 11 healthy male volunteers13 healthy male volunteers

Pharmacological effectPharmacological effects comparable to SC

189187

Smith

and

Maib

ach/

Percu

taneous

Penetratio

nEnhan

cers

2nd

edn

TF2152_c0

14

Fin

alPro

of

pag

e199

12.9

.2005

12:0

9pm

Ion

top

ho

resis

&199


cells might also be inhibited by diffusion of drug around the site of application. The moreselective agent, tranilast, was preferred for these studies (described below in the ‘‘HumanStudies–Dermal Applications’’ section).116

Opioids

Iontophoretic delivery of hydromorphone has been investigated in a cross-over study inpigs, after having first been tested in vitro on porcine and human skin.38 Delivery rateswere determined from plasma drug concentrations and from residual drug analysis in thehydrogel patches used. A good correlation was observed between the plasma hydro-morphone levels during both iontophoresis and constant IV infusion (approximately1 mg/h) (Figure 14.11). Furthermore, the in vitro and in vivo data were also wellcorrelated.38

The iontophoretic delivery of buprenorphine, a potent highly lipophilic and posi-tively charged narcotic analgesic, was superior to passive administration in weanling

0

50

100

150

0 5 10 15 30 45 60 75 90 105

120

Time (min)

Blo

od

pre

ssu

re(m

m H

g)

systolic pressure

diastolic pressure

Iontophoresis

Figure 14.10 Effect of iontophoretic delivery of metoprolol on blood pressure in hypertensiverabbits (n 5 4). (Adapted from Zakzewski, C.A. and Li, J.K.J., J. Control. Release, 17, 157, 1991.)

0

0

5

10

20

25

15

5 10 15Time (h)

Ion

top

ho

reti

c an

d IV

pla

sma

con

cen

trat

ion

s (m

g L

-1)

Iontophoresis (0.8 mA)

IV infusion (0.94 mg h−1)

Figure 14.11 Comparison between iontophoretic and IV delivery of hydromorphone in thesame pig. (Adapted from Padmanabhan, R.V., Phipps, J.B., Lattin, G.A., and Sawchuk, R.J.,J. Control. Release, 11, 123, 1990.)




Yorkshire swine. The efficacy of iontophoretic delivery (I¼ 0.2 mA/cm2) was comparedto IM administration. Steady-state levels were achieved rapidly, and therapeutic amountsof buprenorphine were reported to be delivered.117

Antiinflammatory Agents

The effect of lipophilicity on the iontophoretic delivery of nonsteroidal antiinflammatoryagents (NSAIDs) in vivo has also been investigated.118 Four NSAIDs — salicylic acid,ketoprofen, naproxen, and indomethacin — were delivered to Sprague–Dawley rats bycathodal iontophoresis. A positive correlation was observed between lipophilicity andskin concentrations of NSAIDs, while plasma concentrations decreased with increasinglipophilicity.118 Iontophoresis of salicylic acid resulted in reasonable levels in the skinand the highest transfer to the cutaneous vein,118 as was the case for the hydrophilic beta-blocker pindolol.109

Miscellaneous Nonpeptidic Drugs

The iontophoretic delivery of ropinirole, a new dopamine agonist used for Parkinson’sdisease therapy has been investigated in vitro (piglet skin)39 and in vivo (hairless rats).119

In both cases, iontophoretic transport was independent of the donor drug concentration,in the absence of competing ions in the formulation, as observed for lidocaine32 andhydromorphone hydrochloride38 in vitro.

Iontophoresis has been proposed for delivering antisense oligonucleotides, which inview of their physicochemical properties — high MW (~3000 Da and higher), relativeinstability and negative charge at physiological pH — would not appear to be suitablecandidates for this technique.120 However, recent studies performed on animals havedemonstrated a physiological response after iontophoretic delivery. For example, deliv-ery of an oligonucleotide targeted to the cytochrome P 450–3A2 mRNA translational startsite in rats (0.5 mA/cm2 for 3.5 h) resulted in metabolic changes.121 Iontophoresis of anantisense oligonucleotide directed against the 3’-untranslated region of mouse IL-10mRNA induced an inhibitory effect on the production of IL-10, which is involved in thepathogenesis of atopic dermatitis and an improvement of the skin lesions was observedin treated mice.122 An ophthalmic study in rats reported the iontophoretic (0.3 mA for 5min) penetration of oligonucleotide into all the corneal layers, without any detectableocular damage.123

Protein and Peptide Drugs

One of the greatest challenges for noninvasive protein and peptide delivery concernsinsulin. Iontophoretic delivery of insulin has been extensively studied, both in vitro124–129

and in vivo, in small animals.126,130–134 However, the physicochemical properties ofinsulin are not suited to iontophoretic delivery: the insulin monomer is approximatelya 6000 Da negatively charged peptide with an isolectric point (pI ) approximately of 5.4,that is, it is negatively charged at a pH greater than 5.4, but positively charged at a pHlower than 5.4. Unfortunately (for insulin transport), the skin also presents a pH gradientand is endowed with an excellent buffering capacity: its pH increases approximately from5 (outermost layers) to 7.4 (inner layers). This means that when insulin is delivered as acation from the anode, it will tend to become ‘‘neutral’’ upon contact with the skin before


Iontophoresis & 201


becoming negatively charged within the skin (pH > 5.4) — hindering its anodal transport.Conversely, when it is delivered as an anion from the cathode, cathodal delivery willnot only be opposed by electroosmosis, but insulin will take on an increasingly cationiccharacter in the upper layers of the skin (Figure 14.12). Hence, its pI plays an importantrole in its iontophoretic transport, and explains why both anodal and cathodal deliveriesgive poor results.135

Significant barrier impairment is usually necessary to deliver sufficient insulin todecrease blood glucose levels even in small animals. For example, (cathodal) insuliniontophoresis subsequent to stratum corneum removal in alloxan-diabetic rabbits pro-duced a reduction in blood glucose, which persisted even after termination of current.133

Iontophoresis of bovine insulin lowered the plasma glucose level of diabetic rats, butonly when the skin had been pretreated with depilatory cream.130 The iontophoreticdelivery of insulin analogs may be more favorable; iontophoresis of a monomeric insulinanalog in diabetic rats (with intact skin) has been reported to reduce plasma glucoselevels.130 It is thus evident that the successful delivery of therapeutic insulin doses tohumans is an immense challenge, and that even the basal insulin input required betweenmeals is unfeasible without unacceptable levels of barrier perturbation.

The iontophoresis of arginine–vasopressin and its analog, 1-deamino-8-D-argininevasopressin has been investigated in rats in vivo, where the antidiuretic effectwas estimated from the volume of urine collected.136,137 A minor antidiuretic effectwas observed after iontophoretic treatment, while it was enhanced when the cutaneousenzyme inhibitor, camostat mesilate, was introduced to the formulation. Theenzymatic barrier was purported to reside in the dermis and attributed to the presence

An

od

eC

ath

od

e

pH < pI

pH > pI

Ins+

Ins-

pH < pI pH = pI

Ins

pH > pI

INCREASING pH

Ins-Ins+

Figure 14.12 When delivered by anodal iontophoresis (donor pH < pI), the insulin cationbecomes progressively more negatively-charged in the skin as the local pH increases. In contrast,the insulin anion, when delivered from the cathode (donor pH > pI), will become more positive.These changes in the ionization state as a function of local pH hinder insulin electromigrationacross the skin.




of (predominantly) aminopeptidases and trypsin.137 Such enzymatic activity has alsobeen described in human cadaver skin in vitro.138 Despite the significant metabolicactivity of skin enzymes, therapeutically relevant concentrations might still be achievedconsidering that 40% of peptide remained intact after 12 h.

The pharmacological effect of pulsed desmopressin iontophoresis in rats with dia-betes insipidus was monitored by measuring urinary osmotic pressure. The desmopressindelivery rate was independent of pulse frequency but the response was prolonged uponincreasing the duty cycle.139 This is reasonable since increased duty cycle would implygreater time of current application and hence increased peptide delivery.

Calcitonin is a 32-amino acid peptide indicated in the treatment of Paget’s disease,in the therapy of postmenopausal osteoporosis, and in malignant hypercalcemia. It isusually delivered parenterally, the other alternative being a nasal spray, which suffersfrom low bioavailability.140 A number of studies, both in vitro and in vivo, have inves-tigated the iontophoretic delivery of this peptide, which carries a positive charge atphysiological pH.134,141–148

Iontophoretic (pulsed current) delivery of human calcitonin was able to inducesimilar hypocalcemic effects to that of an IV injection (7 mg/kg) in rats.142 Other studieshave been performed with salmon calcitonin, which is approximately 40-fold morepotent than the human form.144–147 Pulsatile iontophoresis of salmon calcitonin, using adry disc reservoir system, in vivo in rabbits,144,145 induced an equivalent hypocalcemia tothat observed after IV administration (10 IU/kg).144

A pulse-depolarization iontophoretic system, which is thought to decrease skinirritation by enabling skin depolarization, has also been evaluated using salmon calcito-nin. In a study in rats, no significant difference in the hypocalcemic effect was observedupon increasing the dose, suggesting that a dose–response plateau had been reached.147

This was also observed by Santi et al., who demonstrated that increasing the IV dose ofsalmon calcitonin from 10 to 25 IU/kg did not produce a significant increase in thehypocalcemic effect.144 Finally, a cutaneous first-pass effect during salmon calcitonindelivery has been proposed.141,147,148 The enzymatic inhibitors, aprotinin and camostatmesilate, have been demonstrated to enhance the hypocalcemic effect of salmon calci-tonin in rats,141 although aprotinin was not found to modify human calcitonin deliverykinetics across hairless rat skin in vitro.142

Human parathyroid hormone (hPTH) is an 84-amino acid peptide, which can haveeither an anabolic or a catabolic effect on bones, depending on its input kinetics (e.g.,pulsatile delivery favors its anabolic and antiosteoporotic effects). Pulsatile iontophoreticadministration of hPTH (1–34), a pharmacologically active fragment, to ovariectomizedSprague–Dawley rats, produced an increase in bone mineral density equivalent to dailysubcutaneous injections.149 A mechanistic investigation in Sprague–Dawley rats, hairlessrats, and beagle dogs, demonstrated a linear relationship between the absorption ratesand the ratio of hair follicles to epidermal thickness. Based on these results, the maintransport route for hPTH (1–34) during iontophoresis was suggested to be via the hairfollicles, suggesting that absorption in man might be intermediate between that inhairless rats and beagle dogs.150

Iontophoretic delivery of the luteinizing hormone-releasing hormone (LHRH) inYorkshire pigs in vivo, was monitored via the plasma levels of follicle stimulatinghormone (FSH) and luteinizing hormone (LH), showing that pharmacologically activepeptide had been delivered.151 As is the case for hPTH, LHRH activity depends onthe input profile, and its pulsatile delivery is indicated for treating hypogonadotropichypogonadism. Recent in vitro studies with human epidermis have suggested that a


Iontophoresis & 203


pulsed direct current profile may be more efficient than a simple constant currentapplication for delivering LHRH and its analog nafarelin, and also improve the stabilityof LHRH.152

Growth hormone releasing hormone (GHRH; sermorelin) is a 44-amino acidendogenous peptide used in the therapy of children with growth hormone(GH)deficiency. Iontophoretic delivery in hairless guinea pigs in vivo achieved steady-state plasma levels comparable to those after intravenous and subcutaneous injections(10 mg/kg).153GH-releasing peptide (GHRP), a hexapeptide analog of met-enkephalinethat stimulates the secretion of GH, was successfully delivered in vivo in rats by pulsatileiontophoresis; extrapolation of transport data to human requirements suggested thattherapeutic levels could be achieved.154

Octreotide, a synthetic octapeptide analog of somatostatin used in the therapy ofacromegaly, carcinoid syndrome, and relief from diarrhea associated with vasoactiveintestinal peptide secreting tumors, was delivered by iontophoresis in rabbits in vivo. Aproportional increase in delivery was observed upon increasing current density (50 to150 mA/cm2); however, the relationship between donor concentration and delivery wasparabolic.155

Human Studies

Dermal Applications

Although iontophoresis would seem to be an appropriate approach for topical therapy,applications of this technique in dermatology are rather limited.3 Gangarosa et al. havereviewed the dermatological applications of iontophoresis, especially those concerningantiviral drugs.41,99 Iontophoresis of the antiviral idoxuridine was shown to be clinicallyeffective against Herpes Simplex Virus (HSV) orolabialis.41,99,156 Subsequent studiesfocused on new antiviral agents presenting fewer side-effects than idoxuridine.41,99 In adouble-blind, placebo-controlled, clinical study, Ara-AMP and acyclovir (ACV) weredelivered by iontophoresis to 27 volunteers (9 subjects per treatment) who had devel-oped HSV vesicular orolabial lesions within the 48 h preceding the study. Their lesionswere treated either with Ara-AMP or ACV or NaCl for 6 to 8 min at a current intensity of0.5 to 0.7 mA. The results showed that Ara-AMP induced a decrease in viral titer after 24 h,whereas the ACV and NaCl results were not significantly different. Furthermore, thenumber of days from onset to dry crust formation was significantly decreased by theAra-AMP treatment. However, Ara-AMP-treated lesions did not heal significantly fastercompared to those receiving ACV or NaCl.41,157 The lack of effectiveness of ACV wasattributed to the high solution pH (10.6).99 In retrospect, cathodal iontophoresis of ACVwould have also been opposed by electroosmosis in the anode-to-cathode direction. Theeffect of pH on ACV delivery has been investigated in vitro using nude mouse skin158 andhuman skin159: ACV could be delivered by anodal iontophoresis at pH 3 (20% proto-nated; electromigration contributes to transport) and at pH 7.4 (neutral, hence transportdue exclusively to electroosmosis).

Gangarosa et al. have also described several clinical iontophoretic studies for thetreatment of postherpetic neuralgia (PHN). The initial protocol consisted of anodaliontophoresis of lidocaine and epinephrine (2 mA for 6 to 8min) followed by cathodaldelivery of methylprednisolone sodium succinate (2 mA for 15 min). Of 1250 patientsreceiving this treatment from 1982 to 1995, 60 to 70% reported significant and long-lastingrelief that also had a rapid onset. Adverse effects were limited, although small localized




burns, attributed to the presence of skin defects, were noted approximately in 1.5% of thepatients.99

Iontophoresis has also been successfully evaluated for the treatment of skin cancer,an approach that can overcome the risks of scarring associated with surgical inter-vention and of long-term complications associated with radiation therapy.160 Ionto-phoretic delivery of cisplatin has been performed on basal cell carcinomas (BCC) andsquamous cell carcinomas (SCC).160–162 In a study involving 12 patients presentinga total of 15 lesions of either BCC or SCC, 11 out of the 15 lesions showed either acomplete or a 50% decrease in lesion area after iontophoresis. The treatment ofchoice appeared to be a daily 20 to 30 min current application for 5 days followed bya 2-week rest period. Epinephrine was simultaneously delivered during this protocol,in order to induce vasoconstriction, and localize the effect of cisplatin at the target site.160

Subsequently, this protocol was successfully applied (four cycles) to a 67-year-oldman with BCC.162 Iontophoretic delivery of cisplatin has also produced partial remissionwhen applied to patients with BCC or SCC lesions on the eyelids and periorbitaltissues.161

The delivery of vinblastine is problematic because (i) its systemic administration iscontraindicated in myelosuppression, (ii) the subcutaneous route may cause phlebitisand necrosis, and (iii) intralesional administration is painful. The iontophoresis of vin-blastine was, therefore, tested on four healthy volunteers before being evaluated onhuman immunodeficiency virus-type 1 (HIV-1)-infected patients over 6 months for thetreatment of Kaposi’s sarcoma. The protocol involved the ‘‘preiontophoresis’’ of a lido-caine or epinephrine solution (to induce local anesthesia) followed by the iontophoresisof a 1% vinblastine solution at 4 mA for 10 to 90 min. Although this resulted in a partialto complete clearing of all the 31 lesions treated,163 no follow-up studies have beenpublished.

Twenty-six patients with Bowen’s disease (intraepithelial SCC) received a 4-weekiontophoretic treatment with 5-fluorouracil. Only one patient showed any histologicalsign of disease 3 months post-treatment, suggesting that the approach was effective.164

The topical administration of 5-aminolevulinic acid (ALA) in conjunction with photo-dynamic therapy (PDT) has been investigated with the aim of treating skin cancer. ALAis the precursor of the endogenous fluorescent photosensitizer, protoporphirin IX(PpIX), involved in heme synthesis, and generates singlet oxygen when activated byvisible light, inducing cell damage in the host. However, passive ALA administrationfrom simple cream formulations requires several hours of application to achieve pene-tration of this highly polar molecule.165 Iontophoretic delivery of ALA and its esters hasbeen studied and has produced promising results in healthy volunteers.165–167 Rhodeset al. delivered ALA by iontophoresis (0.2 mA) to 13 healthy volunteers and quantified

Different periods of iontophoresis and irradiation doses were tested. It wasconcluded that PpIX synthesis was ALA-dose dependent (exponential relationship),and that phototoxicity could be predicted from the ALA and irradiation doses. Theauthors estimated that sufficient amounts of ALA for inducing tumor necrosis could bedelivered, and argued that PDT should be more efficient on nonmelanoma skin tumorsthan on normal skin.165 The iontophoretic delivery (0.2 mA) of ALA (a zwitterion) hasbeen compared with that of two cationic esters, ALA-n-butyl and ALA-n-hexylester,in healthy volunteers. ALA-n-hexylester iontophoresis resulted in greater PpIX forma-tion and lower phototoxicity relative to the other ester and the parent molecule.A linear correlation between the logarithm of prodrug dose and PpIX fluorescence or


Iontophoresis & 205


its permeation through measurement of PpIX fluorescence and phototoxicity (Figure14.13).

phototoxicity was observed for the three compounds.166 This relationship was alsoobserved with ALA-n-pentyl ester, which induced more phototoxicity than ALA in aniontophoretic study involving 20 healthy volunteers, possibly due to a more favorablelocalization of PpIX in the tissue.167

Methotrexate, an antineoplastic agent also used in the treatment of psoriasis, hasrecently been delivered by cathodal iontophoresis to a patient presenting recalcitrantpalmer psoriasis. The treatment consisted of 15 min of iontophoresis at 12 to 15 mA(corresponding to 0.6 mA/cm2) once a week during 4 weeks. A significant improvement(>75%) was recorded at the end of the 4-week treatment period.168

Cathodal pulsatile iontophoresis of tranilast, an agent used in the treatment of keloidsand hypertrophic scars, has been evaluated in vivo in healthy volunteers as well as inpatients. Iontophoretic delivery from an ethanol/water mixture (80:20) was effective inrelieving the pain and itching of keloid and hypertrophic scars at a lower dose than thatrequired orally and without side-effects.116

Opioids

As discussed earlier, an iontophoretic device for delivering fentanyl is currentlyunder evaluation by the U.S. FDA. Other opioid analgesics investigated include mor-phine HCl, which has also been successfully delivered iontophoretically postsurgery,resulting in the use of significantly less patient-controlled analgesia (PCA) compared toa control group. Although the morphine serum levels achieved in the treated patientswere in the therapeutic range, a local ‘‘wheal and flare’’ typical of histamine releasewas observed.169

0.03 mC 6 mC 12 mC 24 mC 60 mC 120 mC

0.5

1.0

2.0

1.5

ALA dose

PpI

X fl

uore

scen

ce (

AU

C)

Figure 14.13 PpIX fluorescence was dependent on the amount of ALA iontophoresed. Fluores-cence decreased immediately upon irradiation (preirradiation levels: open columns, immediatepostirradiation: solid columns) and gradually recovered postirradiation (up to 3 h time-point:cross-hatched columns). (From Rhodes, L.E., Tsoukas, M.M., Anderson, R.R., and Kollias, N.,J. Invest. Dermatol., 108, 87, 1997. With permission from Blackwell Publishing Ltd.)




Antiinflammatory Agents

Iontophoresis of steroids and NSAIDs has been used to treat joint pain.35 The transdermaldelivery of NSAIDs is of interest because of the potential to avoid gastrointestinal side-effects observed when administered orally. Thus, the goal is to achieve high localconcentrations (and ‘‘targeting’’) while minimizing systemic exposure to the drug.NSAIDs are generally of low molecular weight and negatively-charged at physiologicalpH, and are delivered, therefore, by cathodal iontophoresis. Although the iontophoreticdelivery of aspirin was first reported in 1903, few published clinical studies are avail-able.170,171 In a double-blind study, pirprofen and ‘‘lysine soluble aspirin’’ (lysine acet-ylsalicylate) were delivered by iontophoresis (10 20-min sessions over 2 weeks) to 40patients suffering from a variety of rheumatic conditions. The results were reported to beeither good or excellent in approximately 75% of the patients, and no side-effects wereobserved. Furthermore, only local penetration of the drug in the inflamed area wasnoted, without high systemic levels.170 However, reports of a systemic reaction toiontophoretically delivered aspirin172 and diclofenac173 suggest that circulating NSAIDlevels may not always be negligible. Ketoprofen174 and ketorolac175 have also beensuccessfully delivered by iontophoresis to human volunteers; the latter was reported tobe efficient in treating pain due to rheumatic disease.175 Diclofenac iontophoresis hasalso been used clinically,176 but side-effects including a systemic adverse reaction173 andallergic contact dermatitis177 have been reported. Cutaneous bioavailability of piroxicamfollowing passive and iontophoretic delivery from a commercially available gel formula-tion was compared in healthy volunteers.178 Quantification of drug levels in the stratumcorneum revealed that piroxicam was better delivered via iontophoresis.

Prednisolone was delivered by anodal iontophoresis to the skin and nail of healthyvolunteers. While the plasma concentrations observed were three times lower than thoseobtained after oral ingestion of a 10 mg dose, a reservoir effect was observed in thestratum corneum (as radiolabelled prednisolone was still detectable after 2 weeks in twoout of four patients), which was not the case after oral delivery. However, whether thismeans that therapeutic levels can be achieved by iontophoresis without significantsystemic exposure remains to be seen.179

Miscellaneous Nonpeptidic Drugs

Alniditan, a serotonin receptor (5 HT1B/1D) agonist for the treatment of migraine, hasbeen evaluated for iontophoretic delivery in a phase I clinical trial.180 Its physicochemicalproperties (MW~300 Da and a charge of þ2 at physiological pH) are suited to iontophor-esis. Therapeutic concentrations, albeit at the lower limit, were achieved during two30-min current application periods.

The iontophoretic delivery of R-apomorphine, a potent dopamine agonist,181–183 hasbeen evaluated in an in vivo trial in patients with Parkinson’s disease. The plasma levelsachieved with current densities of 250 and 375 mA/cm2 were, however, either subther-apeutic or at the lower end of the therapeutic concentration range.182,184

Tacrine, a cholinesterase inhibitor used perorally for Alzheimer’s disease, was deliv-ered to healthy volunteers by iontophoresis. Clinically relevant plasma profiles wereobserved, and were in a similar range to that obtained after oral delivery.185

Therapeutic amounts of metoclopramide, an antiemetic, were delivered by ionto-phoresis to healthy volunteers.186 The coiontophoresis of hydrocortisone inhibited ery-thema and oedema, but did not impact on metoclopramide delivery kinetics.186


Iontophoresis & 207


Protein and Peptide Drugs

The only peptide drug that has been delivered iontophoretically in vivo in humans is theLHRH analog leuprolide.187–189 The acetate salt was administered once a week over 3weeks to healthy volunteers, using a current of 0.2 mA over 10 to 12 h.189 Pharmaco-logically-active levels of leuprolide were delivered as deduced from the elevation ofserum LH concentrations, and steady-state levels were achieved after 30 min of ionto-phoresis. However, increasing the donor concentration of the peptide resulted in de-creased transport, suggesting that leuprolide inhibits its principal mechanism ofelectrotransport, that is, electroosmosis.189

Conclusions

Successes and Opportunities for the Future

The GlucoWatch1 Biographer for glucose monitoring (FDA approval: 2001), the Lido-Site1 device for topical lidocaine anesthesia (FDA approval: 2004), and the Ionsysfentanyl delivery system for patient-controlled postoperative analgesia demonstrate thationtophoresis has become a mature, viable drug delivery platform and, in this respect, issome way ahead of other competing transdermal technologies. The extensive studiesconducted during the last 20 to 30 years clearly show that, from a technical perspective,a wide range of drug molecules with applications in different therapeutic areas can bedelivered by iontophoresis. And yet, technical feasibility alone is not sufficient to take amolecule to the marketplace. A successful iontophoretic product must demonstrate clearadvantages over existing therapies. Both the LidoSite1 and Ionsys systems respond to aclear unmet need: they provide rapid onset, significantly reducing the ‘‘lag-time’’ beforeperception of a therapeutic effect by the patient.

It is clear that physicochemical and pharmacokinetic properties must play a role in theselection of candidates for administration by transdermal iontophoresis. However, tech-nical feasibility alone cannot justify the development of an iontophoretic product; it mustbe coupled with a clinical (and market) need. The key feature that distinguishes ionto-phoresis is the ability to control drug delivery kinetics. Therefore, future applicationsshould focus on indications where control of drug dose is a key element to therapy: forexample, drugs with a narrow therapeutic index, or where the dose must be modified(and/or customized), for example, in neurodegenerative conditions, as the illness pro-gresses, or where there is a need for pulsatile delivery. In simplistic terms, drug candidatescan be divided into two broad categories — the readily feasible (low MW cations with goodaqueous solubility, for example, lidocaine and fentanyl) and the highly desirable but not sofeasible (low and medium MW peptides). Future work will gravitate to the former groupbut the potential rewards of developing a system for noninvasive peptide delivery willinevitably mean that considerable effort will be spent in trying to achieve this objective.

Limitations

Fundamental physicochemical constraints limit the number of drugs that can be realis-tically considered for transdermal iontophoretic administration. The iontophoretic deliv-ery efficiency of a drug depends on its mobility and concentration — which, in turn,depends on its solubility — with respect to those of competing ions. For a cationic drug,




even under the most favorable conditions, that is, in the absence of competing cations,the current will be principally carried by highly mobile chloride ions, entering the anodefrom the interior of the body where they are present at a concentration of 133 mM. In viewof this, the maximum amount of drug that can be delivered per day from a reasonablysized patch is probably in the range of 20 to 30 mg; with this constraint, a large number ofdrugs with more onerous dosing requirements can be dismissed as potential candidates.Since mobility, and frequently solubility decrease with MW, there is probably a MWcut-off, above which delivery is not feasible. To date, proteins of up to 4000 Da (calci-tonin and hPTH (1–34)) have been successfully delivered to animals and have elicitedpharmacological effects. In principle, it is unlikely that the delivery of high MW species,including larger proteins, will be achievable by iontophoresis.

Remaining Challenges

Once appropriate therapeutic areas and suitable drug candidates have been identified,the major challenge is to design an iontophoretic patch system. At the very least, this willinvolve the development of formulations in which the drug will be stable while in contactwith the other patch components, for periods of up to 18 months. More realistically, thiswill involve the development of different types of patch platforms that will each beoptimized for the delivery of drugs with specific physicochemical and stability properties.For example, the delivery of peptides, which are susceptible to degradation when insolution, may require the development of ‘‘dry’’ patches where the peptide is hydratedimmediately prior to use.190 Thus, the development of a therapeutically effective productwill require a multidisciplinary effort with teams bringing together scientists and engin-eers with backgrounds in pharmaceutics, materials science, and electrochemistry to takea promising iontophoretic candidate from benchtop to market.

References1. Tyle, P., Iontophoretic devices for drugs delivery, Pharm. Res., 3, 318, 1986.2. Banga, A.K. and Chien, Y.W., Iontophoretic delivery of drugs: fundamentals, developments

and biomedical applications, J. Control. Release, 7, 1, 1988.3. Kalia, Y.N., Naik, A., Garrison, J., and Guy, R.H., Iontophoretic drug delivery, Adv. Drug

Deliv. Rev., 56, 619, 2004.4. Naik, A., Kalia, Y.N., and Guy, R.H., Transdermal drug delivery: overcoming the skin’s barrier

function, Pharm. Sci. Technol. Today, 3, 318, 2000.5. Burton, H.S.J., Iontophoresis, in Percutaneous Penetration Enhancers, Smith, E.W. and

Maibach, H.I., Eds, CRC Press, Boca Raton, Florida, 1995, 351.6. Guy, R.H. and Hadgraft, J., Prediction of drug disposition kinetics in skin and plasma

following topical administration, J. Pharm. Sci., 73, 883, 1984.7. Potts, R.O. and Guy, R.H., Predicting skin permeability, Pharm. Res., 9, 663, 1992.8. Guy, R.H., Current status and future prospects of transdermal drug delivery, Pharm. Res., 13,

1765, 1996.9. Merino, V., Alberti, I., Kalia, Y.N., and Guy, R.H., Transdermal and skin-targeted drug

delivery, J. Cutan. Med. Surg., 2, 108, 1997.10. Aguilella, V., Belaya, M., and Levadny, V., Passive transport of small ions through human

stratum corneum, J. Control. Release, 44, 11, 1997.11. Barry, B.W., Novel mechanisms and devices to enable successful transdermal drug delivery,

Eur. J. Pharm. Sci., 14, 101, 2001.


Iontophoresis & 209


12. Geinoz, S., Guy, R.H., Testa, B., and Carrupt, P.A., Quantitative structure–permeation rela-tionships (QSPeRs) to predict skin permeation: a critical evaluation, Pharm. Res., 21, 83, 2004.

13. Prausnitz, M.R., Mitragotri, S., and Langer, R., Current status and future potential of transder-mal drug delivery, Nat. Rev. Drug. Discov., 3, 115, 2004.

14. Phipps, J.B., Padmanabhan, R.V., and Lattin, G.A., Iontophoretic delivery of model inorganicand drug ions, J. Pharm. Sci., 78, 365, 1989.

15. Phipps, J.B. and Gyory, J.R., Transdermal ion migration, Adv. Drug Deliv. Rev., 9, 137, 1992.16. Sage, B.H.J. and Riviere, J.E., Model systems in iontophoresis–transport efficacy, Adv. Drug

Deliv. Rev., 9, 265, 1992.17. Delgado-Charro, M.B. and Guy, R.H., Transdermal iontophoresis for controlled drug delivery

and non-invasive monitoring, S.T.P. Pharma Sci., 11, 403, 2001.18. Kasting, G.B. and Keister, J.C., Application of electrodiffusion theory for a homogeneous

membrane to iontophoretic transport through skin, J. Control. Release, 8, 195, 1989.19. Rein, H., Experimental electroendosmotic studies on living human skin, Zeitschrift fuer

Biologie, 81, 125, 1924.20. Gangarosa, L.P., Park, N.H., Fong, B.C., Scott, D.F., and Hill, J.M., Conductivity of drugs used

for iontophoresis, J. Pharm. Sci., 67, 1439, 1978.21. Hill, J.M., Gangarosa, L.P., and Park, N.H., Iontophoretic application of antiviral chemother-

apeutic agents, Ann. N.Y. Acad. Sci., 284, 604, 1977.22. Gangarosa, L.P., Park, N.H., Wiggins, C.A., and Hill, J.M., Increased penetration of none-

lectrolytes into mouse skin during iontophoretic water transport (iontohydrokinesis),J. Pharmacol. Exp. Ther., 212, 377, 1980.

23. Pikal, M.J., Transport mechanisms in iontophoresis. 1. A theoretical model for the effect ofelectroosmotic flow on flux enhancement in transdermal iontophoresis, Pharm. Res., 7, 118,1990.

24. Pikal, M.J. and Shah, S., Transport mechanisms in iontophoresis. 2. Electroosmotic flow andtransference number measurements for hairless mouse skin, Pharm. Res., 7, 213, 1990.

25. Pikal, M.J. and Shah, S., Transport mechanisms in iontophoresis. 3. An experimental study ofthe contributions of electroosmotic flow and permeability change in transport of low andhigh molecular weight solutes, Pharm. Res., 7, 222, 1990.

26. Pikal, M.J., The role of electroosmotic flow in transdermal iontophoresis, Adv. Drug Deliv.Rev., 9, 201, 1992.

27. Marro, D., Guy, R.H., and Delgado-Charro, M.B., Characterization of the iontophoreticpermselectivity properties of human and pig skin, J. Control. Release, 70, 213, 2001.

28. Kim, A., Green, P.G., Rao, G., and Guy, R.H., Convective solvent flow across the skin duringiontophoresis, Pharm. Res., 10, 1315, 1993.

29. Peck, K.D., Srinivasan, V., Li, S.K., Higuchi, W.I., and Ghanem, A.H., Quantitative descriptionof the effect of molecular size upon electroosmotic flux enhancement during iontophoresisfor a synthetic membrane and human epidermal membrane, J. Pharm. Sci., 85, 781, 1996.

30. Lin, R.Y., Ou, Y.C., and Chen, W.Y., The role of electroosmotic flow on in vitro transdermaliontophoresis, J. Control. Release, 43, 23, 1997.

31. Bath, B.D., White, H.S., and Scott, E.R., Visualization and analysis of electroosmotic flow inhairless mouse skin, Pharm. Res., 17, 471, 2000.

32. Marro, D., Kalia, Y.N., Delgado-Charro, M.B., and Guy, R.H., Contributions of electromigra-tion and electroosmosis to iontophoretic drug delivery, Pharm. Res., 18, 1701, 2001.

33. Cullander, C., Rao, G., and Guy, R.H., Why silver/silver chloride? Criteria for iontophoresiselectrodes, in Prediction of Percutaneous Penetration, vol. 3B, Brain K.R., James V.J., andWalters K.A., Eds, STS Publishing, Cardiff, 1993, 381.

34. Gay, C.L., Green, P.G., Guy, R.H., and Francoeur, M.L., Iontophoretic delivery of piroxicamacross the skin in vitro, J. Control. Release, 22, 57, 1992.

35. Ledger, P.W., Skin biological issues in electrically enhanced transdermal delivery, Adv. DrugDeliv. Rev., 9, 289, 1992.




36. Rodriguez Bayon, A.M. and Guy, R.H., Iontophoresis of nafarelin across human skin in vitro,Pharm. Res., 13, 798, 1996.

37. Hirvonen, J. and Guy, R.H., Iontophoretic delivery across the skin: electroosmosis and itsmodulation by drug substances, Pharm. Res., 14, 1258, 1997.

38. Padmanabhan, R.V., Phipps, J.B., Lattin, G.A., and Sawchuk, R.J., In vitro and in vivo evalu-ation of transdermal iontophoretic delivery of hydromorphone, J. Control. Release, 11, 123,1990.

39. Luzardo-Alvarez, A., Delgado-Charro, M.B., and Blanco-Mendez, J., Iontophoretic delivery ofropinirole hydrochloride: effect of current density and vehicle formulation, Pharm. Res., 18,1714, 2001.

40. Burnette, R.R. and Marrero, D., Comparison between the iontophoretic and passive transportof thyrotropin releasing hormone across excised nude mouse skin, J. Pharm. Sci., 75, 738,1986.

41. Gangarosa, S. and Hill, J.M., Modern iontophoresis for local drug delivery, Int. J. Pharm., 123,159, 1995.

42. Huang, Y.Y., Wu, S.M., Wang, C.Y., and Jiang, T.S., Response surface method as an approachto optimization of iontophoretic transdermal delivery of pilocarpine, Int. J. Pharm., 129, 41,1996.

43. Gibson, L.E. and Cooke, R.E., A test for concentration of electrolytes in sweat in cystic fibrosisof the pancreas utilizing pilocarpine by iontophoresis, Pediatrics, 23, 545, 1959.

44. Stern, R.C., The diagnosis of cystic fibrosis, N. Engl. J. Med., 336, 487, 1997.45. Singh, P. and Maibach, H.I., Iontophoresis in drug delivery: basic principles and applications,

Crit. Rev. Ther. Drug Carrier Syst, 11, 161, 1994.46. Wang, L. and Freedman, S.D., Laboratory tests for the diagnosis of cystic fibrosis, Am. J. Clin.

Pathol., 117 Suppl, S109–S115, 2002.47. Ichihashi, T., Effect of drugs on the sweat glands: Cataphoresis and an effective method for

suppression of local sweating. Observation on the effects of diaphoretics and adiaphoretics,J. Orient. Med., 25, 101, 1936.

48. Holzle, E. and Alberti, N., Long-term efficacy and side effects of tap water iontophoresis ofpalmoplantar hyperhidrosis — the usefulness of home therapy, Dermatologica, 175, 126,1987.

49. Thomas, I., Brown, J., Vafaie, J., and Schwartz, R.A., Palmoplantar hyperhidrosis: a thera-peutic challenge, Am. Fam. Physician, 69, 1117, 2004.

50. Karakoc, Y., Aydemir, E.H., Kalkan, M.T., and Unal, G., Safe control of palmoplantar hyperhi-drosis with direct electrical current, Int. J. Dermatol., 41, 602, 2002.

51. Togel, B., Greve, B., and Raulin, C., Current therapeutic strategies for hyperhidrosis: a review,Eur. J. Dermatol., 12, 219, 2002.

52. Dolianitis, C., Scarff, C.E., Kelly, J., and Sinclair, R., Iontophoresis with glycopyrrolate for thetreatment of palmoplantar hyperhidrosis, Australas. J. Dermatol., 45, 208, 2004.

53. Holzle, E. and Ruzicka, T., Treatment of hyperhidrosis by a battery-operated iontophoreticdevice, Dermatologica, 172, 41, 1986.

54. Hritzo, G., Management of postsurgical hyperhidrosis with direct current and tap water: a casereport, J. Hand Ther., 17, 427, 2004.

55. Karakoc, Y., Aydemir, E.H., and Kalkan, M.T., Placebo-controlled evaluation of directelectrical current administration for palmoplantar hyperhidrosis, Int. J. Dermatol., 43, 503,2004.

56. Connolly, M. and de Berker, D., Management of primary hyperhidrosis: a summary of thedifferent treatment modalities, Am. J. Clin. Dermatol., 4, 681, 2003.

57. Nyamekye, I.K., Current therapeutic options for treating primary hyperhidrosis, Eur. J. Vasc.Endovasc. Surg., 27, 571, 2004.

58. Sato, K., Timm, D.E., Sato, F., Templeton, E.A., Meletiou, D.S., Toyomoto, T., Soos, G., andSato, S.K., Generation and transit pathway of Hþ is critical for inhibition of palmar sweatingby iontophoresis in water, J. Appl. Physiol., 75, 2258, 1993.


Iontophoresis & 211


59. Wahlgren, C.F. and Quiding, H., Depth of cutaneous analgesia after application of a eutecticmixture of the local anesthetics lidocaine and prilocaine (EMLA ceam), J. Am. Acad. Derma-tol., 42, 584, 2000.

60. Gangarosa, L.P., Sr., Iontophoresis for surface local anesthesia, J. Am. Dent. Assoc., 88, 125,1974.

61. Comeau, M., Brummett, R., and Vernon, J., Local anesthesia of the ear by iontophoresis, Arch.Otolaryngol., 98, 114, 1973.

62. Russo, J., Jr., Lipman, A.G., Comstock, T.J., Page, B.C., and Stephen, R.L., Lidocaine anesthe-sia: comparison of iontophoresis, injection, and swabbing, Am. J. Hosp. Pharm., 37, 843,1980.

63. Riviere, J.E., Sage, B.H., and Monteiro-Riviere, N.A., Transdermal lidocaineiontophoresisin isolated perfused porcine skin, J. Toxicol. Cutaneous Ocul. Toxicol., 8, 493,1990.

64. Riviere, J.E., Monteiro-Riviere, N.A., and Inman, A.O., Determination of lidocaine concentra-tions in skin after transdermal iontophoresis: effects of vasoactive drugs, Pharm. Res., 9, 211,1992.

65. Irsfeld, S., Klement, W., and Lipfert, P., Dermal anaesthesia: comparison of EMLA cream withiontophoretic local anaesthesia, Br. J. Anaesth., 71, 375, 1993.

66. Marro, D., Kalia, Y.N., Delgado-Charro, M.B., and Guy, R.H., Optimizing iontophoretic drugdelivery: identification and distribution of the charge-carrying species, Pharm. Res., 18, 1709,2001.

67. Costello, C.T. and Jeske, A.H., Iontophoresis: applications in transdermal medication deliv-ery, Phys. Ther., 75, 104–113/563, 1995.

68. Meyer, D.R., Linberg, J.V, and Vasquez, R.J., Iontophoresis for eyelid anesthesia, OphthalmicSurg., 21, 845, 1990.

69. Zempsky, W.T., Sullivan, J., Paulson, D.M., and Hoath, S.B., Evaluation of a low-doselidocaine iontophoresis system for topical anesthesia in adults and children: a randomized,controlled trial, Clin. Ther., 26, 1110, 2004.

70. Tamada, J.A., Bohannon, N.J.V., and Potts, R.O., Measurement of glucose in diabetic subjectsusing noninvasive transdermal extraction, Nat. Med., 1, 1198, 1995.

71. Rao, G., Glikfeld, P., and Guy, R.H., Reverse iontophoresis: development of a noninvasiveapproach for glucose monitoring, Pharm. Res., 10, 1751, 1993.

72. Rao, G., Guy, R.H., Glikfeld, P., LaCourse, W.R., Leung, L., Tamada, J., Potts, R.O., and Azimi,N., Reverse iontophoresis: noninvasive glucose monitoring in vivo in humans, Pharm. Res.,12, 1869, 1995.

73. Tamada, J.A., Garg, S., Jovanovic, L., Pitzer, K.R., Fermi, S., and Potts, R.O., Noninvasiveglucose monitoring: comprehensive clinical results. Cygnus Research Team, JAMA, 282, 1839,1999.

74. Garg, S.K., Potts, R.O., Ackerman, N.R., Fermi, S.J., Tamada, J.A., and Chase, H.P., Correlationof fingerstick blood glucose measurements with GlucoWatch biographer glucose results inyoung subjects with type 1 diabetes, Diabetes Care, 22, 1708, 1999.

75. Tierney, M.J., Tamada, J.A., Potts, R.O., Jovanovic, L., and Garg, S., Clinical evaluation of theGlucoWatch biographer: a continual, non-invasive glucose monitor for patients with dia-betes, Biosens. Bioelectron., 16, 621, 2001.

76. Eastman, R.C., Chase, H.P., Buckingham, B., Hathout, E.H., Fuller-Byk, L., Leptien, A.,Van Wyhe, M.M., Davis, T.L., Fermi, S.J., Pechler, H., Sahyun, G., Lopatin, M., Wang, B.Y.,Wei, C., Bartkowiak, M., Ginsberg, B.H., Tamada, J.A., and Pitzer, K.R., Use of the Gluco-Watch biographer in children and adolescents with diabetes, Pediatr. Diabetes, 3, 127, 2002.

77. Sieg, A., Guy, R.H., and Delgado-Charro, M.B., Reverse iontophoresis for noninvasive glu-cose monitoring: the internal standard concept, J. Pharm. Sci., 92, 2295, 2003.

78. Sieg, A., Guy, R.H., and Delgado-Charro, M.B., Noninvasive glucose monitoring by reverseiontophoresis in vivo: application of the internal standard concept, Clin. Chem., 50, 1383,2004.




79. Tamada, J.A., Davis, T.L., Leptien, A.D., Lee, J., Wang, B., Lopatin, M., Wei, C., Wilson, D.,Comyns, K., and Eastman, R.C., The effect of preapplication of corticosteroids on skin irritationand performance of the GlucoWatch G2 biographer, Diabetes Technol. Ther., 6, 357, 2004.

80. Guy, R.H., A sweeter life for diabetics? Nat. Med., 1, 1132, 1995.81. Chelly, J.E., Grass, J., Houseman, T.W., Minkowitz, H., and Pue, A., The safety and efficacy of

a fentanyl patient-controlled transdermal system for acute postoperative analgesia: a multi-center, placebo-controlled trial, Anesth. Analg., 98, 427, 2004.

82. Scott, E.R., Phipps, J.B., Gyory, J.R., and Padmanabhan, R.V., Electrotransport systemsfor transdermal delivery: a practical implementation of iontophoresis, in Handbook ofPharmaceutical Controlled Release Technology, Wise, D.L., Ed., Marcel Dekker, New York,2000, 617.

83. Thysman, S., Tasset, C., and Preat, V., Transdermal iontophoresis of fentanyl: delivery andmechanistic analysis, Int. J. Pharm., 101, 105, 1994.

84. Thysman, S. and Preat, V., In vivo iontophoresis of fentanyl and sufentanil in rats: pharma-cokinetics and acute antinociceptive effects, Anesth. Analg., 77, 61, 1993.

85. Ashburn, M.A., Streisand, J., Zhang, J., Love, G., Rowin, M., Niu, S., Kievit, J.K., Kroep, J.R.,and Mertens, M.J., The iontophoresis of fentanyl citrate in humans, Anesthesiology, 82, 1146,1995.

86. Gupta, S.K., Southam, M., Sathyan, G., and Klausner, M., Effect of current density onpharmacokinetics following continuous or intermittent input from a fentanyl electrotransportsystem, J. Pharm. Sci., 87, 976, 1998.

87. Gupta, S.K., Bernstein, K.J., Noorduin, H., Van Peer, A., Sathyan, G., and Haak, R., Fentanyldelivery from an electrotransport system: delivery is a function of total current, not durationof current, J. Clin. Pharmacol., 38, 951, 1998.

88. Gupta, S.K., Sathyan, G., Phipps, B., Klausner, M., and Southam, M., Reproducible fentanyldoses delivered intermittently at different time intervals from an electrotransport system,J. Pharm. Sci., 88, 835, 1999.

89. Viscusi, E.R., Reynolds, L., Chung, F., Atkinson, L.E., and Khanna, S., Patient-controlledtransdermal fentanyl hydrochloride vs. intravenous morphine pump for postoperative pain.A randomized controlled trial, JAMA, 291, 1333, 2004.

90. Panchagnula, R., Pillai, O., Nair, V.B., and Ramarao, P., Transdermal iontophoresis revisited,Curr. Opin. Chem. Biol., 4, 468, 2000.

91. Gudeman, S.D., Eisele, S.A., Heidt, R.S., Jr., Colosimo, A.J., and Stroupe, A.L., Treatment ofplantar fasciitis by iontophoresis of 0.4% dexamethasone. A randomized, double-blind,placebo-controlled study, Am. J. Sports Med., 25, 312, 1997.

92. Li, L.C., Scudds, R.A., Heck, C.S., and Harth, M., The efficacy of dexamethasone iontophoresisfor the treatment of rheumatoid arthritic knees: a pilot study, Arthritis Care Res., 9, 126, 1996.

93. Roques, C.-F., Agents physiques antalgiques: donnees cliniques actuelles: analgesic physicaltherapy: present clinical data, Ann. Readapt. Med. Phys., 46, 565, 2003.

94. Glass, J.M., Stephen, R.L., and Jacobson, S.C., The quantity and distribution of radiolabeleddexamethasone delivered to tissue by iontophoresis, Int. J. Dermatol., 19, 519, 1980.

95. Neeter, C., Thomee, R., Silbernagel, K.G., Thomee, P., and Karlsson, J., Iontophoresis with orwithout dexamethasone in the treatment of acute Achilles tendon pain, Scan. J. Med. Sci.Sports, 13, 376, 2003.

96. Psaki, C.G. and Carroll, J., Acetic acid ionization; a study to determine the absorptive effectsupon calcified tendinitis of the shoulder, Phys. Ther. Rev., 35, 84, 1955.

97. Kahn, J., Acetic acid iontophoresis for calcium deposits, Phys. Ther., 57, 658, 1977.98. Leduc, B.E., Caya, J., Tremblay, S., Bureau, N.J., and Dumont, M., Treatment of calcifying

tendinitis of the shoulder by acetic acid iontophoresis: a double-blind randomized controlledtrial, Arch. Phys. Med. Rehabil., 84, 1523, 2003.

99. Gangarosa, L.P., Sr., Ozawa, A., Ohkido, M., Shimomura, Y., and Hill, J.M., Iontophoresis forenhancing penetration of dermatologic and antiviral drugs, J. Dermatol., 22, 865, 1995.


Iontophoresis & 213


100. Wilson, J.M., Fry, B.W., Walton, R.E., and Gangarosa, L.P., Fluoride levels in dentin afteriontophoresis of 2% sodium fluoride, J. Dent. Res., 63, 897, 1984.

101. Puapichartdumrong, P., Ikeda, H., and Suda, H., Facilitation of iontophoretic drug deliverythrough intact and caries-affected dentine, Int. Endod. J., 36, 674, 2003.

102. Voigt, M., Kralinger, M., Kieselbach, G., Chapon, P., Anagnoste, S., Hayden, B., and Parel, J.,Ocular aspirin distribution: a comparison of intravenous, topical, and coulomb-controllediontophoresis administration, Invest. Ophthalmol. Vis. Sci., 43, 3299, 2002.

103. Von Sallmann, L. and Meyer, K., Penetration of penicillin into the eye, Arch. Ophthalmol., 31,1, 1944.

104. Halhal, M., Renard, G., Courtois, Y., BenEzra, D., and Behar-Cohen, F., Iontophoresis: fromthe lab to the bed side, Exp. Eye. Res., 78, 751, 2004.

105. Monti, D., Saccomani, L., Chetoni, P., Burgalassi, S., and Saettone, M.F., Effect of iontophor-esis on transcorneal permeation in vitro of two [beta]-blocking agents, and on cornealhydration, Int. J. Pharm., 250, 423, 2003.

106. Echols, D.F., Norris, C.H., and Tabb, H.G., Anesthesia of the ear by iontophoresis of lidocaine,Arch. Otolaryngol., 101, 418, 1975.

107. Greminger, R.F., Elliott, R.A., Jr., and Rapperport, A., Antibiotic iontophoresis for the man-agement of burned ear chondritis, Plast. Reconstr. Surg., 66, 356, 1980.

108. Harris, R., Iontophoresis, in Therapeutic Electricity and Ultraviolet Radiation, 2nd ed., Licht,S., Ed., Waverly Press, Baltimore, MD, 1967, 156.

109. Tashiro, Y., Sami, M., Shichibe, S., Kato, T., Hayakawa, E., and Itoh, K., Effect of lipophilicityon in vivo iontophoretic delivery. II. Beta-blockers, Biol. Pharm. Bull., 24, 671, 2001.

110. Conjeevaram, R., Chaturvedula, A., Betageri, G.V., Sunkara, G., and Banga, A.K., Iontophore-tic in vivo transdermal delivery of beta-blockers in hairless rats and reduced skin irritation byliposomal formulation, Pharm. Res., 20, 1496, 2003.

111. Zakzewski, C.A. and Li, J.K.J., Pulsed mode constant current iontophoretic transdermalmetoprolol tartrate delivery in established acute hypertensive rabbits, J. Control. Release,17, 157, 1991.

112. Kanikkannan, N., Singh, J., and Ramarao, P., Transdermal iontophoretic delivery of timololmaleate in albino rabbits, Int. J. Pharm., 197, 69, 2000.

113. Zakzewski, C.A., Amory, D.W., Jasaitis, D.K., and Li, J.K., Iontophoretically enhanced trans-dermal delivery of an ACE inhibitor in induced hypertensive rabbits: preliminary report,Cardiovasc. Drugs Ther., 6, 589, 1992.

114. Shigeki, S., Murakami, T., Kiyonaka, G., Yata, N., and Ikuta, Y., Transdermal iontophoreticdelivery of triamcinolone acetonide: a preliminary study in hairless rats, Scand. J. Plast.Reconstr. Surg. Hand Surg., 30, 177, 1996.

115. Shigeki, S., Uda, H., Ikuta, Y., Murakami, T., and Yata, N., Evaluation of iontophoretictransdermal delivery for the treatment of keloids and hypertrophic scars using triamcinoloneacetonide and tranilast, Drug Deliv. Syst., 11, 43, 1996.

116. Shigeki, S., Murakami, T., Yata, N., and Ikuta, Y., Treatment of keloid and hypertrophic scarsby iontophoretic transdermal delivery of tranilast, Scand. J. Plast. Reconstr. Surg. Hand Surg.,31, 151, 1997.

117. DeNuzzio, J., Boericke, K., Sutter, D., McFarland, A., Dey, D., Cesarini, R., Monty, E., Colville,D., and Bock, R., Iontophoretic delivery of buprenorphine, in Proceedings of the Inter-national Symposium on Controlled Release of Bioactive Materials, Vol. 23, Controlled ReleaseSociety, Inc., Minneapolis, MN, 1996, 285.

118. Tashiro, Y., Shichibe, S., Kato, Y., Hayakawa, E., and Itoh, K., Effect of lipophilicity on in vivoiontophoretic delivery. I. NSAIDs, Biol. Pharm. Bull., 24, 278, 2001.

119. Luzardo-Alvarez, A., Delgado-Charro, M.B., and Blanco-Mendez, J., In vivo iontophoreticadministration of ropinirole hydrochloride, J. Pharm. Sci., 92, 2441, 2003.

120. van der Geest, R., Hueber, F., Szoka, F.C., and Guy, R.H., Iontophoresis of bases, nucleosidesand nucleotides, Pharm. Res., 13, 553, 2000.




121. Brand, R.M., Hannah, T.L., Norris, J., and Iversen, P.L., Transdermal delivery of antisenseoligonucleotides can induce changes in gene expression in vivo, Antisense Nucleic Acid DrugDev., 11, 1, 2001.

122. Sakamoto, T., Miyazaki, E., Aramaki, Y., Arima, H., Takahashi, M., Kato, Y., Koga, M., andTsuchiya, S., Improvement of dermatitis by iontophoretically delivered antisense oligonu-cleotides for interleukin-10 in NC/Nga mice, Gene Ther., 11, 317, 2004.

123. Berdugo, M., Valamanesh, F., Andrieu, C., Klein, C., BenEzra, D., Courtois, Y., and Behar-Cohen, F., Delivery of antisense oligonucleotide to the cornea by iontophoresis, AntisenseNucleic Acid Drug Dev., 13, 107, 2003.

124. Langkjaer, L., Brange, J., Grodsky, G.M., and Guy, R.H., Iontophoresis of monomeric insulinanalogues in vitro : effects of insulin charge and skin pretreatment, J. Control. Release, 51, 47,1998.

125. Banga, A.K. and Chien, Y.W., Characterization of in vitro transdermal iontophoretic deliveryof insulin, Drug Dev. Ind. Pharm., 19, 2069, 1993.

126. Chien, Y.W., Siddiqui, O., Sun, Y., Shi, W.M., and Liu, J.C., Transdermal iontophoretic deliveryof therapeutic peptides/proteins-I: Insulin, in Biological approaches to the Controlled Deliveryof Drugs, Juliano, R.L., Ed., The New York Academy of Sciences, New York, 1987, 32.

127. Pillai, O., Borkute, S.D., Sivaprasad, N., and Panchagnula, R., Transdermal iontophoresis ofinsulin II. Physicochemical considerations, Int. J. Pharm., 254, 271, 2003.

128. Pillai, O. and Panchagnula, R., Transdermal iontophoresis of insulin V. Effect of terpenes,J. Control. Release, 88, 287, 2003.

129. Pillai, O., Nair, V., and Panchagnula, R., Transdermal iontophoresis of insulin: IV. Influence ofchemical enhancers, Int. J. Pharm., 269, 109, 2004.

130. Kanikkannan, N., Singh, J., and Ramarao, P., Transdermal iontophoretic delivery of bovineinsulin and monomeric human insulin analog, J. Control. Release, 59, 99, 1999.

131. Liu, J.C., Sun, Y., Siddiqui, O., Chien, Y.W., Shi, W.M., and Li, J., Blood glucose controlin diabetic rats by transdermal iontophoretic delivery of insulin, Int. J. Pharm., 44, 197,1988.

132. Siddiqui, O., Sun, Y., Liu, J.C., and Chien, Y.W., Facilitated transdermal transport of insulin,J. Pharm. Sci., 76, 341, 1987.

133. Kari, B., Control of blood glucose levels in alloxan-diabetic rabbits by iontophoresis ofinsulin, Diabetes, 35, 217, 1986.

134. Tomohira, Y., Machida, Y., Onishi, H., and Nagai, T., Iontophoretic transdermal absorption ofinsulin and calcitonin in rats with newly-devised switching technique and addition of urea,Int. J. Pharm., 155, 231, 1997.

135. Sage, B.H. and Hoke, R.A., Molecules for Iontophoretic Delivery, United States PatentNumber 5, 494, 1996.

136. Iwakura, Y. and Morimoto, K., Transdermal iontophoretic delivery of vasopressin and itsanalog in rats, S. T. P. Pharma Sciences, 1, 387, 1991.

137. Morimoto, K., Iwakura, Y., Miyazaki, M., and Nakatani, E., Effects of proteolytic enzymeinhibitors of enhancement of transdermal iontophoretic delivery of vasopressin and ananalogue in rats, Int. J. Pharm., 81, 119, 1992.

138. Banga, A.K., Katakam, M., and Mitra, R., Transdermal iontophoretic delivery and degradationof vasopressin across human cadaver skin, Int. J. Pharm., 116, 211, 1995.

139. Nakakura, M., Kato, Y., Hayakawa, E., Ito, K., and Kuroda, T., Effect of pulse on iontophoreticdelivery of desmopressin acetate in rats, Biol. Pharm. Bull., 19, 738, 1996.

140. Lee, W.A., Ennis, R.D., Longenecker, J.P., and Bengtsson, P., The bioavailability of intranasalsalmon calcitonin in healthy volunteers with and without a permeation enhancer, Pharm.Res., 11, 747, 1994.

141. Morimoto, K., Iwakura, Y., Nakatani, E., Miyazaki, M., and Tojima, H., Effects of proteolyticenzyme inhibitors as absorption enhancers on the transdermal iontophoretic delivery ofcalcitonin in rats, J. Pharm. Pharmacol., 44, 216, 1992.


Iontophoresis & 215


142. Thysman, S., Hanchard, C., and Preat, V., Human calcitonin delivery in rats by iontophoresis,J. Pharm. Pharmacol., 46, 725, 1994.

143. Thysman, S., Hanchard, C., and Preat, V., Transdermal delivery of calcitonin by iontophoresis,J. Control. Release, 29, 395, 1994.

144. Santi, P., Volpato, N.M., Bettini, R., Catellani, P.L., Massimo, G., and Colombo, P., Transdermaliontophoresis of salmon calcitonin can reproduce the hypocalcemic effect of intravenousadministration, Farmaco, 52, 445, 1997.

145. Santi, P., Colombo, P., Bettini, R., Catellani, P.L., Minutello, A., and Volpato, N.M., Drugreservoir composition and transport of salmon calcitonin in transdermal iontophoresis,Pharm. Res., 14, 63, 1997.

146. Chang, S.L., Hofmann, G.A., Zhang, L., Deftos, L.J., and Banga, A.K., Transdermal ionto-phoretic delivery of salmon calcitonin, Int. J. Pharm., 200, 107, 2000.

147. Nakamura, K., Katagai, K., Mori, K., Higo, N., Sato, S., and Yamamoto, K., Transdermaladministration of salmon calcitonin by pulse depolarization–iontophoresis in rats, Int.J. Pharm., 218, 93, 2001.

148. Burke, B., Mulryan, G., Foley, D., Meehan, E., McKenna, B., Devane, J., and Hession, P.,Factors affecting transdermal delivery of salmon calcitonin, J. Control. Release, 46, 180, 1997.

149. Suzuki, Y., Nagase, Y., Iga, K., Kawase, M., Oka, M., Yanai, S., Matsumoto, Y., Nagakawa, S.,Fukuda, T., Adachi, H., Higo, N., and Ogawa, Y., Prevention of bone loss in ovariectomizedrats by pulsatile transdermal iontophoretic administration of human PTH(1–34), J. Pharm.Sci., 91, 350, 2002.

150. Suzuki, Y., Iga, K., Yanai, S., Matsumoto, Y., Kawase, M., Fukuda, T., Adachi, H., Higo, N.,and Ogawa, Y., Iontophoretic pulsatile transdermal delivery of human parathyroid hormone(1–34), J. Pharm. Pharmacol., 53, 1227, 2001.

151. Heit, M.C., Williams, P.L., Jayes, F.L., Chang, S.K., and Riviere, J.E., Transdermal iontophoreticpeptide delivery: in vitro and in vivo studies with luteinizing hormone releasing hormone,J. Pharm. Sci., 82, 240, 1993.

152. Raiman, J., Koljonen, M., Huikko, K., Kostiainen, R., and Hirvonen, J., Delivery and stability ofLHRH and Nafarelin in human skin: the effect of constant/pulsed iontophoresis, Eur.J. Pharm. Sci., 21, 371, 2004.

153. Kumar, S., Char, H., Patel, S., Piemontese, D., Malick, A.W., Iqbal, K., Neugroschel, E., andBehl, C.R., In vivo transdermal iontophoretic delivery of growth hormone releasing factorGRF (1–44) in hairless guinea pigs, J. Control. Release, 18, 213, 1992.

154. Ellens, H., Lai, Z., Marcello, J., Davis, C.B., Cheng, H.Y., Oh, C.K., and Okabe, K., Transdermaliontophoretic delivery of [3H]GHRP in rats, Int. J. Pharm., 159, 1, 1997.

155. Lau, D.T.W., Sharkey, J.W., Petryk, L., Mancuso, F.A., Yu, Z., and Tse, F.L.S., Effect of currentmagnitude and drug concentration on iontophoretic delivery of octreotide acetate (Sandos-tatin) in the rabbit, Pharm. Res., 11, 1742, 1994.

156. Gangarosa, L.P., Merchant, H.W., Park, N.H., and Hill, J.M., Iontophoretic application ofidoxuridine for recurrent herpes labialis: report of preliminary clinical trials, Methods Find.Exp. Clin. Pharmacol., 1, 105, 1979.

157. Gangarosa, L.P., Sr., Hill, J.M., Thompson, B.L., Leggett, C., and Rissing, J.P., Iontophoresis ofvidarabine monophosphate for herpes orolabialis, J. Infect. Dis., 154, 930, 1986.

158. Volpato, N.M., Santi, P., and Colombo, P., Iontophoresis enhances the transport of acyclovirthrough nude mouse skin by electrorepulsion and electroosmosis, Pharm. Res., 12, 1623,1995.

159. Volpato, N.M., Nicoli, S., Laureri, C., Colombo, P., and Santi, P., In vitro acyclovir distributionin human skin layers after transdermal iontophoresis, J. Control. Release, 50, 291, 1998.

160. Chang, B.K., Guthrie, T.H., Jr., Hayakawa, K., and Gangarosa, L.P., Sr., A pilot study ofiontophoretic cisplatin chemotherapy of basal and squamous cell carcinomas of the skin,Arch. Dermatol., 129, 425, 1993.

161. Luxenberg, M.N. and Guthrie, T.H., Jr., Chemotherapy of basal cell and squamous cellcarcinoma of the eyelids and periorbital tissues, Ophthalmology, 93, 504, 1986.




162. Bacro, T.R., Holladay, E.B., Stith, M.J., Maize, J.C., and Smith, C.M., Iontophoresis treatment ofbasal cell carcinoma with cisplatin: a case report, Cancer Detect. Prev., 24, 610, 2000.

163. Smith, K.J., Konzelman, J.L., Lombardo, F.A., Skelton, H.G., III, Holland, T.T., Yeager, J.,Wagner, K.F., Oster, C.N., and Chung, R., Iontophoresis of vinblastine into normal skin andfor treatment of Kaposi’’s sarcoma in human immunodeficiency virus-positive patients. TheMilitary Medical Consortium for Applied Retroviral Research, Arch. Dermatol., 128, 1365,1992.

164. Welch, M.L., Grabski, W.J., McCollough, M.L., Skelton, H.G., Smith, K.J., Menon, P.A., andAnderson, L.L., 5-Fluorouracil iontophoretic therapy for Bowen’s disease, J. Am. Acad.Dermatol., 36, 956, 1997.

165. Rhodes, L.E., Tsoukas, M.M., Anderson, R.R., and Kollias, N., Iontophoretic delivery of ALAprovides a quantitative model for ALA pharmacokinetics and PpIX phototoxicity in humanskin, J. Invest. Dermatol., 108, 87, 1997.

166. Gerscher, S., Connelly, J.P., Griffiths, J., Brown, S.B., MacRobert, A.J., Wong, G., and Rhodes,L.E., Comparison of the pharmacokinetics and phototoxicity of protoporphyrin IX metabol-ized from 5-aminolevulinic acid and two derivatives in human skin in vivo, Photochem.Photobiol., 72, 569, 2000.

167. Gerscher, S., Connelly, J.P., Beijersbergen Van Henegouwen, G.M.J., MacRobert, A.J., Watt, P.,and Rhodes, L.E., A quantitative assessment of protoporphyrin IX metabolism and photo-toxicity in human skin following dose-controlled delivery of the prodrugs 5-aminolaevulinicacid and 5-aminolaevulinic acid-n-pentylester, Br. J. Dermatol., 144, 983, 2001.

168. Tiwari, S.B., Kumar, B.C.R., Udupa, N., and Balachandran, C., Topical methotrexate deliveredby iontophoresis in the treatment of recalcitrant psoriais — a case report, Int. J. Dermatol., 42,157, 2003.

169. Ashburn, M.A., Stephen, R.L., Ackerman, E., Petelenz, T.J., Hare, B., Pace, N.L., and Hofman,A.A., Iontophoretic delivery of morphine for postoperative analgesia, J. Pain SymptomManage., 7, 27, 1992.

170. Garagiola, U., Dacatra, U., Braconaro, F., Porretti, E., Pisetti, A., and Azzolini, V., Iontophore-tic administration of pirprofen or lysine soluble aspirin in the treatment of rheumatic diseases,Clin. Ther., 10, 553, 1988.

171. Puccetti, G., Galassi, A., Gigli, P., Pierallini, F.M., Pieraccini, A., and Scalabrino, A., Iontophor-esis of tenoxicam in the therapy of osteoarthrosis, Clin. Ter., 140, 239, 1992.

172. Macchia, L., Caiaffa, M.F., Vacca, A., and Tursi, A., General adverse reaction to aspirinadministered by transdermal iontophoresis, Eur. J. Clin. Pharmacol., 58, 641, 2002.

173. Macchia, L., Caiaffa, M.F., di Gioia, R., and Tursi, A., Systemic adverse reaction to diclofenacadministered by transdermal iontophoresis, Allergy, 59, 367, 2004.

174. Panus, P.C., Campbell, J., Kulkarni, S.B., Herrick, R.T., Ravis, W.R., and Banga, A.K., Trans-dermal iontophoretic delivery of ketoprofen through human cadaver skin and in humans,J. Control. Release, 44, 113, 1997.

175. Saggini, R., Zoppi, M., Vecchiet, F., Gatteschi, L., Obletter, G., and Giamberardino, M.A.,Comparison of electromotive drug administration with ketorolac or with placebo in patientswith pain from rheumatic disease: a double-masked study, Clin. Ther., 18, 1169, 1996.

176. Vecchini, L. and Grossi, E., Ionization with diclofenac sodium in rheumatic disorders: adouble-blind placebo-controlled trial, J. Int. Med. Res., 12, 346, 1984.

177. Foti, C., Cassano, N., Conserva, A., and Vena, G.A., Allergic contact dermatitis due todiclofenac applied with iontophoresis, Clin. Exp. Dermatol., 29, 91, 2004.

178. Curdy, C., Kalia, Y.N., Naik, A., and Guy, R.H., Piroxicam delivery into human stratumcorneum in vivo: iontophoresis versus passive diffusion, J. Control. Release, 76, 73, 2001.

179. James, M.P., Graham, R.M., and English, J., Percutaneous iontophoresis of prednisolone — apharmacokinetic study, Clin. Exp. Dermatol., 11, 54, 1986.

180. Jadoul, A., Mesens, J., Caers, W., de Beukelaar, F., Crabbe, R., and Preat, V., Transdermalpermeation of alniditan by iontophoresis: in vitro optimization and human pharmacokineticdata, Pharm. Res., 13, 1348, 1996.


Iontophoresis & 217


181. van der Geest, R., Danhof, M., and Bodde, H.E., Iontophoretic delivery of apomorphine. I:in vitro optimization and validation, Pharm. Res., 14, 1798, 1997.

182. Danhof, M., van der Geest, R., Van Laar, T., and Bodde, H.E., An integrated pharmacokinetic–pharmacodynamic approach to optimization of R-apomorphine delivery in Parkinson’s dis-ease, Adv. Drug Deliv. Rev., 33, 253, 1998.

183. Li, G.L., Danhof, M., and Bouwstra, J.A., Iontophoretic delivery of apomorphine in vitro:physicochemic considerations, Pharm. Res., 18, 1509, 2001.

184. van der Geest, R., Van Laar, T., Gubbens-Stibbe, J.M., Bodde, H.E., and Danhof, M., ionto-phoretic delivery of apomorphine II: an in vivo study in patients with Parkinson’s disease,Pharm. Res., 14, 1804, 1997.

185. Kankkunen, T., Sulkava, R., Vuorio, M., Kontturi, K., and Hirvonen, J., Transdermal ionto-phoresis of tacrine in vivo, Pharm. Res., 19, 704, 2002.

186. Cormier, M., Chao, S.T., Gupta, S.K., and Haak, R., Effect of transdermal iontophoresiscodelivery of hydrocortisone on metoclopramide pharmacokinetics and skin-induced reac-tions in human subjects, J. Pharm. Sci., 88, 1030, 1999.

187. Meyer, B.R., Kreis, W., Eschbach, J., O’Mara, V., Rosen, S., and Sibalis, D., Successful trans-dermal administration of therapeutic doses of a polypeptide to normal human volunteers,Clin. Pharmacol. Ther., 44, 607, 1988.

188. Meyer, B.R., Kreis, W., Eschbach, J., O’Mara, V., Rosen, S., and Sibalis, D., Transdermal versussubcutaneous leuprolide: a comparison of acute pharmacodynamic effect, Clin. Pharmacol.Ther., 48, 340, 1990.

189. Lu, M.F., Lee, D., Carlson, R., Rao, G.S., Hui, H.W., Adjei, L., Herrin, M., Sundberg, D., andHsu, L., The effects of formulation variables on iontophoretic transdermal delivery of leu-prolide to humans, Drug Dev. Ind. Pharm., 19, 1557, 1993.

190. Green, P.G., Iontophoretic delivery of peptide drugs, J. Control. Release, 41, 33, 1996.191. Wearley, L., Liu, J.C., and Chien, Y.W., Iontophoresis-facilitated transdermal delivery of

verapamil. I. In vitro evaluation and mechanistic studies, J. Control. Release, 8, 237, 1989.192. Wearley, L., Liu, J.C., and Chien, Y.W., Iontophoresis-facilitated transdermal delivery of

verapamil. II. Factors affecting the reversibility of skin permeability, J. Control. Release, 9,231, 1989.

193. Wearley, L.L. and Chien, Y.W., Iontophoretic transdermal permeation of verapamil (III): effectof binding and concentration gradient on reversibility of skin permeation rate, Int. J. Pharm.,59, 87, 1990.

194. Jadoul, A., Hanchard, C., Thysman, S., and Preat, V., Quantification and localization offentanyl and trh delivered by iontophoresis in the skin, Int. J. Pharm., 120, 221, 1995.

195. Huang, Y.Y., Wu, S.M., and Wang, C.Y., Response surface method: a novel strategy tooptimize iontophoretic transdermal delivery of thyrotropin-releasing hormone, Pharm.Res., 13, 547, 1996.

196. Miller, L.L., Kolaskie, C.J., Smith, G.A., and Rivier, J., Transdermal iontophoresis of gonado-tropin releasing hormone (LHRH) and two analogues, J. Pharm. Sci., 79, 490, 1990.

197. Kumar, S., Char, H., Patel, S., Piemontese, D., Iqbal, K., Malick, A.W., Neugroschel, E., andBehl, C.R., Effect of iontophoresis on in vitro skin permeation of an analogue of growthhormone releasing factor in the hairless guinea pig model, J. Pharm. Sci., 81, 635, 1992.

198. Laneri, S., Sacchi, A., Abignente di Frassello, E., Luraschi, E., Colombo, P., and Santi, P.,Ionized prodrugs of dehydroepiandrosterone for transdermal iontophoretic delivery, Pharm.Res., 16, 1818, 1999.

199. Jadoul, A. and Preat, V., Electrically enhanced transdermal delivery of domperidone, Int.J. Pharm., 154, 229, 1997.

200. Wearley, L. and Chien, Y.W., Enhancement of the in vitro skin permeability of azidothymidine(AZT) via iontophoresis and chemical enhancer, Pharm. Res., 7, 34, 1990.

201. Oh, S.Y., Jeong, S.Y., Park, T.G., and Lee, J.H., Enhanced transdermal delivery of AZT(Zidovudine) using iontophoresis and penetration enhancer, J. Control. Release, 51, 161,1998.




202. Hirvonen, J., Kontturi, K., Murtomaki, L., Paronen, P., and Urtti, A., Transdermal iontophor-esis of sotalol and salicylate; the effect of skin charge and penetration enhancers, J. Control.Release, 26, 109, 1993.

203. Hirvonen, J., Murtomaki, L., and Kontturi, K., Experimental verification of the mechanisticmodel for transdermal transport including iontophoresis, J. Control. Release, 56, 169, 1998.

204. Denet, A.R., Ucakar, B., and Preat, V., Transdermal delivery of timolol and atenolol usingelectroporation and iontophoresis in combination: a mechanistic approach, Pharm. Res., 20,1946, 2003.

205. Ruland, A., Rohr, U., and Kreuter, J., Transdermal delivery of the tetrapeptide hisetal (mela-notropin (6–9)) and amino acids: their contribution to the elucidation of the existence of an‘aqueous pore’ pathway, Int. J. Pharm., 107, 23, 1994.

206. Craan-van Hinsberg, W.H.M., Bax, L., Flinterman, N.H.M., Verhoef, J., Junginger, H.E., andBodde, H.E., Iontophoresis of a model peptide across human skin in vitro; effects ofiontophoresis protocol, pH, and ionic strength on peptide flux and skin impedance,Pharm. Res., 11, 1296, 1994.

207. Marconi, B., Mancini, F., Colombo, P., Allegra, F., Giordano, F., Gazzaniga, A., Orecchia, G.,and Santi, P., Distribution of khellin in excised human skin following iontophoresis andpassive dermal transport, J. Control. Release, 60, 261, 1999.

208. Sung, K.C., Fang, J.Y., and Yoa-Pu Hu, O., Delivery of nalbuphine and its prodrugs acrossskin by passive diffusion and iontophoresis, J. Control. Release, 67, 1, 2000.

209. Nugroho, A.K., Li, G., Grossklaus, A., Danhof, M., and Bouwstra, J.A., Transdermal ionto-phoresis of rotigotine: influence of concentration, temperature and current density in humanskin in vitro, J. Control. Release, 96, 159, 2004.

210. Nugroho, A.K., Li, G.L., Danhof, M., and Bouwstra, J.A., Transdermal iontophoresis ofrotigotine across human stratum corneum in vitro: influence of ph and NaCl concentration,Pharm. Res., 21, 844, 2004.

211. Adachi, H., Higo, N., Sato, S., Naik, A., Kalia, Y.N., and Guy, R.H., Anti-microorganism agentdosing device and assembly, JP 2002–124855, 2003.

212. Stephen, R., Miotti, D., Bettaglio, R., Rossi, C., and Bonezzi, C., Electromotive administrationof a new morphine formulation: morphine citrate, Artif. Organs, 18, 461, 1994.

213. Riviere, J.E., Williams, P.L., Hillman, R.S., and Mishky, L.M., Quantitative prediction oftransdermal iontophoretic delivery of arbutamine in humans with the in vitro isolatedperfused porcine skin flap, J. Pharm. Sci., 81, 504, 1992.


Iontophoresis & 219


Chapter 15

Electroporation

Babu M. Medi and Jagdish Singh

CONTENTS

Introduction .................................................................................................................................... 221Mechanisms of Percutaneous Penetration Enhancement ............................................................. 222

Expansion of Preexisting Pathways ........................................................................................... 223Creation of New Pathways......................................................................................................... 223Thermal Effects Due to Electroporation .................................................................................... 224

Factors Influencing Percutaneous Penetration Enhancement by Electroporation ...................... 224Electrical Parameters................................................................................................................... 224Physicochemical Factors............................................................................................................. 226

Effects of Electroporation on Skin ................................................................................................. 228Potential Applications..................................................................................................................... 232Conclusions and Future Prospects................................................................................................. 234Acknowledgment ........................................................................................................................... 234References....................................................................................................................................... 235

Introduction

The administration of drugs to skin is being practiced for centuries to treat local diseases,but this route is being used recently for systemic delivery of therapeutic agents. This routeof administration is of special interest because of the advantages offered over other routesincluding avoidance of gastric degradation and first-pass metabolism in addition tosuperior patient compliance. However, the major limitation of this route of drug admin-istration is that the skin is permeable to only small lipophilic drugs and is highlyimpermeable to hydrophilic and macromolecular drugs. This barrier property is mainlyattributed to the largely lipophilic outermost layer of the skin, stratum corneum (SC) [1].The successful transdermal or topical formulation of a drug depends on the permeationrate of the drug across the skin or into the skin to achieve therapeutic levels. As manyof the drugs lack favorable physicochemical properties for percutaneous absorption,


221


percutaneous penetration enhancers are promising in the development of transdermalformulations. An ideal percutaneous penetration enhancer should promote the transportof drugs across or into the skin in a predictable way without any irreversible effects on theskin barrier properties. Several investigations delved into this aspect and studied differentenhancement methods including chemical and physical methods to overcome the barrierproperties of skin [2–5]. The present chapter focuses on electroporation, an electricalmethod to enhance the transport of drug molecules across or into the skin by overcomingthe barrier of the SC.

The use of electrical current for enhancing the percutaneous penetration of drugs,that are otherwise impermeable, was known for long time. Nearly a century ago,Leduc demonstrated the transdermal delivery of strychnine and potassium cyanide intorabbits using a low voltage electric current, known as iontophoresis [6]. In contrast,electroporation is a relatively new technique of percutaneous penetration enhance-ment [7], which makes use of high voltage electric current. Both iontophoresis andelectroporation use electric current to enhance the percutaneous absorption of drugsand macromolecules, with the difference being that the iontophoresis acts primarily onthe drug molecule while electroporation acts on the skin structure as well as, to someextent, on the drug molecules. Electroporation is a physical method of percutaneouspenetration enhancement using electrical pulses. It involves application of controlled,high voltage electric pulses of very short duration (microsecond–millisecond) to enhancethe tissue permeability reversibly [8]. Electroporation may not show significant differenceover the other enhancement methods for the transdermal delivery of small ions ormolecules but shows dramatically higher fluxes of macromolecules in comparisonto other enhancement methods. The major advantage of this technique is that themacromolecules such as peptide and gene-based drugs could become potentialcandidates for transdermal delivery [9–11]. The other potential advantages are thatwe can have a better control over the amount of drug delivered and kinetics of drugdelivery [12].

Mechanisms of Percutaneous Penetration Enhancement

Electroporation of cell membrane has been studied extensively and used since 1970s forDNA transfection of the cells by reversibly permeabilizing the cell membranes with theapplication of brief electric pulses [13–15]. Dielectic breakdown of a lipid bilayer and acell membrane occur at a transmembrane potential of about 0.5 and 1 V, respectively [16].The nature of SC makes it even attractive for electroporation. It is 10 to 40 mm thick and iscomposed of approximately 100 layers of flattened corneocytes with intercellular lipidbilayers. A transdermal voltage of higher than 50 V is required to electroporate multi-lipidbilayers present in the SC [17]. Although the detailed molecular mechanism of electro-poration is still not completely understood, application of strong electric field pulses tocells and tissue is known to cause some type of structural rearrangement of the cellmembrane. Many theoretical models have been put forward to explain the mechanismsof electroporation. However, there is a general agreement in the literature that theapplied field induces some sort of metastable structural defect in the membrane, whichserves as a pathway for macromolecular entry [7, 18, 19]. The main idea behind usingelectroporation for percutaneous penetration enhancement is to perturb the barrierproperty of SC to enhance the transport of drugs.




Expansion of Preexisting Pathways

A schematic drawing of the SC with the possible pathways of transport during electro-poration is shown in Figure 15.1. At relatively low voltages (<30 V across the skin), thedrop in skin resistance and enhanced transdermal transport can be mainly attributed toelectroporation of the appendageal ducts present in the skin [20]. The propagation of theelectric pulse through these ducts forces water and expands these pathways causingappendageal electroporation. This phenomenon does not show any dramatic increase inthe transdermal transport of molecules [21]. At higher voltages (>30 V), electroporationof the lipid–corneocyte matrix leads to an additional drop in skin resistance, which allowsdramatic increase of the drug transport [21, 22].

Creation of New Pathways

Application of electric pulses to skin results in dramatic increase of transdermal transportassociated with reversible structural changes in the skin [7]. Skin electroporation involvesthe exposure of tissue to short, high voltage electric pulses that are shown to causepermeabilization of skin for macromolecular entry. The permeabilization of SC is gener-ally believed to occur through the formation of aqueous pathways across the lipidbilayers of the SC. Electroporation alters lipid bilayers when transient electric fieldleads to the formation of nonlamellar lipid phases: a pore, also called localized transportregion (LTR). These new aqueous pathways are thought to be formed when water fromboth sides of the membrane meets due to the electric field force [23]. The pore mechan-ism for the enhanced transdermal transport is generally accepted. In addition to electro-poration, the local electric field also provides driving force for the small ions and watersoluble molecules to traverse the skin through these newly created pathways [24].

A

B C D

Figure 15.1 Schematic drawing of the stratum corneum with the possible pathways (preexistingand new) of transport during electroporation. Preexisting pathways including (A) via hair follicle;(B) intercellular, involving the gaps between corneocytes; (C) via sweat ducts; and (D) newlycreated pathway that goes through the corneocytes and lipid bilayers due to electroporationwere shown.


Electroporation & 223


Thermal Effects Due to Electroporation

When an electrical energy passes through a resistance, it is transformed into heat.According to the first law of thermodynamics, the electrical energy released into a systemwill increase the energy level of the sample. Thus, the heat production due to electricalenergy dissipation results in an increase in the sample temperature. However, theincreases in the temperature are not too drastic as the whole circuit is involved in theenergy dissipation, not just the sample [25]. Nevertheless, caution should be observedwhen multiple pulses of longer duration are applied. It is hypothesized that the applica-tion of electric pulses causes temperature rise in the SC during electroporation, whichmight further contribute to permeation enhancement. It was estimated by computersimulation that for a peak voltage of 70 V exponential decay pulse across the SC, thetemperature rise would be 198C [26]. This temperature rise occurs within localizedregions surrounding the LTRs and are called localized dissipation regions (LDRs)[27, 28]. The morphological changes studied using time-resolved freeze-fracture electronmicroscopy following electroporation revealed the formation of multilamellar vesicles of0.1 to 5.5 nm in diameter in the intercellular lipid bilayers of the SC [29]. These vesicleswere similar to those formed when the SC is heated to 658C, suggesting that thesechanges are related to the heating effect of the electric pulses [29]. The temperature risewithin the SC may alter the structure and prolong the recovery of the skin barrier afterelectroporation [30].

Factors Influencing Percutaneous Penetration Enhancementby Electroporation

There are several parameters influencing the extent of percutaneous penetration en-hancement of drug molecules using electroporation. These include both electrical param-eters associated with the pulses and physicochemical properties of the molecules to bedelivered.

Electrical Parameters

Type of the pulses: Two different types of pulses (wave forms), square wave [10, 31]and exponentially decaying [7, 32], are being investigated for percutaneous penetration

using fast switches. Basically, the power supply set to generate a given voltage isconnected to a square wave pulse generator, which closes the circuit at t¼ 0 and opensit at a defined time point later. Thus, the theoretical shape of the wave is as shown inFigure 15.2. Square wave pulses do not rely on capacitor discharge into the circuit as inthe case of exponential decay pulses. Skin electroporation using exponentially decaying

[33]. Reproducibility of exponential decay pulse might be problematic while this is not anissue with square wave pulses [34].

Pulsing parameters: The pulsing parameters, such as pulse amplitude, pulse length,number of pulses, and the interval between each pulse, can have dramatic effects on thetransport of drugs through the skin during electroporation. Pulse amplitude is reported tobe a critical parameter, which has a profound effect on the transdermal delivery of drugs.




enhancement (Figure 15.2). Square wave pulse electroporators generate a voltage pulse

pulses was shown to be more effective (Figure 15.3) than square wave electroporation

Sharma et al. [35] reported that the transport of terazosin hydrochloride through hairlessrat skin was enhanced linearly with pulse amplitude using exponentially decaying pulses.In another study, human parathyroid hormone (1–34), hPTH (1–34), delivery was shown

The pulse length and number of pulses also affect the extent of transdermal delivery as

an important role as a large number of pulses with a big time gap between them may notbe useful, since it allows the recovery of skin barrier before the application of next pulse.

Time

Vol

tage

Time

Vol

tage

(a)

(b)

Figure 15.2 Schematic drawing of the (a) square wave and (b) exponential decay pulses. Theelectrical pulse parameters, especially the duration of the pulses, can be controlled better in thecase of square wave pulse than in exponentially decaying pulses.

Cumulative fentanyl transported (ng/cm2)

600

500

400

300

200

100

00 2 4 6

Time (h)

Figure 15.3 Effect of the type of electroporation pulses applied on cumulative transport offentanyl through full thickness hairless rat skin. Key: (&) passive diffusion; () 5X (100 V–60 msec)square wave pulses; (^) 5X (250 V–60 msec) square wave pulses; (4) 5X (100 V–125 msecexponentially decaying pulses; (&) 5X (250 V–125 msec) exponentially decaying pulses. Fenta-nyl 40 mg/ml was introduced in a citrate buffer pH 5 (0.01 M). (Reproduced from Vanbever, R.,Boulenge, E.L., and Preat, V., Pharm. Res., 13, 559, 1996. With permission of Kluwer.)




to depend linearly on the pulse amplitude using square wave pulses (Figure 15.4) [10].

shown in Figure 15.5 and Figure 15.6, respectively. The pulsing frequency might also play

Physicochemical Factors

Molecular size and charge of the permeant: Both the size and charge of the drugmolecule play an important role in percutaneous absorption. Electroporation has been

shows the effect of molecular weight of permeant on the transdermal transport usingelectroporation. Electroporation has shown to increase transport of calcein [7], Leutiniz-

R 2 = 0.9767

0

10

20

30

40

0 100 200 300 400

Voltage applied (V)

hP

TH

(1-

34)

flu

x(n

mo

l/cm

2 / h)

1

09

Figure 15.4 Effect of electroporation pulse voltages on the flux of hPTH (1–34) throughdermatomed porcine skin. Twenty square wave pulses of 100 msec pulse length with 1 msecinterval between each pulse and of different voltage were applied at the beginning. Pulses wereapplied to 0.785 cm2 area of the skin. Values are shown as the mean+ SD of three determin-ations. (Reproduced from Medi, B.M. and Singh, J., Int. J. Pharm., 263, 25, 2003. With permis-sion of Elsevier.)

250

n = 5

200

150

100

50

0Control 10 20

Pulse length (ms)

Am

ou

nt

of

TR

Z d

eliv

ered

in s

kin

(mg

) ±

SE

30 40

Figure 15.5 Effect of pulse length on terazosin hydrochloride delivery. Twenty pulses at Uskin,0

88 V were delivered using small-area electrode. (Reproduced from Sharma, A., Kara, M., Smith,F.R., and Krishnan, T.R., J. Pharm. Sci., 89, 528, 2000. With permission of Wiley-Liss, Inc., asubsidiary of John Wiley & Sons, Inc.)




shown to enhance transdermal transport of a broad range of drug molecules. Figure 15.7

ing Hormone Releasing Hormone [36], heparin [37], oligonucleotides [38], FITC-Dextran[39], insulin [40] and hPTH (1–34) [10]. These studies suggest that electroporation can beuseful for the transdermal delivery of macromolecules that could not be transportedusing other enhancement methods.

pH of the formulation: The pH of the formulation is also an important factor that caninfluence the barrier properties of the skin in addition to its influence on the ionic state of

200

150

100

50

0

Control

n = 5

1 5 10Number of pulses

Am

ou

nt

of

TR

Z d

eliv

ered

in s

kin

(mg

) ± S

E

2015 25

Figure 15.6 Effect of number of pulses on terazosin hydrochloride delivery. The Uskin,0 was setat 88 V and the pulse length was set at 40 msec. (Reprinted from Sharma, A., Kara, M., Smith,F.R., and Krishnan, T.R., J. Pharm. Sci., 89, 528, 2000. With permission of Wiley-Liss, Inc., asubsidiary of John Wiley & Sons, Inc.)

800 FITCFD 4,4FD 12FD 38

600

400

Pm

ol/c

m2

200

00 1 2 3

Time (h)4 5 6

Figure 15.7 Effect of molecular weight of permeant on cumulative transdermal transport usingelectroporation. (Reproduced from Lombry, C., Dujardin, N., and Preat, V., Pharm. Res., 17, 32,2000. With permission of Kluwer.)




the drug. Enhanced percutaneous penetration of water was reported at pH lower than 4and higher than 10 due to the extraction of insoluble fraction of keratin [41]. It was alsoshown to decrease skin impedance at a pH lower than 3 and higher than 9 [42, 49].Murthy et al. [43] showed pH dependence of the electroporation-enhanced transportusing glucose and fluorescein isothiocyanate labeled dextran. They reported that thetransport of glucose across porcine epidermis was increased gradually when the formu-lation pH was increased from pH 5 to 7.5 (Figure 15.8), which might be due to theprolonged postpulse permeability state of the skin.

Effect of electrolytes: The presence of monovalent electrolytes such as NaF, NaCl,

2 and especially CaCl2

transport of calcein across the skin [44]. The presence of CaCl2 was shown to prolong thepostpulse recovery of the skin, which might be the reason for the enhanced transport incomparison to electroporation alone [45]. This suggests the possibility of further enhan-cing the skin permeation of drugs using electroporation.

Effect of temperature: It is well known that temperature affects the permeability ofthe diffusing drug molecules through skin [46, 47]. Recently, it was shown that an increasein temperature above 408C results in enhanced transport of molecules with electropor-

electroporation [30].

Effects of Electroporation on Skin

In order to be useful clinically, the permeabilization of SC should be reversible as it isthe primary barrier between the body and environment besides playing a critical role

00

10

20

30

40

50

60

2 4 6 8 10pH

Cu

mu

lati

ve t

ran

spo

rt (

mg)

Figure 15.8 Electroporation transport of glucose across porcine epidermis at different donorpH values (60 pulses at 100 V; 1 msec, 1 Hz 1 15 min postpulse duration). (Reproduced fromMurthy, S.N. et al., J. Control. Release, 93, 49, 2003. With permission of Elsevier.)




NaBr, NaI (Figure 15.9) and divalent electrolytes such as MgCl(Figure 15.10), was shown to have synergistic effect on the electroporation-enhanced

ation (Figure 15.11), which is likely due to the delayed recovery of the skin following

0.00 2 4 6

0.1

0.2

0.3

0.4

Time (h)

Cal

cein

per

mea

ted

(n

mo

l / c

m2 )

Figure 15.9 Effect of various monovalent electrolytes on electroporation-enhanced permeationof calcein through excised hairless rat skin. Symbols: , control (passive diffusion); ., distilledwater; ^, NaF; &, NaCl; &, NaBr; ~, NaI. (Reproduced from Tokudome, Y. and Sugibayashi, K.,J. Control. Release, 92, 93, 2003. With permission of Elsevier.)

1.0

0.8

0.6

Cal

cein

per

mea

ted

(n

mo

l / c

m2 )

0.4

0.2

0.00 2

Time (h)

4 6

Figure 15.10 Effect of various divalent electrolytes on electroporation-enhanced permeation ofcalcein through excised hairless rat skin. Symbols: , control (passive diffusion); ., distilledwater; &, CaCl2; ^, MgCl2; ~, CuCl2; 4, ZnCl2. Each point represents the mean+ SE of threeto five determinations. (Reproduced from Tokudome, Y. and Sugibayashi, K., J. Control. Release,92, 93, 2003. With permission of Elsevier.)




in regulating the homeostatic reactions. Unlike electroporation of simple lipid bilayersthat anneal immediately after ceasing the pulses, the complex lipid matrix of the SChas a slower return to normal permeability state [48]. Although not completely under-stood, application of strong electric field pulses to cells and tissue is known to causesome type of structural rearrangement of the cell membrane [49]. Sensitization andpain were reported with skin electroporation due to direct excitation of the underlyingnerves and muscles [50]. The clinical acceptability of this technique depends onthe demonstration of safety when used for transdermal drug delivery. The presentsection delineates the studies carried out so far to address the safety issues usingelectroporation.

Biophysical changes: Biophysical methods allow investigators to study the changesof SC lipids and protein in addition to the SC water content. Different methods includingFTIR, DTA and x-ray diffraction have been used to investigate these changes followingelectroporation treatment. ATR–FTIR studies show an increase in the water content of SC[51], which was also confirmed by thermogravimetric studies [52]. A dramatic perturb-ation in the lamellar ordering of the intercellular lipid has been reported after highvoltage pulsing using differential thermal analysis and freeze-fracture electron micro-scopy [52]. Small angle x-ray scattering studies carried out about 5min after electropor-ation pulsing provided further evidence for a general perturbation of interlamellar andintralamellar lipid packing order [53].

Histological changes: Histological examination of the skin after electroporationshowed intraepidermal edema and vacuolization [54]. An increased detachment of SC

poration pulse voltage [10]. Freeze-fracture electron microscopic studies revealed severe

00

0.5

1.5

2.5

3.5

1

2

3

10 20 30 40 50

Temperature (C)

Tot

al tr

ansp

ort o

f FD

10K

(µg

/cm

2 )

Figure 15.11 Total electroporation transport of FD10K across porcine epidermis at differenttemperatures. Porcine epidermis samples were subjected to 60 pulses, each of 1-msec durationat 100 V, 1 Hz. FD10K (5 mg/mL) was present in the donor chamber during the pulse applicationand for 15 min after pulsing. (Reprinted from Murthy, S.N. et al., J. Control. Release, 93, 49,2003. With permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.)




and an amorphous epidermis (Figure 15.12) were reported with an increase in electro-

distortion of the lamellar structure of the SC lipids [52]. Another study using time-resolvedfreeze-fracture electron microscopy revealed the formation of multilamellar vesicles of0.1 to 5.5 nm in diameter in the SC that could be related to the heating effect ofelectroporation [29].

Figure 15.12 Effect of electroporation on skin: (a) microscopic section of control porcine skin(without any electric pulses), 1003; (b) microscopic section of skin sample electroporated withpulses of 100 V; (c) microscopic section of skin treated with pulses of 200 V; (d) microscopicsection of skin treated with pulses of 300 V. (Reproduced from Medi, B.M. and Singh, J., Int.J. Pharm., 263, 25, 2003. With permission of Elsevier.)




Macroscopic barrier and skin irritation: The barrier property of the skin is criticalto prevent the entry of exogenous toxic chemicals into the body and also to avoid the lossof internal body components, particularly water [55]. The effect of electroporation onmacroscopic barrier property of the skin was studied by measuring transepidermal waterloss (TEWL) following electroporation in vivo in rats and rabbits [56, 57]. The studiesreported a reversible increase in TEWL following electroporation. It is shown to causemild, transient erythema and edema in New Zealand White rabbits [57]. Skin irritation wasmeasured at different time points following the visual scoring method of Draize et al. [58].It is suggested that the use of iontophoresis followed by electroporation pulses mightreduce the skin irritation [59, 60]. This may be due to the creation of new pathwayswith electroporation, which results in more even distribution of the iontophoreticcurrent [61].

Potential Applications

Transdermal or topical drug delivery: Since the demonstration of the electroporationfor enhanced transdermal delivery [7], numerous studies reported the delivery of severalmolecules. It can improve transdermal/topical delivery of drugs ranging from smallmolecules to macromolecules such as peptide drugs and nucleic acids (oligonucleotidesand genes). Enhanced transdermal delivery of macromolecules of at least up to 40 kDa

in vitro percutaneous penetration enhancement of different drugs using electroporation.Furthermore, skin electroporation in combination with other physical and chemical

percutaneous penetration enhancers has been explored for transdermal delivery. Theaim of combining chemical enhancers and electroporation is either to enlarge thepathways created by electroporation or to prolong the reversal of these pathwaysbut not to disrupt lipids [34]. Conventional chemical penetration enhancers may notbe useful for this purpose. Polysaccharides such as heparin [62], dextran [63], andanionic phospholipids [64] were found to enhance transdermal delivery by electropor-ation. The application of ultrasound along with electroporation is not expected to have adramatic effect as both these techniques have similar mechanisms of action [34]. How-ever, the application of iontophoresis in combination with electroporation is anticipatedto increase the transdermal transport synergistically due to the different mechanismsof action of these methods. Medi and Singh [10] reported that the combination ofelectroporation and iontophoresis synergistically enhances the flux of hPTH (1–34)

Gene delivery: Skin electroporation has the ability to permeabilize the cells of varioustissues, including keratinocytes of the skin [73, 76] and was investigated for skin-targetedgene delivery. Skin is an attractive target site for somatic gene transfer due to its large size,easy accessibility [77]. Gene transfer to skin can be potentially useful for the treatmentof local skin disorders and also for systemic disorders as it can produce and releasepolypeptides into systemic circulation [78–80]. Application of electrical pulses afterinjecting the DNA intrademally or topical application proved to be useful [81–84]. Topicaldelivery of a reporter plasmid, pEGFP-N1, using square wave electroporation pulses of1000 V having 10 msec enhanced the expression of the plasmid by fourfold in comparisonto the passive delivery [85]. The plasmid DNA entered the epidermis within minutes after




(Figure 15.13).

was shown to be feasible with electroporation. Table 15.1 provides a summary of the

Table 15.1 Summary of In Vitro Percutaneous Penetration Enhancement of Drugs Using Electroporation

Drug Electroporation Protocol Membrane Enhancement Reference(s)

5-Fluorouracil Twenty ED pulses of 300 V, 200 msec Nude mice 25-fold [65]Alniditan Five ED pulses 0f 100 V, 603 m sec Hairless rat 100-fold [66]Buprenorphine Twenty ED pulses of 500 V, 10 msec Porcine skin Several fold [67]Calcein Different parameters Human skin Up to 10,000-fold [7]Calcitonin ED pulses for 4 h (300 V, 1 ppm) in combination

with iontophoresis (5 mA/cm2)Human epidermis 2-fold [32]

Cyclosporin A Twenty five pulses of 200 V, 10 msec Hairless rat 60-fold [68]Heparin Different parameters Human Sskin Up to 100-fold [37]hPTH (1–34) Twenty SW pulses of 100 V, 100 msec Porcine skin 7-fold [10]

Twenty SW pulses of 200 V, 100 msec Porcine skin 16.5-fold [10]Twenty SW pulses of 300 V, 100 msec Porcine skin 20-fold [10]

Insulin SW pulses of 100–105 V (1 msec at 1 Hz) inthe presence of DMPS

Porcine epidermis 20-fold [40]

LHRH Single ED pulse of 1000 V, 5 msec followedby iontophoresis (0.5 mA/cm2)

Human skin 5- to 10-fold [36]

Single ED pulse of 500 V, 5 msec followed by30 min iontophoresis (0.5 mA/cm2)

IPPSF 2-fold [54]

Single ED pulse of 500 V, 5 msec every 10 minand 30 min iontophoresis (0.5 mA/cm2)

IPPSF 3-fold [54]

Mannitol Five ED pulses of 250 V, 330 msec Hairless rat Up to 100-fold [69]Metoprolol ED pulses of different voltages Hairless rat Several fold [70]Nalbuphine Twenty ED pulses of 500 V, 200 msec Nude mice 5-fold [71]Oligonucleotides Different parameters Human skin Up to 10-fold [38,72,73]Terazosin HCl Twenty pulses of 88 V (Uskin), 20 msec Hairless rat 14-fold [35]Tetracaine Forty pulses min1 for 10 min of 130 V, 0.4 sec Rat 6-fold [74]Timolol and Atenolol Ten SW pulses of 400 V, 10 msec followed by iontophoresis Human SC Several fold [75]Water Five ED pulses of 250 V, 330 msec Hairless rat 100-fold [69]

Note: Enhancement, a ratio of the transdermal flux or the amount of drug transported for electroporated skin versus a control value for skin not exposed to electroporation; DMPS,1,2-dimyristoyl-3-phosphatidylserine; ED, exponentially decaying; hPTH, human parathyroid hormone; IPPSF, isolated perfused porcine skin flap mode; LHRH, leuteinizinghormone releasing hormone; SW, square wave; Uskin, transdermal voltage.

Smith

and

Maib

ach/

Percu

taneous

Penetratio

nEnhan

cers

2nd

edn

TF2152_c0

15

Fin

alPro

of

pag

e233

8.9

.2005

10:4

5pm

Electro

po

ration

&233


electroporation and entered the keratinocytes cytoplasm within hours. However, topicalelectroporation using five square wave electric pulses of 300 V and 10 msec pulse lengthafter intradermal injection of a reporter plasmid (gWiz b-Gal) resulted in over 100-foldenhancement of gene expression compared to passive injection [86]. This is shown to beparticularly effective for the genetic immunizations using plasmid DNA based vaccines[82, 87, 88]. This opens up a whole new area of application for the skin-targeted deliveryof gene-based therapeutics.

Conclusions and Future Prospects

Percutaneous penetration enhancement using electroporation offers a better way toenhance the rate of transport of macromolecular drugs through skin. A large number ofdrugs have been investigated for the feasibility of transdermal delivery. The other areathat holds great promise is the electroporation assisted delivery of gene-based therapeut-ics. Understanding the mechanisms involved in addition to the effects on the skin isimportant for electroporation to debut in the clinic. Rapid advances in analytical andmicroscopic techniques allowed deducing some of these effects. Generalization of thetransport using electroporation cannot be done for all the drugs using models. Eachindividual drug has to be studied separately because the transdermal or topical deliveryusing electroporation depends on the electrical parameters and physicochemical prop-erties of the drug. In addition to this, the design of safe electrodes and the developmentof miniaturized versions of the electroporator, probably operating on a battery, areneeded for advancing this technique for routine use.

Acknowledgment

We acknowledge the financial support from National Institutes of Health grant #HD41372.

0

5

10

15

20

P I 100 V 100 V+I 300 V 300 V+I

Treatment

hPT

H (

1-34

) flu

x(n

mol

/cm

2 /h)

1

08

Figure 15.13 Flux of hPTH (1–34) with electroporation and iontophoresis through porcineskin: (P), passive; (I), iontophoresis at 0.2 mA/cm2, (100 V), electroporation pulses of 100 V,(100 V 1 I), electroporation pulses of 100 V followed immediately by iontophoresis of 0.2 mA/cm2, (300 V), electroporation pulses of 100 V and (300 V 1 I), electroporation pulses of 300 Vfollowed immediately by iontophoresis of 0.2 mA/cm2. All the values are shown as themean+ SD of three determinations. (Reproduced from Medi, B.M. and Singh, J., Int. J. Pharm.,263, 25, 2003. With permission of Elsevier.)




References1. Flynn, G.L., Mechanism of percutaneous absorption from physicochemical evidence, in Per-

cutaneous Absorption, Mechanisms–Methodology–Drug Delivery, Bronaugh, R.L. and Maibach,H.I., Eds, Marcel Dekker, Inc., New York, 1989, pp. 27–51.

2. Bronaugh, R.L. and Maibach, H.I. (Eds), Percutaneous Absorption, Mechanisms–Methodology–Drug Delivery, Marcel Dekker, Inc., New York, 1989.

3. Singh, S. and Singh, J., Transdermal drug delivery by passive diffusion and iontophoresis:a review, Med. Res. Rev., 13, 569, 1993.

4. Prausnitz, M.R., Mitragotri, S., and Langer R., Current status and future potential of transdermaldrug delivery, Nat. Rev. Drug Discov., 3, 115, 2004.

5. Barry, B.W., Drug delivery routes in skin: a novel approach, Adv. Drug Deliv. Rev., 54 (Suppl. 1),S31, 2002.

6. Tyle, P., Iontophoretic devices for drug delivery, Pharm. Res., 3, 318–26, 1986.7. Prausnitz, M.R. et al., Electroporation of mammalian skin: a mechanism to enhance transder-

mal drug delivery, Proc. Natl Acad. Sci. USA, 90, 10504, 1993.8. Neumann, E., Kakorin, S., and Toensing, K., Principles of membrane electroporation and

transport of macromolecules, in Electrochemotherapy, Electrogenetherapy, and TransdermalDrug Delivery, Jaroszeski, M.J., Heller, R., and Gilbert, R., Eds, Humana Press, Totowa, 2000,pp. 1–35.

9. Barry, B.W., Novel mechanisms and devices to enable successful transdermal drug delivery,Eur. J. Pharm. Sci., 14, 101, 2001.

10. Medi, B.M. and Singh, J., Electronically facilitated transdermal delivery of human parathyroidhormone (1–34), Int. J. Pharm., 263, 25, 2003.

11. Dujardin, N. and Preat, V., Delivery of DNA to skin by electroporation, Methods Mol. Biol., 245,215, 2004.

12. Potts, R.O., Transdermal peptide delivery using electroporation, Proceedings of the Third TDSTechnology Symposium: Polymers and Peptides in Transdermal Delivery, Nichon Toshi Center,Tokyo, 47–67, 1993.

13. Neumann, E. and Rosenheck, K., Permeability changes induced by electric impulses invesicular membranes, J. Membr. Biol., 10, 279, 1972.

14. Auer, D., Brandner, G., and Bodemer, W., Dielectric breakdown of the red blood cellmembrane and uptake of SV 40 DNA and mammalian cell RNA, Naturwissenschaften, 63,391, 1976.

15. Kinosita, K. Jr. and Tsong, T.Y., Formation and resealing of pores of controlled sizes in humanerythrocyte membrane, Nature, 268, 438, 1977.

16. Hui, S.W., Low voltage Electroporation of skin, or is it iontophoresis? Biophys. J., 74, 679,1998.

17. Banga, A.K. and Prausnitz, M.R., Assessing the potential of skin electroporation for the deliveryof protein-and gene-based drugs, Trends Biotechnol., 16, 408, 1998.

18. Hristova, N.I., Tsoneva, I., and Neumann, E., Sphingosine-mediated electroporative DNAtransfer through lipid bilayers, FEBS Lett., 415, 81, 1997.

19. Golzio, M., Teissie, J., and Rols, M.P., Direct visualization at the single-cell level of electricallymediated gene delivery, Proc. Natl Acad. Sci. USA., 99, 1292, 2002.

20. Chizmadzhev, Y.A. et al., Electrical properties of skin at moderate voltages: contribution ofappendageal macropores, Biophys. J., 74, 843, 1998.

21. Chen, T. et al., Skin electroporation: rapid measurements of the transdermal voltage and flux offour fluorescent molecules show a transition to large fluxes near 50 V, J. Pharm. Sci., 87, 1368,1998.

22. Weaver, J.C., Vaughan, T.E., and Chizmadzhev, Y., Theory of electrical creation of aqueouspathways across skin transport barriers, Adv. Drug Deliv. Rev., 35, 21, 1999.

23. Pliquett, U., Mechanistic studies of molecular transdermal transport due to skin electropor-ation, Adv. Drug Deliv. Rev., 35, 41, 1999.




24. Vaughan, T.E. and Weaver, J.C., Mechanism of transdermal drug delivery by electroporation, inElectrochemotherapy, Electrogenetherapy and Transdermal Drug Delivery, Jaroszeski, M.J.,Heller, R., and Gilbert, R., Eds, Humana Press, Totowa, 2000, pp. 187–211.

25. Lurquin, P.F., Gene transfer by electroporation. Mol. Biotechnol., 7, 5, 1997.26. Martin, G.T., Pliquett, U.F., and Weaver, J.C., Theoretical analysis of localized heating in human

skin subjected to high voltage pulses, Bioelectrochemistry, 57, 55, 2002.27. Weaver, J.C., Vaughan, T.E., and Chizmadzhev, Y., Theory of skin electroporation: implications

of straight-through aqueous pathway segments that connect adjacent corneocytes, J. Invest.Dermatol. Symp. Proc., 3, 143, 1998.

28. Pliquett, U.F., Martin, G.T., and Weaver, J.C., Kinetics of the temperature rise within humanstratum corneum during electroporation and pulsed high-voltage iontophoresis, Bioelectro-chemistry, 57, 65, 2002.

29. Gallo, S.A. et al., Time-dependent ultrastructural changes to porcine stratum corneum follow-ing an electric pulse, Biophys. J., 76, 2824, 1999.

30. Murthy, S.N. et al., Temperature influences the postelectroporation permeability state of theskin, J. Pharm. Sci., 93, 908, 2004.

31. Denet, A.R. and Preat, V., Transdermal delivery of timolol by electroporation through humanskin, J. Control. Release, 88, 253, 2003.

32. Chang, S.L. et al., The effect of electroporation on iontophoretic transdermal delivery ofcalcium regulating hormones, J. Control. Release, 66, 127, 2000.

33. Vanbever, R., Boulenge, E.L., and Preat, V., Transdermal delivery of fentanyl by electroporationI. Influence of electrical factors, Pharm. Res., 13, 559, 1996.

34. Denet, A.R., Vanbever, R., and Preat, V., Skin electroporation for transdermal and topicaldelivery, Adv. Drug Deliv. Rev., 56, 659, 2004.

35. Sharma, A., Kara, M., Smith, F.R., and Krishnan, T.R., Transdermal drug delivery using electro-poration. I. Factors influencing in vitro delivery of terazocin hydrochloride in hairless rats,J. Pharm. Sci., 89, 528, 2000.

36. Bommannan, D.B. et al., Effect of electroporation on transdermal iontophoretic delivery ofluteinizing hormone releasing hormone (LHRH) in vitro, Pharm. Res., 11, 1809, 1994.

37. Prausnitz, M.R., Edelman, E.R., Gimm, J.A., Langer, R., and Weaver, J.C. Transdermal deliveryof heparin by skin electroporation, Biotechnology (NY), 13(11), 1205–9, 1995.

38. Regnier, V., Le Doan, T., and Preat, V., Parameters controlling topical delivery of oligonucleo-tides by electroporation, J. Drug Target., 5(4), 275–89, 1998.

39. Lombry, C., Dujardin, N., and Preat, V., Transdermal delivery of macromolecules using skinelectroporation, Pharm. Res., 17, 32, 2000.

40. Sen, A., Daly, M.E., and Hui, S.W., Transdermal insulin delivery using lipid enhanced electro-poration. Biochim. Biophys. Acta., 1564, 5, 2002.

41. Matoltsy, A.G., Downes, A.M., and Sweeney, T.M., Studies of the epidermal water barrier,J. Invest. Dermatol., 50, 19, 1968.

42. Allenby, A.C. et al., The effect of heat, pH and organic solvents on the electrical impedanceand permeability of excised human skin, Br. J. Dermatol., 81 (Suppl. 4), 31, 1969.

43. Murthy, S.N. et al., pH influences the postpulse permeability state of skin after electroporation,J. Control. Release, 93, 49, 2003.

44. Tokudome, Y. and Sugibayashi, K., The synergic effects of various electrolytes and electro-poration on the in vitro skin permeation of calcein, J. Control. Release, 92, 93, 2003.

45. Tokudome, Y. and Sugibayashi, K., Mechanism of the synergic effects of calcium chloride andelectroporation on the in vitro enhanced skin permeation of drugs, J. Control. Release, 95, 267,2004.

46. Oh, S.Y. et al., Effect of current, ionic strength and temperature on the electrical properties ofskin, J. Control. Release, 27, 115, 1993.

47. Peck, K.D., Ghanem, A.H., and Higuchi W.I., The effect of temperature upon the perme-ation of polar and ionic solutes through human epidermal membrane, J. Pharm. Sci., 84, 975,1995.




48. Riviere, J.E. and Heit, M.C., Electrically-assisted transdermal drug delivery, Pharm. Res., 14,687, 1997.

49. Weaver, J.C., Electroporation theory. Concepts and mechanisms, Methods Mol. Biol., 48, 3,1995.

50. Prausnitz, M.R., The effects of electric current applied to skin: a review for transdermal drugdelivery, Adv. Drug Deliv. Rev., 18, 395, 1996.

51. Jadoul, A., Bouwstra, J., and Preat, V., Effects of iontophoresis and electroporation on thestratum corneum. Review of the biophysical studies, Adv. Drug Deliv. Rev., 35, 89, 1999.

52. Jadoul, A. et al., Electroperturbation of human stratum corneum fine structure by high voltagepulses: a freeze-fracture electron microscopy and differential thermal analysis study, J. Invest.Dermatol. Symp. Proc., 3, 153, 1998.

53. Jadoul, A. et al., X-ray scattering analysis of human stratum corneum treated by high voltagepulses, Pharm. Res., 14, 1275, 1997.

54. Riviere, J.E. et al., Pulsatile transdermal delivery of LHRH using electroporation: drug deliveryand skin toxicology, J. Control. Release, 36, 229, 1995.

55. Roberts, M.S. and Walters, K.A., The relationship between structure and barrier function ofskin, in Dermal Absorption and Toxicity Assessment, Roberts, M.S. and Walters, K.A., Eds,Marcel Dekker, New York, 1998, p. 1.

56. Dujardin, N. et al., In vivo assessment of skin electroporation using square wave pulses,J. Control. Release, 79, 219, 2002.

57. Singh, J. and Medi, B.M., Effect of electroporation on skin barrier function, irritation andviability in vivo. AAPS PharmSci., 5, Abstract T2200, 2003.

58. Draize, J.H., Woodward, G., and Cavlery, H.O., Methods for the study of irritation and toxicityof substances applied topically to the skin and mucous membranes, J. Pharmacol. Exp. Ther.,82, 377, 1944.

59. Prausnitz, M.R. et al., Methods for in vivo tissue electroporation using surface electrodes, Drug.Deliv., 1, 125, 1993.

60. Vanbever, R. et al., In vivo noinvasive evaluation of hairless rat skin after high-voltage pulseexposure, Skin Pharmacol. Appl. Physiol., 11, 23, 1998.

61. Singh, J. and Maibach, H.I., Transdermal delivery and cutaneous reactions, in Dermato-logical and transdermal formulations, Walters, K.A., Ed., Marcel Dekker, New York, 2002,p. 529,

62. Weaver, J.C. et al., Heparin alters transdermal transport associated with electroporation,Biochem. Biophys. Res. Commun., 234, 637, 1997.

63. Vanbever, R., Prausnitz, M.R., and Preat, V., Macromolecules as novel transdermal transportenhancers for skin electroporation, Pharm. Res., 14, 638, 1997.

64. Sen, A., Zhao, Y.L., and Hui, S.W., Saturated anionic phospholipids enhance transdermaltransport by electroporation, Biophys. J., 83, 2064, 2002.

65. Fang, J.Y., Hung, C.F., Fang, Y.P., and Chan, T.F., Transdermal iontophoresis of 5-fluorouracilcombined with electroporation and laser treatment., Int. J. Pharm., 11, 270(1–2), 241–9,2004.

66. Jadoul, A. et al., Transdermal alniditan delivery by skin electroporation. J. Control. Release, 14,54, 265–72, 1998.

67. Bose, S., Ravis, W.R., Lin, Y.J., Zhang, L., Hofmann, G.A., and Banga, A.K., Electrically-assistedtransdermal delivery of buprenorphine, J Control. Release, 15, 73(2–3), 197–203, 2001.

68. Wang, S., Kara, M., and Krishnan, T.R., Transdermal delivery of cyclosporin-A using electro-poration, J. Control. Release, 50, 61, 1998.

69. Vanbever, R., Leroy, M.A., and Preat, V., Transdermal permeation of neutral molecules by skinelectroporation, J. Control. Release, 54, 243, 1998.

70. Vanbever, R., Lecouturier N, and Preat V., Transdermal delivery of metoprolol by electropor-ation, Pharm. Res., 11(11), 1657–62, 1994.

71. Sung, K.C. et al., Transdermal delivery of nalbuphine and its prodrugs by electroporation, Eur.J. Pharm. Sci., 18, 63–70, 2003.




72. Zewert, T.E., Pliquett, U.F., Langer, R., and Weaver, J.C. Transdermal transport of DNAantisense oligonucleotides by electroporation. Biochem. Biophys. Res. Commun., 17, 212(2),286–92, 1995.

73. Regnier, V., De Morre, N., Jadoul, A., and Preat, V., Mechanisms of a phosphorothioateoligonucleotide delivery by skin electroporation, Int. J. Pharm., 20, 184(2), 147–56, 1999.

74. Hu, Q. et al., Enhanced transdermal delivery of tetracaine by electroporation, Int. J. Pharm.,202, 121, 2000.

75. Denet, A.R., Ucakar, B., and Preat, V., Transdermal delivery of timolol and atenolol usingelectroporation and iontophoresis in combination: a mechanistic approach, Pharm. Res.20(12), 1946–51, 2003.

76. Mir, L.M. and Orlowski, S., Mechanisms of electrochemotherapy, Adv. Drug Deliv. Rev., 35,107, 1999.

77. Khavari, P.A., Therapeutic gene delivery to the skin. Mol. Med. Today, 3, 533–8, 1997.78. Cao, T., Wang, X.J., and Roop, D.R., Regulated cutaneous gene delivery: the skin as a

bioreactor, Hum. Gene Ther., 11, 2297–300, 2000.79. Cao, T. et al., The epidermis as a bioreactor: topically regulated cutaneous delivery into the

circulation, Hum. Gene Ther., 13, 1075–80, 2002.80. Spirito, F. et al., Cutaneous gene transfer and therapy: the present and the future, J. Gene Med.,

3, 21–31, 2001.81. Heller, R. et al., Intradermal delivery of interleukin-12 plasmid DNA by in vivo electroporation,

DNA Cell Biol., 20, 21, 2001.82. Medi, B.M. and Singh, J., In vivo cutaneous delivery of hepatitis B DNA vaccine using

electroporation, AAPS PharmSci., 5, Abstract T3112, 2003.83. Zhang, L. et al., Depth-targeted efficient gene delivery and expression in the skin by pulsed

electric fields: an approach to gene therapy of skin aging and other diseases, Biochem.Biophys. Res. Commun., 220, 633, 1996.

84. Zhang, L. et al., Enhanced delivery of naked DNA to the skin by non-invasive in vivoelectroporation, Biochim. Biophys. Acta, 1572, 1, 2002.

85. Dujardin, N., VanDer, P., and Preat, V., Topical gene transfer into rat skin using electroporation,Pharm. Res., 18, 61, 2001.

86. Medi, B.M. and Singh, J., In vivo cutaneous gene delivery using electroporation, Proc. ControlRelease Soc., Abstract # 611, 2003.

87. Babiuk, S. et al., Electroporation improves the efficacy of DNA vaccines in large animals,Vaccine, 20, 3399, 2002.

88. Choi, M.J. and Maibach, H.I., Topical vaccination of DNA antigens: topical delivery of DNAantigens, Skin Pharmacol. Appl. Skin Physiol., 16, 271, 2003.




Chapter 16

Microneedles

Mark R. Prausnitz, John A. Mikszta, and Jennifer Raeder-Devens

CONTENTS

Introduction .................................................................................................................................... 239Microneedle Fabrication................................................................................................................. 240

Solid Silicon Microneedles ......................................................................................................... 240Solid Metal Microneedles ........................................................................................................... 241Solid Polymer Microneedles....................................................................................................... 242Hollow Metal Microneedles ....................................................................................................... 244Hollow Silicon Microneedles ..................................................................................................... 244

Microneedle Properties .................................................................................................................. 244Insertion into Skin ...................................................................................................................... 244Avoidance of Pain....................................................................................................................... 246

Transdermal Delivery Using Microneedles.................................................................................... 247In Vitro Delivery of Model Compounds.................................................................................... 247In Vivo Delivery of Peptides and Proteins ................................................................................ 249In Vivo Delivery of Genetic Material ......................................................................................... 251

Discussion and Conclusions .......................................................................................................... 252Manufacturing ............................................................................................................................. 252Optimization of Microneedle Design......................................................................................... 252Strengths and Limitations of Microneedles................................................................................ 253

References....................................................................................................................................... 254

Introduction

Conventional drug delivery using pills or injection is often not suitable for new protein,DNA, and other therapies.1–3 An attractive alternative involves transdermal delivery froma patch, which avoids (i) degradation in the gastrointestinal tract and first-pass effects ofthe liver associated with oral delivery and (ii) the pain and inconvenience of intravenousinjection.4–7 Delivery across skin also offers the possibility to continuously control thedelivery rate, in contrast to conventional methods that deliver a large, discrete bolus.Despite these advantages, transdermal drug delivery is severely limited by the poor


239


permeability of human skin; most drugs do not cross skin at therapeutic rates. Chemical,8

electrical,9 ultrasonic,10 and other methods have been developed to increase rates oftransdermal transport, but have made only limited clinical impact to date.

To capture the delivery advantages of needles and the convenience and safetyadvantages of transdermal patches, a novel hybrid approach has been proposed, calledmicroneedles. The concept employs an array of microscopic needles that are sufficientlylarge to deliver drug effectively, but small enough to avoid causing pain. Needles ofmicroscopic dimensions can be efficacious because the rate-limiting barrier to transder-mal delivery is the skin’s outer layer of stratum corneum, which is just 10 to 20 mm thick.11

Thus, microneedles that penetrate past stratum corneum can deposit drug in the viableepidermis or dermis, where drug can diffuse rapidly for local delivery to skin or systemicdistribution via uptake by dermal capillaries.

Microneedles of the proposed dimensions can be made using microfabrication tech-nology.12 Although microfabrication has historically employed silicon processing bylithography and plasma etching, the field is being expanded to include other materials,such as metals and polymers, that are fabricated by laser cutting, molding, chemicaletching, and other techniques. By leveraging technologies beyond those of the micro-electronics industry, methods to make microneedles should provide inexpensive andreproducible mass production; we predict that manufacturing costs can be less than onedollar and, in some cases, less than five to ten cents per needle array.

Microneedle Fabrication

Microneedles can be fabricated in a number of different ways using a variety of materialsto produce a range of different geometries, including both solid and hollow needles.Initial studies emphasized solid needles made of silicon, but more recent efforts haveshifted to (i) hollow needles made of metal to perform minimally invasive injections orinfusions and (ii) solid needles made of polymer or metal to pierce the skin for drug

This section provides a representative overview of microneedles that have beenfabricated for transdermal drug delivery. This overview is not exhaustive and emphasizeswork from the authors’ laboratories to serve as examples. Additional information aboutthe large number of microneedle designs that have been developed can be found ina recent review of the rapidly growing field,13 as well as in the primary literature.14–25

Solid Silicon Microneedles

strength than hollow needles and because processing methods for silicon have beenextensively studied in the microelectronics industry. Most solid silicon microneedles havebeen symmetric in design, having the shape of a spike or elongated pyramid. Fortransdermal drug delivery, solid silicon needles can be used to pierce the skin andthereby increase its permeability.15,23,26 A drug patch can then be placed over thepermeabilized skin for enhanced transdermal delivery. Alternatively, disruption of skinbarrier function and delivery of a drug substance can occur simultaneously,27 for ex-




The largest number of microneedle designs has been for solid silicon needles (Figure

release from a patch or to deliver drug from a coating on the needle itself.

16.1). This is because solid needles are easier to fabricate and have greater mechanical

ample, through the use of drug-coated needles or reservoir-containing delivery systems.Advantages of this approach include adaptation of well-established silicon processingmethods and a simple drug delivery method. Disadvantages include the expense ofsilicon wafers and cleanroom processing and the risk of needles breaking due to thebrittleness of silicon.

Solid Metal Microneedles

Solid microneedles have also been fabricated out of metal, such as stainless steel,

been as well developed for metals as for silicon, metal can be fabricated into needlegeometries similar to silicon. In addition, metal needles can be manufactured outside the

Figure 16.1 Solid silicon microneedles. (a) Array of sharp-tipped microneedles, each measuring150 mm tall. (From Henry, S., McAllister, D.V., Allen, M.G., and Prausnitz, M.R., J Pharm Sci 87(8), 922, 1998. With permission.) (b) Close-up view of a microneedle tip with a radius ofcurvature of approximately 1 mm. (From Henry, S., McAllister, D.V., Allen, M.G., and Prausnitz,M.R., J Pharm Sci 87 (8), 922, 1998. With permission.) (c) A 400-microneedle array measuring3 3 3 mm resting on a U.S. penny (Courtesy of Georgia Institute of Technology.) (d) Array ofmesa-tipped microneedles each measuring approximately 200 mm tall (From Mikszta, J.A.,Alarcon, J.B., Brittingham, J.M., Sutter, D.E., Pettis, R.J., and Harvey, N.G., Nat Med 8 (4),415, 2002. With permission.)


Microneedles & 241


titanium, and nickel iron (Figure 16.2). Although microfabrication technology has not

costly cleanroom environment, for example, by electroplating onto molds or etchingmetal sheets using laser or chemical methods. Metal needles offer the further advantageover silicon that most metals are mechanically stronger and many metals are alreadyknown to be safe in FDA-approved devices. Solid metal needles can be used in a mannersimilar to solid silicon needles to pierce the skin prior to applying a patch or as a carrier ofdrug coated on the needles.28–32

Solid Polymer Microneedles

Solid polymer microneedles have been fabricated out of both engineering plastics, suchas polycarbonate and paralyne, and biodegradable polymers, including polylactic and/or

Figure 16.2 Solid metal microneedles. (a) Array of arrowhead microneedles, each measuring200 mm tall. (From Cormier, M., Johnson, B., Ameri, M., Nyam, K., Libiran, L., Zhang, D.D., andDaddona, P., J Control Release 97 (3), 503, 2004. With permission.) (b) Symmetrically taperedmicroneedle measuring 120 mm tall. (From McAllister, D.V., Wang, P.M., Davis, S.P., Park, J.-H.,Canatella, P. J., Allen, M. G., and Prausnitz, M. R., Proc Natl Acad Sci USA 100, 13755, 2003.With permission.) (c) Array of microneedles, each measuring 1000 mm tall, shown next to thetip of a 27-gauge hypodermic needle. (From Martanto, W., Davis, S., Holiday, N., Wang, J., Gill,H., and Prausnitz, M., Pharm Res 21, 947, 2004. With permission.)




polyglycolic acid (Figure 16.3). While polymer needles can be fabricated by adaptingcleanroom fabrication methods, they can also be manufactured using molding or othermicroreplication techniques. Polymer microneedles have the potential to be highly costeffective, given the low cost of many bulk polymers and the possibility of mass produc-tion by adapting existing molding and surface modification processes for continuous,high-speed fabrication. However, polymer microneedles are generally weaker than thoseprepared from silicon or metal, so that selection of the appropriate polymer and micro-needle design are critical to making needles that penetrate the skin but do not break.

While polymer microneedles can be used in ways similar to other solid microneedles,biodegradable microneedles offer special capabilities. In addition to improved safety in

Figure 16.3 Solid polymer microneedles: (a) Array of tapered microneedles, each measuring1500 mm tall. (From Park, J.-H., Allen, M.G., and Prausnitz, M.R., J Control Release, 104, 51,2005. With permission.) (b) Flat-bevel tipped microneedles measuring 400 mm tall. (FromMcAllister, D.V., Wang, P.M., Davis, S.P., Park, J.-H., Canatella, P.J., Allen, M.G., and Prausnitz,M.R., Proc Natl Acad Sci USA 100, 13755, 2003. With permission.) (c) Array of flat-topmicroneedles, each measuring 250 mm tall. (Courtesy of 3M. With permission.) (d) Curved-bevel tipped microneedles measuring 600 mm tall. (From McAllister, D.V., Wang, P.M., Davis,S. P., Park, J.-H., Canatella, P.J., Allen, M. G., and Prausnitz, M.R., Proc Natl Acad Sci USA 100,13755, 2003. With permission.)


Microneedles & 243


the event a microneedle breaks off in the skin, biodegradable needles can be used toencapsulate drugs within the needle matrix for rapid or controlled release drug deliv-ery.33 In this way, the polymer microneedles become a hybrid between a conventionalhypodermic needle and well-known polymeric drug delivery systems, such as micro-spheres that release drug, typically after injection or implantation.

Hollow Metal Microneedles

Hollow microneedles offer additional functionality over solid microneedles because theyhave a hollow internal bore. Hollow microneedles made of metal have received attentionand have been fabricated out of, for example, stainless steel, palladium, and nickel, byadapting microfabrication methods involving micromolding as well scaling down con-

transdermal patch, which can be applied to the skin as a single step for drug deliverythrough the microneedles over time.34 This eliminates the two-step process of needleinsertion/removal followed by patch application sometimes used with solid needles. Thisapproach may also improve control over drug delivery because the needle bore providesa known, predictable, and unchanging pathway for drug transport.

As an active system, hollow microneedles can be used like a hypodermic needle forbolus injection35 or like an indwelling catheter for slow infusion over time.23 However,these advantages of hollow microneedles come at a price. First, hollow microneedles aregenerally more difficult to fabricate, which may increase manufacturing complexity andcost. They are also inherently weaker than solid needles, which limits choice of materialsand needle design.36 Their small geometry can restrict flow rates for rapid injections.24,37

Finally, practical application of hollow microneedles may be more challenging, due togreater difficulty inserting needles into skin, preventing needle clogging, and achievingdesirable flow rates into skin.

Hollow Silicon Microneedles

Hollow microneedles have also been made out of silicon using cleanroom lithography

discussed above, hollow silicon needles have similar capabilities to hollow metalneedles.

Microneedle Properties

Before microneedles can be used for drug delivery, mechanical and other propertiesassociated with inserting needles into the skin need to be addressed and optimized.Specifically, microneedles need to be designed to insert into the skin without breakingand, upon insertion, should not cause pain.

Insertion into Skin

The mechanics of microneedle insertion into skin have been studied by comparingthe force required for needle insertion and the force required for needle fracture.36




ventional hypodermic needle manufacturing methods by laser cutting metal tubes (Fig-ure 16.4). As a passive system, hollow microneedles can be mounted on the base of a

and etching (Figure 16.5). Notwithstanding the limitations of silicon materials properties

subjects can be measured until a drop in skin resistance indicates needle penetration. Inthis way, the insertion force can be found. As shown in Figure 16.6b, the force applied toa needle pressed against a rigid surface can be measured until that force suddenly drops,which indicates that the needle broke. In this way, the fracture force can be found. Aseries of these experiments showed that penetration force depended strongly on needletip area (i.e., sharpness) and fracture force depended on both needle tip area and needlelength.36 A separate study showed that needle vibration further reduced insertionforces.38 Using these data, Figure 16.6c shows that the safety margin, defined as the

Figure 16.4 Hollow metal microneedles. (a) Straight-walled microneedle measuring 200 mm tall.(From McAllister, D. V., Wang, P. M., Davis, S. P., Park, J.-H., Canatella, P. J., Allen, M. G., andPrausnitz, M. R., Proc Natl Acad Sci USA 100, 13755, 2003. With permission.) (b) Array oftapered microneedles, each measuring 500 mm tall, shown next to the tip of a 26-gaugehypodermic needle. (From McAllister, D. V., Wang, P. M., Davis, S. P., Park, J.-H., Canatella,P. J., Allen, M. G., and Prausnitz, M. R., Microfabricated needles for transdermal delivery ofmacromolecules and nanoparticles: Fabrication methods and transport studies, Proc Natl Acad SciUSA 100, 13755, 2003. With permission.) (c) Tapered microneedle measuring 500 mm tall. (FromMcAllister, D. V., Wang, P. M., Davis, S. P., Park, J.-H., Canatella, P. J., Allen, M. G., and Prausnitz,M. R., Proc Natl Acad Sci USA 100, 13755, 2003. With permission.) (d) Stainless steel micro-needle measuring 1000 mm in length penetrating porcine skin, with hair follicles shown for sizecomparison. (From Mikszta, J., Sullivan, V., Dean, C., Waterston, A., Alarcon, J., Dekker, J.,Brittingham, J., Huang, J., Hwang, C., Ferriter, M., Jiang, G., Mar, K., Saikh, K., Stiles, B., Roy,C., Ulrich, R., and Harvey, N., J Infect Dis, 191, 278, 2005. With permission.) (e) Single micro-needle with multiple output ports. (From Brazzle, J., Papautsky, I., and Frazier, A. B., IEEE Eng MedBiol Mag 18, 53,1999. With permission.)


Microneedles & 245


As shown in Figure 16.6a, the force applied to a needle pressed against the skin of human

ratio of fracture force to insertion force, had a value much greater than unity for manymicroneedle geometries. In some cases, the fracture force was up to an order of magni-tude larger than the insertion force. Under these conditions, microneedles insert into theskin without breaking.

Other studies have also addressed microneedle mechanical properties. The forcerequired to break microneedles has been modeled and measured using silicon24 andpolymer33 microneedles in the context of failure due to buckling, shearing, and bending.The performance of blunt-ended, pyramid-like plastic microneedles has been evaluatedand shown to be structurally robust upon repeated preclinical use.35

Avoidance of Pain

Microneedles will be most useful for drug delivery if they are perceived by patients aspainless. This can be achieved physiologically by making microneedles truly painless orat least stimulate much less pain than a standard hypodermic needle. This can also beachieved psychologically by making needles too small for patients to see and therebyreduce fear and anxiety. To address the physiological effects of microneedles on pain, theskin of human subjects was pierced with an array of microneedles in a blinded fashion.39

which was indistinguishable from the negative control of a smooth surface pressedagainst the skin. In contrast, piercing the skin with a standard 26-gauge hypodermicneedle was much more painful.

A separate study addressed the use of blunt microneedle arrays scraped across theskin surface to breach the skin barrier.27 As shown in Figure 16.7b, this study similarlyconcluded that microneedle treatment caused only weak to very mild sensation thatincreased with microneedle length. Skin irritation measured in this study was found to benegligible to slight and also increased with needle length.

Figure 16.5 Hollow silicon microneedles. (a) Array of asymmetrically tapered microneedlesmeasuring 350 mm tall. (From Gardeniers, J. G. E., Luttge, R., Berenschot, J. W., de Boer, M. J.,Yeshurun, Y., Hefetz, M., van ‘t Oever, R., and van den Berg, A., J MEMS 6 (12), 855, 2003. Withpermission.) (b) A single polysilicon microneedle with a 6 mm long shaft. (From Zahn, J. D.,Deshmukh, A., Pisano, A. P., and Liepmann, D., Biomed Microdevices 6 (3), 183, 2004. Withpermission.)




As shown in Figure 16.7a, the pain score elicited by microneedles was essentially zero,

Transdermal Delivery Using Microneedles

In Vitro Delivery of Model Compounds

Initial studies with microneedles for transdermal delivery examined model compoundsin vitro. The first use of microneedles to increase skin permeability was demonstratedwith solid silicon needles to deliver a small fluorescent tracer, calcein, across human

15 Inserting needles into the skin and leaving them in placeprovided an annular pathway between the needle and surrounding skin that increasedskin permeability on the order of 1000-fold. Subsequently removing the needles, andthereby unplugging the holes, increased skin permeability on the order of 10,000-fold.

Additional experiments using insulin, bovine serum albumin, and latex nano-spheres showed similar levels of skin permeability, which confirmed the hypothesis

0

1

2

3

4

0 10 20 30 40 50

2550

75 5 10 15 200

5

10

15

20

25

0

100

200

300

400

500

0 200 400 600 800Displacement (µm) Displacement (µm)

Res

ista

nce

(kΩ

)

Saf

tey

mar

gin

Tip radius (µm) Wall thickness (µm)

0.0

0.2

0.4

0.6

0.8

1.0

For

ce (

N)

For

ce (

N)

(a)

(c)

(b)

Figure 16.6 Safety of microneedle insertion into skin without breaking. (a) Representativemeasurement of needle force (*) and skin resistance (&) during microneedle displacementand insertion into the skin of a human subject. The point of insertion is identified by the suddendecrease in skin electrical resistance and the discontinuity in applied force. (b) Representativemeasurement of needle force during microneedle displacement and fracture while being pressedagainst a rigid surface. The discontinuity marks the fracture of the microneedle. (c) Safety marginbetween fracture and insertion forces of the microneedles, where margins of safety greater thanone indicate needles that insert without breaking. (From Davis, S.P., Landis, B.J., Adams, Z.H.,Allen, M.G., and Prausnitz, M.R., J Biomech 37, 1155, 2004. With permission.)


Microneedles & 247


cadaver skin (Figure 16.8).

Prim

ary

derm

al Ir

ritat

ion

inde

x

0.0

0.2

0.4

0.6

0.8

1.0

2.0

1.2

1.4

1.6

1.8

50µm 100µm 200µm150µm SandpaperECG

FibrousECG

0

2

4

6

8

10

12

14

16

18

20

Gra

cely

pai

n sc

ore

Slight Irritant

NegligibleIrritant

PDI scoreGracely score

PDIScale

Weak

Mild

GracelyScale

MEA height

0

10

20

30

40

Pai

n sc

ale

Microneedles

Hypodermicneedle

Smoothsurface

(a) (b)

Figure 16.7 Avoidance of pain by microneedles in human subjects. (a) Box plot showing visualanalog pain scores from a blinded comparison between (i) a smooth silicon surface, (ii) a 400-

75th, and 95th percentiles are shown. (From Kaushik, S., Hord, A. H., Denson, D. D., McAllister,D. V., Smitra, S., Allen, M. G., and Prausnitz, M. R., Anesth Analg 92, 502, 2001. Withpermission.) (b) Skin irritation and perception of microneedles (Figure 16.1d) scraped acrossthe skin. Irritation was scored according to an eight-point PDI scale (grey bars) and perceptionwas scored on a 21-point Gracely scale (black diamonds). (From Mikszta, J. A., Alarcon, J. B.,

permission.)

Ski

n pe

rmea

bilit

y (c

m/h

)

1 10 100

insulinBSA

nanospheres

10−1

10−2

10−3

10−4

25 nm 50 nm

calcein

Molecular radius (nm)

Figure 16.8 Transdermal delivery in vitro using solid microneedles. The permeability of humancadaver epidermis was increased by orders of magnitude with a 400-needle array (Figure 16.1A)inserted (&) and after the array was removed (.) for calcein, insulin, bovine serum albumin, andlatex nanospheres of 25 and 50 nm radius. In the absence of microneedles, permeability to allcompounds was below their detection limits on the order of 1026 to 1024 cm/h (data not shown).Predictions are shown for needles inserted (dashed line) and needles removed (solid line) using atheoretical model. (From McAllister, D. V., Wang, P. M., Davis, S. P., Park, J.-H., Canatella,P. J., Allen, M. G., and Prausnitz, M. R., Proc Natl Acad Sci USA 100, 13755, 2003. Withpermission.)




microneedle array (Figure 16.1a) and (iii) a 26-gauge hypodermic needle (large needle in Figure

Brittingham, J. M., Sutter, D. E., Pettis, R. J., and Harvey, N. G., Nat Med 8 (4), 415, 2002. With

16.4c) inserted into the forearm of human subjects. For each treatment, the 5th, 25th, 50th,

that microneedles can dramatically increase skin permeability to a broad range ofcompounds, including macromolecules.23 In a separate study, solid microneedles wereused in a similar manner to increase transport of DNA complexes and nanospheres acrossin vitro skin culture tissues.26

In Vivo Delivery of Peptides and Proteins

Building off in vitro observations, a number of investigators have carried out in vivoexperiments to deliver peptides and proteins. Insulin has been delivered to diabetichairless rats using both solid and hollow microneedles (Figure 16.9). Solid needleswere pierced into the skin and either left in place or removed.32 After placing an insulinsolution on the skin surface for 4 h, blood glucose levels were monitored over time.Blood glucose levels dropped by 60% over the 4-h delivery period and by 80% thereafter(Figure 16.9a). This pharmacodynamic response was bounded by subcutaneous deliveryof 0.05 and 0.5 U insulin by hypodermic injection. No adverse effects were reported forthe anesthetized animals used in these studies.

Hollow microneedles have also been used to deliver insulin. In Figure 16.9b, insulinwas infused into the skin using a hollow glass microneedle, which dropped bloodglucose levels by up to 70%.23 Similar reductions in blood glucose levels have also

0

25

50

75

100

−1 0 1 2 3 4 5 6 7 8Time (h)

Nor

mal

ized

blo

od g

luco

se le

vel (

%)

Nor

mal

ized

blo

od g

luco

se le

vel (

%)

0−2 −1 0 1 2 3 4 5

20

40

60

80

100

120

Time (h)

**

(a) (b)

Figure 16.9 In vivo delivery of insulin to diabetic hairless rats using microneedles. (a) Blood

hypodermic injection of 0.05 U (^), 0.5 U (&) or 1.5 U (*) of insulin, or passive delivery acrossuntreated skin (3). Microneedles were inserted into skin for 10 min and then removed. Insulinsolution was applied to the skin immediately after microneedle insertion and left on the skin for4 h. Subcutaneous injections took a few seconds to perform. (From Martanto, W., Davis, S.,Holiday, N., Wang, J., Gill, H., and Prausnitz, M., Pharm Res 21, 947, 2004. With permission.)(b) Blood glucose levels before and after microinjection of insulin solution at 10 psi (.) or 14 psi(*) for 30 min (shaded region) using hollow microneedles. As a negative control, microinjectionof saline did not cause significant changes in blood glucose levels (data not shown). (FromMcAllister, D.V., Wang, P.M., Davis, S.P., Park, J.-H., Canatella, P.J., Allen, M.G., and Prausnitz,M.R., Proc Natl Acad Sci USA 100, 13755, 2003. With permission.)


Microneedles & 249


glucose level after insulin delivery using solid microneedles (~) (Figure 16.2c), subcutaneous

been shown using arrays of hollow silicon microneedles coupled to an insulin pump22

and hollow metal microneedles coupled to an insulin ‘‘patch’’ for passive delivery.34

Other peptides and proteins have been delivered to animals too. Desmopresin wascoated onto the tips of solid metal microneedles and administered to hairless guinea pigsto an insertion depth generally between 50 and 150 mm into the skin.31 Delivery lag timewas less than 5 min and up to 85% bioavailability was reported with pharmacokineticssimilar to intravenous delivery. Human growth hormone has also been delivered tohairless guinea pigs after pretreating the skin with solid metal microneedles and thendriving the protein across the permeabilized skin by iontophoresis.30 Transdermal fluxwas on the order of 10 mg/cm2h, which was much larger than the iontophoretic fluxacross untreated skin.

Vaccine delivery using microneedles is attractive because delivery to the shallow skincan target the immune-stimulatory antigen presenting cells in the epidermis.40,41 In onestudy, hairless guinea pigs were administered ovalbumin as a model protein antigen fromsolid metal microneedles at rates up to 20 mg in 5 s for a total of up to 80 mg.29 At smalldoses, microneedles achieved ovalbumin-specific antibody response levels up to 50-foldgreater than subcutaneous or intramuscular injection of the same dose.

In another study, an anthrax vaccine based on the recombinant protective antigen(rPA) of Bacillus anthracis was administered using both solid and hollow micronee-dles.35 After challenging rabbits with a lethal aerosol dose of anthrax spores, rabbitsimmunized using hollow microneedles were completely protected and those via topicaladministration to skin pretreated with a solid microneedle array were partially protected(Figure 16.10). This discrepancy is likely due, at least in part, to differences in dosageefficiency between the two methods.

0

20

40

60

80

100

Per

cent

sur

viva

l

Needle IM Hollowmicroneedle

Solidmicroneedle

Topical(no device)

Figure 16.10 Microneedle-based delivery of anthrax rPA vaccine. The percent of rabbits thatsurvived aerosol challenge with 100 LD50 of Ames strain anthrax spores is shown for rabbitsimmunized with rPA by IM injection using a conventional needle (Needle IM), with a hollowstainless steel microneedle (Hollow microneedle), with a solid plastic microneedle array (SolidMicroneedle), or topical vaccine application to the skin without using a microneedle device(Topical). (From Mikszta, J., Sullivan, V., Dean, C., Waterston, A., Alarcon, J., Dekker, J.,Brittingham, J., Huang, J., Hwang, C., Ferriter, M., Jiang, G., Mar, K., Saikh, K., Stiles, B., Roy,C., Ulrich, R., and Harvey, N., J Infect Dis, 191, 278, 2005. With permission.)




In Vivo Delivery of Genetic Material

Genetic material, including DNA and oligonucleotides, has also been delivered to ani-mals in vivo using microneedles. To induce local expression of a firefly luciferase reportergene, silicon needles were dipped into a naked plasmid DNA solution and then rubbedacross the skin to deposit the DNA in microscopic troughs scraped into the skin (Figure16.11a).27 Gene transfer enhanced by microneedles increased gene expression by up to2,800-fold above intact skin. Guided by these results, naked plasmid DNA encoding theHepatitis B Surface Antigen (HBsAg) was delivered to mice in a similar manner to serveas a model DNA vaccine (Figure 16.11b).27 After delivery using microneedles, humoraland cellular immune responses to HBsAg were stronger and less variable than when

NeedleInjections (number of lateral passes)

MEA

MEA

(pre

ss)

Topi

cal

101

105

104

103

102<10 2

102

103

104

105

Ser

um a

ntib

ody

titer

(1/d

ilutio

n)

ID MEA TopicalIM

IM ID x12 x10 x6 x4 x2 x1

RLU

/mg

prot

ein

(a)

1 2 3 1 2 3 1 2 3 1 2 3

(b)

Number of immunizations

Figure 16.11 In vivo gene transfer and genetic immunization. (a) Plasmid DNA encoding fireflyluciferase was administered by injection (i.m. or i.d.) or using microneedles (MEA), varying thenumber of lateral passes. Alternatively, the microneedles were placed in contact with the DNAsolution on the skin and pressed six times (press). Controls received DNA topically withoutmicroneedles. Expression levels in individual mice are indicated by closed symbols. (b) Micewere given a total of three immunizations with DNA by needle injections (i.m. or i.d.), micro-needle-based (MEA) delivery (six passes) or topical application to untreated skin. Hepatitis Bsurface antigen-specific serum antibody titers are indicated by bars with responses of individualmice indicated by open symbols. (From Mikszta, J.A., Alarcon, J.B., Brittingham, J.M., Sutter,D.E., Pettis, R.J., and Harvey, N.G., Nat Med 8 (4), 415, 2002. With permission.)


Microneedles & 251


administered via injection using conventional needles. Moreover, full seroconversion wasachieved after fewer immunizations using microneedles.

In a separate study, oligonucleotides were administered to hairless guinea pigs usingmicroneedles by first piercing the skin with solid microneedles and then applyingiontophoresis to drive oligonucleotides into the skin.28 Optimized transdermal fluxeswere on the order of 100 mg/cm2 h, which was at least tenfold greater than iontophoresisacross untreated skin.

Discussion and Conclusions

Manufacturing

A great deal of research has addressed novel methods to fabricate microneedles. Someproduce complex designs that involve many fabrication steps using state-of-the-art clean-room technology. Such designs will probably be costly to produce and may not be robustenough for practical applications. Other approaches, including those emphasized by theauthors, have prioritized simplicity of both design and fabrication methods. Microneedlesthat are inexpensive enough to be single-use, disposable devices are much more likely tofind widespread clinical application and commercial success. Although acceptable micro-needle manufacturing cost will depend strongly on the application, we believe thatmicroneedles costing much less than one dollar, and preferably less than five or tencents, have the greatest potential.

Many microneedles developed by the microfabrication research community havebeen silicon-based. These approaches adapt microelectronics cleanroom technology todefine needle shape using lithography and then employing plasma, chemical, and otheretching methods to sculpt needles out of silicon wafers.12 Although the microelectronicsindustry has extensively developed this technology, cleanroom equipment and operationare still expensive.

Microneedles made of metals and polymers not only offer functional advantages asdrug delivery devices, but can also be produced using inexpensive methods that do notrely heavily or at all on silicon processing infrastructure. Metal microneedles can befabricated using inexpensive electroplating technology more than a century old, incombination with micromolds produced by more sophisticated methods.23 Metal micro-needles can also be produced by scaling down conventional methods used to producehypodermic needles. Polymer microneedles can also be molded by modified injectionmolding techniques23,33 or adapting commercial embossing methods currently used tocreate micropatterned surfaces.

Optimization of Microneedle Design

Optimal design of microneedles requires balancing a number of different constraints.Microneedles need to be made using cost-effective and large-scale manufacturing that iscapable of making the desired needle geometry. Microneedles also need to be made froma material that is strong enough to insert into skin without breaking, safe for the skin,nondamaging to the drug, and manufacturable. Microneedles should have a geometrysmall enough to avoid causing pain, but large enough to be easily inserted into the skinwithout expert handling. Microneedles should be integrated into a device that facilitates




drug storage, needle insertion, and patient convenience. Finally, microneedles mustdeliver the correct drug dose with the correct pharmacokinetics.

Given these multiple constraints, there is not a single design that simultaneouslyoptimizes all aspects of microneedles. The best compromise will be governed by thespecific needs of the drug therapy and the capabilities of available technology. Forexample, bolus delivery of a sub-milligram drug dose may be most easily achievedusing solid polymer or metal microneedles coated with drug. Because the payload of aneedle coating is limited, the small drug dose lends itself to using solid needles, which arestronger and easier to manufacture than hollow needles. In contrast, delivery of a largerdose, perhaps modulated over an extended time, might be best achieved by infusionthrough hollow metal microneedles. For the appropriate indication, the clinical valueof this type of system would justify the extra expense and design difficulty of hollowneedles coupled with a pump.

Strengths and Limitations of Microneedles

Transdermal drug delivery using microneedles presents an exciting opportunity tocouple the efficacy of needle-based delivery with the patient-acceptance of transdermalmethods. When solid or hollow microneedles are used to increase skin permeability incombination with a drug reservoir patch, transdermal delivery can be expanded from notonly small, lipophilic drugs delivered by a conventional patch to now include delivery ofmacromolecules, such as proteins and DNA. Delivery from microneedles with coated orencapsulated drug can similarly administer small doses of high molecular-weight drugs.When used in an injection or infusion scenario, hollow microneedles permit delivery thatis more rapid than traditional adhesive patches and can be modulated over time via activedelivery controlled by hand or pump.

Despite the power of the microneedles approach, more work remains to be donebefore FDA approval and commercialization are realized. In addition to optimizingneedle design and manufacturing discussed above, more work will be needed to developcomplete microneedle-based systems that perform reliably in the hands of untrainedclinicians and by patients themselves in situations where self-administration may bedesirable. Moreover, the effects of this new route of administration on drug performanceand drug safety will need to be comprehensively evaluated in the preclinical and clinicalphases of a product development program. Although microneedles themselves areexpected to be safe, careful safety studies are needed.

Some limitations may be inherent to microneedles, such as an upper limit to the totaldose that can be administered and adverse effects caused by delivery of certain drugs tothe skin. In addition, the inherent immune-stimulatory properties of the skin, whiledesirable for vaccines, may pose new challenges for conventional drugs where animmune response could potentially neutralize the effects of the drug.

In conclusion, microneedles provide a minimally invasive access port into the bodythat can be utilized in a variety of ways. As a device that sits on the skin surface,microneedle-based systems have greater flexibility in design, size, and other propertiescompared to, for example, injected, ingested, or implanted drug delivery systems. Be-cause microneedles may provide the enabling technology to deliver macromoleculesacross the skin, microneedles could represent the scientific advance needed to allowtransdermal drug delivery to reach its true potential.


Microneedles & 253


References1. Crystal, R.G., Transfer of genes to humans: early lessons and obstacles to success, Science 270,

404, 1995.2. Park, K., Controlled Drug Delivery: Challenges and Strategies, American Chemical Society,

Washington, D.C. 1997.3. Langer, R., Drug delivery and targeting, Nature 392, 5, 1998.4. Amsden, B.G. and Goosen, M.F.A., Transdermal delivery of peptide and protein drugs: an

overview, AIChE J 41, 1972, 1995.5. Prausnitz, M.R., Mitragotri, S., and Langer, R., Current status and future potential of transdermal

drug delivery, Nat Rev Drug Discov 3 (2), 115, 2004.6. Purdon, C.H., Azzi, C.G., Zhang, J., Smith, E.W., and Maibach, H.I., Penetration enhancement

of transdermal delivery — current permutations and limitations, Crit Rev Ther Drug CarrierSyst 21 (2), 97, 2004.

7. Bronaugh, R.L. and Maibach, H.I., Percutaneous Absorption, 4th ed. Marcel Dekker, New York,2005.

8. Williams, A.C. and Barry, B.W., Penetration enhancers, Adv Drug Deliv Rev 56 (5), 603, 2004.9. Banga, A.K., Electrically-Assisted Transdermal and Topical Drug Delivery, Taylor & Francis,

London, 1998.10. Merino, G., Kalia, Y.N., and Guy, R.H., Ultrasound-enhanced transdermal transport, J Pharm

Sci 92 (6), 1125, 2003.11. Champion, R.H., Burton, J.L., Burns, D.A., and Breathnach, S.M., Textbook of Dermatology,

Blackwell Science, London, 1998.12. Madou, M.J., Fundamentals of Microfabrication: The Science of Miniaturization, 2nd ed. Marc

J. Madou, ed. CRC Press, Boca Raton, FL, 2002.13. Reed, M.L., and Lye, W.-K., Microsystems for drug and gene delivery, Proc IEEE 92, 56, 2004.14. Chen, J. and Wise, K.D., A multichannel neural probe for selective chemical delivery at the

cellular level, IEEE Trans Biomed Eng 44, 760, 1997.15. Henry, S., McAllister, D.V., Allen, M.G., and Prausnitz, M.R., Microfabricated microneedles:

a novel approach to transdermal drug delivery, J Pharm Sci 87 (8), 922, 1998.16. Reed, M.L., Wu, C., Kneller, J., Watkins, S., Vorp, D.A., Nadeem, A., Weiss, L.E., Rebello, K.,

Mescher, M., Smith, A.J.C., Rosenblum, W., and Feldman, M.D., Micromechanical devices forintravascular drug delivery, J Pharm Sci 87 (11), 1387, 1998.

17. Brazzle, J., Papautsky, I., and Frazier, A.B., Micromachined needle arrays for drug delivery orfluid extraction, IEEE Eng Med Biol Mag 18, 53,1999.

18. Liepmann, D., Pisano, A.P., and Sage, B., Microelectromechanical systems technology todeliver insulin, Diabetes Technol Ther 1 (4), 469, 1999.

19. Lin, L. and Pisano, A.P., Silicon processed microneedles, J MEMS 8, 78, 1999.20. Smart, W.H. and Subramanian, K., The use of silicon microfabrication technology in painless

blood glucose monitoring, Diabetes Technol Ther 2 (4), 549, 2000.21. Griss, P., Enoksson, P., Tolvanen-Laakso, H.K., Merilainen, P., Ollmar, S., and Stemme, G.,

Micromachined electrodes for biopotential measurements, J MEMS 10 (1), 10, 2001.22. Gardeniers, J.G.E., Luttge, R., Berenschot, J.W., de Boer, M.J., Yeshurun, Y., Hefetz, M., van ‘t

Oever, R., and van den Berg, A., Silicon micromachined hollow microneedles for transdermalliquid transport, J MEMS 6 (12), 855, 2003.

23. McAllister, D.V., Wang, P.M., Davis, S.P., Park, J.-H., Canatella, P.J., Allen, M.G., and Prausnitz,M.R., Microfabricated needles for transdermal delivery of macromolecules and nanoparticles:fabrication methods and transport studies, Proc Natl Acad Sci USA 100, 13755, 2003.

24. Zahn, J.D., Talbot, N.H., Liepmann, D., and Pisano, A.P., Microfabricated polysilicon micro-needles for minimally invasive biomedical devices, Biomed Microdevices 2 (4), 295, 2000.

25. Kim, K., Park, D.S., Lu, H.M., Che, W., Kim, K., Lee, J.-B., and Ahn, C.H., A tapered hollowmetallic microneedle array using backside exposure of SU-8, J Micromech Microeng 14, 597,2004.




26. Chabri, F., Bouris, K., Jones, T., Barrow, D., Hann, A., Allender, C., Brain, K., and Birchall, J.,Microfabricated silicon microneedles for nonviral cutaneous gene delivery, Br J Dermatol 150(5), 869, 2004.

27. Mikszta, J.A., Alarcon, J.B., Brittingham, J.M., Sutter, D.E., Pettis, R.J., and Harvey, N.G.,Improved genetic immunization via micromechanical disruption of skin-barrier function andtargeted epidermal delivery, Nat Med 8 (4), 415, 2002.

28. Lin, W., Cormier, M., Samiee, A., Griffin, A., Johnson, B., Teng, C., Hardee, G.E., and Daddona,P., Transdermal delivery of antisense oligonucleotides with microprojection patch (Macroflux)technology, Pharm Res 18 (12), 1789, 2001.

29. Matriano, J.A., Cormier, M., Johnson, J., Young, W.A., Buttery, M., Nyam, K., and Daddona,P. E., Macroflux microprojection array patch technology: a new and efficient approach forintracutaneous immunization, Pharm Res 19 (1), 63, 2002.

30. Cormier, M. and Daddona, P.E., Macroflux technology for transdermal delivery of therapeuticproteins and vaccines, in Modified-Release Drug Delivery Technology, Rathbone, M.J.,Hadgraft, J., and Roberts, M.S. eds, Marcel Dekker, New York, 2003, p. 589.

31. Cormier, M., Johnson, B., Ameri, M., Nyam, K., Libiran, L., Zhang, D.D., and Daddona, P.,Transdermal delivery of desmopressin using a coated microneedle array patch system,J Control Release 97 (3), 503, 2004.

32. Martanto, W., Davis, S.P., Holiday, N.R., Wang, J., Gill, H.S., and Prausnitz, M.R., Transdermaldelivery of insulin using microneedles in vivo, Pharm Res 21, 947, 2004.

33. Park, J.-H., Allen, M.G., and Prausnitz, M.R., Biodegradable polymer microneedles: fabrication,mechanics and transdermal drug delivery, J Control Release 104, 51, 2005.

34. Davis, S.P., Martanto, W., Allen, M.G., and Prausnitz, M.R., Transdermal insulin delivery todiabetic rats through microneedles, IEEE Transact Biomed Eng 52, 909, 2005.

35. Mikszta, J., Sullivan, V., Dean, C., Waterston, A., Alarcon, J., Dekker, J., Brittingham, J., Huang,J., Hwang, C., Ferriter, M., Jiang, G., Mar, K., Saikh, K., Stiles, B., Roy, C., Ulrich, R., and Harvey,N., Protective immunization against inhalational anthrax: a comparison of minimally-invasivedelivery platforms, J Infect Dis, 191, 278, 2005.

36. Davis, S.P., Landis, B.J., Adams, Z.H., Allen, M.G., and Prausnitz, M.R., Insertion of micro-needles into skin: measurement and prediction of insertion force and needle fracture force, JBiomech 37, 1155, 2004.

37. Martanto, W., Smith, M.K., Baisch, S.M., Costner, E.A., and Prausnitz, M.R., Fluid dynamics inconically tapered microneedles, AICHE J 51, 1599, 2005.

38. Yang, M. and Zahn, J.D., Microneedle insertion force reduction using vibratory actuation,Biomed Microdevices 6 (3), 177, 2004.

39. Kaushik, S., Hord, A.H., Denson, D.D., McAllister, D.V., Smitra, S., Allen, M.G., and Prausnitz,M.R., Lack of pain associated with microfabricated microneedles, Anesth Analg 92, 502, 2001.

40. Babiuk, S., Baca-Estrada, M., Babiuk, L., Ewen, C., and Foldvari, M., Cutaneous vaccination:the skin as an immunologically active tissue and the challenge of antigen delivery, J ControlRelease 66, 199, 2000.

41. Partidos, C., Beignon, A.-S., Mawas, F., Belliard, G., Briand, J.-P., and Muller, S., Immunityunder the skin: potential application for topical delivery of vaccines, Vaccine 21, 776, 2003.

42. Zahn, J.D., Deshmukh, A., Pisano, A.P., and Liepmann, D., Continuous on-chip micropumpingfor microneedle enhanced drug delivery, Biomed Microdevices 6 (3), 183, 2004.


Microneedles & 255


Chapter 17

Vesicles under Voltage

Michael C. Bonner and Brian W. Barry

CONTENTS

Introduction .................................................................................................................................... 257Iontophoresis .................................................................................................................................. 258Iontophoresis in Drug Delivery ..................................................................................................... 258Pathways of Iontophoretic Transport ............................................................................................ 258Electroporation ............................................................................................................................... 259Liposomes for Dermal and Transdermal Delivery ........................................................................ 259Transfersomes1 and Analogs — Ultradeformable Vesicles.......................................................... 260Combined Liposomal and Iontophoretic Delivery ....................................................................... 261Liposomes and Electroporation ..................................................................................................... 263Concluding Remarks ...................................................................................................................... 265References....................................................................................................................................... 265

Introduction

Historically, the skin was believed to have evolved as a control barrier to the outwardtransport ofwater and the inward movement of topically contacting agents [1, 2]. In contrastto this evolutionary standpoint, over the last three decades, the use of intact skin as apotential site for local and continuous systemic administration of drugs has receivedconsiderable attention [3, 4]. A transdermal mode of drug delivery offers several advantagesover more conventional methods. These include avoidance of problems with enzymaticdeactivation during gastrointestinal passage, bypassing the first-pass hepatic metabolism,better patient compliance, and ease of termination in problematic cases [1, 5, 6].

Due to the lipophilic nature of the skin’s permeability barrier that resides within thehorny layer of the stratum corneum (SC), transdermal administration is currently notfeasible for drugs that are charged, have high molecular weights, or are strongly hydro-philic. Many approaches have been considered so as to increase percutaneous absorp-tion of such compounds. These can be broadly subdivided into chemical and physicalstrategies.


257


Chemical enhancement techniques include the use of liposomes, which may bedefined as lipid vesicles enclosing an aqueous core [7, 8] and chemical penetrationenhancers [9, 10] while physical strategies include the use of ultrasound or phonophor-esis [11, 12] and the enhancement of penetration by electrical methods known as ionto-phoresis and electroporation.

Several groups have investigated the synergy of various combinations of these strat-egies including iontophoresis with chemical penetration enhancers [13–16], with ultra-sound [17, 18], or combination between chemical enhancers and ultrasound [19].

Few reports, however, exist on the combined use of liposomes and electrical pene-tration enhancement and this combination is the main focus of this chapter. Before welook at their combined use, however, it may be helpful to consider each technique in turn.

Iontophoresis

The main principle underlying iontophoresis is based on a general rule of electricity, thatis, like charges repel each other, while opposite charges attract. In theory, when a voltageis applied to the skin, it yields a potential difference primarily across the SC [20]. If a drugin an appropriate formulation is then placed under an electrode of the same polarity asthe charge on the drug and applied to the skin, it will be driven into the tissue byelectrostatic repulsion [21].

Transdermal delivery of neutral molecules can also be enhanced from the anode,though the mechanism for this augmentation is believed to be linked with the process ofelectroosmosis [20, 22, 23]. Electroosmosis, or convective flow, occurs in the same direc-tion as the flow of counterions, when a voltage difference is applied across a chargedporous membrane. Since skin is a cation selective membrane and carries a net negativecharge at the physiological pH [24], it induces a net solvent flow during iontophoresis,which is in the direction of positive ions transport [25–27]. This flow carries otherdissolved molecules with it, increasing transport of cations and neutral species, whileretarding that of anions.

Iontophoresis in Drug Delivery

Over the past two decades, various drugs have been investigated for transdermal deliverywith iontophoresis. These include antihypertensive agents, synthetic narcotics, and localanaesthetics. Work is continuing on iontophoretic devices, and the first prefilled ionto-phoretic patch system known as Lidosite (containing a formulation of lidocaine andnoradrenaline for local analgesia prior to superficial dermatological procedures) isunder development in the United States [28, 29]. Interest also lies in iontophoreticdelivery of peptides and proteins such as vasopressin, leutinizing hormone releasinghormone, and insulin and several groups have reported varying degrees of success withthese molecules.

Pathways of Iontophoretic Transport

Because ions tend to penetrate through the skin via pathways of lowest resistance, appen-dageal pores are generally believed to be the major route for iontophoretic transport.




This was elegantly demonstrated by Burnette and Ongpipattanakul [24], who showed thatfluorescein, anegatively chargeddyedrivenunder an iontophoretic current (0.16mA/cm2),penetrated through excised human skin and appeared on the dermal side as spots at poresites. Cullander and Guy [30] used a vibrating probe electrode to identify appendagealpathways as carrying most of the current through hairless mouse skin in vitro. Directvisualization of penetration pathways during iontophoresis has also resulted in yet moreevidence for an important appendageal role [31, 32].

Although, appendages are considered to play a major role in iontophoretic transport,it is inappropriate to conclude that all such penetration occurs through these pores [20].As well as the electro-osmotic effect, several studies have suggested the existence of anonappendageal pore pathway in parallel with the appendageal route. A small portion ofthe current has been shown to pass into the skin at locations other than appendages [30].In addition, some authors have suggested that current flows through artificial shunts,which occurs as a result of temporary disruption of highly organized SC structure [30, 31].

Electroporation

In electroporation, application of short (ms–ms), high-voltage (30 to 1000 V) pulsesprovides a way to disrupt the multilamellar lipid bilayer system in the SC. Overallunderstanding of molecular transport mechanisms through electroporated skin remainsrelatively inferior compared to iontophoresis. However, it is generally believed thatelectroporation involves the creation of transient aqueous pathways (pores) in the SC[33]. This mechanism, however, remains controversial to some extent, as these pores havenot yet been identified in any microscopic studies. This is believed to be due to theirsmall radius (10 nm), sparse distribution (<0.1% of the surface area) and transient nature(ms–s) [34].

Various electrical [35–37] and transport studies [36, 38–41] have provided evidence insupport of an ‘‘aqueous pore’’ theory. In addition, fluorescence microscopy, employed tovisualize transdermal transport pathways during electroporation, has also indicated that,unlike iontophoresis or passive diffusion, the sites of transport during high-voltageexposures represent pathways which travel straight through the bulk of SC, with noevidence for tortuosity [42–44].

The mechanism of molecular transport during electroporation is expected to involvepassive diffusion and/or electrically driven transport, but unlike iontophoresis, the in-volvement of electroosmosis is controversial and believed to be of much less significance[36, 40, 45]. As with iontophoretic transport, some studies have indicated that high-voltagepulsing-induced transdermal delivery of charged or even neutral drugs could be con-trolled by an appropriate use of the electric pulse parameters, that is, pulse voltage, pulsewidth, and pulse number [39, 40, 45].

Liposomes for Dermal and Transdermal Delivery

The potential advantages of liposomes as a transdermal drug delivery system includeenhancement of drug delivery, solubilization of poorly soluble drugs, formation of localdepot for the sustained release of topically applied drugs, reduction of side effects orincompatibilities, or as a rate-limiting barrier for the modulation of systemic absorption[8]. Examples of liposome delivery to the skin have appeared in the literature over the last


Vesicles under Voltage & 259


two decades. Liposome formulations of various lipid compositions, sizes, charges, andtypes all resulted in significantly higher flux of triamicinoline acetonide through rat skinthan did a commercially available ointment [46]. Diphylline liposomes have also beendelivered to the skin for potential use as a topical treatment for psoriasis [47].

The means by which liposomes enhance dermal penetration remain controversial.Some researchers reported that phospholipids administered on the skin as aggregates(i.e., vesicles) first disintegrate and then diffuse through the barrier in the form of smallfragments or lipid monomers [48]. Others believe that some vesicles are deformableenough to pass the intact SC and reach systemic circulation as intact structures [49, 50]or may accumulate in the channel-like regions in the SC [51–53].

Transfersomes1 and Analogs — Ultradeformable Vesicles

Cevc and Blume [49] reported that certain types of vesicles (Transfersomes) can penetrateintact to the deep layers of the skin and may progress far enough to reach the systemiccirculation. Transfersomes (also called ultradeformable liposomes) are specially opti-mized vesicular particles consisting of at least one inner aqueous compartment sur-rounded by a lipid bilayer with specially tailored properties due to the incorporation ofso-called edge activators. Surfactants were suggested as examples of such edge activator[54]; sodium cholate and sodium deoxycholate [55–61], Span 80 [57], and polysorbate 80[58, 62] have been used for this purpose.

The available information about Transfersomes indicates that the main component isphosphatidylcholine (PC). The edge activator is added in a range of 10 to 24%, althoughsome have been reported to have a higher surfactant concentration where polysorbate 80was incorporated in a concentration of up to 45% w/w relative to PC [62]. In addition, 3 to7% v/v of pure ethanol is included. The total lipid concentration of the Transfersomesuspension ranges from 4 to 10% w/v [50, 54, 55, 63]. Only at an optimal balance betweenthe amount of edge activator and the amount of the bilayer-forming lipid are the vesicleshighly deformable.

In a study by Cevc and Blume [49], after 8 h of Transfersome application in mice,significant amounts were found in blood (6 to 8% of the applied dose) and liver (20% ofthe recovered dose). The authors stated that 50 to 90% of the dermally deposited lipidcould be transported beyond the level of the SC. Although there is no doubt that theirTransfersomes have clear advantages with respect to increasing the transport of activematerial across the skin, in these studies it was claimed that the highly deformablevesicles penetrated intact through the SC and viable epidermis into the blood circulation.

The inventors of these vesicles suggest that such vesicles, being ultradeformable (upto 105 times that of unmodified liposomes), squeeze through pores in SC less than one-tenth of the vesicle’s diameter. Thus, vesicles with sizes up to 200 to 300 nm can penetrateintact skin. To do that, a hydration gradient was suggested to be the key to initiate skinpenetration. The transdermal hydration gradient operating from the relatively dry skinsurface towards the viable tissues, with their higher water content, drives Transfersomesthrough the horny layer. When applied nonocclusively, excess water from the testsuspension of vesicles will evaporate and the system will partially dehydrate, leavingdeposited vesicles on the skin surface. As phospholipids tend to avoid dry surroundings,for vesicles to remain maximally swollen, they must follow the local hydration gradientand penetrate into more hydrated and deeper skin layers of SC where some may fusewith skin lipid [49, 64, 65].




In the first published study on drugs associated with Transfersomes by Planas et al.[55], the vesicle suspension contained relatively high amounts of lidocaine or tetracaine.The Transfersomes were tested in vivo on rat and human and in both studies ultrade-formable liposomes appeared to be more efficient than traditional liposomes or solu-tions, although applied under occluded conditions. However, the differences found usingcorticosteroids were somewhat less pronounced [56]. In another report, diclofenacappeared to be better deposited in skin in vivo when applied in Transfersomes, com-pared with commercial hydrogel [66].

Large molecules such as insulin were also associated with Transfersomes, when bloodglucose levels could be lowered after about 3 h application. A clear difference in glucoselevel in blood was observed in vivo after application of insulin associated with micelles,traditional liposomes, or Transfersomes. The last significantly lowered glucose levels.This effect was first observed in mice, but later in humans [63]. Two groups [67, 68] havesuccessfully used Transfersomes for the noninvasive administration of protein antigen.

El Maghraby and co-workers measured in vitro drug delivery from ultradeformableliposomes (PC-cholate vesicles) and traditional vesicles (with no edge activator) throughhuman epidermal membranes under open and occluded applications, using estradiol and5-fluorouracil as model drugs. Both liposome types improved maximum flux and skindeposition compared to saturated aqueous drug solution (maximum thermodynamicactivity) under hydration gradient conditions, with a superior effect obtained fromthe ultradeformable type [57–61]. However, such enhancement was less dramatic thanthat reported by the group of Cevc for the in vivo situation, as only 1 to 3% of drugwas delivered. The data also revealed a possible penetration enhancing effect forphospholipids. There was also evidence of an improved drug uptake by the skin.However, their results did not confirm that intact liposomes penetrated through thehorny layer [57].

Combined Liposomal and Iontophoretic Delivery

The first published information on the combined use of iontophoresis and liposomesreported that iontophoretic delivery of neutral colchicine encapsulated in positivelycharged liposomes increased the iontophoretic flux by two to threefold [69, 70]. It wasalso demonstrated that the combined use of liposomes and iontophoresis could reducedegradation and enhance transdermal delivery of enkephalin in vitro [71].

Fang et al. [72] investigated the effect of different liposomal formulations on theiontophoretic transport of enoxacin through rat skin in vitro. The iontophoretic permea-bility of enoxacin increased as the fatty acid chain length of the phospholipid decreased,due to the fall in the phase transition temperature of the lipid.

The effects of L-alpha-PC, cationic lipid (stearylamine), and the penetration enhancerazone on the iontophoretic transdermal flux of mannitol through human skin in vitrowere investigated [73]. The skin was pretreated with the lipid suspensions and azonesolution, all containing 32% ethanol, prior to iontophoresis. For the lipid suspensions,only PC increased mannitol flux compared to control (without pretreatment). Interest-ingly, the authors found that the combination of PC and electric current increasedmannitol flux almost as effectively as the potent penetration enhancer (azone), suggest-ing a synergistic effect between PC and electric current.

The combination of iontophoresis and surfactant-based elastic vesicles (composedof polyoxyethylene ester PEG-8L, L 595, and cholesterol sulfate) was investigated by




Li et al. [74]. Human skin pretreatment with the vesicles increased by about 1.4-fold theapomorphine iontophoretic flux, compared to control (buffer-treated skin).

A more recent study by Essa et al. [75] examined the enhancement of estradiolpenetration by iontophoresis of a liposomal formulation. Iontophoretic (cathodal)investigations involved 6 h application of constant current of densities 0.2, 0.5, and0.8 mA/cm2, to ultradeformable liposomes and a saturated aqueous solution. To probethe role of iontophoresis in improving drug delivery, passive penetration (i.e., currentdensity of 0.0 mA/cm2) for the same time length was also studied.

Plots of cumulative amount versus time for estradiol penetrated from solution and

increased estradiol penetration from solution and liposomes compared to their ownpassive delivery. Penetration plots show an expected rise in the cumulative amountpenetrated with increasing current density. Fluxes were calculated from regressionlines of each penetration profile at different current density for the two formulations.

In spite of being driven against electro-osmotic water drift, cathodic iontophoresisresulted in the passage of a considerable amount of estradiol from aqueous solutionthrough the membrane. The improved drug penetration correlated with the appliedcurrent density. This increase was not significantly different from passive (P> 0.05) fluxwhen the lowest electric field (0.2 mA/cm2) was used. However, there were about twoand fourfold increases in flux following the use of 0.5 and 0.8 mA/cm2 current densities,respectively.

For ultradeformable liposomes, at zero current (passive) estradiol penetrated well, asshown by 7.4-fold enhancement in comparison to control. During iontophoresis, estra-diol penetrated the skin in a linear fashion. Compared with their passive flux, there wereabout two-, four- and eightfold increase in estradiol fluxes after using 0.2, 0.5, and0.8 mA/cm2, respectively (Figure 17.2). Such enhancement in flux indicates that theelectrostatic repulsion force between the negatively charged vesicles and the negativecathode overcomes the skin’s cation permselectivity and also the reverse electro-osmotic

00

0.2

0.4

0.6

0.8

1 2 3 4 5 6

Time (h)

0.0 mA/cm2

0.5 mA/cm2

0.2 mA/cm2

0.8 mA/cm2

Figure 17.1 Iontophoretic estradiol delivery through human epidermal membranes from sat-urated aqueous solution at different current densities (n 5 5).




liposomes are shown in Figure 17.1 and Figure 17.2, respectively. Current application

flow. The repulsion force between the applied current and the negatively chargedliposome moved more liposomes towards the skin surface. As phospholipids releasedfrom liposomes can adhere to and fuse with the SC, altering the lipid barrier propertiesand producing a more permeable structure [76, 77], the presence of liposomes in highconcentration due to iontophoresis may increase the effect. At the same time, current isknown to disorganize the lipid layer stacking [78, 79]. Consequently, phospholipids andelectric current could synergistically enhance the transdermal drug flux, agreeing withKirjavainen et al. [73], with a probability of some intact vesicular penetration through theskin.

Liposomes and Electroporation

Though a combination could be potentially synergistic, combining electroporation withliposomes has gained little attention. Badkar et al. [70] investigated the in vitro humanskin permeation of colchicine encapsulated in positive liposomes under electroporation,where disappointing results were reported.

More recently, Essa et al. [80] examined the electroporation of estradiol in solution andentrapped within ultradeformable liposomes. After high-voltage pulsing, passive estra-diol permeation through the skin was monitored for 8 h. The total amount penetratedafter pulsing was 16-fold higher than for passive diffusion, reflecting the capability ofhigh-voltage pulses to enhance the transdermal penetration of neutral molecules.

Surprisingly, the electroporation only slightly increased liposomal estradiol penetra-tion, over that of simple passive diffusion. The total amount penetrated was only 1.3-foldhigher than for passive treatment. Such a low estradiol penetration was unexpected fromultradeformable liposomes, as the authors felt it reasonable to assume that the combin-ation of the two promoting mechanisms (phospholipids acting as chemical enhancer andelectroporation as a physical force) would markedly enhance penetration. The group

00

2

4

6

8

10

12

1 2 3 4 5 6Time (h)

0.0 mA/cm2

0.5 mA/cm2

0.2 mA/cm2

0.8 mA/cm2

Figure 17.2 Iontophoretic estradiol delivery through human epidermal membranes from ultra-deformable liposomes at different current densities (n 5 5).




then devised an experiment to probe further the effect of vesicles on electroporatedepidermal membranes.

A 5% w/v suspension of empty ultradeformable liposomes (containing no drug) wasprepared. As most of the liposomes reported in the literature are simple phospholipidvesicles, a pure PC vesicle, that is, with neither drug nor edge activator, was alsoinvestigated.

The idea of the experiment was to produce the same number of pores within the skin,by applying the same pulsing protocol for all diffusion cells with water in the donorchambers. Then, half of the cells (treated cells) were dosed with empty liposomesuspension (with or without edge activator); the remainder acted as control. After30 min of skin treatment with empty vesicles (with or without edge activator), thesuspension was removed from the treated cells and skin was repeatedly washed withwater. For control cells, water in the donor chambers was dried off using soft tissuepaper. Then, penetration of estradiol from saturated aqueous solution (150 ml) throughcontrol and liposome-treated skin was monitored for 2.5 h (Stage I). To delineate furtherthe possible recovery of the skin barrier, another run of pulses was then applied while thetarget molecule remained in the donor (Stage II), and penetration was followed foranother 2 h.

Cumulative amounts of estradiol penetrated with time from control and treated cellsare shown in Figure 17.3a. These graphs show an enhancement in estradiol penetrationduring Stage I, with higher results from the control. After the second pulsing (Stage II),the penetration once again increased, with the control illustrating a marked rise com-pared to the liposome-treated cells.

Expanded Stage I graphs (Figure 17.3b) clearly reveal higher estradiol penetrationfrom control cells relative to lipid-treated epidermal membranes. The plots also display asteady increase in estradiol amount penetrated during that stage, suggesting that thedamage introduced to the skin by the high-voltage pulses persisted for at least severalhours after pulse cessation. In spite of the fact that estradiol solution was applied 30 min

00 0

0

1

2

3

1 2 31 2 3 4 5

10

20

30

40

50

60

70

80

Time (h) Time (h)

Drug

Second pulsing

First pulsing

Stage IIStage I

Control

Liposome-PC+cholate

Liposome-PCExpanded Stage I

(a) (b)

Figure 17.3 (a) Two stage human epidermal penetration of estradiol from saturated aqueoussolution (n 5 6) after two sets of pulses (5 3 100 V, 100 ms and 1 min spacing) through control(water-treated) and empty liposome-treated skin (with or without cholate), (b) expanded stage Iplots.




after pulsing, estradiol penetrated the skin well. This implies that the main pathway of theneutral lipophilic estradiol penetrating under electroporation is mainly through partition-ing into, and diffusing through, the highly impaired intercellular skin lipid structure orthrough the electrically created aqueous pathways.

The overall results provided evidence that the presence of phospholipid after pulsingpromoted considerable repair of the altered skin barrier. Nevertheless, cells treated withultradeformable liposomes showed higher estradiol penetration in Stage I compared tothose treated with pure PC vesicles. This could be explained by the presence of sodiumcholate (edge activator) in the former formulation with its known penetration enhancingeffect [81]. However, the enhancing effect of edge activator did not overcome thepenetration retarding effect of the released PC monomers. Additionally, a minor effectmay have been due to the 14% lower PC content in this lipid suspension compared tothe pure PC suspension. However, both formulations gave essentially the same results(P> 0.05) during Stage II, where a considerable amount of estradiol came through theskin. This may be due to the incomplete recovery of some of the highly damaged areas ofthe intercellular lipids. As a small lipophilic molecule, estradiol managed to penetratewell through these unrepaired skin defects.

Concluding Remarks

The combined use of chemical and physical enhancement strategies has shown promisefor the delivery of molecules both dermally and transdermally, although the studies arestill largely investigative, with few signs of clinical testing. However, the combination mayyet prove attractive for delivery of more challenging, larger molecules (including pep-tides and proteins). The recent finding of a ‘‘barrier repair’’ effect of phospholipid onelectrically-treated skin may provide a caveat to future formulators when combininglipids with electrical enhancement.

References1. Barry, B.W. Dermatological Formulations: Percutaneous absorption. Marcel Dekker, New

York and Basel, 1983, Chapter 1.2. Potts, R.O., Bommannan, D.B., and Guy, R.H. Percutaneous absorption. In: Mukhtar, H. (ed.),

Pharmacology of the Skin: Pharmacology and Toxicology. CRC Press, Boca Raton, Florida,1992, 13.

3. Friend, D.R. Transdermal delivery of contraceptives. Crit. Rev. Ther. Drug Carrier Syst., 7, 149,1990.

4. Cleary, G.W. The first two decades of transdermal drug delivery systems and a peek into the1990s. In: Gurney, R. and Teubner, A. (eds), Dermal and Transdermal Drug Delivery: NewInsights and Perspectives. Wissenchaftliche Verlagsgesell Schaft mbh, Stuttgart, 1993, 13.

5. Guy, R.H. and Hadgraft, J. Transdermal drug delivery: the ground rules are emerging. Pharm.Int., 6, 112, 1985.

6. Knepp, V.M., Hadgraft, J., and Guy, R.H. Transdermal drug delivery: problems and possibil-ities. Crit. Rev. Ther. Drug Carrier Syst., 4, 13, 1987.

7. Egbaria, K., Ramachandran, C., and Weiner, N. Liposomes as topical drug delivery systems.Adv. Drug Del. Rev., 5, 287, 1990.

8. Weiner, N. et al. Liposomes: a novel topical delivery system for pharmaceutical and cosmeticapplication. J. Drug Target., 2, 405, 1994

9. Barry, B.W. Mode of action of penetration enhancers in human skin. J. Control. Rel., 6, 85, 1987.




10. Williams, A.C. and Barry, B.W. Skin absorption enhancers. Crit. Rev. Ther. Drug Carrier Syst., 9,305, 1992.

11. Tyle, P. and Agrawala, P. Drug delivery by phonophoresis. Pharm. Res., 6, 355, 1989.12. McElnay, J.C. et al. The use of ultrasound in skin penetration enhancement. In: Walters, K.A.

and Hadgraft, J. (eds), Pharmaceutical Skin Penetration Enhancement. Marcel Dekker, NewYork and Basel, 1993, 293.

13. Bhatia, K.S. and Singh, J. Mechanism of transport enhancement of LHRH through porcineepidermis by terpenes and iontophoresis: permeability and lipid extraction studies. Pharm.Res., 15, 1857, 1998.

14. Bhatia, K.S. and Singh, J. Synergistic effect of iontophoresis and a series of fatty acid on LHRHpermeability through porcine skin. J. Pharm. Sci., 87, 462, 1998.

15. Choi, E.H. et al. The pretreatment effect of chemical skin penetration enhancers in transdermaldrug delivery using iontophoresis. Skin Pharmacol. Appl. Skin Physiol., 12, 326, 1999.

16. Wang, Y.M., Allen, L.V., and Li, L.C. Effect of sodium dodecyl sulphate on iontophoresis ofhydrocortisone across hairless mouse skin. Pharm. Dev. Technol., 5, 533, 2000.

17. Le, L., Kost, J. and Mitragotri, S. Combined effect of low frequency ultrasound and iontophor-esis: Application for transdermal heparin delivery. Pharm. Res., 17, 1151, 2000.

18. Mitragotri, S. and Kost, J. Low frequency sonophoresis: a noninvasive method of drug deliveryand diagnostics. Biotech. Prog., 16, 488, 2000.

19. Tiwari, S.B. and Udupa, N. Investigation into the potential of iontophoresis facilitated deliveryof ketorolac. Int. J. Pharm., 260 (1), 93, 2003.

20. Burnette, R.R. Iontophoresis. In: Hadgraft, J. and Guy, R.H. (eds), Transdermal Drug Delivery:Developmental Issues and Research Initiatives. Marcel Dekker, New York and Basel, 1989, 247.

21. Green, P.G. Iontophoretic delivery of peptide drugs. J. Control. Rel., 41, 33, 1996.22. Riviere, J.E. and Heit, M.C. Electrically assisted transdermal drug delivery. Pharm. Res., 14, 687,

1997.23. Pikal, M.J. The role of electroosmotic flow in transdermal iontophoresis. Adv. Drug Del. Rev.,

46, 281–305, 2001.24. Burnette, R.R. and Ongpipattanakul, B. Characterisation of the pore transport properties and

tissue alteration of excised human skin during iontophoresis. J. Pharm. Sci., 77, 132, 1988.25. Pikal, M.J. Transport mechanisms in iontophoresis. I. A theoretical model for the effect of

electroosmotic flow on flux enhancement in transdermal iontophoresis. Pharm. Res., 7, 118,1990.

26. Pikal, M.J. and Shah, S. Transport mechanisms in iontophoresis. II. Electroosmotic flow andtransference number measurements for hairless mouse skin. Pharm. Res., 7, 213, 1990.

27. Pikal, M.J. and Shah, S. Transport mechanisms in iontophoresis. III. An experimental study ofthe contributions of electroosmotic flow and permeability change in transport of low and highmolecular weight solutes. Pharm. Res., 7, 222, 1990.

28. Kalia, Y.N. et al. Iontophoretic drug delivery. Adv. Drug Del. Rev., 56, 619, 2004.29. Bryan, J. Transdermal drug delivery may be a common technique in the future. Pharm. J., 273,

292, 2004.30. Cullander, C. and Guy, R.H. Sites of iontophoretic current flow into the skin: identification and

characterisation with the vibrating probe electrode. J. Invest. Dermatol., 97, 55, 1991.31. Monteiro-Riviere, N.A. Identification of the pathway of transdermal iontophoretic drug deliv-

ery: ultrastructural studies using mercuric chloride in vivo in pigs. Pharm. Res., 8, S-141, 1991.32. Scott, E.R. et al. Transport of ionic species in skin: contribution of pores to the overall skin

conductance. Pharm. Res., 10, 1699, 1993.33. Weaver, J.C., Vaughan, T.E., and Chizmadzhev, Y. Theory of electrical creation of pathways

across skin transport barriers. Adv. Drug. Del. Rev., 35, 21, 1999.34. Prausnitz, M.R. A practical assessment of transdermal drug delivery by skin electroporation.

Adv. Drug Del. Rev., 35, 61, 1999.35. Pliquett, U.F., Langer, R., and Weaver, J.C. Changes in the passive electrical properties of

human stratum corneum due to electroporation. Biochim. Biophys. Acta, 1239, 111, 1995.




36. Pliquett, U.F. and Weaver, J.C. Electroporation of human skin: simultaneous measurement ofchanges in the transport of two fluorescent molecules and in the passive electrical properties.Bioelectrochem. Bioenerget., 39, 1, 1996.

37. Vanbever, R. et al. Comparison of the effects of short, high-voltage and long, medium voltagepulses on skin electrical and transport properties. J. Control. Rel., 69, 35, 1999.

38. Prausnitz, M.R. et al. Electroporation of mammalian skin: a mechanism to enhance transdermaldrug delivery. Proc. Natl. Acad. Sci. USA, 90, 10504, 1993.

39. Vanbever, R., Lecoutturier, N., and Preat, V. Transdermal delivery of metoprolol by electro-poration. Pharm. Res., 11, 1657, 1994.

40. Vanbever, R., Le Boulenge, E., and Preat, V. Transdermal delivery of fentanyl by electropor-ation I. Influence of electrical factors. Pharm. Res., 13, 559, 1996.

42. Pliquett, U.F. et al. Imaging of fluorescent molecules and small ion transport through humanstratum corneum during high voltage pulsing: localised transport regions are involved. Bio-phys. Chem., 58, 185, 1996.

43. Prausnitz, M.R. et al. Imaging regions of transport across human stratum corneum during high-voltage and low-voltage exposure. J. Pharm. Sci., 85, 1363, 1996.

44. Chen, T., Langer, R., and Weaver, J.C. Skin electroporation causes molecular transport acrossthe stratum corneum through localised regions. J. Invest. Dermatol. Symp. Proc., 3, 159, 1998.

45. Vanbever, R., Leroy, M.A., and Preat., V. Transdermal permeation of neutral molecules byelectroporation. J. Control. Rel., 54, 243, 1998.

46. Yu, H.Y. and Liao, H.M. Triamcinolone permeation from different liposome formulationsthrough rat skin in vitro. Int. J. Pharm., 127, 1, 1996.

47. Touitou, E. et al. Diphylline liposomes for delivery to the skin. J. Pharm. Sci., 81, 131, 1992.48. Schreier H. and Bouwstra J. Liposomes and niosomes as topical drug carriers: dermal and

transdermal drug delivery. J. Control. Rel., 30, 1, 1994.49. Cevc, G. and Blume, G. Lipid vesicles penetrate into intact skin owing to the transdermal

osmotic gradient and hydration force. Biochim. Biophys. Acta, 1104, 226, 1992.50. Cevc, G., Schatzlein, A., and Richardsen, H. Ultradeformable lipid vesicles can penetrate the

skin and other semi-permeable barriers un-fragmented. Evidence from double label CLSMexperiments and direct size measurements. Biochim. Biophys. Acta, 1564, 21, 2002.

51. Honeywell-Nguyen, P.L. et al. The in vivo and in vitro interactions of elastic and rigid vesicleswith human skin. Biochim. Biophys. Acta, 1573, 130, 2002.

52. Honeywell-Nguyen, P.L. et al. Transdermal delivery of pergolide from surfactant-based elasticand rigid vesicles: characterisation and in vitro transport studies. Pharm. Res., 19, 991, 2002.

53. Honeywell-Nguyen, P.L. et al. The in vivo transport of elastic vesicles into human skin: effectsof occlusion, volume and duration of application. J. Control. Rel., 90, 243, 2003.

54. Cevc, G. et al. Ultra-high efficiency and peptide transfer through the intact skin by means ofnovel drug-carriers, transfersomes. In: Brain, K.R. and Walters K.A. (eds), Prediction of Percu-taneous Penetration, 3rd International Conference, 14–16 April, Cardiff, 1993, 226.

55. Planas, M. et al. Noninvasive percutaneous induction of topical analgesia by a new type ofdrug carrier, and prolongation of local pain insensitivity by anesthetic liposomes. Aneth.Analg., 75, 615, 1992.

56. Cevc, G., Blume, G., and Schatzlein, A. Transfersome-mediated transepidermal delivery im-proves the regio-specificity and biological activity of corticosteroids in vivo. J. Control. Rel., 45,211, 1997.

57. El Maghraby, G.M.M., Williams, A.C., and Barry, B.W. Skin delivery of oestradiol from deform-able and traditional liposomes: mechanistic studies. J. Pharm. Pharmacol., 51, 1123, 1999.

58. El Maghraby, G.M.M., Williams, A.C., and Barry, B.W. Oestradiol skin delivery from ultrade-formable liposomes: refinement of surfactant concentration. Int. J. Pharm., 196, 63, 2000.

59. El Maghraby, G.M.M., Williams, A.C., and Barry, B.W. Skin delivery of estradiol from lipidvesicles: importance of liposome structure. Int. J. Pharm., 204, 159, 2000.

60. El Maghraby, G.M.M., Williams, A.C., and Barry, B.W. Skin delivery of 5-flurouracil fromultradeformable and standard liposomes in vitro. J. Pharm. Pharmacol., 53, 1069, 2001.




61. El Maghraby, G.M.M., Williams, A.C., and Barry, B.W. Skin hydration and possible shunt routepenetration in controlled estradiol delivery from ultradeformable and standard liposomes.J. Pharm. Pharmacol., 53, 1311, 2001.

62. Cevc, G. and Blume, G. Biological activity and characteristics of triamcinolone–acetonideformulated with the self-regulating drug carriers, Transfersomes1. Biochim. Biophys. ActaBiomembr., 1614, 156, 2003.

63. Cevc, G. et al. Ultraflexible vesicles, Transfersomes, have an extremely low pore penetrationresistance and transport therapeutic amount of insulin across intact mammalian skin. Biochim.Biophys. Acta, 1368, 201, 1998.

64. Cevc, G. Rationale for the production and dermal application of lipid vesicles. In: Falco, O.B.,Korting, H.C., and Maibach, H.I. (eds), Liposome Dermatics, Springer-Verlag, Berlin, 1992, 82.

65. Cevc, G. et al. The skin: a pathway for systemic treatment with patches and lipid-based agentcarriers. Adv. Drug Deliv. Rev., 18, 349, 1996.

66. Cevc, G. and Blume, G. New, highly efficient formulation of diclofenac for topical, transdermaladministration in ultradeformable drug carriers, Transfersomes. Biochim. Biophys. Acta, 1514,191, 2001.

67. Paul, A. and Cevc, G. Non-invasive administration of protein antigens: transdermal immun-ization with bovine serum albumin in transfersomes. Vaccine Res., 4, 145, 1995.

68. Paul, A., Cevc, G., and Bachhawat, B. K. Transdermal immunization with large proteins bymeans of ultradeformable drug carriers. Eur. J. Immunol., 25, 5321, 1995.

69. Kulkarni, S.B., Banga, A.K., and Betageri, G.V. Transdermal iontophoretic delivery of colchi-cine encapsulated in liposomes. J. Drug Del., 3, 245, 1996.

70. Badkar A.V. et al. Enhancement of transdermal iontophoretic delivery of a liposomal formu-lation of colchicine by electroporation. Drug Deliv., 6, 111, 1999.

71. Vutla, N.B., Betageri, G.V., and Banga, A.A. Transdermal iontophoretic delivery of enkephalinformulated in liposomes. J. Pharm. Sci., 85, 5, 1996.

72. Fang, J. et al. Transdermal iontophoretic delivery of enoxacin from various liposome-encapsulated formulations. J. Control. Rel., 60, 1, 1999.

73. Kirjavainen, M. et al. Influence of lipid on the mannitol flux during transdermal iontophoresisin vitro. Eur. J. Pharm. Sci., 10, 97, 2000.

74. Li, G.L., Danhof, M., and Bouwstra, J.A. Effect of elastic liquid state vesicle on apomorphineiontophoresis transport through human skin in vitro. Pharm. Res., 11, 1627, 2001.

75. Essa, E.A., Bonner, M.C., and Barry, B.W. Iontophoretic estradiol skin delivery and tritiumexchange in ultradeformable liposomes. Int. J. Pharm., 240, 55, 2002.

76. Blume, A. et al. Interaction of phospholipid liposomes with lipid model mixture of stratumcorneum lipids. Int. J. Pharm., 99, 219, 1993.

77. Hofland, H.E. J. et al. Interaction between liposomes and human stratum corneum in vitro:freeze fracture electron microscopical visualization and small angle x-ray scattering studies. Br.J. Dermatol., 132, 853, 1995.

78. Jadoul, A. et al. Modification induced on stratum corneum after in vitro iontophoresis: ATR-FTIR and x-ray scattering studies. Int. J. Pharm., 42, 165, 1996.

79. Jadoul, A., Bouwstra, J. and Preat, V. Effect of iontophoresis and electroporation on the stratumcorneum. Adv. Drug Deliv. Rev., 35, 89, 1999.

80. Essa, E.A., Bonner, M.C., and Barry, B.W. Electroporation and ultradeformable liposomes;human skin repair by phospholipid. J. Control. Rel., 92, 163, 2003.

81. Almog, S. and Lichtenberg, D. Effect of calcium on kinetic and structural aspects of dilution-induced micellar to lamellar phase transformation in PC-cholate mixture. Biochemistry, 27,873, 1988.




ASSESSMENT OF

PENETRATION

ENHANCEMENT

IV



Chapter 18

Mechanistic Studiesof Permeation Enhancers

S. Kevin Li and William I. Higuchi

CONTENTS

Introduction .................................................................................................................................... 272Methods .......................................................................................................................................... 273

Animal Model.............................................................................................................................. 273Transport Experiments ............................................................................................................... 273

Permeability Coefficient Determination................................................................................. 273Reversibility Study................................................................................................................... 274Model Description and Analysis of Experimental Data ........................................................ 275Permeant Solubility Determination........................................................................................ 276Determinations of Partition Coefficients in Bulk Organic Solvent/PBS Systems ................. 276

Partition Experiments ................................................................................................................. 276n-Heptane Treatment and SC Preparation............................................................................. 276HMS SC Delipidization ........................................................................................................... 276Partition Experiments with Heptane-Treated and Delipidized HMS SC .............................. 277

Permeant Partitioning into the Transport Rate-Limiting Domain andEquilibrium Permeant Partitioning into the Stratum Corneum Intercellular Lipids ............. 278

Results and Discussion ................................................................................................................... 278Isoenhancement Concentrations and Enhancer Effects ........................................................... 278Effects of Alkyl Chain Length..................................................................................................... 279Effects of Polar Head Functional Groups .................................................................................. 281Effects of Hydrocarbon Chain Carbon–Carbon Double Bond ................................................. 281Effects of Branched Alkyl Chain ................................................................................................ 283Equilibrium Partition Enhancement of ES into SC Intercellular Lipids .................................... 286Transport Rate-Limiting Domain and Equilibrium Partitioning Domain.................................. 286Effects of Permeation Enhancement on Permeants of Different Molecular Sizes ................... 287Permeation Enhancers in a Nonaqueous System in Transdermal Drug Delivery ................... 289


271


Conclusions .................................................................................................................................... 290Acknowledgment ........................................................................................................................... 290References....................................................................................................................................... 290

Introduction

Most of the chemical permeation enhancer studies over the past decades have beenaimed at gaining better insights into the relationship between the nature of the enhancersand their effectiveness in permeation enhancement. In typical in vitro studies of chemicalpermeation enhancers, the enhancer in question is usually applied with a drug in solutionor suspension to one side of the membrane, and the effectiveness of the enhancercompared to a control is determined by the rate of transport of the drug. Under thisapproach, the different relationships among the enhancer molecular structures and theireffects as permeation enhancers have been observed (e.g., reviewed in Lee et al., 1991;Smith and Maibach, 1995; Hadgraft, 2001).

Our laboratory has been studying themechanismof action of permeation enhancers formore than a decade. A different experimental approach has been employed in thesestudies (Kim et al., 1992; Yoneto et al., 1995; Warner et al., 2003; He et al., 2004). First, ifmechanistic insight is to be collected directly, a symmetric and equilibrium configuration(with respect to the enhancer) should be used. In the symmetric configuration, theenhancer is present at equal concentrations in both the donor and receiver chambers ofa side-by-side diffusion cell and in equilibrium with the membrane. Under these condi-tions, the complications arising from enhancer concentration (or activity) gradients acrossthe membrane (Liu et al., 1991, 1992) can be avoided. These enhancer gradients wouldotherwise lead to a situation in which the local permeation enhancement varies with theposition within the membrane and make mechanistic data analysis difficult. With thesymmetric configuration, the permeability coefficients obtained for the permeants can beused directly to determine the effectiveness of an enhancer in enhancing transdermaltransport (enhancer potency). Second, good enhancers are usually lipophilic and relativelywater-insoluble. Because they are water-insoluble, well absorbed, and need to be solubil-ized for effective presentation, they cannot be systematically investigated conveniently forstructure–enhancement activity. Establishing equilibrium between the enhancer and themembrane is therefore necessary in obtaining mechanistic insights into the action ofpermeation enhancers and for establishing a structure–enhancement relationship. Third,model analyses to separate the effects of permeation enhancement on transport across thelipoidal and pore transport pathways using model permeants of different polarity havebeen used. The effects of the enhancers upon the intercellular lipoidal and pore pathwaytransport have been delineated in order to understand the mechanism of action of theenhancers. Last, in the assessment of skin permeation enhancement, changes in thechemical potential (activity) of the permeant in the enhancer solution with respect to thatin the buffer solution (the control) are corrected for to account for the effects of permeantactivity alteration upon transport in the presence of the enhancers. Consideration of theissues discussed above has allowed us to develop a research strategy to gain mechanisticinsights into the effects of enhancers upon transport across membranes, to determine theintrinsic potencies of the enhancers, and to establish a quantitative structure–enhancementrelationship between the enhancers and permeation enhancement. This chapter is areview of our studies employing this strategy and the experimental approaches.




Methods

Animal Model

All experiments were conducted with freshly separated hairless mouse skin (HMS)obtained from the abdomen region and freed from adhering fat and other visceral debris.HMS was selected as the model for human skin for the following reasons. HMS hasrelatively constant lipid content (Yoneto et al., 1998), and a large body of HMS data isavailable in the literature allowing the direct comparisons of our results with those inprevious studies. HMS stratum corneum (SC) lipid composition (Grubauer et al., 1989) isalso similar to that of human skin (Lampe et al., 1983). In certain cases, such as forexperiments requiring long-term skin stability in aqueous solution, HMS is not a goodmodel of human skin (Lambert et al., 1989); however, in the investigation of chemicalpermeation enhancers for the lipoidal pathways and where relatively short experimentaltimes are involved, HMS has been found to be an adequate quantitative model forhuman skin (Kim et al., 1992; Li et al., 1997). There is no direct evidence of significantdiscrepancy between the mechanism of action of permeation enhancers in HMS andhuman skin.

Transport Experiments

Permeability Coefficient Determination

The permeability experiments were carried out as previously described with a two-chamber side-by-side diffusion cell in phosphate buffered saline (PBS) at 378C (Yonetoet al., 1995; Warner et al., 2003). Each compartment has a 2-mL volume and an effectivediffusional area of 0.67 cm2. The skin membrane was sandwiched between the two halfcells and an enhancer solution in PBS was pipetted into both chambers. A list of the

with the HMS, the enhancer solution in both diffusion cell chambers was replaced untilthe SC was essentially in equilibrium with the enhancer solution (enhancer/PBS). Withhighly enhancers (e.g., 1-dodecyl-2-pyrrolidone), due to the extensivedepletion of the enhancers in the aqueous phase that precluded achieving an equilibriumof the enhancers between the aqueous phase and the SC, an aqueous reservoir system(enhancer solubilizing system of micelles or cyclodextrin) that neither significantlyinteracts with the SC nor acts as a permeation enhancer, was used (Warner, 2003 Shakeret al., 2003). The concentrations of the enhancers in the diffusion cell chamber werefrequently checked by HPLC or GC. The loss or depletion of all the enhancers was lessthan 5% in most cases and less than 15% in the extreme case at the end of the transportexperiments. Corticosterone (CS) was the main model permeant. Estradiol (ES) andhydrocortisone (HC) were the two other steroidal permeants tested in the studies.Other permeants used will be discussed in the section on Effects of Permeation Enhance-ment on Permeants of Different Molecular Sizes. The steroidal permeant at radiotracerlevels and at concentrations far below its solubility was added to the donor chamberfollowing enhancer equilibration. Samples were withdrawn from the donor and receiverchambers at predetermined time intervals and analyzed. Permeability experiments withthe model ionic polar permeant (TEA) were conducted in essentially the same manner.The total permeability coefficients (PT) were determined from data obtained understeady-state conditions (around three to five times longer than the lag times). The


Mechanistic Studies of Permeation Enhancers & 273

enhancers investigated is shown in Figure 18.1. To attain equilibrium of the enhancer

hydrophobic


permeability coefficient of the dermis–epidermis combination (PD/E) was obtained in thesame manner but the abdominal skin was stripped 30 times with 3M Scotch tape prior toassembly into the diffusion cell.

Reversibility Study

Diffusion cells were assembled with full thickness HMS as described above for a typicalpermeation experiment, and equilibrium between the membrane and the enhancersolution was allowed to take place. However, in this protocol, both chambers of thediffusion cell were then rinsed with PBS to remove the enhancer equilibrated in themembrane. Following the PBS rinsing regime, transport studies with PBS in both cham-bers were carried out. The permeability coefficients obtained with PBS after pretreatmentwith enhancers were then compared with those obtained with pretreatment with PBSonly. All enhancers were tested for reversibility at E ¼ 10 and their effects upon perme-ation across SC were shown to be essentially reversible (permeability coefficients in PBS

1-alkyl-2-pyrrolidones (AP) 1-alkyl-2-piperidinones (API) 1-alkyl-2-azacycloheptanones (AZ)

R

OO

2-(1-alkyl)-2-methyl-1,3-dioxolanes (MD) 1,2-alkanediols (AD) 1-alkanols (AL)

n-alkyl-b -D-glucopyranosides (AG) N,N-dimethyl alkanamides (AM)

OH

OHO

O

R

R OHOH

R

1,2-dihydroxypropyl alkanoates (MG) trans-3-alken-1-ols (TAL) cis-3-alken-1-ols (CAL)

R'OH

R

branched alkanols (bAL)

N

O

RN

O

RN

O

R

HO RHO

OH

R

N

O

R

O

O R

HO

HO

HO HO

Figure 18.1 Chemical structures of the enhancers. R, R’5 alkyl chain.




after enhancer pretreatment were within a factor of two of those in PBS without thepretreatment).

Model Description and Analysis of Experimental Data

The permeability coefficients (P) for the probe permeants were calculated according toEquation (18.1) (Warner et al., 2001):

P ¼ 1

ACD

dQ

dt(18:1)

where A is the diffusional area of the diffusion cell, CD is the concentration in the donorchamber, and dQ/dt is the slope of the linear region of the amount of permeant inreceiver chamber (Q) vs time plot.

Total permeability coefficient expression for full-thickness skin is written as follows:

PT ¼ 11PSC

þ 1PD=E

(18:2)

where PSC is the permeability coefficient for the SC and PD/E is the permeability coeffi-cient for the epidermis–dermis combination (D/E) and can be obtained from experimentsfrom tape-stripped skin. PSC can be further divided into parallel lipoidal and porepathway components in SC via the following equation:

PSC ¼ PL þ PP (18:3)

where PL and PP are the permeability coefficients for the lipoidal pathway and the porepathway (TEA is used as the probe permeant for estimating the magnitude of PP),respectively, in the SC. The intercellular lipid domain in SC is generally accepted as thelipoidal transport pathway across SC. Substituting Equation (18.3) into Equation (18.2)yields

PT ¼ 11

PLþPPþ 1

PD=E

(18:4)

Based on the results from previous studies, the use of CS as the probe permeant allowsEquation (18.4) to be approximated by

PT PL (18:5)

For other steroidal permeants, PL can be calculated by Equation (18.4) with PD/E and PPvalues obtained from transport experiment with stripped skin and TEA. The equation forthe lipoidal pathway transport enhancement factor (E ) is

E ¼ PL;X

PL;O

SXSO

(18:6)

where PL, X and PL,O are the permeability coefficients for the lipoidal pathway whenthe solvent is enhancer/PBS and PBS, respectively, and SX and SO are the permeant




solubilities in enhancer/PBS and in PBS, respectively. The solubility ratio corrects for anyactivity coefficient differences between the activity coefficient in PBS and that in theenhancer solution. Use of the solubility ratio assumes that Henry’s Law is obeyed to CSsaturation in both PBS and enhancer solutions (Kim et al., 1992).

Permeant Solubility Determination

The solubilities of the steroidal permeants in PBS and the enhancer solutions weredetermined by adding excess crystals of the permeant into the enhancer solution inPyrex culture tubes. The drug suspension was shaken for 72 h at 378C. The culturetubes were then centrifuged for 15 min at 3500 rpm and the clear supernatants wereanalyzed for permeant concentrations with HPLC.

Determinations of Partition Coefficients in Bulk OrganicSolvent/PBS Systems

Organic solvent/PBS partition coefficients were obtained at the aqueous enhancer con-centrations corresponding to E ¼ 10 and at one tenth of the E ¼ 10 concentration, thelatter to test whether Henry’s law is obeyed in the two liquid phases. The two-phasesystems were maintained at 378C for 72 h. Both the organic and aqueous phases werecentrifuged, and aliquots were carefully withdrawn from both phases and appropriatelydiluted for subsequent analysis using HPLC or GC.

Partition Experiments

n-Heptane Treatment and SC Preparation

Before SC preparation, HMS was rinsed with heptane for 3 10 sec to remove the SCsurface lipids. This rinsing protocol (the number of rinses and the rinse time) was shownto remove approximately 20% of the SC lipids but did not disrupt the SC barrier (He et al.,2003). Similar treatments with nonpolar organic solvent were also shown to remove skinsurface lipids (e.g., Abrams et al., 1993; Nicolaides, 1974). SC was then prepared accord-ing to the method described by kligman and Christophers (1963) and Yoneto et al. (1998).Briefly, the skin was placed, dermis side down on a filter paper (quantitative filter paperNo. 1, Whatman1) mounted on a Petri dish. The Petri dish was filled with 0.2% trypsin inPBS solution up to the surface of SC. The Petri dish was covered and maintained at 378Cfor 16 h. When the skin membrane was placed in distilled water after the trypsintreatment, the dermis and viable epidermal layers would separate and fall away fromthe SC. The SC was then rinsed with distilled water several times and swabbed withKimwipe1 tissue paper to remove excess water. Then, the SC was placed on aluminumfoil and dried at room temperature. After drying, the SC was kept in a freezer for later use.

HMS SC Delipidization

Heptane-treated HMS SC samples were prepared as described in the previous section.The delipidized HMS SC was prepared according to the method described previously(Yoneto et al., 1998). Briefly, dried n-heptane-treated SC samples (about 1 to 2 mg) were




weighed and transferred into 5 mL CHCl3/MeOH (2:1) mixture and equilibrated for 48 hat room temperature. The residue of SC was then rinsed several times with fresh CHCl3/MeOH (2:1) mixture and dried under room temperature for 24 h. The dried residue wascarefully weighed and used for the partition experiments.

Partition Experiments with Heptane-Treated and Delipidized HMS SC

Partition experiments were carried out to determine the uptake amounts of the chemicalpermeation enhancer and of estradiol (ES) into n-heptane-treated or delipidized HMS SC.Two different partition experimental setups have been used in our laboratory. The oldsetup has used a Franz diffusion cell (Yoneto et al., 1998) and will not be discussed here.The following is a brief description of the new method (Chantasart et al., 2004). SC (about1 to 2 mg) or delipidized SC sample was carefully weighed and equilibrated in about20mL of enhancer solution (E ¼ 10 concentration) containing trace amounts of 3H-ES in ascrew capped glass vial. The vial was sealed with parafilm to prevent enhancer evapor-ation and put in a thermostatted water bath with shaking at 37+ 0.18C for 12 h. The 12-hincubation period was chosen because preliminary studies showed that equilibrium ofenhancer and 3H-ES with the SC sample took place in less than 12 h and that a longerincubation period might result in too fragile a membrane sample for the partitioningexperiments. After 12 h, the SC sample was then taken out from the solution by tweezersand blotted by Kimwipe tissue paper. The enhancer and ES concentrations of the solutionin the screw-capped glass vial were checked. The wet sample was carefully weighed in asnap-capped glass bottle. Then, 5 mL of 100% ethanol was added into the bottle to extractthe enhancer and ES for 48 h at room temperature with occasional gentle agitation. Theextracted solution was then transferred to a screw-capped Pyrex test tube. The test tubewas centrifuged at 3500 rpm for 15 min. The supernatant was analyzed for the enhancerby GC or HPLC and for ES by a scintillation counter.

The uptake amount of enhancer in the heptane-treated SC or delipidized SC wascalculated as follows:

Acorrected, i ¼Aextracted, i

Wdry (Wwet Wdry)

Ci

Wdry(18:7)

where Aextracted, i is the amount of enhancer extracted from heptane-treated or delipidizedHMS SC, Wdry is the dried heptane-treated or delipidized SC weight, and the subscript irepresents the enhancer. A correction for the enhancer in the aqueous compartment(s) ofthe SC was calculated according to the wet weight of SC (Wwet) and the concentration ofthe enhancer in aqueous bulk phase (Ci). The partition coefficient of ES for partitioningfrom the aqueous phase into n-heptane-treated SC or delipidized SC (KES) was calculatedas follows:

KES ¼ [A0extracted (Wwet Wdry)C

0i ]=Wdry

C 0i

S 0XS 0O

(18:8)

where A0extracted is the amount of extracted 3H-ES. C 0

i is the concentration of the 3H-ES inaqueous bulk phase. S 0X and S 0O are the solubilities of ES in enhancer solution and in PBS,respectively. The solubility ratio corrects for any activity coefficient differences betweenthe activity coefficient in PBS and that in the enhancer solution.




Permeant Partitioning into the Transport Rate-Limiting Domainand Equilibrium Permeant Partitioning into the Stratum CorneumIntercellular Lipids

An important question raised in the above transport and equilibrium partition studieswas: is the equilibrium partition enhancement data of ES a direct correlate of the partitionenhancement of ES in SC permeation? To accomplish this, the partition enhancement inSC permeation was to be determined in skin transport experiments and a non-steady-state transport analysis (He et al., 2005). However, a direct comparison of the partitionenhancement data obtained in transport experiments and those data obtained in equi-librium partitioning experiments of ES was not practical due to the D/E layer being asignificant barrier for ES permeation across HMS. Furthermore, significant ES metabolismwas observed in ES transdermal penetration. Because of these difficulties, non-steady-state ES transport analysis was complicated, and it was decided to employ CS as thesurrogate permeant for ES in the following study. The strategy here was to examine therelationship between the transport partitioning enhancement of CS and the equilibriumpartitioning enhancement of ES, with the assumption that ES and CS should likely behavesimilarly. This assumption was considered to be reasonable because previous studies hadshown similar permeability coefficient enhancement effects of chemical enhancers withES and CS for permeation across the lipoidal pathway of HMS SC (Yoneto et al., 1995).

The skin transport model (He et al., 2005) is a two-layer numerical transport simula-tion with a least squares-fitting software Scientist (MicroMath, Salt Lake City, UT). Thismodel divides the SC and D/E into a sufficient number of layers characterized bypartition, diffusion, and dimension parameters. The permeant concentration in thedonor chamber was assumed constant, which was true in all transport experimentscarried out in the study. The receiver chamber concentration was kept at sink conditions.The transport data of full-thickness HMS was analyzed using the model to obtain thepartition coefficient (KSC) and diffusion coefficient (DSC) of SC. The reduced parametersKSC’ and DSC’ of SC were then calculated:

DSC0 ¼ DSC=L

2 (18:9)

KSC0 ¼ KSCL (18:10)

where L is the effective path length across SC. These reduced parameters KSC’ and DSC’were defined (Okamoto et al., 1988) to avoid the difficulty and uncertainty in assigningthe L value and to minimize the number of parameters for least square fitting in modelanalysis of the experimental transport data. The enhancement of KSC’ and DSC’ (EK, SC andED, SC, respectively) was calculated by dividing the KSC’ and DSC’ parameters obtained withthe enhancers at E ¼ 10 by those with PBS control.

Results and Discussion

Isoenhancement Concentrations and Enhancer Effects

concentration for estradiol (ES), corticosterone (CS), and hydrocortisone (HC) permeationacross the SC lipoidal pathway with the 1-butyl-2-pyrrolidone, 1-hexyl-2-pyrrolidone,



Figure 18.2 shows a representative plot of enhancement factor vs. aqueous enhancer


1-octyl-2-pyrrolidone as permeation enhancers (Yoneto et al., 1995). Similar enhancementfactor vs. aqueous enhancer concentration plots were observed for all the enhancers

are essentially the same for the steroidal permeants of different lipophilicity, suggesting thesame mechanism of permeation enhancement for these steroidal permeants.

Effects of Alkyl Chain Length

The isoenhancement concentrations at E ¼ 10 for more than 20 different enhancers are

were interpolated from the E vs. aqueous enhancer concentration plots similar to those inFigure 18.2. Figure 18.3 shows the relationship between the E ¼ 10 enhancer concentra-tion and the carbon number of the enhancer n-alkyl group. The major conclusiondeduced from the data in Figure 18.3 is a slope of around 0.55 found for each enhancerseries (enhancers have the same polar head functional group but different alkyl chainlength) in the figure. The value of 0.55 translates into an around 3.5-fold increase inpotency per methylene group for the enhancers. In other words, the aqueous concen-tration required to induce E ¼ 10 (at constant permeant thermodynamic activity) in-creases 3.5-fold when the alkyl chain length of the enhancer decreases by onemethylene group. The constant slope of 0.55 for the different enhancer series suggestsa hydrophobic effect involving the transfer of the methylene group from the aqueousphase to a relatively nonploar organic phase (e.g., Tanford, 1980).

The results of the equilibrium partition experiments with the enhancers con-ducted to determine the amount of enhancers in the SC intercellular lipids under the

0

5

10

15

0.01 0.1 1 10

Enhancer concentration (% w/w)

Enh

ance

men

t fac

tor

(E)

ES BP CS BP HC BP

ES HP CS HP HC HP

ES OP CS OP HC OP

Figure 18.2 Transport enhancement factors of estradiol (ES), corticosterone (CS), andhydrocortisone (HC) across the SC lipoidal pathway in the presence of 1-butyl-2-pyrrolidone(BP), HP, and OP. The transport enhancement factors were calculated using Equations (18.1)to (18.6).



presented in Figure 18.3 (Warner et al., 2003). These isoenhancement concentrations

studied. The enhancement factor profiles at increasing aqueous enhancer concentrations


isoenhancement E ¼ 10 conditions (He et al., 2003, 2004) are shown in Figure 18.4. Notethat the scale of the y-axis in Figure 18.4 is the same as that in Figure 18.3. The data inFigure 18.4 suggest that there was little effect of the enhancer alkyl chain length upon theenhancer potency based on the concentrations of the enhancers in the intercellular lipidlamellae (relative to that based on the E ¼ 10 aqueous enhancer concentrations in

0.00001

0.0001

0.001

0.01

0.1

1

2 4 6 8 10 12 14Carbon number

Aqu

eous

con

cent

ratio

n (M

)

AP AL API AZ AD

AM AG MD MG

Figure 18.3 Relationships between the aqueous E5 10 isoenhancement concentrations of theenhancers and the carbon number of the enhancer alkyl chain. Each data point represents theaverage value without showing the standard deviation because the error bar generally lies within

0.001

0.01

0.1

1

10

100


Mem

bran

e co

ncen

trat

ion

(μm

ol/m

g m

embr

ane)

AP AL API AZ

AD AM AG

Figure 18.4 Relationship between the enhancer concentrations in the intercellular lipid domainof the SC membrane at the E5 10 isoenhancement conditions and the carbon number of theenhancer alkyl chain. Each data point represents the average value. (Enhancer abbreviations areprovided in Figure 18.1.)



the symbol in the plot. (Enhancer abbreviations are provided in Figure 18.1.)


action is relatively independent of their alkyl group chain length and lipophilicity.

Effects of Polar Head Functional Groups

The data in Figure 18.3 also show that some enhancer polar functional groups are moreeffective (more potent) than the others in inducing permeation enhancement. For ex-ample, the E ¼ 10 isoenhancement concentrations of 1-alkyl-2-azacycloheptanones areconsistently around tenfold lower than those of 1-alkyl-2-pyrrolidones at the samen-alkyl chain length, suggesting that the azacycloheptanone group makes the 1-alkyl-2-azacycloheptanones more effective as permeation enhancers compared with the pyrro-lidone group of the 1-alkyl-2-pyrrolidones based on their concentration in the aqueousphase in the donor and receiver chamber. However, the relative constant concentration

little effect of the enhancer polar head functional group upon the enhancer potencybased on the concentrations of the enhancers at their site of action. This is an interestingfinding because studies using conventional experimental methods in the literature havedemonstrated the influence of the polar head functional group of an enhancer upon itseffectiveness in transdermal permeation enhancement (e.g., Smith and Maibach, 1995).In particular, it has been suggested that the azacycloheptanone functional group is morepotent than other polar head functional groups in general due to specific interactionsbetween the functional group and the ceramide lipid matrix (e.g., Brain et al., 1993;Hadgraft et al., 1996). The data in Figure 18.4, however, imply that the polar head andalkyl groups of the enhancers act only to transfer the enhancers from the aqueous phaseto the hydrocarbon phase of the lipid bilayer and make available the enhancers for theiraction in the transport rate-limiting domain.

enhancement relationship for the enhancers (Warner et al., 2003). The correlation be-tween the E ¼ 10 isoenhancement concentrations and the octanol/water partition coef-ficients of the enhancers with a slope of around 1 suggests that the potencies of theenhancers for the steroidal permeants are related to the enhancer lipophilicities. Togetherwith the data analysis in Figure 18.3 and Figure 18.4, it is reasonable to hypothesize that(a) permeation enhancement is related to the ability of the permeation enhancer topartition into the transport rate-limiting domain, (b) the polar head group assists thetranslocation of the enhancer to the site of action through a free energy of transfer fromthe bulk aqueous phase to the transport rate-limiting domain, and (c) the transport rate-limiting domain has a microenvironment with polarity similar to the polarity of bulkoctanol. These hypotheses and the results of Figure 18.5 have been discussed in detail in

of the enhancer site of action’’).

Effects of Hydrocarbon Chain Carbon–Carbon Double Bond

The effects of substituting a single carbon–carbon bond on the n-alkyl chain of anenhancer with a carbon–carbon double bond have been investigated, and the E ¼ 10isoenhancement concentrations of the cis- and trans 3-alken-1-ols (closed symbols) are

(He et al., 2004). The data for the n-alkanols reported in Figure 18.3 are also included



Figure 18.3), thus suggesting that the intrinsic potency of the enhancers at their site of

of enhancer uptake into the SC lipid domain at E ¼ 10 in Figure 18.4 reveals that there is

Figure 18.5 is a replot of the data shown in Figure 18.3 to demonstrate a structure-

plotted against the carbon numbers of the enhancer hydrocarbon chains in Figure 18.6

Chapter 3 (‘‘Quantitative structure–enhancement relationship and the microenvironment


in Figure 18.6 for comparison. The E ¼ 10 isoenhancement concentrations of the cis- andtrans-3-alken-1-ols and of the n-alkyl enhancers vs. their octanol/PBS partition coeffi-cients (Koctanol/PBS

obtained from Figure 18.5. In Figure 18.6, it is seen that the E ¼ 10 isoenhancementconcentrations of the cis- and trans-3-alken-1-ols are two to three times higher than thoseof the corresponding n-alkanols (open squares). Based on the criterion that the E ¼ 10

−5

−4

−3

−2

−1

0

0.5 1.5 2.5 3.5 4.5 5.5

Log Koctanol/PBS

Log(

aque

ous

conc

entr

atio

n [M

])

AP AL API AZ AD

AM AG MD MG

Figure 18.5 Correlation between the aqueous E 5 10 isoenhancement concentrations of theenhancer and its octanol/PBS partition coefficient (Koctanol/PBS). The slope of the line is 21. Eachdata point represents the average value without showing the standard deviation because theerror bar generally lies within the symbol in the plot. (Enhancer abbreviations are provided in

0.0001

0.001

0.01

0.1

1

2 4 6 8 10 12Carbon number

Aqu

eous

con

cent

ratio

n (M

)

AL

CAL

TAL

Figure 18.6 Relationships between the aqueous E 5 10 isoenhancement concentrations of theenhancers and the carbon number of the enhancer alkyl chain. Each data point represents theaverage value without showing the standard deviation because the error bar generally lies withinthe symbol in the plot. (Enhancer abbreviations are provided in Figure 18.1.)



Figure 18.1.)

) are shown in Figure 18.7. The line in Figure 18.7 is the correlation line


isoenhancement (aqueous) concentration is a measure of the enhancer potency, the plots

n-alkanols by a factor of 2 to 3, but the cis- and trans-3-alken-1-ol data fall closely onthe regression line in Figure 18.7 when the lipophilicity of the enhancers is taken intoconsideration. The correlation between the E ¼ 10 isoenhancement concentration andoctanol/PBS partition coefficient here continues to be consistent with the demonstrated

In equilibrium partition experiments of the cis- and trans-3-alken-1-ols, the concen-trations of enhancers in the SC intercellular lipid domain under the isoenhancement

The substitution of a single

effect upon the enhancer potency based on the concentrations of the enhancers in the SCintercellular lipid domain. This is somewhat surprising because unsaturated enhancersare expected to be more potent than saturated enhancers based on molecular modelingof skin permeation and previous experimental results (Golden et al., 1987; Aungst, 1989;Brain et al., 1993; Tenjarla et al., 1999).

Effects of Branched Alkyl Chain

Branched chain alkanols (2-alkanols, 3-alkanols, 4-alkanols, and 5-nonanol) were alsoinvestigated as another group of skin permeation enhancers to provide new insights intothe mechanism of enhancement action of both n-alkyl and branched chain enhancers.The 2-alkanols, 3-alkanols, and 4-alkanols are 2-hexanol, 2-heptanol, 2-octanol, 2-non-anol, 3-hexanol, 3-heptanol, 3-octanol, 3-nonanol, 4-heptanol, 4-octanol, and 4-nonanol,respectively. In concentrations at E ¼ 10 of thebranched alkanols (closed symbols) are plotted against their Koctanol/PBS values.

−5

−4

−3

−2

−1

0

0.5 1.5 2.5 3.5 4.5 5.5Log K octanol/PBS

Log

(aqu

eous

con

cent

ratio

n [M

])

AP AL API AZ

AD AM AG MDMG CAL TAL

Figure 18.7 Correlation between the aqueous E 5 10 isoenhancement concentrations of theenhancer and its octanol/PBS partition coefficient (Koctanol/PBS). Each data point represents theaverage value without showing the standard deviation because the error bar generally lies withinthe symbol in the plot. The slope of the line is 21. (Enhancer abbreviations are provided in



Figure 18.1.)

of Figure 18.6 would suggest the cis- and trans-3-alken-1-ols are less potent than the

structure–enhancement relationship for the n-alkyl enhancers in Figure 18.5.

E ¼ 10 conditions are essentially constant (Figure 18.8).carbon–carbon bond with a carbon–carbon double bond on the alkyl chain here has little

isoenhancementtheFigure 18.9,


Again, the data of the n-alkyl enhancers (including the n-alkanols) are also included inFigure 18.9, and the straight line shown in the figure is a best fit line based on the data for

the n-alkyl enhancers (crosses), the branched chain alkanols (closed symbols) showmodest but consistent positive deviations (in the direction of lower potency) from the

0.001

0.01

0.1

1

10


Mem

bran

e co

ncen

trat

ion

(μm

ol/m

g m

embr

ane)

AP AL API AZ AD

AM AG CAL TAL

Figure 18.8 Relationship between the enhancer concentrations in the intercellular lipid domainof the SC membrane at the E 5 10 isoenhancement conditions and the carbon number of theenhancer alkyl chain. Each data point represents the average value. (Enhancer abbreviations are

−3.5

−3

−2.5

−2

−1.5

-1

1 1.5 2 2.5 3 3.5 4

Log K octanol/PBS

Log

(aqu

eous

con

cent

ratio

n [M

])

2-alkanols

3-alkanols

4-alkanols

5-nonanol

n-alkyl enhancers

Figure 18.9 Correlation between the aqueous E 5 10 isoenhancement concentrations of theenhancer and its octanol/PBS partition coefficient (Koctanol/PBS). Each data point represents theaverage value without showing the standard deviation because the error bar generally lies withinthe symbol in the plot. The slope of the line is 21.



provided in Figure 18.1.)

more than 20 n-alkyl enhancers in Figure 18.5. Different from the random deviations for


quantitative structure–enhancement correlation (the straight line). These deviations sup-port the view, based on the assumption that Koctanol/PBS is a valid predictor of enhancerpotency, that the branched chain alkanols are slightly less potent than the n-alkylenhancers. The lower potencies based on the E ¼ 10 aqueous concentrations of thebranched chain alkanols are a result of decreasing intrinsic potency and increasingeffective hydrophilicity of the enhancers when the hydroxyl group moves from theterminal end towards the center of the enhancer alkyl chain. Nevertheless, the resultsof the branched chain alkanols continue to support the hypothesis previously establishedfor the n-alkyl enhancers that the potency of an enhancer based on its aqueous concen-tration increases with enhancer lipophilicity.

Figure 18.10 presents the concentrations of the branched chain alkanols and n-alkylenhancers in the SC intercellular lipids under the isoenhancement conditions of E ¼ 10.Whereas the intrinsic potencies of the n-alkyl enhancers are essentially the same andindependent of their alkyl chain length, branching of the alkyl chain decreases theintrinsic potencies of the enhancers; the concentrations of the branched alkanols in theSC intercellular lipid domain (closed symbols) required to induce the E ¼ 10 conditionsare generally higher than those of the n-alkyl enhancers (crosses in the figure). This result

Despite the observed deviation of the branched chain alkanols from the n-alkyl chainenhancers, it should be noted that the correlation between the logarithm of the enhancerpartition coefficient from the aqueous phase to the SC intercellular lipid phase (log KSC

lipid/PBS) and log Koctanol/PBS continues to hold for the branched chain alkanols. Themicroenvironment of the enhancer site of action remains essentially the same andindependent of alkyl-chain branching; the n-alkanols, branched chain alkanols, and all

volume).

0.001

0.01

0.1

1

10

0.5 1.5 2.5 3.5 4.5 5.5Log K octanol/PBS

Mem

bran

e co

ncen

trat

ion

(μm

ol/m

g m

embr

ane)

2-alkanols3-alkanols4-alkanols5-nananoln-alkyl enhancers

Figure 18.10 Relationship between the enhancer concentrations in the intercellular lipiddomain of the SC membrane at the E 510 isoenhancement conditions and the octanol/PBSpartition coefficient (Koctanol/PBS) of the enhancers. Each data point represents the average value.



is consistent with the relatively lower intrinsic potency of the branched chain alkanolssuggested with the data in Figure 18.9.

other studied enhancers fall on the same regression line (Figure 3.9 in Chapter 3 of this


Equilibrium Partition Enhancement of ES into SC Intercellular Lipids

In addition to the equilibrium partition experiments with the enhancers, experimentswere also conducted with a model steroidal compound ES. The goal here is to determinethe enhancement of the partitioning of a lipophilic permeant into the SC intercellularlipids under the isoenhancement E ¼ 10 conditions. Figure 18.11 shows the plots of thepartition coefficients of ES from the aqueous phase into the SC intercellular lipid domain(KES) under the E ¼ 10 conditions of more than 20 different enhancers vs. the Koctanol/PBS

values of the enhancers. The KES values were determined with Equation (18.8). Thedotted line represents the KES value in PBS control. As can been seen in the figure,approximately the same enhancement of KES (four to sevenfold) was induced under theisoenhancement conditions of E ¼ 10 for all the enhancers studied. This constant four tosevenfold enhancement in permeant partitioning suggests that (a) the same target site inthe SC lipid lamellae is fluidized by the studied enhancers, (b) the uptake domain probedin these partitioning studies is at the same time the transport rate-limiting domain and theenhancer site of action, and (c) the tenfold permeation enhancement corresponds toaround a 4- to 7- and 1.5- to 2.5-fold enhancement in permeant partitioning and diffusion,respectively, in the transport rate-limiting domain.

Transport Rate-Limiting Domain and Equilibrium Partitioning Domain

It would be inappropriate to conclude that the uptake domain probed in the equilibriumpartitioning experiments is at the same time the transport rate-limiting domain andthe enhancer site of action with only the KES data above. Comparison of the parti-tion enhancement in transport across the SC rate-limiting domain and the partitionenhancement in the equilibrium partition experiments is required. Consistent enhancereffects upon transport and equilibrium partitioning would suggest that the intercellular

0

100

200

300

400

500

600

1 2 3 4 5Enhancer Log K octanol/PBS

ES

par

titio

n co

effic

ient

AP AZ AL AG API

AM AD CAL TAL bAL

PBS

Figure 18.11 Relationship between the partition coefficients of ES (KES) for partitioning fromthe aqueous phase into the SC intercellular lipid domain and the enhancer octanol/PBS partitioncoefficient (Koctanol/PBS). The dotted line represents the KES value in PBS control. Each data pointrepresents the average and its standard deviation (n>3). The standard deviations of Log Koctanol/

PBS are not shown because the error bars generally lie within the symbol in the plot. (Enhancer



abbreviations are provided in Figure 18.1.)


lipid domain probed in the partitioning experiments is the same as the transport rate-limiting domain for permeation across HMS SC.

As described in the Experimental section, the cumulative amount of CS transportedacross HMS vs. time profiles in CS transport experiments were analyzed with a transportmodel to obtain the least squares-fitting KSC’ and DSC’ of CS in PBS, 1-octyl-2-pyrrolidone(OP), and 1-hexyl-2-azacycloheptanone (HAZ) (He et al., 2005). The least squares-fittingsof the CS transport data were satisfactory and the results show that the enhancement ofpermeant partitioning into the transport rate-limiting domain of HMS is significantlyhigher than the enhancement of permeant diffusion coefficient in the domain. Whenthe total flux enhancement (E) was 12 for OP, EK,SC was 6.0+ 1.9 and ED,SC was 1.8+ 0.9(mean+ SD, n 3). For HAZ with E of 11, EK,SC was 7.9+ 2.8 and ED,SC was 1.3+ 0.6(mean+ SD, n 3). This suggests that the transport enhancement of CS was mainlydriven by partition enhancement in the rate-limiting domain of SC. The consistencybetween the partitioning enhancement of transport found with the SC rate-limitingdomain (EK, SC around 6 to 8) and the equilibrium partitioning enhancement of ES withthe intercellular lipids of HMS SC (in the range of 5 to 7) is believed to be quite important.This finding provides quantitative evidence that the rate-limiting domain for the transportof the model permeants through the lipoidal pathway of HMS SC and the intercellularlipid ‘‘phase’’ probed in the equilibrium partitioning experiments have similar propertiesregarding the partitioning enhancement effects of chemical permeation enhancers uponthe lipophilic model permeants and therefore that these domains are likely to be the

Effects of Permeation Enhancement on Permeants ofDifferent Molecular Sizes

Most of the work so far presented in this chapter was based on the data of a single modelsteroidal permeant CS. Two other steroidal permeants HC and ES were also employed toexamine the generality of the transport enhancement results, and essentially the samepermeation enhancement was observed with all three steroidal permeants (e.g.,

transport of the permeant across SC. For example, it is general knowledge that there maybe a steep-permeant molecular size dependence in permeation across lipid bilayers (e.g.,Stein, 1986; Xiang and Anderson, 1994), and when enhancers fluidize the SC lipids, theincrease in the bilayer free volume can have different consequences regarding transportenhancement of permeants with different molecular sizes.

To examine the effects of permeant molecular size upon transport enhancement,transport experiments were conducted using permeants of different molecular sizes

1-Hexyl-2-pyrrolidone (HP) and OP were the model permeation enhancers in this study.

lipoidal pathway vs. the molecular weight of the permeants under the isoenhancementconditions: E ¼ 10 for steroidal permeants. The enhancement factors are calculated usingEquations (18.4) and (18.6) and with the assumption that the presence of the enhancersdid not affect the thermodynamic activity of the permeant in the aqueous solution.As discussed earlier, the enhancement factors for ES, CS, and HC are essentially the

upon permeation enhancement. This strong molecular weight dependence is consistent



same. This result also supports the conclusions presented in Chapter 3.

Figure 18.2). However, the physiochemical properties of a permeant can influence the

Figure 18.13 presents the results of the enhancement factors of transport across the SC

same in Figure 18.13. However, there is a strong permeant molecular weight dependence

and lipophilicities (Figure 18.12) under the E ¼ 10 conditions for CS (Warner, 2003).


with an enhancer-induced increase in the free volume of the SC intercellular lipids,which favors the transport enhancement of permeants of large molecular sizes. Theeffects of permeant lipophilicity upon permeation enhancement were minimal, and nosignificant dependency between permeation enhancement and permeant lipophilicitywas observed among the studied permeants. Given the results in Figure 18.13, some

OH OH OH

Ethanol Propanol Butanol

OHHO H2N O

Phenol Benzyl alcohol Benzamide

NHO

OH

O

O

HO

OHO

O

OH

OH

HO

2-Acetamidophenol 4-Ipomeanol Kaempferol

O

HO

O

OH

OH

H H

H

H H

H

OH

OH

Hydrocortisone Estradiol

Figure 18.12 Chemical structures of the probe permeants.

0

2

4

6

8

10

12

10 100 1000Permeant molecular weight

Enh

ance

men

t fac

tor

(E)

HP

OP

Figure 18.13 Relationship between the transport enhancement factors and the molecularweight of the permeants under the E5 10 condition for the steroidal permeants with HP orOP as the enhancers. Each data point represents the average and its standard deviation (n>3).




caution needs to be taken in generalizing the results presented in this chapter topermeants of different physicochemical properties. Further investigation on this subjectis required.

Permeation Enhancers in a Nonaqueous System inTransdermal Drug Delivery

Although the studies presented in this chapter did not include nonaqueous vehicles orconventional co-solvents, the conclusions derived from these experiments are expectednot to be limited only to the aqueous system. First, unless the vehicle is able to partitioninto the SC intercellular lipid ‘‘phase’’ and itself behaves as an enhancer, a nonaqueoussystem should not affect the intrinsic potency of the enhancer. In this scenario, thenonaqueous vehicle or co-solvent may only alter the thermodynamic activity of theenhancer in a transdermal patch and alter the partitioning tendency of the enhancerfrom the patch vehicle into the SC. The concentration of the enhancer at its site of actionmay therefore be lowered or raised, but this effect can be predicted from thermodynam-

Another important issue is the symmetric situation with the enhancers in equilibriumwith skin in our study of permeation enhancers. In transdermal drug delivery, enhancerpermeation occurs across the SC from the transdermal patch to the blood sink, thisresulting in an asymmetric enhancer situation with an enhancer concentration gradientin the SC. This enhancer concentration gradient is related to the permeability coefficientsPSC and PD/E of the enhancer. For illustrative purposes Figure 18.14 qualitativelyshows the SC concentration gradients of two enhancers with different lipophilicitiesand permeability coefficients across the SC. As can be seen in Figure 18.14, the

CD, membrane

Blood sink Combinedviableepidermis andsome dermis

SC

Enhancerconcentration

Enhancer A

Enhancer B

Ci

Patch

Figure 18.14 Enhancer concentration profiles in SC in transdermal drug delivery (see Equation(18.2)): dotted line, Enhancer A; dashed line, Enhancer B. PD/E for Enhancers A and B are thesame, log Koct/PBS for Enhancer A > Enhancer B, and PSC for Enhancer A >> Enhancer B. Thisanalysis assumes SC is homogenous and does not account for (a) enhancer-induced variation inlocal enhancement (permeation and partition enhancement) at different locations within the SCand (b) enhancer-induced enhancement for the enhancer. These will affect the enhancer con-centration profile in SC and lead to nonlinear profiles.



ics. This is discussed in Chapter 3 of this volume.


absorption of the more lipophilic enhancer (Enhancer A) is largely dermis-controlled andtherefore exhibits a relatively constant concentration across the SC compared with that ofthe other enhancer (Enhancer B). For Enhancer B, due to its relatively low permeabilityacross the SC, a much more significant concentration drop across the SC is observed.Thus, a large portion of the SC is not affected by Enhancer B and this region of the SCbecomes the rate-limiting barrier for drug transport. The relative constant concentrationof Enhancer A in the SC would suggest that lipophilic enhancers are likely to be moreeffective in providing uniform transport enhancement over the entire SC and a highoverall flux enhancement of drug transport across SC. However, simply applying themost lipophilic enhancer does not guarantee success. The solubility of the enhancer anddepletion of the enhancer in the transdermal patch are other factors that need to beconsidered.

Conclusions

New insights into the factors influencing the effectiveness of chemical permeationenhancers for the lipoidal pathway of the SC have been obtained. The present studysupports the view that (a) the potency of an n-alkyl enhancer (based on its aqueousconcentration) is related to the enhancer lipophilicity, this being the case because of thelipophilic nature of the enhancer site of action, which is well mimicked by liquidn-octanol; (b) the intrinsic potency of the enhancer (as represented by its concentrationat the target site of action) is relatively independent of its lipophilicity; (c) the substitutionof a carbon–carbon single bond on the hydrocarbon chain of the enhancer with acarbon–carbon double bond does not significantly affect its intrinsic potency; and (d)with modest effects, branching of the n-alkyl chain of the enhancer generally reduces theintrinsic potency of the enhancer. To date, we have not encountered any enhancercandidates that are inconsistent with this view. However, skin penetration retardershave been reported (e.g., Hadgraft et al., 1996). This suggests that further studies areneeded for greater generalizations of the present findings. Nevertheless, the presentstudy has demonstrated useful concepts and effective methodologies for mechanisticstudies of permeation enhancers.

Acknowledgment

The authors thank Kevin S. Warner, Ning He, and Doungdaw Chantasart for theircontributions in the project and the financial support by NIH Grants GM 043181 andGM 063559.

ReferencesAbrams K, Harvell JD, Shriner D, Wertz P, Maibach H, Maibach HI, Rehfeld SJ, Effect of organic

solvents on in vitro human skin water barrier function. J Invest Dermatol, 101, 609–613, 1993.Aungst BJ, Structure/effect studies of fatty acid isomers as skin penetration enhancers and skin

irritants. Pharm Res, 6, 244–247, 1989.Brain KR, Walters KA, Molecular modeling of skin permeation enhancement by chemical agents.

In: Pharmaceutical Skin Penetration Enhancement, Walters KA, Hadgraft J (Eds), 1993, MarcelDekker, New York. p. 389–416.




Chantasart D, Li SK, He N, Warner KS, Prakongpan S, Higuchi WI, Mechanistic studies of branched-chain alkanols as skin permeation enhancers. J Pharm Sci, 93, 762–779, 2004.

Golden GM, McKie JE, Potts RO, Role of stratum corneum lipid fluidity in transdermal drug flux.J Pharm Sci, 76, 25–28, 1987.

Grubauer G, Feingold KR, Harris RM, Elias PM, Lipid content and lipid type as determinants of theepidermal permeability barrier. J Lipid Res, 30, 89–96, 1989.

Hadgraft J, Modulation of the barrier function of the skin. Skin Pharmacol Appl Skin Physiol, 14Suppl 1, 72–81, 2001.

Hadgraft J, Peck J, Williams DG, Pugh J, Allan G, Mechanisms of action of skin penetrationenhancers/retarders: azone and analogues. Int J Pharm, 141, 17–25, 1996.

He N, Li SK, Suhonen TM, Warner KS, Higuchi WI, Mechanistic study of alkyl azacycloheptanonesas skin permeation enhancers by permeation and partition experiments with hairless mouseskin. J Pharm Sci, 92, 297–310, 2003.

He N, Warner KS, Chantasart D, Shaker DS, Higuchi WI, Li SK, Mechanistic study of chemical skinpermeation enhancers with different polar and lipophilic functional groups. J Pharm Sci, 93,1415–1430, 2004.

He N, Warner KS, Higuchi WI, Li SK, Model analysis of flux enhancement across hairless mouseskin induced by chemical permeation enhancers. Int J Pharm, 297, 9–21, 2005.

Kligman AM, Christophers E, Preparation of isolated sheets of human stratum corneum. ArchDermatol, 88, 702–705, 1963.

Kim YH, Ghanem AH, Higuchi WI, Model studies of epidermal permeability. Semin Dermatol, 11,145–156, 1992.

Lambert WJ, Higuchi WI, Knutson K, Krill SL, Effects of long-term hydration leading to thedevelopment of polar channels in hairless mouse stratum corneum. J Pharm Sci, 78, 925–928,1989.

Lampe MA, Williams ML, Elias PM, Human epidermal lipids: characterization and modulationsduring differentiation. J Lipid Res, 24, 131–140, 1983.

Lee VHL, Yamamoto A, Kompella UB, Mucosal penetration enhancers for facilitation of peptideand protein drug absorption. Crit Rev Ther Drug Carrier Syst, 8, 91–192, 1991.

Li SK, Ghanem A-H, Yoneto K, Higuchi WI, Effects of 1-alkyl-2-pyrrolidones on the lipoidalpathway of human epidermal membrane: a comparison with hairless mouse skin. Pharm Res,14, S-303, 1997.

Liu P, Higuchi WI, Song WQ, Kurihara-Bergstrom T, Good WR, Quantitative evaluation of ethanoleffects on diffusion and metabolism of beta-estradiol in hairless mouse skin. Pharm Res, 8,865–872, 1991.

Liu P, Higuchi WI, Ghanem A-H, Kurihara-Bergstrom T, Good WR, Assessing the influence ofethanol in simultaneous diffusion and metabolism of estradiol in hairless mouse skin for the‘‘asymmetric’’ situation in vitro. Int J Pharm, 78, 123–136, 1992.

Nicolaides N, Skin lipids: their biochemical uniqueness. Science, 186, 19–26, 1974.Okamoto H, Hashida M, Sezaki H, Structure–activity relationship of 1-alkyl or 1-alkenylazacy-

cloalkanone derivatives as percutaneous penetration enhancers. J Pharm Sci, 77, 418–424, 1988.Shaker DS, Ghanem AH, Li SK, Warner KS, Hashem FM, Higuchi WI, Mechanistic studies of the

effect of hydroxypropyl-beta-cyclodextrin on in vitro transdermal permeation of corticosteronethrough hairless mouse skin. Int J Pharm, 253, 1–11, 2003.

Smith EW, Maibach HI, Percutaneous Penetration Enhancers. 1995, CRC Press, Boca Raton,Florida.

Stein W, Transport and Diffusion Across Cell Membranes. 1986, Academic Press, New York.Tanford C, The Hydrophobic Effect: Formation of Micelles and Biological Membranes. 2nd edition.

1980, John Wiley & Sons, New York.Tenjarla SN, Kasina R, Puranajoti P, Omar MS, Harris WT, Synthesis and evaluation of N-acetylpro-

linate esters — novel skin penetration enhancers. Int J Pharm, 192, 147–158, 1999.Warner KS, Mechanistic Aspects of Chemical Skin Permeation Enhancers, PhD thesis, University of

Utah, 2003.




Warner KS, Li SK, Higuchi WI, Influences of alkyl group chain length and polar head group onchemical skin permeation enhancement. J Pharm Sci, 90, 1143–1153, 2001.

Warner KS, Li SK, He N, Suhonen TM, Chantasart D, Bolikal D, Higuchi WI, Structure–activityrelationship for chemical skin permeation enhancers: probing the chemical microenvironmentof the site of action. J Pharm Sci, 92, 1305–1322, 2003.

Xiang TX, Anderson BD, The relationship between permeant size and permeability in lipid bilayermembranes. J Membr Biol, 140, 111–122, 1994.

Yoneto K, Ghanem AH, Higuchi WI, Peck KD, Li SK, Mechanistic studies of the 1-alkyl-2-pyrrolidones as skin permeation enhancers. J Pharm Sci, 84, 312–317, 1995.

Yoneto K, Li SK, Higuchi WI, Shimabayashi S, Influence of the permeation enhancers 1-alkyl-2-pyrrolidones on permeant partitioning into the stratum corneum. J Pharm Sci, 87, 209–214,1998.




Chapter 19

Penetration EnhancerAssessment byCorneoxenometry

Claudine Pierard-Franchimont, Frederique Henry, Emmanuelle Uhoda,Caroline Flagothier, and Gerald E. Pierard

CONTENTS

Introduction .................................................................................................................................... 293Corneoxenometry........................................................................................................................... 294Corneoxenometry and Dose–Response Effect of Chemical Penetration Enhancers................... 295Corneoxenometry and Organic Solvents ...................................................................................... 296Conclusion ...................................................................................................................................... 297References....................................................................................................................................... 297

Introduction

One of the most important functions of the epidermis is the formation of a well-structuredbarrier between the body and the ingress of potentially hostile xenobiotics. The lattercompounds are as various as environmental contaminants, toxins, microorganisms, andothers. Its function as a barrier is also vital to maintain constant the internal milieu. Muchresearch has been undertaken to understand the skin barrier function which resides inthe stratum corneum (SC). In some instances, however, chemical penetration enhancers(absorption enhancers or accelerants) represent an attractive potential in order to over-come the barrier efficacy and to increase the penetration of drugs through the SC.Penetration enhancers induce a temporary and reversible decrease in the skin barrierproperties. They act in a number of ways. Some of them alter the solubility propertiesor disrupt the ordered nature of the epidermal lipids [1]. Other molecules alter thecohesiveness between corneocytes.


293


The desirable attributes for penetration enhancers are varied [1–3]. The compoundshould be pharmacologically inert with no action at receptor sites within the body. Therisk for irritation, allergy, and toxicity should be minimal. The enhancer should becompatible, both chemically and physically, with drugs and vehicles in the dosageform. It should possess a rapid onset of action with a predictable duration of activity.In addition, the effects should be completely and rapidly reversible upon removal of thematerial from the skin. Furthermore, the effects should be unidirectional, allowing onlythe ingress of specific xenobiotics without loss of any endogenous compound from thebody. Ideally, the penetration enhancer should be cosmetically acceptable, spreadingwell on the skin with a suitable ‘‘feel.’’ It should be odorless, inexpensive, tasteless,and colorless.

Despite the wide range of compounds proposed as penetration enhancers, no chem-ical combines all of the desirable attributes. Many compounds have been assessed aschemical skin penetration enhancers. Some are chemicals specifically designed for thispurpose such as 1-dodecylazacycloheptan-2-one (laurocapram or Azone1). Others aremore common constituents of topical formulations such as surfactants and solvents.The efficacies of enhancers toward various drugs have been largely explored andcompared [4].

The two classes of penetration enhancers, namely the solvent type and the lipidfluidizer type, can be combined to reach synergistic effects [5, 6]. In complex formula-tions, each component can act in many different ways, precluding the determinationof the different interactions that are possible. Binary and ternary mixtures have beenreported to be better than single penetration enhancers [7]. The exact combinationactivity of the chemicals is, however, difficult to ascribe until a more precise knowledgeof the mechanisms of action has been ascertained [1]. There is a need for accurateassessments of the SC permeability alterations because when the effect of penetrationenhancers can be measured, safe, reliable, and effective formulations can be made [8].This chapter will focus on the value of corneoxenometry in predicting the value ofchemical penetration enhancers.

Corneoxenometry

In vivo testing with penetration enhancers has been performed safely by some re-searchers in contrast to others who reported severe cell damage in the epidermis andeven skin necrosis [9]. Such hazards call for ex vivo predictive bioassays on human skin orSC [9, 10]. The corneoxenometry bioassay named after corneocyte, xenobiotic, and metryhas been introduced as a convenient approach to explore the effect of some xenobioticson human SC [11]. It is a variant of corneosurfametry which was specifically designed totest diluted surfactants [12–16].

Corneoxenometry has been used for testing a series of chemicals harmful to the SC[11, 17–19]. The bioassay entails collection of cyanoacrylate skin surface strippings (CSSS)from normal human skin. The harvested SC sheet which is uniform in thickness issubjected to the ex vivo action of the selected xenobiotics. CSSS covered in excess withtheir respective chemicals are kept for 2 h at room temperature in a closed environmentto prevent evaporation of the test solution. They are then thoroughly rinsed underrunning tap water, air dried, and stained with a toluidine blue-basic fuschin solution atpH 3.45 for 3 min. Lipid removal and protein denaturation induce increased dye binding




on corneocytes (Figure 19.1). It has been shown that harsh compounds to the skin

considerably increase the intensity of staining of the CSSS [11–19]. After placing thesamples on a white reference tile, reflectance colorimetry (Chroma Meter CR200 Minolta,Osaka, Japan) is used to derive the L* and Chroma C* values. Colorimetric data are usedto quantify the corneoxenometry bioassay. The colorimetric index of mildness (CIM) iscalculated as previously defined [11, 15–18] following: CIM¼ L*-Chroma C*. The relativeindex of irritancy (RII) is calculated following: [RII¼ 1[(CIM product) (CIM water)1]. Itis evident that RII is not a direct measure of the barrier function. However, it correlateswith clinical signs of irritancy and transepidermal water loss increase when surfactantsare tested [20]. In fact the bioassay explores the combined effects of lipid removal anddisorganization, and of protein denaturation as well. Hence, any RII increase is a clue forSC damage responsible for barrier function impairment.

Corneoxenometry and Dose–Response Effect of ChemicalPenetration Enhancers

A dose–response effect was searched for ethanol and laurocapram using the corneox-enometry bioassay [19]. In the same study, other assessments were performed using a gelformulation (propylene carbonate, hydroxypropylcellulose, butylhydroxytoluene, etha-nol, and glycerol) containing 10% propylene glycol and a combination of three otherenhancers, namely N-acetyl-L-cysteine (NAC), urea, and salicylic acid (SA). The threelatter penetration enhancers were incorporated in various proportions keeping the sumof their respective concentrations at the 20% level.

Data from corneoxenometry were reproducible and sensitive enough to disclosesignificant differences between formulations. Both the nature and concentration ofpenetration enhancers affected the RII values. For each tested formulation, the interindi-divudal variability was reasonably low. Linear dose–effect responses were obtained with

Figure 19.1 Corneoxenometry. Aspect of a cyanoacrylate skin surface stripping stained by atoluidine blue-basic fuschin after contact with a penetration enhancer. The staining of corneo-cytes is uneven and indicates where the damages take place.


Penetration Enhancer Assessment by Corneoxenometry & 295


ethanol in the range 0 to 100%, and laurocapram in the range 0 to 5%. The 10% propyleneglycol-based gel exhibited a wide range in RII values when supplemented with NAC,urea, and SA. In the bioassay, NAC exhibited a moderate effect. RII increased withincreasing amounts of urea replacing NAC. The RII increase was more striking whensupplementing with SA instead of urea. The combination of SA and urea always provedto be more active than SA alone.

Corneoxenometry and Organic Solvents

The effects of organic solvents have been studied in many instances [21, 22]. They werecompared using corneoxenometry [17]. Series of CSSS were immersed for 1, 5, 10, 30, 60,or 120 min in vials containing deionized water or an organic solvent including chloro-form, ethanol, hexane, methanol, chloroform:methanol (2:1, v/v), hexane:ethanol (2:3,v/v), and hexane:methanol (2:3, v/v). After contact with the selected solvent for thepredetermined time, CSSS were thoroughly rinsed under running tap water for 20 s, air-dried, and stained for 3 min with toluidine blue-basic fuschin dyes.

The ranking from the least to the most aggressive product according to the mean CIMwas as follows: hexane (40.7), ethanol (26.5), methanol (23.5), hexane–ethanol (23.3),chloroform (20.8), chloroform–methanol (15.5), and hexane–methanol (7.8). CIM valuesshowed that the effect of hexane–methanol on SC was significantly higher ( p < 0.01) thanthose of all other solvents with the exception of chloroform–methanol. There was nosignificant difference between ethanol, methanol, and hexane–ethanol, but each of themwas significantly ( p < 0.05) more active than hexane.

The influence of exposure time between solvents and the SC showed some differ-ences according to the solvents. However, all correlation were significant ( p < 0.01) andbest fitted a logarithmic relationship. It appeared that most of the changes in CIM werereached within 10 min for each solvent.

The tested organic solvents are known to extract lipids [9, 10, 23–26]. In addition,alterations in the SC other than pure lipid extraction are likely [10]. Large interindividualdifferences in CIM were found for each solvent or mixture [17] reflecting the variability inthe overall lipid extraction by these compounds [9]. The alterations induced in the humanSC by solvents (corneoxenometry bioassay) were indeed reported to be more variable inextent than those induced by diluted surfactants (corneosurfametry bioassay) in normalsubjects [14, 17, 18]. Despite interindividual inconsistencies in corneocyte alterations, sig-nificant differences were reported among solvents using the corneoxenometry bioassay[17]. Hexane–methanol and chloroform–methanol were the mixtures strongly altering theSC structure. Chloroform–methanol is well known as the most potent extraction mixturefor lipids in biological samples. However, it did not reach the top rank using the corneox-enometry bioassay [17]. Such a finding further illustrated the fact that organic solvents mayalter other biological components [8], which in turn affect the corneoxenometry data.

The corneoxenometry bioassay allows evaluation of the influence of the contact timebetween the SC and the solvents. In previous studies [17] the time range between 1 and120 min was chosen following available information about the kinetics of lipid extractionfrom human SC [9]. The corneoxenometry data were in line with the handful of previousexperiments using other methodological approaches [9, 10, 23, 26]. However, it does notexplore the effects of solvents on the living epidermis and on the nature and intensity ofinflammation that could result in irritant dermatitis.




Conclusion

Corneoxenometry appears as a relevant and predictive bioassay to assess the overalleffect of single and combined penetration enhancers. It is cheap, rapid, minimallyinvasive, and relevant to human skin. In addition, the reproducibility, specificity, andsensibility are quite high. Corneoxenometry is therefore a valuable screening test pro-posed as an alternative to animal testing.

References1. Hadgraft, J. and Walters, K.A. Skin penetration enhancement. J. Dermatol. Treat., 5, 43–47,

1994.2. Hadgraft, J. Penetration enhancers in percutaneous absorption. Pharm. Int., 5, 252, 1984.3. Woodford, R. and Barry, B.W. Penetration enhancers and the percutaneous absorption of

drugs: an update. J. Toxicol. Cut. Occular Toxicol., 5, 165, 1986.4. Williams, A.C. and Barry, B.W. Skin absorption enhancers. Crit. Rev. Ther. Drug Carrier Syst., 9,

305, 1992.5. Wotton, P.K. et al. Vehicle effect on topical drug delivery. Effect of azone on the cutaneous

penetration of metronidazole and propylene glycol. Int. J. Pharmacol., 24, 19–26, 1985.6. Ward, A.J.I. and Du Reau, C. The essential role of lipid bilayers in the determination of stratum

corneum permeability. Int. J. Pharm., 74, 137–146, 1991.7. Rojas, J. et al. Optimization of binary and ternary solvent systems in the percutaneous

absorption of morphine base. STP Pharmacol. Sci., 1, 70–75, 1991.8. Diembeck, W. et al. Test guidelines for in vitro assessment of dermal absorption and percu-

taneous penetration of cosmetic ingredients. Food Chem. Toxicol., 37, 191–205, 1999.9. Lavrijsen, A.P.M. et al. Validation of an in vivo extraction method for human stratum corneum

ceramides. Arch. Dermatol. Res., 286, 495–503, 1994.10. Abrams, K. et al. Effect of organic solvents on in vitro human skin water barrier function.

J. Invest. Dermatol., 101, 609–613, 1993.11. Goffin, V. et al. Sodium hypochlorite, bleaching agents and the stratum corneum. Ecotoxicol.

Environ. Safe, 37, 199–202, 1997.12. Pierard, G.E., Goffin, V., and Pierard-Franchimont, C. Squamometry and corneosurfametry

in rating interactions of cleansing products with stratum corneum. J. Soc. Cosmet. Chem., 45,269–277, 1994.

13. Goffin, V., Paye, M., and Pierard, G.E. Comparison of in vitro predictive tests for irritationinduced by anionic surfactants. Contact Dermatitis, 33, 38–41, 1995.

14. Goffin, V., Pierard-Franchimont, C., and Pierard, G.E. Sensitive skin and stratum corneumreactivity to household cleaning products. Contact Dermatitis, 34, 81–85, 1996.

15. Pierard, G.E. and Pierard-Franchimont, C. Drug and cosmetic evaluations with skin stripping.In: Maibach, H.I. (ed.). Dermatologic Research Techniques. CRC Press, Boca Raton, Florida,133–149, 1996.

16. Uhoda, E., Goffin, V., and Pierard, G.E. Responsive corneosurfametry following in vivo pre-conditioning. Contact Dermatitis, 49, 292–296, 2003.

17. Goffin, V., Letawe, C., and Pierard, G.E. Effect of organic solvents on normal human stratumcorneum. Evaluation by the corneoxenometry bioassay. Dermatology, 195, 321–324, 1997.

18. Goffin, V., Pierard-Franchimont, C., and Pierard, G.E. Shielded corneosurfametry and corneox-enometry: novel bioassays for the assessment of skin barrier products. Dermatology, 196,434–437, 1998.

19. Goffin, V. et al. Penetration enhancers assessed by corneoxenometry. Skin Pharmacol. Appl.Skin Physiol., 13, 280–284, 2000.


Penetration Enhancer Assessment by Corneoxenometry & 297


20. Pierard, G.E. et al. Surfactant induced dermatitis. A comparison of corneosurfametry withpredictive testing on human and reconstructed skin. J. Am. Acad. Dermatol., 33, 462–469,1995.

21. Peck, K.D., Ghanem, A.H., and Higuchi, W.I. Hindered diffusion of polar molecules throughand effective pore radii estimates of intact and ethanol treated human epidermal membrane.Pharm. Res., 11, 1306–1314, 1994.

22. Garcia, N. et al. Use of reconstructed human epidermis cultures to assess the disrupting effectof organic solvent on the barrier function of excised human skin. In Vitro Mol. Toxicol., 13,159–171, 2000.

23. Bligh, E.G. and Dyer, W.J. A rapid method of total lipid extraction and purification. Can.J. Biochem. Physiol., 37, 911–917, 1959.

24. Deffond, D. et al. In vivo measurements of epidermal lipids in man. Bioeng. Skin, 2, 71–85,1986.

25. Scheuplein, R. and Ross, L. Effect of surfactants and solvents on the permeability of epidermis.J. Soc. Cosmet. Chem., 21, 853–873, 1970.

26. Imokawa, G. et al. Selective recovery of deranged water-holding properties by stratumcorneum lipids. J. Invest. Dermatol., 87, 758–761, 1986.




Chapter 20

Assessment of Vehicle Effectsby Skin Stripping

Carryn H. Purdon, Carolina Pellanda, Christian Surber, and Eric W. Smith

CONTENTS

Introduction .................................................................................................................................... 299Investigation of the Delivery Vehicle and Skin Reservoir Effect .................................................. 300Investigation of Stratum Corneum Lipid Biochemistry................................................................. 300Bioequivalence Assessments.......................................................................................................... 302Lateral Drug Distribution................................................................................................................ 303Corneocyte Quantification ............................................................................................................. 303Follicular Penetration ..................................................................................................................... 304Limitations of Tape Stripping......................................................................................................... 304Conclusions .................................................................................................................................... 305References....................................................................................................................................... 306

Introduction

Tape stripping is a technique that has been found useful in dermatopharmaceuticalresearch for selectively and exhaustively removing the skin’s outermost layer, the stratumcorneum (SC). This technique uses an adhesive film to remove the corneocyte aggregatesof the SC in a stepwise manner. The sequential removal of cells from the SC by adhesivetape is one method by which the relationship between xenobiotic and cell abundance oneach removed strip may be assessed. Quantitative information about the drug concen-tration and the respective amount of corneocytes per tape may be used to describe thelocal distribution of the substance within the depth profile of the SC [1–4], and to describeinfluences that penetration enhancer chemicals may have on the drug penetrationkinetics. Furthermore, many in vivo methods for measuring dermal absorption ofchemicals are invasive (e.g., blood samples are collected) or slow (e.g., urine samples


299


are collected for extended periods). On the other hand, tape stripping of the SC is a fastand relatively noninvasive technique for measuring dermal absorption [5, 6].

Skin surface stripping with adhesive tape has been widely used to examine thelocalization and distribution of substances within the SC, and to provide informationabout the kinetics of transdermal drug delivery (dermatopharmacokinetic analyses)[7–12]. Several factors can influence the quantity of SC removed by a piece of tape;these include the skin properties of the volunteer [13–16], the applied formulation [6, 17],the physical technique of tape stripping, the hydration of the skin, cohesion betweencells (which increases with depth in the SC), body site and inter-individual differences[18, 19]. The technique has been used for a variety of purposes in dermatologicalresearch; the major uses are reviewed below:

Investigation of the Delivery Vehicle and Skin Reservoir Effect

The observation that the skin may serve as a reservoir for topically-administered chemicalswas originally reported by Malkinson and Ferguson in 1955 [20]. The localization of thisreservoir within the SC was later demonstrated for corticosteroids by Vickers in 1963 [21]and has been confirmed by others [22–25]. The use of the tape stripping method toinvestigate the reservoir and barrier function of the skin significantly expanded theexperimental tools available in this spectrum of skin research [12, 26]. Delivery vehiclemodifications that improve or retard percutaneous absorption may also result in analteration in the magnitude of the SC reservoir formed [21, 25]. Data from tape strippingexperiments may therefore be related to (1) chemical penetration into skin, (2) chemicalpermeation through skin, (3) chemical elimination from the skin, (4) pharmacodynamicparameters, and (5) clinical parameters. Skin stripping methodologies are therefore espe-cially useful in assessing the influence of the delivery vehicle composition, or the presenceof penetration enhancers, on the magnitude and location of the SC reservoir formed.

Dupuis, Lotte, Rougier, and co-workers [27–31] report that the stripping method is ableto determine the concentration of chemical in the SC at the end of a short application period(30 min). They found a linear relationship between the SC reservoir content and in vivopercutaneous absorption (total amount of drug permeated in 4 days) using the standardurinary excretion method [32–34]. They could also show, for a variety of simple pharma-ceutical vehicles, that percutaneous absorption of benzoic acid is vehicle dependent andcan be predicted from the amount of drug within the SC at 30 min after application. Theystated that the major advantages of their validated tape stripping protocol are the subse-quent elimination of urinary and fecal excretion assessments to determine absorption, andthe applicability to nonradiolabeled determination of percutaneous absorption becausethe skin strips contain adequate chemical concentrations for nonlabeled assay methodolo-gies. Despite the fact that the assay provides a reliable prediction of total absorption for agroup of selected compounds, comprehensive mechanistic interpretations of the data arestill rare. Auton [35] presented an initial mathematical approach based on the initial data ofRougier and co-workers, which may help to explain some of the above observations.

Investigation of Stratum Corneum Lipid Biochemistry

Intercellular lipids in the SC are responsible for the barrier function of mammalian skin.The main components of the SC lipids are ceramides, cholesterol, and free fatty acids; as




established by thin-layer chromatographic analysis of lipids extracted from human andmammalian SC. Initial research had analyzed the lipid fraction in extracts of the entire SC,which gave little information on the change in lipid biochemistry at different strataldepths. The optimization of the tape stripping technique held much promise for enablingthe fractionation of lipids at different depths. However, the use of the tape strippingtechnique for this purpose was hampered by the contamination of lipid extracts withcompounds that were co-extracted from the tape when organic solvents were used.Weerheim and Ponec [36] established a suitable analytical method for the determinationof the local SC lipid composition. SC samples were collected by sequential strippingwith Leukoplex1 tape in five healthy volunteers, the lipids were extracted with ethylacetate:methanol mixture (20:80) and separated by means of HPTLC. The resultsrevealed that the free fatty acid level is highest, and the cholesterol and ceramide levelslowest, in the uppermost SC layers. The levels remained unchanged in the underlyingSC layers, where the ceramide level was about 60% and the free fatty acid and choles-terol levels were about 20%, respectively. Ceramides could be separated into sevendifferent fractions and the relative amounts of individual ceramide fractions did notsignificantly change with the SC depth. Cholesterol sulfate levels were about 5% oftotal cholesterol and did not change with the SC depth, except for the first strip wherethe level was about 1%. This methodology makes it possible to study the differences inthe SC lipid profile in healthy and diseased human skin, with relation to the SC lipidorganization and to the skin barrier function in vivo. Therefore, stripping technologymay also be useful for assessing the effects that penetration enhancers or retarders haveon SC lipids.

Redoules et al. [37] described a method for the assay of five enzymatic activitiesinvolved in establishing the SC permeability barrier: b-glucocerebrosidase, acid phos-phatase, phospholipase A2 (PLA2), and two serine proteases: chymotrypsin and itsactivator in the SC, trypsin. The method was applied to the pathological situation of anoneczematous, dry atopic dermatitis. Several reasons motivated the authors to quantifytheir in vitro activities using the tape stripping technique: (1) to have a more accuratepicture of the contribution of each of the groups of enzymes in normal permeabilitybarrier function; (2) to develop a tool for acquiring precise information on the causes ofpathologies; (3) to determine the conditions in which the various activities can be used torelease compounds suspected to have a beneficial effect on the epidermis (for instancean antioxidant activity), or even a therapeutic effect. The stripping technique enabled theaccurate assay of five distinct enzyme activities. Pooling three tapes when conducting theenzymatic analysis [38] made it possible to assay the activities of b-glucocerebrosidase,acid phosphatase, PLA2, trypsin and chymotrypsin with an error of below 5%. The firsttwo enzymatic activities, b-glucocerebrosidase and acid phosphatase, which are involvedin the terminal differentiation of the keratinocyte, presented the highest values of the fiveactivities studied. The activities seem to be stable and to resist proteolysis since theyremain roughly the same during the 10-day migration from the deep layers of the SCup to the surface of the skin.

The SC differs in the composition of the lipoidal phase. A simple methodology that isable to correlate the differences in the SC composition with the drug amounts detectablewithin the SC is desirable. Wagner et al. [39] carried out penetration experiments inves-tigating several incubation times with three different skin flap models and the lipophilicdrug flufenamic acid. The drug amounts within the SC were obtained with the tapestripping technique, while the drug amounts present in the deeper skin layers wereachieved by cryosectioning. The SC/water-partition coefficient was determined with the


Assessment of Vehicle Effects by Skin Stripping & 301


same three skin flaps to characterize the lipoidal SC phase in general, and the differenceswere attributed to the different amounts of ceramides and sterols. In addition, a directlinear correlation was found between the SC/water-partition coefficients and flufenamicacid amounts that penetrated into the SC for all investigated time intervals. The describedmethodology represents a tool to predict drug amounts in the intact SC based on theknowledge of the SC/water-partition coefficient for the drug of interest, a parameter thatmay be influenced greatly by the co-administration of penetration enhancer chemicalsfrom different chemical classes.

Potard et al. [40] compared the in vitro compartmental distribution and absorption offive UV filters after exposure times of 30 min and 16 h, using a tape stripping techniqueon human skin. The washing procedure and the stripping technique were emphasized asthese aspects are fundamental when the aim is to compare different experiments in termsof the distribution of different chemical products in the SC. The UV filters (octyl methox-ycinnamate, benzophenone 4, benzophenone 3, octyl triazone and octocrylene) wereincorporated separately in a simple oil-in-water emulsion. The affinity for each skin level[SC, viable epidermis, dermis, and receptor fluid] was found to be different according tothe test substance used. Some substances accumulated in the SC, whereas others passedthrough the skin very quickly and were quantified in the receptor fluid. The strippingtechnique demonstrated that more than 94% of the chemical compound in the SC couldbe found in the first eight tapes. The problem of individual values below the limit ofdetection was raised, a correlation between the two exposure times was found and aclassification of products according to their affinity for the SC was determined. Thisresearch exemplifies the vital role that tape stripping can play in assessing vehicledelivery effects.

Bioequivalence Assessments

Drug uptake is usually assessed by applying test and reference products simultaneouslyto multiple skin sites in each study subject. SC samples are obtained at sequentiallyincreasing time intervals by a tape tripping technique. In a similar manner, to assessdrug elimination, test and reference products are applied for a specific period of time atmultiple sites and then cleaned from the sites. The SC samples are collected at sequen-tially increasing times after drug formulation removal by a tape tripping technique.Additionally, drug elimination studies after the drug concentration has reached a plateauin the SC have been proposed [41, 42].

Considerable literature exists comparing the potency and bioavailability of topicallyapplied corticosteroids in vivo using clinical efficacy and vasoconstriction monitoring,also known as skin blanching [43]. These studies have proven useful for ranking cortico-steroid potency and for distinguishing between commercial formulations. However, thevisual quantification of skin blanching is subjective. An analytical method, which object-ively quantifies bioavailability as a function of the amount of drug within the treated SCsite, would provide greater insight into subtle differences among tested formulations.Pershing et al. [44] developed an in vivo technique which simultaneously compared acorticosteroid skin blanching bioassay with drug content in human SC following topicalapplication of four 0.05% betamethasone dipropionate formulations. Bioavailability ofdrug from commercial cream and ointment formulations was assessed by quantificationof drug content in tape stripped SC and skin blanching at the treated skin site.




A correlation between the amount of drug in the treated SC and the corresponding skinblanching score was observed for the four formulations.

Lateral Drug Distribution

Relatively few in vivo studies have demonstrated that the perpendicular drug penetrationof the applied components occurs in conjunction with a competitive lateral spreadingwithin the SC [2, 45–47]. In general, the distribution of the active substance within the SCis expected to be influenced by the formulation used for application [6, 17, 45, 48, 49] andthe physicochemical properties of these substances, and certainly, therefore, by thepresence of penetration enhancer chemicals. The penetration of 4-methylbenzylidenecamphor and 1.5% butyl methoxydibenzoylmethane from an oil-in-water emulsion intothe human SC and the lateral spreading were investigated in vivo by Jacobi et al. [50].Tape stripping in combination with spectroscopic measurements were used [1, 51, 52].The concentration of both UV filters was determined inside and outside the applicationarea by varying the application and tape stripping protocol. A spreading of the topicallyapplied substances from the treated to the untreated areas was observed, whichdepended on the time between application and tape stripping and the size of the treatedskin area. Significant amounts of topically applied substances were found adjacent to theapplication area, which may be due to the lateral spreading that takes place on the skinsurface. In general, the lateral spreading must be considered to be a competitive processwhen studying penetration processes of topically applied substances. It has to be con-sidered during drug treatment of small limited skin areas and for the interpretation ofrecovery rates obtained in penetration studies.

Corneocyte Quantification

In general, the application of the tape stripping procedure in the investigation of topicalpenetration requires the determination of the exact amount of corneocytes fixed to eachtape strip, as a prerequisite for calculation of the SC profile [1, 53]. The amount of SCremoved on each tape strip is influenced by various factors [7, 13, 54, 55], including thecomposition of the vehicle used for topical application [53]. There is a real possibility,therefore, that the inclusion of a penetration enhancer chemical into a topical formulationmay influence the adhesiveness of the corneocytes to the tape strips — this aspect of theexperimental protocol needs to be validated. A direct spectroscopic method for thedetermination of the amount of SC on the tape strips has recently been described [1].This method is based on the determination of the pseudo-absorption of the corneocytesin the visible spectral range, caused by scattering, reflection, and diffraction properties ofthe corneocyte aggregates. Interference of penetration enhancer chemicals in the lightabsorption wavelengths of the drug of interest or the corneocyte measurements may beproblematic in this technique. Essentially the penetration enhancer simply presents as anadditional light absorbing species in the formulation. However, this aspect does notdetract from the usefulness of the visible absorption method, which can be used tostudy both untreated and treated skin with high sensitivity. The major benefit of thistechnique is that the absorbance of the corneocytes is measured at 430 nm, a spectralrange well displaced from the UV absorption bands of typical drugs and penetration




enhancer molecules. The absorption of the adhesive tapes, the emulsions and surfactantsapplied do not usually interfere with the spectra because they have no absorption bandsin the visible spectral range [1].

Follicular Penetration

Follicles and sweat glands account only for approximately 1% of the skin surface area.Therefore, they are not considered to represent significant drug penetration routes.Within the structure of the follicular apparatus, the upper part, the acroinfundibulum, iscovered by a normally structured SC which can be considered as a barrier. However, inthe lower part the infrainfundibulum, the wall has relatively few differentiated corneo-cytes and has to be considered as highly permeable. It is this site that may be amenable topenetration enhancer chemical influence. Lademann et al. [3] investigated the penetrationof coated titanium dioxide microparticles into the SC of human skin by tape stripping incombination with spectroscopic measurements. Small amounts of microparticles werefound in deeper parts of the SC after long-term application of a sunscreen containingtitanium dioxide. These small amounts were clearly located in the follicle orifices whilethe surrounding corneocyte aggregates were free of TiO2. The analysis of biopsy sectionscontaining hair follicle channels shows that small amounts of TiO2 microparticles pene-trated into the acroinfundibulum of follicles without reaching the layer of viable cells.Interestingly, the microparticles were only found in 10% of hair follicles at the treated site.

The follicular penetration process appears to depend on the phase of the hair growthcycle [56]. Absorption of exogenous chemicals appears to take place when hair growth andsebum production are active, and minimal absorption occurs when no hair growth andno sebum production can be measured. These observations further limit the potential forsubstantial drug delivery via the follicular apparatus. In addition, macroscopic furrows inthe SC may cause problems in the interpretation of tape strip results. Van der Molen et al.[57] investigated the efficacy of tape stripping in removing complete cell layers from thehuman SC. A histological section of skin that was tape stripped 20 times clearly showednonstripped skin in the furrows, indicating incomplete corneocytes removal. Replicas oftape stripped skin surface demonstrated that even after removing 40 tape strips, thefurrows were still present. Residual material of the compound under investigation couldtherefore accumulate in furrows, disturbing the interpretation of the tape stripping results.Penetration enhancers may be capable of altering either or both of the follicular and furrowabsorption profiles.

Limitations of Tape Stripping

In a typical tape stripping experiment, an area of skin is exposed to a chemical for a setexposure time and then cleaned. Between 10 and 100 pieces of adhesive tape are appliedto and removed from the dosed area in sequence, and the mass of chemical determinedin each tape [58, 59]. Although the tape stripping procedure is relatively simple toexecute, there are many opportunities for experimental artifacts to develop; for example,tape stripped samples have high surface area-to-volume ratios, and losses by evaporationcan be significant even for chemicals with relatively low volatility. Generally, the tapestripping experiment is unsuitable for volatile chemicals and chemical analysis should becompleted soon after tape stripping removal from the skin [60].




In bioequivalence evaluations, the vehicle components of a test and a referenceproduct may be different and may markedly influence both the adhesive properties ofthe tape and the cohesion of the corneocytes. Hence dermatopharmacokinetic charac-terization may become extremely complex and susceptible to error. The removal ofcorneocytes may be highly dissimilar and dermatopharmacokinetic characterizationbased on a concentration profile within the SC may not be possible. This was demon-strated by Van der Molen et al. [57] who showed that normal tape stripping of human SCyields cell layers that originate from various depths because of furrows in the skin andadhesiveness between the corneocytes. We have confirmed this aspect of the experi-mental protocol using electron microscopy studies. Figure 20.1 clearly shows the‘‘plaques’’ of SC that adhere to the adhesive matrix of the tape and are removed in atypical stripping procedure. There is no uniform sheet of tissue removed across the entirefield of tape application. The presence of furrows or folding of the SC after strippingfurther confounds this analytical process. The inclusion of penetration enhancers maytherefore complicate the dermatopharmacokinetic comparison of test and referencestripping data in the assessment of bioequivalence.

Conclusions

The skin stripping method has a potential for being a specific dermatopharmacokineticmethod that assesses drug concentration in SC as a function of time. Both drug uptakeand drug elimination profiles may be evaluated to determine traditional pharmacokineticmetrics, such as AUC, Cmax, and Tmax. Furthermore, the real advantage is that the subtle

Figure 20.1 Electron micrograph of adhesive side of tape used for skin stripping. The nonuni-formity of adhering stratum corneum sections following in vivo removal is clearly visible.




influences of a penetration enhancer in a topical formulation may be evaluated by thistechnique, in terms of mass of the active in the stratum corneum and the magnitude of theactive reservoir formed after application. To date, skin stripping still appears to be one ofthe most promising techniques for the rapid assessment of the effects of penetrationenhancers in vivo, however full validation of the numerous facets of the experimentalprotocol are obligatory for valid data to be generated.

References1. Weigmann, H., Lademann, J., Meffert, H., Schaefer, H., and Sterry, W. (1999) Determination of

the horny layer profile by tape stripping in combination with optical spectroscopy in thevisible range as a prerequisite to quantify percutaneous absorption. Skin Pharmacol Appl SkinPhysiol, 12, 34–45.

2. Weigmann, H.J., Lademann, J., Schanzer, S., Lindemann, U., von Pelchrzim, R., Schaefer, H.,Sterry, W., and Shah, V. (2001) Correlation of the local distribution of topically appliedsubstances inside the stratum corneum determined by tape-stripping to differences in bio-availability. Skin Pharmacol Appl Skin Physiol, 14 Suppl 1, 98–102.

3. Lademann, J., Weigmann, H., Rickmeyer, C., Barthelmes, H., Schaefer, H., Mueller, G., andSterry, W. (1999) Penetration of titanium dioxide microparticles in a sunscreen formulationinto the horny layer and the follicular orifice. Skin Pharmacol Appl Skin Physiol, 12, 247–256.

4. Tassopoulos, T., Maeder, S., Imanidis, G., Figueiredo, V., Smith, E.W., and Surber, C. (2001)Evaluation of a spectrophotometric, in situ method as a stand-alone method for percutaneousbioavailability studies. Stratum Corneum III Congress, Basel, Switzerland.

5. Stinchcomb, A.L., Pirot, F., Touraille, G.D., Bunge, A.L., and Guy, R.H. (1999) Chemical uptakeinto human stratum corneum in vivo from volatile and non-volatile solvents. Pharm Res, 16,1288–1293.

6. Schwarb, F.P., Gabard, B., Rufli, T., and Surber, C. (1999) Percutaneous absorption of salicylicacid in man after topical administration of three different formulations. Dermatology, 198,44–51.

7. Bommannan, D.B., Potts, R.O., and Guy, R.H. (1990) Examination of stratum corneum barrierfunction in vivo by infrared spectroscopy. J Invest Dermatol, 95, 403–408.

8. Higo, N., Naik, A., Bommannan, D.B., Potts, R.O., and Guy, R.H. (1993) Validation of reflect-ance infrared spectroscopy as a quantitative method to measure percutaneous absorptionin vivo. Pharm Res, 10, 1500–1506.

9. Lotte, C., Wester, R.C., Rougier, A., and Maibach, H.I. (1993) Racial differences in the in vivopercutaneous absorption of some organic compounds — a comparison between black,caucasian and asian subjects. Arch Derm Res, 284, 456–459.

10. Pershing, L.K., Silver, B.S., Krueger, G.G., Shah, V.P., and Skelley, J.P. (1992) Feasibility ofmeasuring the bioavailability of topical betamethasone diproprionate in commercial formula-tions using drug content in skin and a skin blanching bioassay. Pharm Res, 9, 45–51.

11. Rougier, A., Dupuis, D., Lotte, C., Roguet, R., Wester, R.C., and Maibach, H.I. (1986) Regionalvariation in percutaneous absorption in man: measurement by the stripping method. ArchDermatol Res, 278, 465–469.

12. Tojo, K. and Lee, C. (1989) A method for predicting steady-state rate of skin penetration in vivo.Soc Invest Dermatol, 92, 105–108.

13. Anderson, R.L. and Cassidy, B.A. (1973) Variations in physical dimensions and chemicalcomposition of human stratum corneum. J Invest Dermatol, 61, 30–32.

14. Leveque, J.L., Corcuff, P., De Rigal, J., and Agache, P. (1984) In vivo studies of the evolution ofphysical properties of the human skin with age. Int J Dermatol, 23, 322–329.

15. Feldmann, R.J. and Maibach, H.I. (1967) Regional variation in percutaneous penetration of 14Ccortisol in man. J Invest Dermatol, 48, 181–183.




16. Maibach, H.I., Feldman, R.J., Milby, T.H., and Serat, W.F. (1971) Regional variation in percu-taneous penetration in man. Pesticides. Arch Environ Health, 23, 208–211.

17. Treffel, P. and Gabard, B. (1996) Skin penetration and sun protection factor of ultra-violetfilters from two vehicles. Pharm Res, 13, 770–774.

18. King, C.S., Barton, S.P., Nicholls, S., and Marks, R. (1979) The change in properties of thestratum corneum as a function of depth. Br J Dermatol, 100, 165–172.

19. Marttin, E., Neelissen-Surbnel, M.T.A., De Haan, F.H.N., and Bodde, H.E. (1996) A criticalcomparison of methods to quantify stratum corneum removed by tape stripping. Skin Phar-macol, 9, 69–77.

20. Malkinson, F.D. and Ferguson, E.H. (1955) Percutaneous absorption of hydrocortisone-4-C14in two human subjects. J Invest Dermatol, 25, 281–283.

21. Vickers, C.F. (1963) Existence of reservoir in the stratum corneum. Experimental proof. ArchDermatol, 88, 20–23.

22. Stoughton, R.B. (1965) Dimethylsulfoxide (DMSO) induction of a steroid reservoir in humanskin. Arch Dermatol, 91, 657–660.

23. Carr, R.D. and Wieland, R.G. (1966) Corticosteroid reservoir in the stratum corneum. ArchDermatol, 94, 81–84.

24. Carr, R.D. and Tarnowski, W.M. (1966) The corticosteroid reservoir. Arch Dermatol, 94,639–642.

25. Munro, D.D. (1969) The relationship between percutaneous absorption and stratum corneumretention. Br J Dermatol, 81, 92–97.

26. Tsai, J.C., Cappel, M.J., Flynn, G.L., Weiner, N.D., Kreuter, J., and Ferry, J. (1992) Drug andvehicle deposition from topical applications: use of in vitro mass balance technique withminoxidil solutions. J Pharm Sci, 81, 736–743.

27. Rougier, A., Rallis, M., Krien, P., and Lotte, C. (1990) In vivo percutaneous absorption: a keyrole for stratum corneum/vehicle partitioning. Arch Dermatol Res, 282, 498–505.

28. Rougier, A., Lotte, C., and Maibach, H.I. (1987) The hairless rat: a relevant animal model topredict in vivo percutaneous absorption in humans? J Invest Dermatol, 88, 577–581.

29. Rougier, A., Lotte, C., and Dupuis, D. (1987) An original predictive method for in vivopercutaneous absorption studies. J Soc Cosmet Chem, 38, 397–417.

30. Dupuis, D., Rougier, A., Roguet, R., and Lotte, C. (1986) The measurement of the stratumcorneum reservoir: a simple method to predict the influence of vehicles on in vivo percutan-eous absorption. Br J Dermatol, 115, 233–238.

31. Dupuis, D., Rougier, A., Roguet, R., Lotte, C., and Kalopissis, G. (1984) In vivo relationshipbetween horny layer reservior effect and percutaneous absorption in human and rat. J InvestDermatol, 82, 353–356.

32. Feldmann, R.J. and Maibach, H.I. (1974) Percutaneous penetration of some pesticides andherbicides in man. Toxicol Appl Pharmacol, 28, 126–132.

33. Feldmann, R.J. and Maibach, H.I. (1970) Absorption of some organic compounds through theskin in man. J Invest Dermatol, 54, 399–404.

34. Feldmann, R.J. and Maibach, H.I. (1969) Percutaneous penetration of steroids in man. J InvestDermatol, 52, 89–94.

35. Auton, T.R. (1990) Skin stripping and science: a mechanistic interpretation using mathematicalmodelling of skin deposition as a predictor of total absorption. In: Scott, R.C., Guy, R.H.,Hadgraft, J., and Bodde, H.E. (eds), Prediction of Percutaneous Penetration. IBC TechnicalServices Ltd, London, pp. 558–576.

36. Weerheim, A. and Ponec, M. (2001) Determination of stratum corneum lipid profile by tapestripping in combination with high-performance thin-layer chromatography. Arch DermatolRes, 293, 191–199.

37. Redoules, D., Tarroux, R., Assalit, M.F., and Perie, J.J. (1999) Characteristics and assay of fiveenzymatic activities in the stratum corneum using tape-strippings. Skin Pharmacol Appl SkinPhysiol, 12, 182–192.




38. Odland, G.F. (1991) Structure of the skin. In: Goldsmith, L.A. (ed.), Physiology, Biochemistry,and Molecular Biology of the Skin. Oxford University Press, New York, Oxford, pp. 3–61.

39. Wagner, H., Kostka, K.H., Lehr, C.M., and Schaefer, U.F. (2002) Correlation between stratumcorneum/water-partition coefficient and amounts of flufenamic acid penetrated into thestratum corneum. J Pharm Sci, 91, 1915–1921.

40. Potard, G., Laugel, C., Schaefer, H., and Marty, J.P. (2000) The stripping technique: in vitroabsorption and penetration of five UV filters on excised fresh human skin. Skin PharmacolAppl Skin Physiol, 13, 336–344.

41. Lucker, P.W., Beubler, E., Kukovetz, W.R., and Ritter, W. (1984) Retention time and concen-tration in human skin bifonazole and clotrimazole. Dermatologica, 169, 51–56.

42. Shah, V.P., Flynn, G.L., Yacobi, A., Maibach, H.I., Bon, C., Fleischer, N.M., Franz, T.J., Kaplan,S.A., Kawamoto, J., Lesko, L.J., Marty, J.P., Pershing, L.K., Schaefer, H., Sequeira, J.A., Shrivas-tava, S.P., Wilkin, J., and Williams, R.L. (1998) Bioequivalence of topical dermatological dosageforms — methods of evaluation of bioequivalence. Pharm Res, 15, 167–171.

43. Smith, E.W., Haigh, J.M., and Surber, C. (2002) Quantification of corticosteroid-induced skinvasoconstriction: visual ranking, chromameter measurement or digital imaging analysis.Dermatology, 205, 3–10.

44. Pershing, L.K., Silver, B.S., Krueger, G.G., Shah, V.P., and Skelley, J.P. (1992) Feasibility ofmeasuring the bioavailability of topical betamethasone diproprionate in commercial formula-tions using drug content in skin and a skin blanching bioassay. Pharm Res, 9, 45–51.

45. Ashworth, J., Watson, W.S., and Finlay, A.Y. (1988) The lateral spread of clobetasol-17-propi-onate in the stratum corneum in vivo. Br J Dermatol, 119, 351–358.

46. Weigmann, H., Lademann, J., von Pelchrzim, R., Sterry, W., Hagemeister, T., Molzahn, R.,Schaefer, M., Lindscheid, M., Schaefer, H., and Shah, V.P. (1999) Bioavailability of clobetasolpropionate-quantification of drug concentrations in the stratum corneum by dermatopharma-cokinetics using tape stripping. Skin Pharmacol Appl Skin Physiol, 12, 46–53.

47. Chambin-Remoussenard, O., Treffel, P., Bechtel, Y., and Agache, P. (1993) Surface recoveryand stripping methods to quantity percutaneous absorption of caffeine in humans. J PharmSci, 82, 1099–1101.

48. Dick, I.P., Blain, P.G., and Williams, M. (1997) The percutaneous absorption and skin distri-bution of lindane in man. II. In vitro studies. Human Exp Toxicol, 16, 652–657.

49. Trevithick, J.R. and Mitton, K.P. (1993) Topical application and uptake of vitamin E acetate bythe skin and conversion to free vitamin E. Biochem Mol Biol Int, 31, 869–878.

50. Jacobi, U., Weigmann, H.-J., Baumann, M., Reiche, A.-I., Sterry, W., and Lademann, J. (2004)Lateral spreading of topically applied UV filter substances investigated by tape stripping. SkinPharmacol Appl Skin Physiol, 17, 17–22.

51. Weigmann, H.-J., Lindemann, U., Antoniou, C., Tsikrikas, G.N., Stratigos, A.I., Katsambas, A.,Sterry, W., and Lademann, J. (2003) UV/VIS absorbance allows rapid, accurate, and reprodu-cible mass determination of corneocytes removed by tape stripping. Skin Pharmacol Appl SkinPhysiol, 16, 217–227.

52. Lindemann, U., Weigmann, H.-J., Schaefer, H., Sterry, W., and Lademann, J. (2003) Evaluationof the pseudo-absorption method to quantify human stratum corneum removed by tapestripping using the protein absorption. Skin Pharmacol Appl Skin Physiol, 16, 228–236.

53. Surber, C., Schwarb, F.P., and Smith, E.W. (1999) Tape-stripping technique. In: Bronaugh, R.L.and Maibach, H.I. (eds), Percutaneous Absorption. Drugs–Cosmetics — Mechanisms–Method-ology. Basel, New York, pp. 395–409.

54. Bashir, S.J., Chew, A.-L., Anigbogu, A., Dreher, F., and Maibach, H.I. (2001) Physical andphysiological effects of stratum corneum tape stripping. Skin Res Technol, 7, 40–48.

55. Holbrook, K.A. and Odland, G.F. (1974) Regional differences in the thickness (cell layers) ofthe human stratum corneum: an ultrastructural analysis. J Invest Dermatol, 62, 415–422.

56. Schaefer, H. and Lademann, J. (2001) The role of the follicular penetration. Skin PharmacolAppl Skin Physiol, 14 (suppl 1), 23–27.




57. Van der Molen, R.G., Spies, F., van ‘t Noordende, J.M., Boelsma, E., Mommaas, A.M., andKoerten, H.K. (1997) Tape stripping of human stratum corneum yields cell layers that originatefrom various depths because of furrows in the skin. Arch Dermatol Res, 289, 514–518.

58. Ohman, H. and Vahlquist, A. (1994) In vivo studies concerning a pH gradient in human stratumcorneum and upper epidermis. Acta Dermato-Venereologica (Stockh), 74, 375–379.

59. Sheth, N.V., McKeough, M.B., and Spruance, S.L. (1987) Measurement of stratum corneumdrug reservoir to predict the therapeutic efficacy of topical iododeoxyuridine for herpessimplex. J Invest Dermatol, 89, 598–602.

60. Islam, M.S., Zhao, L., McDougal, J.N., and Flynn, G.L. (1999) Uptake of chloroform by skin onbrief eposures to the neat liquid. Am Ind Hygiene Assoc J, 60, 5–15.




Chapter 21

The Use of Skin Alternativesfor Testing PercutaneousPenetration

Charles Scott Asbill, Gary W. Bumgarner, and Bozena B. Michniak

CONTENTS

Introduction .................................................................................................................................... 311Types of Cultured Skin Models...................................................................................................... 313

Epidermal Models....................................................................................................................... 313Epiderme

ˆ................................................................................................................................ 313

SkinEthic ..................................................................................................................................... 314Full-Thickness Models ................................................................................................................ 315

EpidermFTeˆ

........................................................................................................................... 315Apligraf.................................................................................................................................... 315Orcel1................................................................................................................................. 315Episkine

ˆ................................................................................................................................. 315

Bioengineered-Human Skin Equivalent................................................................................. 316Conclusions .................................................................................................................................... 316References....................................................................................................................................... 317

Introduction

The skin’s potential as an alternative drug delivery route has only been realized duringthe past two decades [1]. Transdermal drug delivery provides advantages over otherroutes of administration by avoiding the first pass effect in hepatic and intestinal tissueand by maintaining steady-state plasma levels [2, 3]. The invasiveness of intravenoustherapies and the diligence required for adhering to multiple oral dosing regimens arewhy many prefer the use of transdermal drug delivery systems [4]. More importantly,there is increased patient compliance when using these devices. Unfortunately, not all


311


drugs can be delivered in transdermal systems due to the innate barrier function of theskin. The stratum corneum or outermost layer of skin, which is composed of nonlivingterminally differentiated stratified keratinocytes, is primarily responsible for this barrier.However, this barrier can sometimes be compromised by the use of permeation en-hancers. Permeation enhancers are compounds that circumvent the permeability barrierby temporarily altering the structure of the stratum corneum [5, 6].

Transdermal systems, permeation enhancers, and topical products are often verydifficult to test in vitro because there is no reliable skin model available that matchesthe barrier function and physiology of human skin. The development of an adequate skinmodel to use for in vitro testing is also challenging. Various types of skin models,particularly nonhuman animal skins, have been utilized for many years. Examplesinclude hairless mouse skin, pig skin, human cadaver skin, rodent skin, and culturedskin alternatives [7]. The problems often associated with the above models are the lackof similarities to human skin in terms of permeability, cell type, lipid composition andorganization, and other physiological features.

Cultured skin alternatives have been under development for the past decade and havebeen validated for a number of in vitro and clinical applications. Skin alternatives havebeen used extensively as permanent skin replacement for burn victims. Victims of severeburns often have intensive scarring, infections, and disfigurement. The most seriouscomplication with burns is the onset of infection from breakdown of the barrier functionof the skin. Cultured skin alternatives can act as a closure, covering the wound site andrestricting the entry of bacteria. Among the more successful approaches for restoringbarrier function in burn victims is the use of culture techniques to expand epidermal cellpopulations for an autologous transplant using the patient’s own healthy keratinocytes [8].

Skin models have also been used to evaluate the permeation, phototoxicity, anddermal irritancy of new investigational drugs and cosmetics. Skin alternatives wereoriginally developed to study the pathophysiology of various skin diseases in vitro [9].Cultured skin models have allowed scientists to examine the differentiation of epidermallayers and determine how substances such as retinoids and hormones affect the epider-mal architecture [10, 11].

Skin alternatives are typically categorized as either epidermal models or full-thicknessmodels. There are structural differences between these two models, as well as differencesin the compositional cell types. Epidermal equivalents consist only of an epidermal layerand full-thickness equivalents have both epidermal and dermal layers. Epidermal equiva-lents were originally used for skin grafting and consist of normal keratinoyctes that arecultured on the surface of a suitable membrane such as polycarbonate [12]. The kerati-nocytes are allowed to proliferate and cover the surface of the membrane. After prolif-eration, the next phase is the induction of differentiation of the keratinocytes, which givesrise to a multilayered stratum corneum [13]. Full-thickness skin equivalents are grownby two primary methods. First, de-epidermized dermis may be used in which the livingdermis has been removed from native intact skin, and secondly, the use of a dermalequivalents consisting of fibroblast cells embedded in a collagen gel. In the latter, as thecultured dermal layer develops, the fibroblasts and collagen fibrils begin to interactand the dermal layer contracts uniformly. The dermal layer will contract into a gel-likestructure that possesses characteristics of native human dermal tissue. Next keratinocytesare seeded on top and allowed to grow at the air–liquid interface in a culture dish. Thesekeratinocytes differentiate and stratify in 5 to 10 days into an epidermis, which closelyresembles the in vivo epidermal layer [14]. The final result is a cell culture derivedepidermis and dermis that approaches innate full-thickness skin.




In recent years, both types of models have been utilized for preclinical studiesincluding both pharmacological studies and as a model to study permeation kinetics oftransdermally and topically applied drugs. In addition, these skin alternatives have beenextremely useful for testing the potential of irritation of topically applied compounds andpharmaceutical additives. While skin alternatives have been extensively explored as asuitable model for permeability testing, presently there is no cultured skin alternative thatapproaches the barrier function of the human skin [13, 15].

Problems associated with the use of skin alternatives are well documented and havelimited the practicality of using these models for permeability studies. Some of thedisadvantages associated with skin alternatives compared to using human and animalskin are significant, inter- and intra-batch variation, reduced barrier function, increasedcost, and the time-consuming requirements for cell culture [16]. The purpose of thischapter is to provide background on the types of skin models that have been developed,examine permeability data derived from experiments utilizing skin alternatives, discussbiological markers, and lipid characteristics of cultured skin models, and to evaluate skinalternatives as a screening tool for skin irritation.

Types of Cultured Skin Models

Epidermal Models

Epidermeˆ

Epidermeˆ

is a cultured epidermal model that was developed by MatTek Corporation(Ashland, MA). This cultured skin model uses normal (nontransformed) donated humancells as cell type and basis for the epidermal layer. Epiderm has been used to study bothdrug permeation and for irritancy testing [17]. There are two types of Epiderm models, amodel used for irritation studies and one used for permeation studies. MatTek’s processfor the culture of Epiderm involves the seeding of normal human keratinocytes on cellculture inserts. After a preselected time these keratinocyte cultures are raised to the air–liquid interface at which time the human keratinocytes begin to differentiate and stratifyforming an intact stratum corneum. Being an epidermal model, Epiderm lacks dermaltissue and the fibroblast that are native to dermal tissue. Full epidermal stratas are presentin Epiderm and the number of viable cell layers ranges from 6 to 8, and in the irritationmodel from 7 to 14. Markers of differentiation (Keratin 1 and 10) are found in bothEpiderm models. The lipid profile contains all of the major lipids found in humanepidermal tissues [18].

The permeability of Epiderm has been characterized by using different model drugs.A 2001 study by Zghoul et al. [19] compared the permeation of Epiderm to humanepidermis using flufenamic acid. In this study, it was shown that Epiderm has a fivetimes higher flux than native human epidermis when using flufenamic acid. Also, in thisstudy, the intra batch permeation of flufenamic acid was found not to be statisticallysignificant. A 2000 study by Asbill et al. [15] investigated the permeation of three modeldrugs caffeine, hydrocortisone, and tamoxifen in several skin alternatives. The results ofthis study suggest that Epiderm had a significantly different permeability than that ofhuman cadaver skin. For example, the flux of hydrocortisone in human cadaver skin wasfound to be 1.8+ 0.2 mg/cm2/h versus a flux of 4.8+ 0.8 mg/cm2/h in Epiderm. Thefluxes of caffeine and tamoxifen were also statistically different between the humancadaver skin and Epiderm. Another study in 2002 looked at permeation of lauric acid,


Skin Alternatives for Testing Percutaneous Penetration & 313


caffeine, and mannitol in several different skin models [16]. The permeation was found tobe lower for all skin models when compared to ex vivo human skin. It is interesting tonote that the rank order of the compounds tested was the same in Epiderm as in humanskin.

Several studies have been documented that explore the use of Epiderm as a model fortesting skin irritancy. A 1994 study by Cannon and Neal [20] compared the data obtainedusing 16 surfactant containing formulations tested in Epiderm and the in vivo datacollected from human skin. The results of this study suggest that the skin irritancy resultsobtained using Epiderm where similar to those found in human skin. A 2002 study byFaller determined the reproducibility of data obtained from Epiderm, Episkine

ˆ, and

SkinEthic. This study employed the use of a 3(4,5-dimethylthiazol-2yl)-2,5-dephenyltetrazolium bromide (MTT) assay to measure cell viability and proliferation. Also, proin-flammatory mediators and cytosolic enzymes where measured after exposure of skintissues to sodium lauryl sulfate. The MTT viability assay results suggested that Epidermwas the most resistant to sodium lauryl sulfate treatment and that Epiderm was the mostreproducible model. The release of the proinflammatory cytokine interleukin-1-a washighly variable in all skin models tested [21].

SkinEthic

SkinEthic (SkinEthic Laboratories, Nice, France), like Epiderm, is also a single layeredepidermal model. The predominant epidermal layers found in native human skin are alsofound in SkinEthic. Also, differentiation markers such as Keratin 1 and 10 and the majorlipid classes found in human skin are found in the SkinEthic model. A 1999 study byGysler et al. [22] addressed the penetration and metabolism of topical glucocorticoids inSkinEthic. Prednicarbate (prednisolone 17-ethylcarbonate, 21-proporionate) and beta-methasone-21-valerate were utilized as agents to study the in situ metabolism of drugs inboth SkinEthic and excised human skin. In this study it was shown that esterase activityin SkinEthic correlated to that of human skin. Also, drug amounts were only 1.7-foldhigher in the SkinEthic as compared to excised human skin. A 2001 study by Schmooket al. [23] compared the permeability of human, pig, and rat skins with two skinalternatives, living skin equivalent and SkinEthic. Four drugs of varying polarity werechosen for this study (salicylic acid, hydrocortisone, clotrimazole, and terbinafine). Theresults of this study suggested that the permeability of these model drugs in SkinEthicwas significantly higher than in human skin. Specifically, the flux of salicylic acid wassevenfold higher in SkinEthic as compared to human skin. Also, the permeation of morehydrophobic compounds, clotrimazole and terbinafine, resulted in flux values that weretremendously higher than those obtained using human skin (up to a factor of 800-fold).

A 2003 study by Coquette et al. [24] examined the use of SkinEthic as an in vitro skinmodel to discriminate between skin sensitizers and skin irritants. The authors usedinterleukin-1 and interleukin-8 as markers of irritation and sensitization. Skin sensitizersinduce very low expression of Interleukin-1 while exhibiting significant expression ofInterleukin-8. In contrast, skin irritants promote significant expression of interleukin-1. Inthis study, the authors were able to distinguish between established skin irritants (ben-zalkonium chloride, benzoic acid, sodium lauryl sulfate) and sensitizers (1-chloro-2,4-dinitrobenzene, nickel sulfate, oxazolone, 2,4-dinitrofluorobenzene, 2,4,6-trinitrobenze-nesulfonic acid), in SkinEthic.




Full-Thickness Models

EpidermFTeˆ

EpidermFTeˆ

Full Thickness Skin Model (MatTek corporation, Ashland, MA) exhibitsfeatures that are found in intact native human skin. EpidermFT consist of two predom-inant layers, an epidermis and dermis comprised of fibroblast in a collagen matrix. Also,two cell types are found in EpidermFT, normal human keratinocytes and normal humandermal fibroblasts. This skin model is predominantly used for toxicity studies and invitro studies that examine fibroblast–keratinocyte cell interactions. There are no pub-lished studies that examine the permeability of EpidermFT.

Apligraf

Apligraf (Organogenesis, Canton, MA) is a full-thickness skin model. Apligraf is approvedby the Food and Drug Administration (FDA) for venous leg ulcers and diabetic foot ulcers[8, 25, 26]. It contains both an epidermis composed of normal human keratinocytes, and adermis comprising normal human dermal fibroblast. The majority of studies with Apligrafhave focused on skin grafting.

Orcel1

Orcel is a full-thickness skin alternative that has been developed by Ortec international(New York, NY). It consists of both epidermal keratinocytes and dermal fibroblastcultured on opposite sides of a collagen sponge. The collagen sponge comprises bovinecollagen type I. This model has not been examined for permeation study and is usedprimarily as a wound closure for burn victims and for epidermolysis bullosa [27, 28].

Episkineˆ

Episkin is currently marketed by the cosmetic company L’Oreal. It is a full-thickness skinmodel with similarities to native human skin [18]. It has been utilized for both penetrationand toxicity studies. The dermal equivalent is comprised of both types I and IV humancollagen. After incubation the dermal equivalent is seeded with normal human keratino-cytes. A 2002 study by Dreher et al. [17] compared the permeation of model compoundscaffeine and a-tocopherol acetate in both Episkin and human skin. Model drugs wereformulated as oil-in-water emulsions, water-in-oil emulsions, hydrogels, and a liposomaldispersion. The permeability of these model drugs was significantly higher in Episkinthan in human skin. However, the rank order of solute permeability was the same in bothEpiskin and human skin. Another 2002 study compared Episkin and human skin per-meability using mannitol as the model permeant. This study demonstrated that thepermeation of mannitol in Episkin was significantly higher than permeation dataobtained using human skin [16].

A prevalidation study that was supported by the European Centre for the Validation ofAlternative Methods during 1999 and 2000, examined the use of Episkine

ˆas an in vitro

alternative model for predicting skin irritation [29]. During phases 1 and 2 of the pre-validation study it was determined that the reproducibility and predictability of Episkinwas found to be insufficient. Protocols and procedures were modified to allow for a




reduction in exposure time between the skin and skin irritants. This new methodprovided higher sensitivity, specificity, and accuracy. As a result of the prevalidationstudy, several cultured skin models are undergoing validation trials as in vitro models forskin irritancy. A 2001 study by Cotovio et al. [30] examined the effects of oxidative stresson Episkin and cultured immortalized human keratinocytes. In this study the authorsused ozone as the oxidative stress and then measured chemical by-products in Episkinand cell cultures. It was found that Episkin was remarkably susceptible to oxidative stressgenerated by ozone.

Bioengineered-Human Skin Equivalent

A 2000 study by Asbill et al. [15]described the development of a full-thickness skin modelthat mimicked human skin in terms of its permeability, lipid profile, and biologicalfeatures. This model comprised both normal human keratinocytes and normal humandermal fibroblasts. A mixture of bovine collagen, tissue culture media, and fibroblast wascultured for several days, which resulted in pseudo-dermal tissue. Normal human kera-tinocytes were then seeded onto the surface of this dermal substitute and the keratino-cytes were allowed to proliferate. Eventually the tissues were cultured at the air–liquidinterface, which resulted in stratification of the epidermal layer. This model contained afully developed epidermal layer grown on a dermal substitute. The major lipid classesfound in native human skin were found in this skin model. However, ceramides 6I and6II were under-represented in this skin model. Trace amounts of acylglucosylceramide(a potent differentiation marker) were also found in the Bioengineered-Human SkinEquivalent (BHS) [15].

The permeability of this skin alternative was compared to that of human cadaverskin, Epiderm, and hairless mouse skin. The model drugs used in this study were(caffeine, hydrocortisone, and tamoxifen). A good indicator of permeability is the cumu-lative amount of drug that has permeated into the receptor compartment after a definedperiod of time (Q). For example, Q24 would indicate the amount of drug thathas permeated into the receptor compartment after 24 hours. The permeation trendsfor the skin models were for hydrocortisone: Q24 Epiderm>hairless mouseskin>BHS>Human cadaver skin. A similar trend was found when using the modeldrugs caffeine and tamoxifen. The BHS more closely mimicked human cadaver skinpermeability than hairless mouse skin or Epiderm.

Conclusions

Information regarding the culturing of viable skin alternatives has grown tremendouslyover the last decade. Currently, skin alternatives are being used on a routine basis for skingrafting, irritation studies, and for permeation testing. While there have been numerousstudies that address the development and the testing of skin alternatives for bioavail-ability studies, there is no skin model currently available that provides permeation datasimilar to that obtained using native intact human skin. Most of the documented researchstudies comparing the permeability of cultured skin models to human skin reveal thatthe skin models are significantly more permeable than human skin. Although there is nocultured skin model comparable to the permeability of human skin, there does appear tobe a correlation of the relative permeability of specific compounds in cultured models




and human skin. While cultured models are lacking in the ability to provide data forin vitro–in vivo correlations with regard to overall permeability, they have generateduseful data with regard to relative permeability of specific compounds. These modelshave also given useful viability data in correlating toxicity and irritation of topicallyapplied compounds when compared to results obtained using human skin. The futurechallenge is to develop cultured skin models that provide identical permeation profilesto human skin. Meanwhile, advances in understanding the statistical correlation of therelative permeability between specific compounds should be enhanced.

References1. Brown, L. and R. Langer, Transdermal delivery of drugs. Annu Rev Med, 1988. 39: p. 221–9.2. Imhof, P.R. et al., Studies of the bioavailability of nitroglycerin from a transdermal therapeutic

system (Nitroderm TTS). Eur J Clin Pharmacol, 1984. 27(1): p. 7–12.3. Tojo, K. and A.C. Lee, A method for predicting steady-state rate of skin penetration in vivo.

J Invest Dermatol, 1989. 92(1): p. 105–8.4. Sinha, V.R. and M.P. Kaur, Permeation enhancers for transdermal drug delivery. Drug Dev Ind

Pharm, 2000. 26(11): p. 1131–40.5. Akerman, B. et al., Penetration enhancers and other factors governing percutaneous local

anaesthesia with lidocaine. Acta Pharmacol Toxicol (Copenh), 1979. 45(1): p. 58–65.6. Williams, A.C. and B.W. Barry, Terpenes and the lipid–protein-partitioning theory of skin

penetration enhancement. Pharm Res, 1991. 8(1): p. 17–24.7. Priborsky, J. and E. Muhlbachova, Evaluation of in-vitro percutaneous absorption across

human skin and in animal models. J Pharm Pharmacol, 1990. 42(7): p. 468–72.8. Curran, M.P. and G.L. Plosker, Bilayered bioengineered skin substitute (Apligraf): a review of

its use in the treatment of venous leg ulcers and diabetic foot ulcers. BioDrugs, 2002. 16(6):p. 439–55.

9. Ponec, M. and J. Kempenaar, Use of human skin recombinants as an in vitro model for testingthe irritation potential of cutaneous irritants. Skin Pharmacol, 1995. 8(1–2): p. 49–59.

10. Verma, A.K. and R.K. Boutwell, An organ culture of adult mouse skin: an in vitro model forstudying the molecular mechanism of skin tumor promotion. Biochem Biophys Res Commun,1980. 96(2): p. 854–62.

11. Regnier, M. and M. Darmon, Human epidermis reconstructed in vitro: a model to studykeratinocyte differentiation and its modulation by retinoic acid. In Vitro Cell Dev Biol, 1989.25(11): p. 1000–8.

12. Prunieras, M., Epidermal cell cultures as models for living epidermis. J Invest Dermatol, 1979.73(2): p. 135–7.

13. Regnier, M. et al., Reconstructed human epidermis: a model to study in vitro the barrierfunction of the skin. Skin Pharmacol, 1992. 5(1): p. 49–56.

14. Parenteau, N.L. et al., Epidermis generated in vitro: practical considerations and applications.J Cell Biochem, 1991. 45(3): p. 245–51.

15. Asbill, C. et al., Evaluation of a human bio-engineered skin equivalent for drug permeationstudies. Pharm Res, 2000. 17(9): p. 1092–7.

16. Lotte, C. et al., Permeation and skin absorption: reproducibility of various industrial recon-structed human skin models. Skin Pharmacol Appl Skin Physiol, 2002. 15 Suppl 1: p. 18–30.

17. Dreher, F. et al., Comparison of cutaneous bioavailability of cosmetic preparations containingcaffeine or alpha-tocopherol applied on human skin models or human skin ex vivo at finitedoses. Skin Pharmacol Appl Skin Physiol, 2002. 15 Suppl 1: p. 40–58.

18. Ponec, M. et al., Characterization of reconstructed skin models. Skin Pharmacol Appl SkinPhysiol, 2002. 15 Suppl 1: p. 4–17.




19. Zghoul, N. et al., Reconstructed skin equivalents for assessing percutaneous drug absorptionfrom pharmaceutical formulations. Altex, 2001. 18(2): p. 103–6.

20. Cannon, C.L. and P.J. Neal, New epidermal model for dermal irritancy testing. Toxicol In Vitro,1994. 8(4): p. 889–91.

21. Faller, C. and M. Bracher, Reconstructed skin kits: reproducibility of cutaneous irritancy testing.Skin Pharmacol Appl Skin Physiol, 2002. 15 Suppl 1: p. 74–91.

22. Gysler, A. et al., Skin penetration and metabolism of topical glucocorticoids in reconstructedepidermis and in excised human skin. Pharm Res, 1999. 16(9): p. 1386–91.

23. Schmook, F.P., J.G. Meingassner, and A. Billich, Comparison of human skin or epidermismodels with human and animal skin in in-vitro percutaneous absorption. Int J Pharm, 2001.215(1–2): p. 51–6.

24. Coquette, A. et al., Analysis of interleukin-1alpha (IL-1alpha) and interleukin-8 (IL-8) expres-sion and release in in vitro reconstructed human epidermis for the prediction of in vivo skinirritation and/or sensitization. Toxicol In Vitro, 2003. 17(3): p. 311–21.

25. Fahey, C., Experience with a new human skin equivalent for healing venous leg ulcers. J VascNurs, 1998. 16(1): p. 11–5.

26. Falanga, V., Apligraf treatment of venous ulcers and other chronic wounds. J Dermatol, 1998.25(12): p. 812–7.

27. Still, J. et al., The use of a collagen sponge/living cell composite material to treat donor sitesin burn patients. Burns, 2003. 29(8): p. 837–41.

28. Bello, Y.M., A.F. Falabella, and W.H. Eaglstein, Tissue-engineered skin. Current status inwound healing. Am J Clin Dermatol, 2001. 2(5): p. 305–13.

29. Portes, P. et al., Refinement of the Episkin protocol for the assessment of acute skin irritation ofchemicals: follow-up to the ECVAM prevalidation study. Toxicol In Vitro, 2002. 16(6): p. 765–70.

30. Cotovio, J. et al., Generation of oxidative stress in human cutaneous models following in vitroozone exposure. Toxicol In Vitro, 2001. 15(4–5): p. 357–62.




Chapter 22

High Throughput Screeningof Transdermal PenetrationEnhancers: Opportunities,Methods, and Applications

Amit Jain, Pankaj Karande, and Samir Mitragotri

CONTENTS

Introduction .................................................................................................................................... 319Ability to Screen a Large Number of Formulations................................................................... 321Use of a Surrogate End Point That Is Quick, Easy, and Independent of the

Physicochemical Properties of the Model Permeant............................................................. 321Low Incubation Times to Further Increase the Throughput and Hence Time Efficiency....... 321Minimal Use of Test Chemicals and Efficient Utilization of Model Membrane such

as Animal Skin ........................................................................................................................ 322Adaptability to Automation to Reduce Human Interference .................................................... 322Use of a Common Model Membrane to Represent Human Skin ............................................. 322Use of Consistent Thermodynamic Conditions for Enhancer Formulations............................ 322

Overview of INSIGHT Screening................................................................................................... 322Skin Impedance–Skin Permeability Correlation............................................................................ 324Validation of INSIGHT with FDC................................................................................................... 327Applications of INSIGHT Screening .............................................................................................. 327References....................................................................................................................................... 330

Introduction

The idea of delivering drugs through the skin is as old as human civilization, butthe excitement has increased in recent times after the introduction of the first transder-mal patch in 1970s. Though transdermal route of drug administration offers several


319


advantages — reduced first-pass drug metabolism, no gastro-intestinal degradation, long-term delivery (>24 h) and control over delivery and termination — only few drugmolecules have been formulated into transdermal patches (Barry, 2001). The cause ofthis imbalance between the benefits of this route and the number of products in themarket lies in the skin itself. The skin’s topmost layer, stratum corneum (SC), forms abarrier against permeation of xenobiotics into the body and water evaporation out of thebody. This barrier must be altered to maximize the advantages of transdermal route ofdrug administration. This has engaged pharmaceutical scientists, dermatologists, andengineers alike in research over the last couple of decades (Mitragotri, 2004). Highresearch activity in this field has led to the introduction of a variety of techniquesincluding formulation-based approaches (Williams and Barry, 2004), iontophoresis(Kalia et al., 2004), electroporation (Prausnitz, 1999; Weaver et al., 1999), acousticalmethods (Mitragotri and Kost, 2004), microneedles (Prausnitz, 2004), jet injection(Hingson and Figge, 1952), and thermal poration (Sintov et al., 2003).

All of the above techniques have their own benefits and specific applications.Formulation-based approaches have a number of unique advantages such as designsimplicity and flexibility, and ease of application over a large area (Prausnitz et al.,2004). The last 20 years have seen extensive research in the field of chemical enhancers,which form the core component of formulation-based strategies for transdermal drugdelivery. More than 200 chemicals have been shown to enhance skin permeabilityto various drugs. These include molecules from a diverse group of chemicals includingfatty acids (Golden et al., 1987; Aungst et al., 1990; Jain and Panchagnula, 2003), fattyesters (Chukwumerije et al., 1989), nonionic surfactants (Lopez et al., 2000), anionicsurfactants (Nokhodchi et al., 2003), and terpenes (Williams and Barry, 1991; Jain et al.,2002). However, identification of potent yet safe permeation enhancers has provedchallenging. To date, only few chemicals are to be found in currently marketed trans-dermal products. These include oleic acid, sorbitan monooleate, and methyl laurateamong others.

Even though individual chemical penetration enhancers (CPEs) have found limitedapplications, combinations of CPEs represent a huge opportunity that has been sparselytapped. Several reports have indicated that combinations of CPEs offer better enhance-ments of transdermal drug transport compared to their individual constituents (Mitragotri,2000; Thomas and Panchagnula, 2003). However, such combinations do not necessarilyyield safer enhancers. It should be feasible, in principle, to use CPEs as building blocks toconstruct new microstructures and novel formulations that offer enhancement withoutirritation. However, the challenge now shifts to screening the potency of enhancercombinations. Random mixtures of CPEs are likely to exhibit additive properties, thatis, their potency and irritancy are likely to be averages of corresponding properties oftheir individual constituents. Occurrence of truly synergistic combinations is likely to berare. In the absence of capabilities to predict the occurrence of such rare mixtures, onehas to rely on a brute force screening approach. Starting with a pool of more than 200individual CPEs, millions of binary and billions of higher order formulations can bedesigned. The screening of these mixtures is a mammoth task.

Screening of chemical enhancers can be performed in vitro as well as in vivo. In vivoexperiments are likely to yield better results; however, several issues including variability,cost, and practicality limit the possibility of screening a large database of enhancers.Accordingly, in vitro screening based on excised tissue (human or animal) presents amore practical alternative (Priborsky and Muhlbachova, 1990). A number of models existto predict in vivo pharmacokinetics based on in vitro data (Naito and Tsai, 1981; Guy




et al., 1982; Ogiso et al., 1989; Takayama and Nagai, 1991; Wu et al., 2000). The useof in vitro models for screening is also supported by the fact that SC, the principle site ofenhancer action, shows similar behavior in vivo and in vitro except for the extentof metabolic activity (Chang et al., 1998). Majority of in vitro studies on transdermaldrug transport have been performed using Franz diffusion cells (FDCs). The throughputof this traditional set-up of diffusion chamber is very low; not more than 10 to 15experiments at a time. These permeation studies are time-consuming and are resource-expensive as analytical methods such as high-pressure liquid chromatography (HPLC) orradio labeled drugs for liquid scintillation counting are expensive. Automated in-line flowthrough diffusion cells have been developed in the last few years to increase thethroughput of skin permeation experiments (Bosman et al., 1996; Cordoba-Diaz et al.,2000). Although these methods have facilitated the experiments, the throughput of thesemethods has not been significantly improved. Furthermore, these methods are also costprohibitive. Accordingly, standard FDCs still dominate the screening of CPEs.

The urgent need to increase experimental throughput has led to the development ofhigh throughput screening methods. Though in early stages, these methods have alreadyshown promise in discovering novel formulations for transdermal drug delivery. A highthroughput assay to be used for screening of transdermal formulations should meet thefollowing requirements:

Ability to Screen a Large Number of Formulations

Increasing the throughput by at least 2 to 3 orders of magnitude would result insignificant reduction in the effort and time spent in the very first stage of formulationdevelopment (Karande and Mitragotri, 2002).

Use of a Surrogate End Point That Is Quick, Easy, and Independentof the Physicochemical Properties of the Model Permeant

Permeation experiments using radio labeled (Rosado et al., 2003), fluorescent (Ogisoet al., 1996), HPLC-detectable (Wu et al., 2000), or RIA/ELISA-detectable (Xing et al.,1998; Magnusson and Runn, 1999) markers necessitate the need of extensive samplehandling and sample analysis. This accentuates the cost of sample analysis and overalltime spent in characterizing the efficacy of formulations. Furthermore, current state of theart fluidics systems put a fundamental limit on the number of samples handled in a giventime. Permeation of a model solute across the skin in the presence of an enhancer isdependent not only on the inherent capacity of the enhancer to permeabilize skin butalso on the physico-chemical interactions of the enhancer with the model solute (Lee andKim, 1987; Takacs-Novak and Szasz, 1999; Auner et al., 2003 a,b). An end point tocharacterize the effect of an enhancer on skin permeability should be able to decouplethese two effects to assure the generality of the results.

Low Incubation Times to Further Increase the Throughput and HenceTime Efficiency

FDC experiments typically use incubation times of 48 to 96 h thereby reducing thethroughput of permeation experiments. Low incubation times favor high turnover fre-quencies for assay utilization.


High Throughput Screening of Transdermal Penetration Enhancers & 321


Minimal Use of Test Chemicals and Efficient Utilization of Model Membranesuch as Animal Skin

FDCs typically require application of 1 to 2 ml of enhancer formulations over about 3 to4 cm2 of skin per experiment. This makes it cost prohibitive to include candidates that areexpensive in the test libraries as well as to screen a large number of formulations.

Adaptability to Automation to Reduce Human Interference

Typical FDC set-up requires manual sampling with little opportunities for process auto-mation (Cordoba-Diaz et al., 2000).

In addition to these requirements of the assay tool, the high throughput screeningmethodology should also satisfy, if possible, the following experimental constraints:

Use of a Common Model Membrane to Represent Human Skin

It is common to find in transdermal literature the use of a variety of different models torepresent human skin such as rat skin (Schmook et al., 2001), pig skin (Sekkat et al.,2002), snake skin (Itoh et al., 1990), excised human skin, etc. While human skin isdifficult to procure on a large scale, animal models show deviations in permeabilitycharacteristics from human skin (Panchagnula et al., 1997; Schmook et al., 2001; Auneret al., 2003 a,b). Also, results on one model cannot be directly translated to a differentmodel.

Use of Consistent Thermodynamic Conditions for Enhancer Formulations

Permeation enhancement efficacy of a CPE is a function of its chemical potential (Fran-coeur et al., 1990; Shokri et al., 2001), temperature (Ongpipattanakul et al., 1991;Narishetty and Panchagnula, 2004), and co-solvent (Yamane et al., 1995; Larrucea et al.,2001) amongst other thermodynamic parameters. These thermodynamic conditions needto be standardized for all the enhancers that are being tested to create direct comparisonof their efficacies in increasing skin permeation.

This chapter focuses on a specific high throughput screening method called INSIGHT,IN vitro Skin Impedance Guided High Throughput screening that was recently intro-duced (Karande et al., 2004). This method is described in detail with respect to itsfundamentals, validation, and outcomes.

Overview of INSIGHT Screening

INSIGHT screening offers greater than 100-fold improvement in screening rates oftransdermal formulations (Karande et al., 2004). This improvement in efficiency comesfrom two factors. First, INSIGHT, in its current version, can perform up to 50 tests persquare inch of skin compared to about more than 2 cm2 of skin per test in the case

Second, INSIGHT screening uses skin impedance as a surrogate marker for skinpermeability.




of FDC (Figure 22.1). About 100 formulations can be screened per INSIGHT array.

Skin impedance has been previously used: (i) to assess the skin integrity for in vitrodermal testing (Lawrence, 1997; Fasano et al., 2002; Davies et al., 2004), (ii) to evaluatethe irritation potential of chemicals in a test known as Skin Integrity Function Test (SIFT)(Heylings et al., 2001), and (iii) to monitor skin barrier recovery in vivo following theapplication of current during iontophoresis (Turner et al., 1997; Curdy et al., 2002). Sinceit is evident from the literature that skin impedance can be used to confirm skin integrity,it is logical to hypothesize that alteration in skin barrier due to chemical enhancers canbe used as an in vitro surrogate marker for permeability. Scattered literature data supportthis hypothesis. A study by Yamamoto and Yamamoto (1976a,b) showed that total skinimpedance reduces gradually with tape stripping and after 15 strippings skin impedanceapproaches the impedance value of deep tissues (Yamamoto and Yamamoto, 1976a,b).However, quantitative relationships between skin impedance and permeability in thepresence of chemical enhancers and their validity for a wide range of markers have notbeen previously documented.

~

(a)

(b)

Donor

Receiver

Skin

Electrode

Electrode

Figure 22.1 Schematic of the INSIGHT screening apparatus. The INSIGHT screen is made up ofa donor array (top) and a symmetric receiver array (bottom). A single screen can screen 100formulations at one time. The skin is sandwiched between the donor (Teflon) and receiver(Polycarbonate) and the formulations contact the SC from the donor array. Conductivity meas-urements are made with one electrode inserted in the dermis and a second electrode movedsequentially in the donor wells. (a) and (b) are the top and side view of the INSIGHT apparatus,respectively.




Skin Impedance–Skin Permeability Correlation

SC is a composite of proteins and lipids in which protein-rich corneocytes are surroundedby lipid bilayers (Madison et al., 1987). Approximately 70 to 100 bilayers are stackedbetween two corneocytes (Elias et al., 1977; Elias, 1983). Because of its architecture theSC is relatively nonconductive and possesses high electrical impedance (Lackermeieret al., 1999). Skin impedance can be measured either by applying a constant current (AC)and measuring the potential across the skin or by measuring transepidermal currentfollowing the application of a constant AC potential. Data reported in this chapter arebased on measurement of transepidermal current following the application of a constantpotential (100 mV rms). Frequency of the applied potential is also an important param-eter. Due to the capacitive components of the skin, the measured electrical impedance ofthe skin decreases with increasing frequency (Yamamoto and Yamamoto, 1976a,b).While the use of higher frequencies facilitates measurements due to decreased imped-ance, the correlation between electrical impedance and solute permeability is stronger atlower frequencies. Thus, an optimal frequency must be chosen. All experiments reportedin chapter were performed at a frequency of 100 Hz.

INSIGHT screening is founded on the relationship between skin’s electrical imped-ance (reciprocal of skin conductance) and solute permeability. There is a dearth ofliterature on skin impedance (conductivity) and permeability relationship and moreoverin most of the studies this relationship was used to elucidate the mechanism of transportof hydrophilic molecules across the skin under the influence of temperature (Peck et al.,1995), hydration (Tang et al., 2002), electric current (Sims et al., 1991; Li et al., 1998) orultrasonic waves (Tang et al., 2001; Tezel et al., 2003). Therefore existing data cannot beused to generalize the relationship between skin impedance and permeability. Accord-ingly, a large dataset was first generated to assess the correlation between skin imped-ance and permeability to small (mannitol) and macromolecule (inulin) hydrophilicsolutes in the presence of different chemical enhancer formulations. A set of 22 enhancerformulations, chosen from the candidate pool was used to validate the relationshipbetween skin conductivity and skin permeability. The candidate pool comprised of 15single enhancer formulations and 7 binary enhancer formulations. To establish thecorrelation between skin impedance and permeability for wide range of chemical en-hancers, formulations were made from different classes of chemicals including cationicsurfactants (CTAB — Cetyl trimethyl ammonium bromide, BDAC — Benzyl Dodecylammonium chloride), anionic surfactants (NLS — N-lauorylsarcosine sodium, SLA —Sodium laureth sulfate, SLS — Sodium lauryl sulfate), zwitterionic surfactant (HPS — N-Hexadecyl-N, N-Dimethyl-3-ammonio-1-propanesulfonate), nonionic surfactant (PEGE— Polyethylene dodecyl glycol ether, S20 — Sorbitan monolaurate, T20 — Polyoxyethy-lene sorbitan monolaurate), fatty acid and their sodium salts (LA — lauric acid, OL —oleic acid, SOS — sodium octyl sulfate, SO — Sodium oleate), fatty acid ester (TET —Tetracaine HCl, IPM — Isopropyl myristate) and others (DMP — N-dodecyl 2-pyrroli-done; MEN — Menthol). Skin impedance and permeability to two model solutes, man-nitol and inulin were measured. Inulin (MW, 5 kDa) was selected as a model solute as itsatisfactorily represents a macromolecular hydrophilic drug. Mannitol (MW, 182.2 Da; logKo/w, 3.1) was used as a representative of small hydrophilic drugs.

A strong correlation was observed between skin impedance and permeability of

The measurements reported in Figure 22.2a and b were performed in FDCs. There is areasonable scatter in these data, which is inherent to biological systems such as skin that




mannitol and inulin for different enhancer formulations (Figure 22.2 to Figure 22.4).

exhibit high variability. Also, measurements reported in Figure 22.2a and b represent anaggregate of experiments performed over several different animals and anatomicalregions. The correlation between skin permeability and impedance was improvedwhen data for individual enhancers were plotted separately; example for inulin (withDMP enhancer r 2¼ 0.85) and mannitol (with OL enhancer r 2

approximately over 5 kV cm2 intervals (inulin r 2¼ 0.86 and for mannitol r 2¼ 0.90).Permeability data of mannitol and inulin with a variety of chemical enhancer formula-tions showed that skin impedance is inversely related to permeability of hydrophilicsolutes, which is in agreement with existing data in literature. Correlation coefficient(inulin r 2¼ 0.86 and mannitol r 2¼ 0.90) of average data for all enhancer formulationsindicates that a remarkable correlation exists between skin permeability and impedancefor single and binary enhancers formulations irrespective of the nature of the formula-tion. These results indicate that skin impedance can be used a parameter to measurethe extent of barrier alteration by chemicals irrespective of their mode of action (which,in most cases, is not precisely known). Specifically, good correlations were observed

410−5

10−6

10−5

10−4

10−3

710−5

410−4

10−4

10 100Skin impedance (kΩ cm2)

10 100Skin impedance (kΩ cm2)

Inul

in s

kin

perm

eabi

lity

(cm

/h)

Man

nito

l per

mea

bilit

y (c

m/h

)

(a) (b)

Figure 22.2 Skin impedance — permeability correlation for (a) inulin and (b) mannitol. Testformulations used in this study (in parenthesis total concentration of chemical enhancer w/v,weight fraction used): (a) ., MEN (1.5% w/v); &, SO (1.5% w/v); ~, PEGE (1.5% w/v); ~, OL(1.5% w/v); u, S20 (1.0% w/v);, DMP (1.5% w/v); !, OL:MEN (1.5% w/v, 0.4:0.6); ^, IPM(1.5% w/v); 3, TET (2.0% w/v); 1, LA (1.5% w/v); ^, NLS (1.0% w/v); (, SOS (2.0% w/v); u,NLS:S20 (1.0% w/v, 0.6:0.4); ), TET:SLS (1.0% w/v, 0.6:0.4); TET:HPS (2.0% w/v, 0.1:0.9), ,MEN:T20 (2.0% w/v, 0.5:0.5); *, DMP:TET (2.0% w/v, 0.4:0.6); G, CTAB (1.0% w/v). (b) G, OL(1.5% w/v); ~, DMP (2.0% w/v); !, DMP-TET (2.0% w/v, 0.4:0.6); ~, PEGE (1.5% w/v); u,TET (2.0% w/v); 3, LA (1.5% w/v); 1, S20 (1.0% w/v); ^, HPS (1.5% w/v); s, NLS (1.0% w/v);&, BDAC (1.5% w/v); ^, MEN (1.5% w/v); ^, DMP-HPS (1.5% w/v, 0.6:0.4); j, NLS-S20(1.0% w/v, 0.6:0.4); ., DMP (1.5% w/v).




¼ 0.86) is given in (Figure22.3a and b). The correlation between skin permeability and impedance can be clearlyseen in Figure 22.4a and b where data in Figure 22.2a and b are replotted after averaging

between permeability and skin impedance for enhancers, which act by lipid extraction(NLS, MEN, BDAC) or by lipid bilayer fluidization (OL, LA, IPM). Note, however, that thenature of these correlations is an integral function of the physico-chemical properties ofthe drug or permeant. Educated discretion must therefore be exercised when selecting adelivery formulation for a particular model permeant or drug of interest.

610−5

810−5

310−4

510−4

10−410−4

10−3

2

(a) (b)

4 6 8 10

Skin impedance (kΩ cm2)

5 6 7 8 9 10 20 30

Skin impedance (kΩ cm2)

Inul

in s

kin

perm

eabi

lity

(cm

/h)

Man

nito

l ski

n pe

rmea

bilit

y (c

m/h

)

Figure 22.3 Skin impedance–permeability correlation for single enhancer. (a) Plot of skinpermeability to inulin vs. skin impedance in presence of DMP (1.5% w/v in 1:1EtOH:PBS); (b)plot of skin permeability to mannitol vs. skin impedance in presence of NLS (1.5% w/v in

1 10 100

Man

nito

l ski

n pe

rmea

bilit

y (c

m/h

)

510−5

210−4

310−4

410−4

610−5710−5810−5910−5

10−4

10−6

10−5

10−4

10−3

10 100

Inul

in s

kin

perm

eabi

lity

(cm

/h)

Skin impedance (kΩ cm2) Skin impedance (kΩ cm2)(a) (b)

Figure 22.4 Skin impedance–permeability correlation for (a) inulin and (b) mannitol. Modifiedplot of permeability impedance data shown in Figure 22.2. Permeability data for differentenhancers is grouped in the bins of 5 kV cm2 along the x-axis representing skin impedance.The correlation is much tighter as compared to the one in Figure 22.2.




1:1EtOH:PBS). A much tighter correlation can be observed compared to Figure 22.2.

Validation of INSIGHT with FDC

Conductivity enhancement ratio (ER) — that is, the ratio of skin impedances at time zeroand 24 h following the application of enhancer formulation — measurements in INSIGHTwere plotted against conductivity enhancement and permeability enhancements in FDCs(Figure 22.5a). Inulin was used as a model permeant in these studies. Results shown inFigure 22.5a reflect that the predictions obtained from INSIGHT on the potency ofenhancer formulations are essentially the same as those obtained from FDCs. However,INSIGHT allows collection of information at a much greater speed (~1000 per day) andless skin utilization (about 0.07 cm2 per experiment as compared to 2 cm2 in a 16 mmdiameter FDC, greater than 25-fold reduction in skin utilization).

Further improvements in INSIGHT screening speed can be obtained by reducing theformulation incubation period. Capabilities of INSIGHT in assessing formulation potencyafter a 4-h incubation are demonstrated in Figure 22.5b where potency rankings of 438single and binary formulations randomly prepared from the enhancer library based on 4-hscreening are compared to those based on 24-h screening. Rank 1 corresponds to mostpotent formulation in the library and rank 438 to the weakest formulation. The predictionsof the potency made in 4 h were consistent with those made after a contact time of 24 h,thus indicating that the efficiency of INSIGHT screening can be further improved.

Applications of INSIGHT Screening

1. Discovery of Rare Formulations: INSIGHT screening can be used to screen hugelibraries of chemicals within a short span of time and without the fear of failure that exists

0

20

40

60

80

100

5

10

15

20

25

30

35

40

45

0 10 20 30 40 50 60 70

FD

C p

erm

eabi

lity

enha

ncem

ent

For

mul

atio

n ra

nk b

ased

on

4 h

incu

batio

n

Formulation rank based on 24 h incubationINSIGHT conductivity ER at 24 h

FD

C conductivity enhancem

ent

1

10

100

1000

1 10 100 1000(a) (b)

Figure 22.5 Validation of INSIGHT predictions with FDC. (a) Plot of conductivity enhancementratios in INSIGHT at 24 h vs. conductivity and permeability enhancement ratios in FDC at 96 hfor 19 enhancer formulations. A strong linear correlation indicates the validity of observations inINSIGHT when compared with those from traditional tools like FDC. The closed circles indicateconductivity enhancement numbers and the filled circles indicate permeability enhancementnumbers in FDC. (b) Plot of 24 h predictions in INSIGHT vs. 4 h predictions in INSIGHT on thepotency of enhancer formulations. A strong correlation indicates that predictions on potency offormulations can be obtained at significantly lower incubation periods of 4 h.




with traditional tools. Many current single enhancers are also potent irritants to the skin atconcentrations necessary to induce meaningful penetration enhancement. Attempts havebeen made to synthesize novel chemical enhancers such as Azone, however, achievingsufficient potency without irritancy has proved challenging, especially for macromol-ecules. A number of studies have shown that formulations made up of combination ofchemical enhancers are more potent than its individual components (Karande et al.,2004; Tezel et al., 2002; Mitragotri, 2000). The addition of components increases thenumber of formulations exponentially. However, the use of INSIGHT screening allowsone to tackle this challenge in a more cost-effective way compared to FDCs. In addition,synergies between CPEs not only lead to new transdermal formulations but also poten-tially offer insight into mechanisms by which CPEs enhance skin permeability. Predictionof synergies from the first principles is challenging. INSIGHT screening offers an effectivetool for identifying synergies (positive or negative) between the CPEs.

To identify synergistic combinations of penetration enhancers (SCOPE) formulations,a library of chemical enhancers was first generated from 32 chemicals chosen from a listof more than 250 chemical enhancers belonging to various categories. Random pairingof CPEs from various categories led to 496 binary chemical enhancers pairs. For each pair,44 distinct chemical compositions were created with the concentration of each chemicalenhancer ranging from 0 to 2% w/v, yielding a library of 25,000 candidate SCOPEformulations. About 20% of this library (5,040 formulations) was screened using INSIGHTthe largest ever cohesive screening study reported in the transdermal literature. Eachformulation was tested at least four times in over 20,000 experiments (Karande et al.,2004). Using the traditional tools for formulation screening, it would have taken over7 years to do these many experiments. With INSIGHT screening, the same task wasaccomplished in about 2 months with screening rate of 500 to 1,000 experiments per day.

Binary formulations exhibited a wide range of enhancements. The percent of randomlygenerated enhancer combinations that exhibit ER above a certain threshold decreases

figure corresponding to high ER values. Less than 0.1% of formulations exhibited morethan 60-fold enhancement of skin conductivity. Discovery of such rare formulations bybrute force experimentation is contingent on the throughput of the experimental tool.INSIGHT screening opens up the possibility of discovering such rare formulations.

One of the formulations discovered by INSIGHT, SLA:PP (Sodium Laureth Sulfate:Phenyl Piperazine) was shown to increase the permeability of macromolecules such asinulin across porcine skin by 80- to 100-fold compared to passive skin permeability ofinulin (Karande et al., 2004). SLA:PP also increased the skin permeability of molecules suchas methotrexate, low molecular weight heparin, leutenizing hormone releasing hormone(LHRH), and oligonucleotides by 50- to 100-fold. Animal experiments in hairless rats alsoconfirmed delivery of a synthetic analog of LHRH, leuprolide acetate in vivo. The amountof leuprolide acetate delivered using a SCOPE formulation (SLA:PP) is significantly morethan that delivered from a control solution and lies in the therapeutic window.

2. Generation of Database for Quantitative Understanding: Looking beyondsearching for potent combinations of enhancers, the sheer volume of information gener-ated via INSIGHT screening on the behavior of a wide variety of penetration enhancers willprovide, for the first time, a platform to build further investigations of the fundamentalaspects of enhancer–skin interactions. Quantitative descriptions of structure–activity rela-tions (QSARs) for CPEs, which have had limited success in the past (Moss et al., 2002;Walker et al., 2003), may lead to better outcomes in light of the availability of large volumes




rapidly with increasing threshold (Figure 22.6A). The inset shows a section of the main

of data collected in a consistent manner. As exemplified in Figure 22.6, this informationshould help in generating hypotheses relating the chemistry of CPEs to their potencies. Forworking hypotheses, this knowledge can then help refine our selection rules for designingnext generation transdermal formulations. Repeating the experiment–hypothesis loopover a vast, but limited, number of candidate penetration enhancers will provide themissing pieces in solving a vast multivariate problem. Also, this knowledge should signifi-cantly reduce the cost and effort of designing therapeutics for use on skin in the future.

Bulkobservableparametrical

INSIGHT

QSAR

Relating activity to structure

Building predictive capabilities

Discretestructuralbehavioral

C

(c) Database for QSAR

(a) Discovery of rare formulations

A

0%

5%

10%

15%

20%

25%

30%

35%

0 10 20 30 40 50 60

Per

cent

of e

nhan

cer

form

ulat

ions

Conductivity enhancement ratio

0.01%

0.1%

1%

10%

30 40 50 60

Per

cent

of E

nhan

cer

For

mul

atio

ns

Conductivity Enhancement Ratio

B

(b) Exploration of synergy

INSIGHT

Fraction of MP0 1.0

Tot

al C

onc.

(%w

/v)

0

2.0

Figure 22.6 Applications of INSIGHT screening. (a) Discovery of rare enhancer formulationsthat are significantly potent in increasing skin permeability. Such formulations are difficult todiscover using the traditional tools like FDC due to their low experimental throughput. Thesuccess rate of discovering these potent formulations is very small (~0.1%) requiring a tool withhigh experimental throughput. (b) INSIGHT screening is used to quantify the extent of inter-actions between the components of CPE mixtures in terms of Synergy. Regions of high synergyalmost always overlap with the regions of high potency. (c) INSIGHT screening can be used togenerate large volumes of data on the interaction of CPEs with skin. The information is usedto relate chemistry of the enhancer to its potency using QSAR.




ReferencesAuner, B. G., C. Valenta et al. (2003a). Influence of lipophilic counter-ions in combination with

phloretin and 6-ketocholestanol on the skin permeation of 5-aminolevulinic acid. Int J Pharm255(1–2): 109–116.

Auner, B. G., C. Valenta et al. (2003b). Influence of phloretin and 6-ketocholestanol on the skinpermeation of sodium-fluorescein. J Control Release 89(2): 321–328.

Aungst, B. J., J. A. Blake et al. (1990). Contributions of drug solubilization, partitioning, barrierdisruption, and solvent permeation to the enhancement of skin permeation of various com-pounds with fatty acids and amines. Pharm Res 7(7): 712–718.

Barry, B. W. (2001). Novel mechanisms and devices to enable successful transdermal drugdelivery. Eur J Pharm Sci 14(2): 101–114.

Bosman, I. J., A. L. Lawant et al. (1996). Novel diffusion cell for in vitro transdermalpermeation, compatible with automated dynamic sampling. J Pharm Biomed Anal 14(8–10):1015–1023.

Chang, P., M. D. Rosenquist et al. (1998). A study of functional viability and metabolic degenerationof human skin stored at 4 degrees C. J Burn Care Rehabil 19(1 Pt 1): 25–28.

Chukwumerije, O., R. A. Nash et al. (1989). Studies on the efficacy of methyl esters of n-alkyl fattyacids as penetration enhancers. J Invest Dermatol 93(3): 349–352.

Cordoba-Diaz, M., M. Nova et al. (2000). Validation protocol of an automated in-line flow-throughdiffusion equipment for in vitro permeation studies. J Control Release 69(3): 357–367.

Curdy, C., Y. N. Kalia et al. (2002). Post-iontophoresis recovery of human skin impedance in vivo.Eur J Pharm Biopharm 53(1): 15–21.

Davies, D. J., Ward, R.J., et al. (2004). Multi-species assessment of electrical resistance as a skinintegrity marker for in vitro percutaneous absorption studies. Toxicol In Vitro 18(3): 351–358.

Elias, P. M. (1983). Epidermal lipids, barrier function, and desquamation. J Invest Dermatol 80Suppl: 44s–49s.

Elias, P. M., N. S. McNutt et al. (1977). Membrane alterations during cornification of mammaliansquamous epithelia: a freeze-fracture, tracer, and thin-section study. Anat Rec 189(4): 577–594.

Fasano, W. J., L. A. Manning et al. (2002). Rapid integrity assessment of rat and humanepidermal membranes for in vitro dermal regulatory testing: correlation of electrical resistancewith tritiated water permeability. Toxicol In Vitro 16(6): 731–740.

Francoeur, M. L., G. M. Golden et al. (1990). Oleic acid: its effects on stratum corneum in relation to(trans)dermal drug delivery. Pharm Res 7(6): 621–627.

Golden, G. M., J. E. McKie et al. (1987). Role of stratum corneum lipid fluidity in transdermal drugflux. J Pharm Sci 76(1): 25–28.

Guy, R. H., J. Hadgraft et al. (1982). A pharmacokinetic model for percutaneous absorption. IntJ Pharm 11: 119–129.

Heylings, J. R., H. M. Clowes et al. (2001). Comparison of tissue sources for the skin integrityfunction test (SIFT). Toxicol In Vitro 15(4–5): 597–600.

Hingson, R. A. and F. H. Figge (1952). A survey of the development of jet injection in parenteraltherapy. Curr Res Anesth Analg 31(6): 361–366.

Itoh, T., J. Xia et al. (1990). Use of shed snake skin as a model membrane for in vitro percutaneouspenetration studies: comparison with human skin. Pharm Res 7(10): 1042–1047.

Jain, A. K. and R. Panchagnula (2003). Transdermal drug delivery of tricyclic antidepressants: effectof fatty acids. Methods Find Exp Clin Pharmacol 25(6): 413–421.

Jain, A. K., N. S. Thomas et al. (2002). Transdermal drug delivery of imipramine hydrochloride. I.Effect of terpenes. J Control Release 79(1–3): 93–101.

Kalia, Y. N., A. Naik et al. (2004). Iontophoretic drug delivery. Adv Drug Deliv Rev 56(5): 619–658.Karande, P., A. Jain et al. (2004). Discovery of transdermal penetration enhancers by

high-throughput screening. Nat Biotechnol 22(2): 192–197.




Karande, P. and S. Mitragotri (2002). High throughput screening of transdermal formulations.Pharm Res 19(5): 655–660.

Lackermeier, A. H., E. T. McAdams et al. (1999). In vivo ac impedance spectroscopy of human skin.Theory and problems in monitoring of passive percutaneous drug delivery. Ann NY Acad Sci873: 197–213.

Larrucea, E., A. Arellano et al. (2001). Combined effect of oleic acid and propylene glycol on thepercutaneous penetration of tenoxicam and its retention in the skin. Eur J Pharm Biopharm52(2): 113–119.

Lawrence, J. N. (1997). Electrical resistance and tritiated water permeability as indicators of barrierintegrity of in vitro human skin. Toxicol In Vitro 11: 241–249.

Lee, S. J. and S. W. Kim (1987). Hydrophobization of ionic drugs for transport through membranes.J Control Release 6: 3–13.

Li, S. K., A. H. Ghanem et al. (1998). Characterization of the transport pathways induced duringlow to moderate voltage iontophoresis in human epidermal membrane. J Pharm Sci 87(1):40–48.

Lopez, A., F. Llinares et al. (2000). Comparative enhancer effects of Span20 with Tween20 andAzone on the in vitro percutaneous penetration of compounds with different lipophilicities.Int J Pharm 202(1–2): 133–140.

Madison, K. C., D. C. Swartzendruber et al. (1987). Presence of intact intercellular lipid lamellae inthe upper layers of the stratum corneum. J Invest Dermatol 88: 714–718.

Magnusson, B. M. and P. Runn (1999). Effect of penetration enhancers on the permeation of thethyrotropin releasing hormone analogue pGlu-3-methyl-His-Pro amide through human epider-mis. Int J Pharm 178(2): 149–159.

Mitragotri, S. (2000). Synergistic effect of enhancers for transdermal drug delivery. Pharm Res17(11): 1354–1359.

Mitragotri, S. (2004). Breaking the skin barrier. Adv Drug Deliv Rev 56(5): 555–556.Mitragotri, S. and J. Kost (2004). Low-frequency sonophoresis: a review. Adv Drug Deliv Rev 56(5):

589–601.Moss, G. P., J. C. Dearden et al. (2002). Quantitative structure–permeability relationships (QSPRs)

for percutaneous absorption. Toxicol In Vitro 16(3): 299–317.Naito, S. I. and Y. H. Tsai (1981). Percutaneous absorption of indomethacin from ointment bases in

rabbits. Int. J. Pharm 8: 263–276.Narishetty, S. T. and R. Panchagnula (2004). Transdermal delivery of zidovudine: effect of terpenes

and their mechanism of action. J Control Release 95(3): 367–379.Nokhodchi, A., J. Shokri et al. (2003). The enhancement effect of surfactants on the penetration of

lorazepam through rat skin. Int J Pharm 250(2): 359–369.Ogiso, T., Y. Ito et al. (1989). A pharmacokinetic model for the percutaneous absorption of

indomethacin and the predication of drug disposition kinetics. J Pharm Sci 78: 319–323.Ogiso, T., N. Niinaka et al. (1996). Mechanism for enhancement effect of lipid disperse system on

percutaneous absorption. J Pharm Sci 85(1): 57–64.Ongpipattanakul, B., R. R. Burnette et al. (1991). Evidence that oleic acid exists in a separate phase

within stratum corneum lipids. Pharm Res 8(3): 350–354.Panchagnula, R., K. Stemmer et al. (1997). Animal models for transdermal drug delivery. Methods

Find Exp Clin Pharmacol 19(5): 335–341.Peck, K. D., A. H. Ghanem et al. (1995). The effect of temperature upon the permeation of polar

and ionic solutes through human epidermal membrane. J Pharm Sci 84(8): 975–982.Prausnitz, M. R. (1999). A practical assessment of transdermal drug delivery by skin electropor-

ation. Adv Drug Deliv Rev 35(1): 61–76.Prausnitz, M. R. (2004). Microneedles for transdermal drug delivery. Adv Drug Deliv Rev 56(5):

581–587.Prausnitz, M. R., S. Mitragotri et al. (2004). Current status and future potential of transdermal drug

delivery. Nat Rev Drug Discov 3(2): 115–124.




Priborsky, J. and E. Muhlbachova (1990). Evaluation of in-vitro percutaneous absorption acrosshuman skin and in animal models. J Pharm Pharmacol 42(7): 468–472.

Rosado, C., S. E. Cross et al. (2003). Effect of vehicle pretreatment on the flux, retention, anddiffusion of topically applied penetrants in vitro. Pharm Res 20(9): 1502–1507.

Schmook, F. P., J. G. Meingassner et al. (2001). Comparison of human skin or epidermismodels with human and animal skin in in-vitro percutaneous absorption. Int J Pharm 215(1–2): 51–56.

Sekkat, N., Y. N. Kalia et al. (2002). Biophysical study of porcine ear skin in vitro and itscomparison to human skin in vivo. J Pharm Sci 91(11): 2376–2381.

Shokri, J., A. Nokhodchi et al. (2001). The effect of surfactants on the skin penetration of diazepam.Int J Pharm 228(1–2): 99–107.

Sims, S. M., W. I. Higuchi et al. (1991). Skin alteration and convective solvent flow effects duringiontophoresis: I. neutral solute transport across human skin. Int J Pharm 69(2): 109–121.

Sintov, A. C., I. Krymberk et al. (2003). Radiofrequency-driven skin microchanneling as a newway for electrically assisted transdermal delivery of hydrophilic drugs. J Control Release 89(2):311–320.

Takacs-Novak, K. and G. Szasz (1999). Ion-pair partition of quaternary ammonium drugs:the influence of counter ions of different lipophilicity, size, and flexibility. Pharm Res 16:1633–1638.

Takayama, K. and T. Nagai (1991). Simultaneous optimization for several characteristics concern-ing percutaneous absorption and skin damage of ketoprofen hydrogels containing Dlinomene.Int J Pharm 74: 115–126.

Tang, H., D. Blankschtein et al. (2002). Prediction of steady-state skin permeabilities of polarand nonpolar permeants across excised pig skin based on measurements of transient diffu-sion: characterization of hydration effects on the skin porous pathway. J Pharm Sci 91(8):1891–1907.

Tang, H., S. Mitragotri et al. (2001). Theoretical description of transdermal transport of hydrophilicpermeants: application to low-frequency sonophoresis. J Pharm Sci 90(5): 545–568.

Tezel, A., A. Sens et al. (2003). Description of transdermal transport of hydrophilic solutes duringlow-frequency sonophoresis based on a modified porous pathway model. J Pharm Sci 92(2):381–393.

Tezel, A., A. Sens et al. (2002). Synergistic effect of low-frequency ultrasound and surfactants onskin permeability. J Pharm Sci 91(1): 91–100.

Thomas, N. S. and R. Panchagnula (2003). Combination strategies to enhance transdermal perme-ation of zidovudine (AZT). Pharmazie 58(12): 895–898.

Turner, N. G., Y. N. Kalia et al. (1997). The effect of current on skin barrier function in vivo:recovery kinetics post-iontophoresis. Pharm Res 14(9): 1252–1257.

Walker, J. D., R. Rodford et al. (2003). Quantitative structure–activity relationships for predictingpercutaneous absorption rates. Environ Toxicol Chem 22(8): 1870–1884.

Weaver, J. C., T. E. Vaughan et al. (1999). Theory of electrical creation of aqueous pathways acrossskin transport barriers. Adv Drug Deliv Rev 35(1): 21–39.

Williams, A. C. and B. W. Barry (1991). Terpenes and the lipid–protein-partitioning theory of skinpenetration enhancement. Pharm Res 8(1): 17–24.

Williams, A. C. and B. W. Barry (2004). Penetration enhancers. Adv Drug Deliv Rev 56(5):603–618.

Wu, P. C., Y. B. Huang et al. (2000). Evaluation of pharmacokinetics and pharmacodynamics ofcaptopril from transdermal hydrophilic gels in normotensive rabbits and spontaneously hyper-tensive rats. Int J Pharm 209(1–2): 87–94.

Xing, Q. F., S. Lin et al. (1998). Transdermal testosterone delivery in castrated Yucatan minipigs:pharmacokinetics and metabolism. J Control Release 52(1–2): 89–98.

Yamamoto, T. and Y. Yamamoto (1976a). Dielectric constant and resistivity of epidermal stratumcorneum. Med Biol Eng 14(5): 494–500.




Yamamoto, T. and Y. Yamamoto (1976b). Electrical properties of the epidermal stratum corneum.Med Biol Eng 14(2): 151–158.

Yamane, M. A., A. C. Williams et al. (1995). Terpene penetration enhancers in propylene glycol/water co-solvent systems: effectiveness and mechanism of action. J Pharm Pharmacol 47(12A):978–989.




Chapter 23

Confocal Laser ScanningMicroscopy: An ExcellentTool for Tracking Compoundsin the Skin

Daya D. Verma and Alfred Fahr

CONTENTS

Abstract ........................................................................................................................................... 336Introduction .................................................................................................................................... 336Confocal Laser Scanning Microscopy (CLSM) ............................................................................... 339

Principles of CLSM...................................................................................................................... 339Major Advantages of CLSM......................................................................................................... 340Major Limitations of CLSM.......................................................................................................... 340Further Perspectives for CLSM Devices Used for Dermatological Applications ...................... 341

CLSM Used for Tracking Liposomal Formulations into the Skin.................................................. 341Tracking of Entrapped and Un-Entrapped Hydrophilic Fluorescent Compounds

in Liposomes into the Skin..................................................................................................... 341Visualization of Marker Substances Encapsulated in Vesicles of Different

Diameters on the Skin Penetration ........................................................................................ 342Synergistic Penetration Enhancement Effect of Ethanol and Phospholipids on

the Topical Drug Delivery ...................................................................................................... 344Visualization of Enhanced Penetration from Nanocarriers Containing Penetration

Enhancers into the Skin.......................................................................................................... 347Penetration Studies Using Rat Abdominal Skin......................................................................... 348Penetration Studies Using Double-Labeled Vesicles ................................................................. 349

Tracking of Fluorescence Labels into Hair Follicles ..................................................................... 350Tracking of Fluorescently Labeled Cyclosporin A into the Rat Hair Follicles.......................... 350Accumulation of Liposomally Entrapped Rho-PE and CF into the Human Hair Follicles ...... 350


335


The Efficacy of Top-Rol1 Dermaroller for Substance Penetration into the Skin ........................ 351Concluding Remarks ...................................................................................................................... 353References....................................................................................................................................... 354

Abstract

The introduction of confocal laser scanning microscopy (CLSM) for visualization of fluor-escent compounds permits simultaneous viewing of multiple fluorophores, thereby in-creasing simplicity and accuracy of location identification of the fluorescent compounds inthe skin. CLSM provides vital information regarding the interactions between nano-scaleddrug carriers, like liposomes, nanospheres, and micelles, with the various cells of the skinand their accumulation, intracellular fate, and mechanism of penetration. This chapterdescribes various applications of CLSM to investigate skin penetration properties andspecific targeting of hair follicles. Here, we will discuss the implication of CLSM in thevisualization of fluorescent compounds encapsulated in lipid-based drug carriers. Theenhanced skin penetration of fluorescent model compounds was confirmed by CLSM.CLSM was also used to visualize the accumulation of fluorescent model compoundsencapsulated in lipid-based delivery systems in the hair follicle region. CLSM was alsoable to show enhanced dermal deposition of fluorescent compounds after treating the skinwith Dermarollers1. Overall, CLSM technique proved to be a potential novel way oftracking compounds in the skin.

Introduction

Confocal laser scanning microscopy (CLSM) has emerged as a sophisticated tool for track-ing and studying transport phenomena of fluorescent compounds with a very high degreeof precision for biological and other specimens. CLSM is a fluorescence-based imagingtechnique that offers greater resolution than conventional fluorescence microscopy be-cause of its point illumination and detection properties. The point detection is attained bythe addition of a pinhole in front of the fluorescence detector. Point illumination is attainedusing lasers rather than fluorescent lamps as a light source. CLSM is often used to evaluatethe transport studies through various biological membranes and cell uptake studies asexamples. Its capacity for getting very high resolution images of substantially thin sectionenables us to visualizeprocesses inside the cell. The foremost technical advantages ofCLSMinclude (i) the ability to obtain images of optical sections within reasonable time (fewseconds to acquire ahighquality image) andwith ahigh resolution in anoninvasivemannerboth under in vitro and in vivo conditions,1 and (ii) visualization of images parallel to thesurface of the sample, at multiple depths, without mechanical sectioning of the sample.2

The most common problem with dermal delivery of various pharmaceuticals is theirinability to deliver a sufficient therapeutic dose at the site of action as to elicit apharmacological response. There have been several methods for quantification of per-cutaneous penetration. These include diffusion experiments,3,4 visualization by electronmicroscopy5–10 and microdialysis11–14 as well as fluoromicrography.15,16 CLSM17–23,microdialysis and diffusion experiments provide information about the amount and therate of penetration of the model compound, but do not give any information about theeffect of the model drug on cells and lipid organization. The visualization by electronmicroscopy provides detailed information about the structure of the cells and lipid




organization in the skin, but lacks in providing information on the penetration pathwaysand penetration depth of the model compound. Fluoromicrographs of skin treated withfluorescently labeled liposomes demonstrated that the fluorescent marker remained inthe stratum corneum16 or penetrated deeper in the epidermis mainly along the hairshaft.15 A disadvantage of the fluoromicrograph technique is that the tissue needs to be(cryo-)fixed, which may change skin lipid organization or may result in redistribution ofthe marker.

CLSM is a new emerging optical microscopic technique, which offers significantadvantages over conventional microscopy. CLSM can be considered as microscopy ofoptical sections. Light, which is emitted from regions other than the focal plane, is cut offby introducing a diaphragm in the beam path. The result is an optical section, which showsmore details because the blurring from out of focus has vanished. It has been repeatedlyused in experimental as well as in diagnostic dermatopathology.24 Using conventionaloptical microscopy to image the sub-surface morphology of intact three-dimensionalmatrices, such as skin epidermal tissue, requires the tissue to undergo an extensivepreparative protocol. This protocol includes fixing, embedding, and physically sectioningof the specimen. The process is slow and subject to sample perturbation, which can resultin image misinterpretation. Unlike this conventional optical microscopy, the CLSMtechnique requires little or no pretreatment or cryo-fixing of the tissue prior to imaging.Hence, the technique is rapid, and the tissue experiences minimal physical perturbation ordamage.17 CLSM provides information about the localization and the permeation pathwayof a fluorescent model compound in the tissue. However, in the case of penetration studieswith liposomes, CLSM does not provide information about the permeation of the entireliposome, but only about the penetration of the fluorescent label.20,21

In the last two decades, CLSM has been extensively used as a tool to visualize thefluorescent model compound in the skin. Zellmer et al. reported that neither the vesiclesnor the fluorophore N-(lissamine rhodamine B sulfonyl)diacylphosphatidylethanolamine(Rho-PE) penetrates into the human skin in detectable amount.25 Simonetti et al. in 1995visualized diffusion pathways across the stratum corenum (SC) of native andin vitro reconstructed epidermis by using CLSM.26 van Kuijk-Meuwissen et al. showedin 1998 that the dye applied nonocclusively in flexible liposomes penetrated deeper intothe skin than after occlusive application.21 Kirjavainen et al. reported that the fluores-cence from liposomal compositions containing dioleylphosphatidyl ethanolamine(DOPE) was able to penetrate deeper into the SC than that from liposomes withoutDOPE. A pretreatment of skin with unlabeled liposomes containing DOPE or lyso-phosphatidylcholine (lyso-PC) enhanced the subsequent penetration of the fluorescentmarkers, N-Rh-PE and sulforhodamine B into the skin, suggesting possible enhanceractivity.18

Boderke et al. used CLSM to show that amino peptidase activity was evenly distrib-uted throughout the viable part of the epidermis, with enhanced fluorescence in theupper layers of the stratum granulosum, while dermis and SC showed considerably lessamino peptidase activity.27 Vardaxis et al. employed CLSM to examine the structure ofporcine skin and concluded that it provides valuable additional morphological informa-tion of material examined by conventional microscopy for wound healing studies.19

Zellmer et al. used CLSM to demonstrate that vesicles made of native human SC lipidsrapidly interact with phosphatidylserine liposomes, weakly with human stratum corneumlipid liposomes and have no effect on PC liposomes.28 Turner and Guy showed thationtophoresis significantly enhanced the delivery of calcein into hairless mouse skin,particularly via follicular structures.29 Kirjavainen et al. used CLSM to demonstrate that


Confocal Laser Scanning Microscopy & 337


pretreatment of the skin with zwitterionic egg lecithive (EPC) increases the iontophoretictransdermal mannitol flux about threefold compared to iontophoretic control withoutpretreatment.30 Touitou and co-workers examined the penetration of fluorescent probesinto fibroblasts and nude mice skin by CLSM and showed that ethosomes facilitated thepenetration of all probes into the cells, as evident from the high-intensity fluorescence ascompared to the hydroethanolic solution or classic liposomes.31 Bouwstra and co-work-ers revealed using CLSM that follicular accumulation increased with the lipophilic dyealone and in surfactants — propylene glycol.32 Recently, on-line CLSM was used tovisualize the diffusion of a dye in a cross-sectional view of fresh unfixed piece of skinincluding subcutaneous fat. The authors claimed that this technique can visualize thediffusion of a dye into the upper hair follicle at different time points.33 CLSM was alsoused to understand the mechanism by which nanoparticulate systems facilitate skintransport. The surface images revealed that (a) polystyrene nanoparticles accumulatedpreferentially in the follicular openings; (b) this effect increased in a time-dependentmanner; and (c) the follicular localization was favored by smaller particle sizes.34

Mezei and his group initiated research in the usage of liposomes for topical skinapplication in the early 1980s. Two in vivo studies in rabbits documented comparisonsbetween liposomal and conventional formulations of triamcinolone acetonide.35,36 In bothstudies, the application of the liposomal preparations was associated with larger steroidconcentrations in the epidermis, as well as dermis, and a lower systemic absorption thanthe regular formulations. Further bio-deposition studies in animals have demonstrated thatliposomal encapsulation can improve the penetration of various molecules. Enhanceddelivery into the skin has been reported for caffeine in hairless rats37 and lidocaine in rats.38

In these systems, liposomal delivery results in the formation of a large drug reservoir in theskin, which can be used for local treatment. The penetration kinetics of molecules fromliposomes has also been assessed using in vitro skin studies. Egbaria and co-workersfound that incorporation into liposomes resulted in the increased uptake of hydrocorti-sone, fluocinolone acetonide and ciclosporin39 into the cornified layer of hairless mice andguinea pigs. Liposomal carriers have been successful in enhancing the clinical efficacy of anumber of drugs. These have included tretinoin for the treatment of acne,40 glucocorti-coids for the treatment of atopic eczema,41 lidocaine, and tetracaine,38,42 as well as othersas reviewed in 1995.43 The first commercial topical liposomal preparation, Pevaryl Lipogel,produced by Cilag AG, became available in Switzerland in 1988. The product contains 1%econazole in liposomes formulated in a gel.

Recent approaches in modulating delivery through the skin are the design of twonovel vesicular carriers: ethosomes and transfersomes. These vesicles have been shownto deliver molecules into the deeper layers and through the skin. The ethosomes are softphospholipid vesicles whose size can be modulated from tens of nanometers to micro-meters.44 These vesicular systems have been found to be very efficient for enhanceddelivery of molecules with different physico-chemical characteristics to or through theskin. Transfersomes have been claimed to be versatile carriers for the local and systemicdelivery of various steroids, proteins, and hydrophilic macromolecules.45 The mechanismproposed by the authors is that these liposomes are highly deformable and this propertyfacilitates their rapid penetration through the intercellular lipid pathway of the SC. Theosmotic gradient, caused by the difference in water concentrations between the skinsurface and skin interior, has been proposed as the major driving force for transfersomespenetration.46 Although the precise mechanism by which vesicular carriers achieve theirenhancement effects has yet to be elucidated, the current data point towards their greatpotential in the design of improved delivery systems.




An important contribution to the understanding of the interactions between vesiclesand human skin was made by Junginger and his group.7,47 These workers employedfreeze fracture electron microscopy and small angle x-ray scattering to study the effectsthat vesicle formulations have on the SC. They identified two types of liposomes–skininteractions: (a) adsorption and fusion of loaded vesicles on the surface of the skinleading to increased thermodynamic activity and enhanced penetration of lipophilicdrugs; (b) interaction of the vesicles within the deeper layers of the SC promotingimpaired barrier functions of these layers for the drug.

Confocal Laser Scanning Microscopy (CLSM)

Principles of CLSM

For imaging of the specimen, a collimated laser beam is reflected by a dichroic mirror andpasses through the objective lens of the microscope in a focussed manner on thespecimen. There, light is emitted at a longer wavelength which is able to pass backthrough the dichroic mirror (being more transparent for longer wavelength light) and isagain focussed at the upper pinhole aperture. Here, out-of-focus light (coming fromplaces of the specimen above or below the focus) is eliminated before the beam hits theelectronic detector (Figure 23.1). The optical resolution obtained is about 0.5 mm in theplane area and about 0.3 mm in the vertical direction.

Photomultiplier(PMT)

Pinhole

Beam-splitting mirror

Objective lens

Specimen

z-Control

Laser

In-focus Out-of-focus

Scanning unit(xy-Control)

Figure 23.1 Schematic diagram of the principle of confocal laser scanning microscopy.




As the laser beam can be moved by a scanner unit in x/y-directions quite precisely, anappreciable area of the specimen can be measured via the detector. A picture can bereconstructed from the coordinates (spatial information) and the corresponding lightintensity (detector) which is fed continuously in a computing system, giving in state-of-the-art devices an image of 2048 2048 pixels. Normal working conditions are 512 512pixels with a frame rate of 5 per second.

Major Advantages of CLSM

One of the main advantages of confocal microscopy is that only one point at thespecimen at its focus is detected at one timepoint. In normal fluorescence microscopy(also called wide field microscopy), out-of-focus fluorescence is adding blurr to theresulting image. However, CLSM reduces blurring of the image from light scatteringand provides clear images because of its point illumination and detection properties;hence, out-of-focus light is excluded from reaching the detector. Therefore, only in-focuslight is detected, hence in-focus images are collected resulting in the increased effectiveresolution. CLSM offers enhanced sensitivity because in confocal microscopic systemsfluorescence from the sample is detected using highly sensitive photomultiplier tubes.Further, because of the optical sectioning possibilities, thicker sections can be imagedand reconstructed in an in-focus three-dimensional manner using image analysis soft-ware. In addition, the magnification can be adjusted electronically and because of thevery low chances of out-of-focus light in the digital images generated by CLSM, moreaccurate quantitation of images is possible. The relative fluorescence levels within animage can be quantified using commercial or freeware image analysis software. Last butnot the least, multiple fluorescence detectors enable simultaneous analysis of multiplecellular properties and fluorescent markers. It is also possible to extend the range ofuseful fluorochromes in instruments by utilizing multiple lasers.

Major Limitations of CLSM

CLSM, however, suffers from a number of limitations. For example, CLSM is still limitedin sensitivity and spatial resolution by background optical noise that results from theremaining out-of-focus fluorescence. The inherent heterogenity of the different skin layerin the sample may also add to the optical aberrations caused by the sample preparationnecessary (see48 for a good introduction into the basic techniques of sample prepar-ation).

In addition, CLSM causes photobleaching and photodamage throughout the illu-minated region. Repeated scans with high-energy optical photons greatly reduce theviability of biological tissues and thereby the available time for studying a given speci-men. Other CLSM limitations include the slow scanning action of the laser for high qualityimages. Point illumination light sources are used in most of the confocal microscopes toexcite the samples. For these systems, maximum scan rates of up to 20 frames per secondcan be achieved. These systems are not suitable for very rapid physiological events.Another limitation of CLSM resides in the range of lasers for which efficient fluorophoreexcitations can be achieved.2 Also, a few seconds’ exposure to the high- intensity laserillumination can be highly destructive to both the viable tissue and fluorophore itself.Autofluorescence from biological tissue is another issue associated with CLSM. Not onlyaromatic amino acids, also structural proteins like keratin49 and collagen as well




as elastin50 as examples cause significant autofluorescence, which will blurr or evenseverely perturb imaging. Ideally, therefore, CLSM optically sections thick tissues that aresufficiently transparent to the laser excitation and fluorescence emission wavelengths,which do not strongly scatter this light, and are relatively free of autofluorescence.1

Quantification of CLSM signals in terms of concentration is only possible, if therelationship between the fluorescent substances and the emission is linear51 andthe fluorescence signal is not attenuated differently for different depth locations of thefluorophore in the sample.

Almost needless to say, those confocal microscopes are more expensive than con-ventional fluorescence microscopes due to the advanced electronics, laser excitationsources, and tailor-made software required for running the hardware.

Further Perspectives for CLSM Devices Used forDermatological Applications

Visualization of the skin using skin autofluorescence has been described recently.52,53

These devices use two-photon techniques for enhanced sensitivity, but until now only afew studies have dealt with this new method.54

Intradermally administered sodium fluorescein was used to visualize in vivo skin ina recent study.55 For this study a miniaturized CLSM device was developed for easyhandling.

These devices certainly need more technical improvements in order to obtain moredetailed images of the investigated skin or for use in in vivo penetration studies.

CLSM Used for Tracking Liposomal Formulations into the Skin

In this chapter, we will discuss the interactions between nanocarriers containing hydro-philic and lipophilic fluorescent models with human and rat skin using CLSM. Ethanolicand hydro-alcoholic solutions of the fluorescent model compounds were used as controlformulations and compared with vesicles containing 1,1’-dioctadecyl-3, 3,3’, 3’-tetra-methylindocarbo-cyanine perchlorate (DiI) as a lipophilic and Alexa Fluor 488 (Alxhy)as a hydrophilic model compounds. Fluorescently labeled cyclosporin A encapsulatedin nanocarriers was also used to understand its penetration behavior across skin. Thepenetration pathway and penetration depth was studied by CLSM after different incuba-tion periods.

Tracking of Entrapped and Un-Entrapped Hydrophilic FluorescentCompounds in Liposomes into the Skin

Liposomes have been extensively studied and suggested as a vehicle for topicaldrug-delivery systems. However, the mechanism of liposomes as drug carriers into theintact skin is not fully understood. The effect of separation of the nonentrappedhydrophilic fluorescent compound carboxyflorescein (CF) from liposomally entrappedCF was investigated by measuring the penetration of CF across human skin undernonocclusive conditions in vitro using Franz diffusion cells and CLSM. We assumedthat the topically applied liposomes, prepared from phospholipids, can carry bothentrapped as well as nonentrapped, hydrophilic drugs into the skin. The fluorescence




dye, CF, was incorporated into the liposomes and applied onto the skin. After 6 h, theamount of CF in the epidermal membrane and full thickness skin was determined by CLSM.Liposomes containing Phospholipon 90, a-tocopherol, sodium cholate, and CF wereprepared by a conventional rotary evaporation method. For preparation of liposomesCFout, first the blank liposomes were made from Phospholipon 90, a-tocopherol, sodiumcholate, and Tris buffer pH 7.0. Then a precalculated amount of CF was added to make upthe final volume. The nonentrapped CF was separated from encapsulated CF by using aMini-Lipoprep dialysis device (Amika Corp. catalog number S81110D) for CF in liposomalformulations showed a sufficiently good polydispersity index below 0.3, which indicatesreasonable size homogeneity of the liposomes. The mean diameter of the various liposo-mal suspensions ranged from 70 to 90 nm with polydispersity indexes between 0.26 and0.29.56 The skin pieces after incubation period of 6 h on Franz diffusion cells were sliced insections of 7-mm thickness by means of a cryomicrotome. These cross-sections wereinvestigated for the amount of CF in the different skin layers by using a laser scanningconfocal imaging system (True Confocal Scanner — Leica TCS 4D, upright microscope —Leitz DM RXE, Laser — Argon Krypton emission wavelengths of 488, 578, and 647, Filters— OG 590 for DiI and BP-FITC for CF).

The penetration study and CLSM images showed that the liposomal formulationcontaining CF both inside and outside exhibited maximum deposition of CF in the SC,whereas the liposomes CFin exhibited a higher penetration into deeper skin layers such

and through the skin to the receiver compartmentof Franz diffusion cell. This study supports our assumption that the liposomes CFin–out arenot under osmotic stress and, therefore, will transfer themselves more easily into the SC.

The results indicated that phospholipid vesicles not only carry the entrapped hydro-philic substance, but also the nonencapsulated hydrophilic substance into the SC andpossibly to the deeper layers of the skin. However, CLSM images do not provide thevisualization of single liposomes, so the penetration of intact liposomes still remains anunsolved question. There may be three mechanisms by which the fluorescence label canpenetrate into skin: (i) the label penetrates associated with the liposomal bilayer (pene-tration of intact vesicles), (ii) the fluorescence label penetrates associated with a liposo-mal bilayer fragment, or (iii) the label penetrates solitary.20

Visualization of Marker Substances Encapsulated in Vesiclesof Different Diameters on the Skin Penetration

In this study, the influence of vesicle size on the penetration of two fluorescently labeledsubstances into the human skin was investigated. For the measurements either a hydro-philic fluorescent compound CF or a lipophilic one (DiI) was encapsulated into vesicles.For this purpose liposomes with a well-defined lipid composition and diameters wereused in an attempt to find the best formulation for topical drug delivery.

CLSM was used to visualize the effect of penetration ability of liposomes containingDiI. Liposomes with a size of 120 nm showed a maximum accumulation of CF in the SC,

and also in the receptor compartment of the Franzdiffusion cell, as compared to larger ones. The liposomes with a size of 120 nm diametershowed statistically enhanced penetration of CF into the skin as compared to largerones.57 The results indicated that the CF penetration was inversely related to the size ofthe liposomes, which was confirmed by the data of the CLSM studies. The maximum DiIfluorescence in the skin was observed with smaller liposomes of 71 nm diameter.




as the viable epidermis (Figure 23.2),

deeper skin layers (Figure 23.3),

skin surface. The images were taken after 3 h of the nonocclusive application of theliposomes containing DiI as lipophilic fluorescent label. In all the images a very highfluorescence was observed in the SC, which is obvious as the fluorescent label DiI ishighly lipophilic. In the case of larger vesicles, that is, 586 nm liposomes, the fluorescencein the viable epidermis and dermis was very weak. However, in the case of liposomeswith an average diameter of 272 nm, there was weak fluorescence observed in viableepidermis and dermis. The small sized liposomes of the average diameter 116 nm hadshown weak to medium fluorescence in the viable epidermis and weak fluorescence inthe dermis. The smallest liposomes with an average diameter of 71 nm had shownmedium fluorescence in the viable epidermis and medium to weak fluorescence in thedermis. There has been great progress in the dermal liposomal delivery, up to now thereis no clear evidence, whether liposomes can pass intact into deeper layers of skin or not.

This CLSM study indicates that the large vesicles with a size greater than or equal to600 nm are not able to deliver their contents into deeper layers of the skin. Theseliposomes stay in or on the SC and after drying they may form a layer of lipid, whichmay further strengthen the barrier properties of SC. The liposomes with size lower thanor equal to 300 nm are able to deliver their contents to some extent into the deeper layers

Figure 23.2 CLSM images of a cross-section of human abdominal skin incubated on a Franzdiffusion cell with different formulations containing CF. The liposomes were applied nonocclu-sively for 6 h. (a) Liposomes CFin , (b) liposomes CFin–out , (c) liposomes CFout formulation, and(d) CF in Tris buffer. Scale bar represents 100 mm.




Figure 23.3 represents CLSM images of the skin cross-sections perpendicular to the

of the skin. However, the liposomes with size lower than or equal to 70 nm seem to bepromising for dermal delivery as they have shown maximum fluorescence both in viableepidermis, as well as in dermis.

This study has shown that the fluorescent model compound can be visualized indeeper layers of the skin by making cross-sections perpendicular to the skin surface or bytaking CLSM images.

Synergistic Penetration Enhancement Effect of Ethanoland Phospholipids on the Topical Drug Delivery

It is generally believed that the vehicle components of a dermatological formulation canappreciably affect the penetration of compounds into and through the skin.58–60 In thisstudy, we investigated the effect of lipid vesicular systems embodying ethanol in rela-tively high concentrations on the percutaneous absorption of CyA (cyclosporin A) usinga standardized skin stripping technique and CLSM using Franz diffusion cell. Ethanolwas used with a commercially available lipid mixture, NAT 8539, to improve the topicaldelivery of CyA.

Figure 23.3 CLSM images of a cross-section of human abdominal skin incubated on a Franzdiffusion cell with liposomes containing the lipophilic fluorescent compound, Dil. The formula-tions were applied non-occlusively for 3h. (a) 71 nm vesicles; (b) 116 nm vesicles; (c) 272 nmvesicles; (d) 586 nm vesicles. Scale bar represents 100 nm




Ethanol has been used as a vehicle in the pharmaceutical and cosmetics industries foryears. Ethanol has been widely reported as an efficient skin penetration enhancer in theconcentration of 5 to 100%.26,61,62 However, due to the interdigitation effect of ethanolon lipid bilayers, it was commonly believed that vesicles could not coexist with highconcentrations of ethanol.63

In a preliminary study using double labeled liposomes, we evaluated the role ofcomposition of lipid bilayer on the skin penetration of hydrophilic (CF) and lipophilic(Rho-PE) fluorescent compounds. Liposomes were prepared as described earlier.57

Figure 23.4 represents the results of this preliminary study. CLSM images revealed thatthe ethanolic solution of Rho-PE and CF produced a fairly homogeneous fluorescencethroughout the SC, but no fluorescence was noticed in the viable epidermis and dermis(Figure 23.4a and b). In contrast, the flexible liposomes (prepared from NAT 8539, acommercially available lipids mixture in 25% wt of ethanol) not only delivered very highfluorescence into the deeper layers of the SC, but also very bright fluorescence wasnoticed in the viable epidermis and dermis for both Rho-PE and CF (Figure 23.4c and d).The PL 90H-Liposomes (containing 90% hydrogenated phosphatidylcholine), also calledhard liposomes, produced a weak fluorescence for CF and negligible fluorescence forRho-PE in the SC; however, no fluorescence was observed in the viable epidermis anddermis (Figure 23.4e and f). Moreover, PL 25-Liposomes (containing 25% phosphatidyl-choline) failed to show any fluorescence both for CF and Rho-PE in the SC as well as inthe viable epidermis and dermis (Figure 23.4g and h). From these preliminary experi-ments, we concluded that not only the amount and the type of phospholipids areimportant for skin penetration enhancement effect but also the amount of ethanol hasa significant role in delivering the fluorescent model compounds into the skin. Theseexperiments led to the assumption that ethanol and phospholipids may have synergisticskin penetration enhancement effects, therefore should be further evaluated.

In order to evaluate this synergistic skin penetration enhancement effect of theethanol and phospholipids, we prepared vesicles composed of NAT 8539, ethanol atdifferent ratios and CyA 0.4% (w/v) in phosphate buffered saline (PBS), pH 7.4. The final

Figure 23.4 CLSM images using vesicles containing different amounts of phosphatidylcholinewith human skin after 12 h. (a and b) Ethanolic solution with Rhodamin-PE (a) and CF (b).(c and d) Flexible liposomes with Rhodamin-PE (c) and CF (d). (e and f ) PL 90H-liposomes withRhodamin-PE (e) and CF (f). (g and h) PL25-liposomes with Rhodamin-PE (g) and CF (h).




lipid concentration was 10% (w/v) for all formulations. In the case of DiI vesicles, theconcentration of DiI was 0.25 mM. This mixture was vortexed for 5 min followed bysonication until a clear transparent solution was obtained. The PBS was added to thismixture with the help of a syringe attached with needle (21G 2’’, 0.8 50) with constantvortex mixing.64 The emulsion was vortexed for an additional 5 min and extrudedthrough polycarbonate membranes of 400, 200, 100, and 50 nm pore size with the helpof an Avestin mini hand extruder.65 The vesicles formed from this solution ranged from56.6 to 100.6 nm in diameter, depending on the amount of ethanol added in theformulation. In vitro skin penetration studies were carried out with Franz diffusion cellusing human abdominal skin.

Skin stripping and cryosectioning showed statistically enhanced deposition of CyAinto the stratum corneum (SC) by CyA vesicles containing 10 and 20% ethanol, ascompared to vesicles prepared without ethanol. CyA vesicles prepared with NAT 8539/ethanol (10/3.3) showed a 2.1-fold, CyA vesicles with NAT 8539/ethanol (10/10) showeda 4.4-fold, and CyA vesicles with NAT 8539/ethanol (10/20) showed a 2.2-fold higherdeposition of CyA into SC, as compared to vesicles made of NAT 8539 without ethanol(NAT 8539/ethanol (10/0)).64

Figure 23.5 depicts the results of the CLSM studies for the DiI vesicles. Ethanolicsolution was able to deliver weak fluorescence into the SC; however, no fluorescence

Figure 23.5 CLSM images of a cross-section of human abdominal skin incubated on a Franzdiffusion cell with different formulations containing DiI. The formulations were applied non-occlusively for 12 h. (a) Ethanolic solution of DiI; (b) NAT8539/ethanol (10/3.3); (c) NAT8539/ethanol (10/10); (d) NAT8539/ethanol (10/20). Scale bar represents 50 mm.




8539/ethanol (10/3.3), produced a fairly homogeneous bright fluorescence throughoutthe SC, but no fluorescence was noticed in the viable epidermis and dermis(Figure 23.5b). The formulation, NAT 8539/ethanol (10/10), produced a bright fluores-cence throughout the SC with very weak to weak fluorescence observed in the viable

duced a fairly homogeneous bright fluorescence throughout the SC and a very weakfluorescence was noticeable in the viable epidermis and dermis (Figure 23.5d). CLSMexperiments have shown that the ethanolic solution of DiI was not even able to deliverfluorescence fairly into the SC. In contrast, all the formulations with NAT 8539 andethanol produced a bright fluorescence homogeneously throughout the SC. Formulationsprepared with NAT 8539 containing 10 and 20% ethanol were also able to show veryweak fluorescence in the viable epidermis and dermis.

Overall, the data presented here clearly indicate that ethanol, together with NAT 8539,has synergistic effects on the delivery of the CyA into the skin. There are several reports,which have suggested that the effect of ethanol on the SC is concentration dependent.66

At low concentrations of ethanol only lipoidal pathways are affected in the SC, while athigher concentrations polar pathways are also affected. The penetration enhancingactivity of ethanol can be attributed to two effects: (a) an increase in thermodynamicactivity, due to evaporation of ethanol, known as ‘‘push effect’’67, and (b) a ‘‘pull effect,’’in which the permeation of the drug molecule is increased, due to reduction in barrierproperties of the SC by ethanol.68 Ethanol–water systems enhance the permeation ofionic solutes through human SC. The increased skin permeation of the ionic permeant bythe ethanol–water systems may be associated with alterations involving the polar path-way. Polar pathway alterations may occur in either or both the lipid polar head andproteinaceous regions of the SC.

Visualization of Enhanced Penetration from Nanocarriers ContainingPenetration Enhancers into the Skin

We have developed a novel type of phosphatidylcholine-based liposomal delivery systemfor topical or follicular drug delivery. Vesicles containing CyA have been selectivelytargeted to the hair follicle and hair shafts using the Dundee experimental bald rat(DEBR) model for the treatment of alopecia areata.69 In this study, the interactions betweenvesicles containing hydrophilic and lipophilic fluorescent model substances with humanand rat skin were investigated in vitro. The effect of incorporation of penetration en-hancers was investigated. Ethanolic and hydro-alcoholic solutions of the fluorescentmodel compounds were used as control formulations. We used two lipophilic and onehydrophilic fluorescent model compounds in this study. The lipophilic fluorescent com-pounds were D-Ala 8 CS-betaaminebenzofurazan (Fl-CyA) and DiI, while the hydrophilicfluorescent model was Alxhy. Vesicles with and without penetration enhancers werecompared with ethanolic solutions as well as hydro-alcoholic solutions of the fluorescentlabels. Vesicles were applied nonocclusively onto the human abdominal and rat skinin vitro using a static Franz diffusion cell. Double-labeled vesicles, i.e. vesicles containingboth DiI and Alxhy, were applied for 6 and 12 h. Penetration of the fluorescent labels wasvisualized by CLSM both in terms of depth, as well as intensities of the fluorescence.

The vesicles investigated here were composed of 10% wt. lipids (PL-80 inethanol (75:25 w/w)), 0 or 1% wt of a mixture of terpenes (PE), fluorescent probe DiI




epidermis and dermis (Figure 23.5c). The formulation, NAT 8539/ethanol (10/20), pro-

was noticed in the viable epidermis and dermis (Figure 23.5a). The formulation, NAT

(650 mg/ml), Fl-CyA (90 mg/ml) or Alxhy (90 mg/ml), and PBS pH 7.4 to 100%. Thelipophilic fluorescent probe was dissolved in 5% v/v ethanol and the hydrophilic fluor-escent probe was dissolved in PBS. The drug (CyA 4 mg/ml) and/or the lipophilicfluorescent probe and PE were dissolved in ethanolic solution of lipids. The mixturewas vortexed for 5 min followed by sonication to form a clear transparent solution. PBS,containing the hydrophilic fluorescent probe, was added to this mixture with the help ofa syringe with constant vortex mixing. The mixture was vortexed for an additional 5 minand then extruded through polycarbonate membranes of different sizes with the helpof mini Avestin hand extruder. The double-labeled vesicles were used to investigate thepenetration of both hydrophilic, as well as lipophilic labels. The formulations werelabeled as Fl-CyA vesicles (vesicles containing Fl-CyA without PE), Fl-CyA vesicles PE1% (vesicles containing Fl-CyA with PE 1%), and double-labeled vesicles (vesicles con-taining DiI and Alxhy with PE 1% and without PE).

Penetration Studies Using Rat Abdominal Skin

The effect of terpenes as penetration enhancers was also investigated using Fl-CyAvesicles with and without PE in rat skin. Figure 23.6 depicts the CLSM images of across-section of rat skin incubated with vesicles prepared with and without PE. Thevesicles were applied nonocclusively for 6 h. Both the vesicles formulations, with andwithout PE, showed restricted fluorescence to the SC only after 6 h of incubation time. Abright fluorescence was observed in the SC of the skin treated with Fl-CyA vesicles(Figure 23.6b), but negligible or no fluorescence was seen in the epidermis or dermis.The skin treated with Fl-CyA vesicles with 1% PE showed medium fluorescence in theSC and very weak fluorescence in the epidermis, suggesting the diffusion of the fluores-cence model from SC to the epidermis (Figure 23.6c). However, the ethanolic solutionof Fl-CyA showed weak fluorescence in the SC only and no fluorescence was seenin epidermis or dermis (Figure 23.6a). Results presented in Figure 23.6 indicate thatpenetration enhancers play an important role in the penetration of fluorescent labelsinto the skin.

Figure 23.6 CLSM images of a cross-section of rat skin incubated with different formulations.The vesicles were applied nonocclusively for 6 h on rat skin. (a) Ethanolic solution of fluores-cently labeled Cyclosporin (a and b) fluorescently labeled Cyclosporin A vesicles, and (c)fluorescently labeled Cyclosporin A vesicles containing penetration enhancers. The bar repre-sents 10 mm.




Penetration Studies Using Double-Labeled Vesicles

Skin penetration studies were also carried out using double-labeled vesicles containingDiI as a lipophilic marker and Alxhy as a hydrophilic marker. Figure 23.7(upper row andlower row) depicted the images of the human skin incubated with double-labeledvesicles with and without PE for 6 and 12 h, respectively.

In all the images the fluorescence was restricted mainly to the SC and, to a smalleror larger extent, to the epidermis. The hydro-alcoholic solution of the labels exhibitedfluorescence for both DiI and Alxhy in the SC only. However, a subsequent increase inintensities of fluorescence for both the labels was observed in the SC after 12 h ofincubation.

When the skin was incubated with double-labeled vesicles for 6 h, the labels were onlyobserved in the SC (Figure 23.7B, upper row). DiI showed a higher penetration in thedeeper SC layers, as compared to Alxhy, which was evenly distributed throughout the SC.A reddish-yellow color was seen in the deeper layers of SC, which represents a mixture ofboth the labels indicating that the vesicles may have penetrated into the deeper SC layersintact. There are published reports, which indicate that elastic vesicles can penetrate intothe deeper SC layers as intact vesicles.10 When these vesicles were incubated for 12 h,

A B C

A B C

Figure 23.7 CLSM images of a cross-section of human abdominal skin incubated with double-labeled vesicles and a hydro-alcoholic solution of DiI and Alexa Fluor 488. The vesicles wereapplied nonocclusively for 6 h (upper row) and 12h (lower row). (A) Hydroalcoholic solution ofDiI and Alexa Fluor 488, (B) double-labeled vesicles, and (C) double-labeled vesicles withpenetration enhancer. The bar represents 10 mm.




both the labels showed bright fluorescence throughout the SC, but in the epidermis a

cence was seen in the dermis.The vesicles containing PE showed a higher fluorescence in the SC when incubated

for 6 h, as compared to the hydro-alcoholic solution and vesicles without PE(Figure 23.7C, upper row). In the epidermis a very weak fluorescence was observedfor Alxhy and weak fluorescence for DiI, indicating higher penetration of the lipophilicmodel compound. After an incubation period of 12 h, a very bright fluorescence wasseen in the SC and medium fluorescence for both DiI and Alxhy was observed in theviable epidermis (Figure 23.7C, lower row). Surprisingly, the reddish fluorescence con-tinued from the epidermis towards the dermis, indicating a diffusion of the lipophilicmarker. Overall, vesicles containing PE have shown a comparatively very high fluores-cence both in epidermis and dermis for DiI, as well as for Alxhy, at 6 and 12 h, ascompared to the vesicles without PE.

It was observed that the vesicles containing terpenes as penetration enhancers wereable to deliver relatively higher fluorescent labels into the SC, epidermis and, to a smallextent into the dermis. Terpenes appeared to have an enhancing effect on penetration oflabels into the skin.

Tracking of Fluorescence Labels into Hair Follicles

Tracking of Fluorescently Labeled Cyclosporin A into the RatHair Follicles

Topical application of the liposomal-based formulation has been observed to result in asignificantly higher accumulation of CF in the pilosebaceous units than the applicationof any of the other nonliposomal formulation.70 The effect of terpenes on targeting hairfollicles was investigated using Fl-CyA vesicles, with and without PE, on rat skin (pene-

also depicts the role of the pilosebaceous unit in the penetration of the substance into theskin. Since the skin used was from a rat possessing a large number of hair follicles, abright fluorescence was observed in the pilosebaceous unit (bright fluorescent spots indermis in Figure 23.6b and c identified as the hair shaft, a part of the pilosebaceous unit)for both formulations, with and without PE. The fluorescence was also visualized in the

The images presented here clearlydemonstrate that the vesicles follow the pilosebaceous unit route to deliver their contentto the hair follicle and possibly to the hair bulb. The ethanolic solution of the Fl-CyAfailed to deliver any fluorescence into the skin by this route. These CLSM results weresupported by our in vivo studies with DEBR models69 and other published reports.70–73

The CLSM investigations enable us to conclude that PE plays an the important role inaccumulation of the substances into the hair follicles.

Accumulation of Liposomally Entrapped Rho-PE and CF into theHuman Hair Follicles

In this study, we encapsulated CF and Rho-PE in the liposomes as explained in the earliersections. We incubated these liposomes with the human skin on the Franz diffusion cells




very weak fluorescence was visualized (Figure 23.7B, lower row). However, no fluores-

tration enhancement effect of PE already explained in the earlier section). Figure 23.6

outer root sheath of the hair shaft (Figure 23.8a).

for 6 h. Figure 23.9 represents the accumulation of Rhodamine encapsulated in liposomesto the human hair follicles.

strated that the presence of the hair follicle plays a significant role in the skin penetrationof compounds.

The Efficacy of Top-RolT Dermaroller for Substance Penetrationinto the Skin

In this study, we demonstrated the efficacy of a novel type of device Dermaroller for thedelivery of the fluorescent model compound DiI, into human skin using CLSM. Three

Figure 23.8 CLSM and transmission image of a section incubated with fluorescently labeledCyclosporin A vesicles showing the role of pilosebaceous units in the penetration of the fluor-escent substance into the skin. The structure (originally in green) was identified as part ofpilosebaceous unit in rat skin. (a) CLSM image, (b) transmission image. The bar represents 10 mm.

Figure 23.9 Delivery of CF (left) and rhodamine (right) encapsulated in liposomes to the hairfollicles.




These images presented in Figure 23.6, Figure 23.8, and Figure 23.9 clearly demon-

types of Dermarollers were tested in this study, namely Dermaroller C8 0.13–158, M81.5–158, and M8 1.5–308. All Dermarollers were able to deposit the lipophilic fluorescentlabel, DiI, into the stratum corneum, epidermis and deeper skin layers. The liposomescontaining DiI were prepared as mentioned in the earlier section. The human skin wasincubated for 3 h nonocclusively after DiI liposomes and dermarollers pretreatment.

Figure 23.10 represents CLSM images of the skin cross-sections perpendicular to theskin surface. In all images, including control and different Dermarollers, a very highfluorescence was observed in the SC. This is obvious as the fluorescence label DiIis highly lipophilic and will be accumulated in the SC. The control formulation showeda higher deposition of the fluorescent label in the SC, followed by a weak fluorescence inthe viable epidermis. In the deeper layers of the skin, we observed only a very weakfluorescence.

The Dermaroller C8 0.13–158 was designed to improve the deposition of drugs in theSC. The application of this Dermaroller resulted in a bright fluorescence of the SC,followed by medium fluorescence in the viable epidermis and a weak fluorescence inthe deeper skin layers. As compared to the control, this Dermaroller showed a signifi-cantly enhanced fluorescence deposition both in the epidermis and in deeper skin layers.It can be concluded from Figure 23.10b that this Dermaroller may have penetrated the

Figure 23.10 CLSM images of a cross-section of human abdominal skin incubated on a Franzdiffusion cell with liposomes containing the lipophilic labels, DiI pretreated with Dermaroller.The liposomes were applied nonocclusively for 3 h. (a) Control; (b) Model C8 0.13–158;(c) Model M8 1.5–158; (d) Model M8 1.5–308.




layers of the SC and perhaps made some holes into the SC, through which the liposomeshad passed. The needle length was long enough to produce pores inside the SC, but wasnot able to generate pores throughout the entire SC, due to the length of the needles andthe physical state of the SC. However, as expected, this Dermaroller did not show the

The Dermaroller M8 1.5–158, which has the same number of needles, but a much largerneedle length (1.5 mm in comparison to 0.13 mm in the case of C8 0.13–158), was designedto deliver the drug in deeper layers of the skin by perforating the whole SC. This Dermar-

expected, this Dermaroller showed the highest deposition of the fluorescent substance inthe deeper skin layers. However, there seems to exist a continuous region beneath the SC,where there was weak to medium fluorescence. A possible explanation of these findingsmay be that the Dermaroller was able to push the liposomes, into the deeper layers duringperforation of the SC at the time of application. Therefore, the DiI liposomes, which hadpenetrated into the deeper layers at the time of the Dermaroller application, were able todiffuse further in the dermis. However, the liposomes, which remained after the Dermar-oller application on the surface, showed a maximum fluorescence in the SC and were ableto penetrate, to a small extent, through the holes made in SC by the Dermaroller, into thedeeper skin layers. The lateral diffusion of the fluorescent label in between the holes wasvery low. This might be the reason for the low fluorescence present in the continuousregion beneath the SC.

In the case of the Dermaroller M8 1.5–308, which possesses only half the number ofneedles as the other Dermarollers, a bright fluorescence was observed in the SC followedby a medium fluorescence in the epidermis and a bright fluorescence in the deeper skinlayers. As compared to the control, this Dermaroller showed an enhanced fluorescence inboth the epidermis and deeper skin layers. It can be concluded from Figure 23.10d thatthere was an area of weak fluorescence beneath the SC. This area of weak fluorescenceunder the SC was lower in the case of the Dermaroller M8 1.5–158 as compared to thisDermaroller. The reason for this large area of weak fluorescence results from the numberof needles and the angle position of the needles. However, the depth and intensity of thefluorescent label was at a maximum with this Dermaroller.

This study has shown that the fluorescent model compound can be visualized indeeper layers of the skin by making cross-sections perpendicular to the skin surface andthen visualizing them with CLSM.

Concluding Remarks

The use of CLSM allows scientists working in the field of topical delivery systems toobtain time and spatial information of the penetration process in relevant skin models.The method allows calculating time–depth-profiles of the penetrating agent, as far as it isfluorescently active. In the case of lipid carrier systems, it is quite feasible to add traceamounts of fluorescent lipids commercially available to the lipid carrier system, whichdo not seem to influence the behavior of the carrier system. Several examples of thesestudies, which have been presented above, demonstrate the usefulness of this method forbasic science and development of topical drug carrier systems.

New technical developments like the discussed miniaturized CLSM devices for in vivodetection of skin and penetrating substances will further increase the application range ofthis exciting technique and may become available in the medical practice.




oller showed a bright fluorescence in the SC, epidermis, and dermis (Figure 23.10c). As

highest deposition of the fluorescence label in the deeper skin layers.

References1. Cullander, C. and Guy, R.H., Visualization of iontophoretic pathway with confocal microscopy

and the vibrating probe electrode, Solid State Ionic, 53(56), 197, 1992.2. Alvarez-Roman, R., Naik, A., Kalia, Y.N., Fessi, H., and Guy, R.H., Visualization of skin

penetration using confocal laser scanning microscopy, Eur J Pharm Biopharm, 58(2), 301,2004.

3. Addicks, W.J., Flynn, G.L., and Weiner, N., Validation of a flow-through diffusion cell for use intransdermal research, Pharm Res, 4(4), 337, 1987.

4. Du Plessis, J., Ramachandran, C., Weiner, N., and Muller, D.G., The influence of particle size ofliposomes on the disposition of drug into the skin, Int J Pharm, 103, 277, 1994.

5. Kanerva, L., Electron microscopy of the effects of dithranol on healthy and on psoriatic skin,Am J Dermatopathol, 12(1), 51, 1990.

6. Hashimoto, K., Kagetsu, N., Taniguchi, Y., Weintraub, R., Chapman-Winokur, R.L., and Kasi-borski, A., Immunohistochemistry and electron microscopy in Langerhans cell histiocytosisconfined to the skin, J Am Acad Dermatol, 25(6 Pt 1), 1044, 1991.

7. Hofland, H.E., Bouwstra, J.A., Bodde, H.E., Spies, F., and Junginger, H.E., Interactions betweenliposomes and human stratum corneum in vitro: freeze fracture electron microscopical visu-alization and small angle x-ray scattering studies, Br J Dermatol, 132(6), 853, 1995.

8. van den Bergh, B.A., Vroom, J., Gerritsen, H., Junginger, H.E., and Bouwstra, J.A., Interactionsof elastic and rigid vesicles with human skin in vitro: electron microscopy and two-photonexcitation microscopy, Biochim Biophys Acta, 1461(1), 155, 1999.

9. Schreiner, V., Gooris, G.S., Pfeiffer, S., Lanzendorfer, G., Wenck, H., Diembeck, W., Proksch, E.,and Bouwstra, J., Barrier characteristics of different human skin types investigated with x-raydiffraction, lipid analysis, and electron microscopy imaging, J Invest Dermatol, 114(4), 654,2000.

10. Bouwstra, J.A. and Honeywell-Nguyen, P.L., Skin structure and mode of action of vesicles, AdvDrug Deliv Rev, 54(Suppl 1), S41, 2002.

11. Fang, J.Y., Hsu, L.R., Huang, Y.B., and Tsai, Y.H., Evaluation of transdermal iontophoresis ofenoxacin from polymer formulations: in vitro skin permeation and in vivo microdialysis usingWistar rat as an animal model, Int J Pharm, 180(2), 137, 1999.

12. Benfeldt, E., In vivo microdialysis for the investigation of drug levels in the dermis and theeffect of barrier perturbation on cutaneous drug penetration. studies in hairless rats and humansubjects, Acta Derm Venereol Suppl (Stockh), 206, 1, 1999.

13. Murakami, T., Yoshioka, M., Yumoto, R., Higashi, Y., Shigeki, S., Ikuta, Y., and Yata, N., Topicaldelivery of keloid therapeutic drug, tranilast, by combined use of oleic acid and propyleneglycol as a penetration enhancer: evaluation by skin microdialysis in rats, J Pharm Pharmacol,50(1), 49, 1998.

14. Schnetz, E. and Fartasch, M., Microdialysis for the evaluation of penetration through thehuman skin barrier — a promising tool for future research? Eur J Pharm Sci, 12(3), 165, 2001.

15. Yarosh, D., Bucana, C., Cox, P., Alas, L., Kibitel, J., and Kripke, K., Localization of liposomescontaining a DNA repair enzyme in murine skin, J Invest Dermat, 103, 461, 1994.

16. Kriwet, K. and Muller-Goymann, C., Diclofenac release from phospholipid drug systems andpermeation through excised human stratum corneum, Int J Pharm, 125(2), 231, 1995.

17. Veiro, J.A. and Cummins, P.G., Imaging of skin epidermis from various origins using confocallaser scanning microscopy, Dermatology, 189(1), 16, 1994.

18. Kirjavainen, M., Urtti, A., Jaaskelainen, I., Suhonen, T.M., Paronen, P., Valjakka-Koskela, R.,Kiesvaara, J., and Monkkonen, J., Interaction of liposomes with human skin in vitro — theinfluence of lipid composition and structure, Biochim Biophys Acta, 1304(3), 179, 1996.

19. Vardaxis, N.J., Brans, T.A., Boon, M.E., Kreis, R.W., and Marres, L.M., Confocal laser scanningmicroscopy of porcine skin: implications for human wound healing studies, J Anat, 190(Pt 4),601, 1997.




20. van Kuijk-Meuwissen, M.E., Mougin, L., Junginger, H.E., and Bouwstra, J.A., Applicationof vesicles to rat skin in vivo: a confocal laser scanning microscopy study, J Control Release,56(1–3), 189, 1998.

21. van Kuijk-Meuwissen, M.E., Junginger, H.E., and Bouwstra, J.A., Interactions between lipo-somes and human skin in vitro, a confocal laser scanning microscopy study, Biochim BiophysActa, 1371(1), 31, 1998.

22. Schatzlein, A. and Cevc, G., Non-uniform cellular packing of the stratum corneum and permea-bility barrier function of intact skin: a high-resolution confocal laser scanning microscopy studyusing highly deformable vesicles (Transfersomes), Br J Dermatol, 138(4), 583, 1998.

23. Betz, G., Imboden, R. and Imanidis, G., Interaction of liposome formulations with human skinin vitro, Int J Pharm, 229(1–2), 117, 2001.

24. Fink-Puches, R., Hofmann-Wellenhof, R., Smolle, J., and Kerl, H., Confocal laser scanningmicroscopy: a new optical microscopic technique for applications in pathology and derma-tology, J Cutan Pathol, 22(3), 252, 1995.

25. Zellmer, S., Pfeil, W., and Lasch, J., Interaction of phosphatidylcholine liposomes with thehuman stratum corneum, Biochim Biophys Acta, 1237(2), 176, 1995.

26. Simonetti, O., Hoogstraate, A.J., Bialik, W., Kempenaar, J.A., Schrijvers, A.H., Bodde, H.E., andPonec, M., Visualization of diffusion pathways across the stratum corneum of native and in-vitro-reconstructed epidermis by confocal laser scanning microscopy, Arch Dermatol Res,287(5), 465, 1995.

27. Boderke, P., Merkle, H.P., Cullander, C., Ponec, M., and Bodde, H.E., Localization of amino-peptidase activity in freshly excised human skin: direct visualization by confocal laser scanningmicroscopy, J Invest Dermatol, 108(1), 83, 1997.

28. Zellmer, S., Reissig, D., and Lasch, J., Reconstructed human skin as model for liposome–skininteraction, J Control Release, 55(2–3), 271, 1998.

29. Turner, N.G. and Guy, R.H., Visualization and quantitation of iontophoretic pathways usingconfocal microscopy, J Invest Dermatol Symp Proc, 3(2), 136, 1998.

30. Kirjavainen, M., Urtti, A., Monkkonen, J., and Hirvonen, J., Influence of lipids on the mannitolflux during transdermal iontophoresis in vitro, Eur J Pharm Sci, 10(2), 97, 2000.

31. Touitou, E., Godin, B., Dayan, N., Weiss, C., Piliponsky, A., and Levi-Schaffer, F., Intracellulardelivery mediated by an ethosomal carrier, Biomaterials, 22(22), 3053, 2001.

32. Grams, Y.Y., Alaruikka, S., Lashley, L., Caussin, J., Whitehead, L., and Bouwstra, J.A., Permeantlipophilicity and vehicle composition influence accumulation of dyes in hair follicles of humanskin, Eur J Pharm Sci, 18(5), 329, 2003.

33. Grams, Y.Y., Whitehead, L., Cornwell, P., and Bouwstra, J.A., Time and depth resolvedvisualisation of the diffusion of a lipophilic dye into the hair follicle of fresh unfixed humanscalp skin, J Contr Rel, 98(3), 367, 2004.

34. Alvarez-Roman, R., Naik, A., Kalia, Y.N., Guy, R.H., and Fessi, H., Skin penetration anddistribution of polymeric nanoparticles, J Control Release, 99(1), 53, 2004.

35. Mezei, M. and Gulasekharam, V., Liposomes — a selective drug delivery system for the topicalroute of administration. lotion dosage form, Life Sci, 26(18), 1473, 1980.

36. Mezei, M. and Gulasekharam, V., Liposomes — a selective drug delivery system for the topicalroute of administration: gel dosage form, J Pharm Pharmacol, 34(7), 473, 1982.

37. Touitou, E., Alhaique, F., Dayan, N., Riccieri, F., and Levi-Schaffer, F., Caffeine liposomalsystems for topical use: preparation, characterization, and skin permeation behaviour,J Pharm Sci, 83, 1189, 1994a.

38. Planas, M.E., Gonzalez, P., Rodriguez, L., Sanchez, S., and Cevc, G., Noninvasive percutaneousinduction of topical analgesia by a new type of drug carrier, and prolongation of local paininsensitivity by anesthetic liposomes, Anesth Analg, 75(4), 615, 1992.

39. Egbaria, K., Ramachandran, C., and Weiner, N., Topical application of liposomally entrappedcyclosporin evaluated by in vitro diffusion studies with human skin, Skin Pharmacol, 4(1), 21,1991.




40. Schafer-Korting, M., Korting, H.C., and Ponce-Poschl, E., Liposomal tretinoin for uncompli-cated acne vulgaris, Clin Invest, 72(12), 1086, 1994.

41. Korting, H.C., Zienicke, H., Schafer-Korting, M., and Braun-Falco, O., Liposome encapsulationimproves efficacy of betamethasone dipropionate in atopic eczema but not in psoriasisvulgaris, Eur J Clin Pharmacol, 29, 349, 1991.

42. Gesztes, A. and Mezei, M., Topical anesthesia of the skin by liposome-encapsulated tetracaine,Anesth Analg, 67(11), 1079, 1988.

43. Imbert, D. and Wickett, R.R., Topical delivery with liposomes, Cosmet Toil, 110, 32, 1995.44. Touitou, E., Dayan, N., Bergelson, L., Godin, B., and Eliaz, M., Ethosomes — novel vesicular

carriers for enhanced delivery: characterization and skin penetration properties, J ControlRelease, 65(3), 403, 2000.

45. Cevc, G., Schatzein, A., and Blume, G., Transdermal drug carriers, basic properties, optimiza-tion and transfer efficiency in the case of epicutaneously applied peptides, J Control Release,36, 3, 1995.

46. Cevc, G. and Blume, G., Lipid vesicles penetrate into intact skin owing to the transdermalosmotic gradients and hydration force, Biochim Biophys Acta, 1104(1), 226, 1992.

47. Hofland, H.E., van der Geest, R., Bodde, H.E., Junginger, H.E., and Bouwstra, J.A., Estradiolpermeation from nonionic surfactant vesicles through human stratum corneum in vitro,Pharm Res, 11(5), 659, 1994.

48. Pawley, J.B., Ed., Handbook of Biological Confocal Microscopy, Kluwer Academic Publishers,New York, 1995.

49. Sterenborg, J.J.C.M., Motamedi, M., Wagner, J.R.F., Duvic, M., Thomsen, S., and Jacques, S.L.,In vivo fluorescence spectroscopy and imaging of human skin tumours, Lasers Med Sci, 9, 191,1994.

50. Gillies, R., Zonios, G., Anderson, R.R., and Kollias, N., Fluorescence excitation spectroscopyprovides information about human skin in vivo, J Invest Dermatol, 115, 704, 2000.

51. Entwistle, A. and Noble, M., The quantification of fluorescent emission from biologicalsamples using analysis of polarization, J Microsc, 165, 347, 1992.

52. Corcuff, P., Bertrand, C., and Leveque, J.L., Morphometry of human epidermis in vivo by real-time confocal microscopy, Arch Dermatol Res, 285, 475, 1993.

53. Masters, B.R. and So, P.T., Confocal microscopy and multi-photon excitation microscopy ofhuman skin in vivo, Opt Express, 8, 2, 2001.

54. Aghassi, D., Anderson, R.R., and Gonzalez, S., Confocal laser microscopic imaging of actinickeratoses in vivo: a preliminary report, J Am Acad Dermatol, 43, 42, 2000.

55. Swindle, L.C., Thomas, S.G., Mfreeman, M., and Delaney, P.M., View of normal human skinin vivo as observed using fluorescent fiber-optic confocal microscopic imaging, J InvestDermatol, 121, 706, 2003.

56. Verma, D.D., Verma, S., Blume, G., and Fahr, A., Liposomes increase skin penetration ofentrapped and non-entrapped hydrophilic substances into human skin: a skin penetrationand confocal laser scanning microscopy study, Eur J Pharm Biopharm, 55(3), 271, 2003.

57. Verma, D.D., Verma, S., Blume, G., and Fahr, A., Particle size of liposomes influences dermaldelivery of substances into skin, Int J Pharm, 258(1–2), 141, 2003.

58. Jimbo, Y., Ishihara, M., Osamura, H., Takano, M., and Ohara, M., Influence of vehicles onpenetration through human epidermis of benzyl alcohol, isoeugenol and methyl isoeugenol,J Dermatol, 10(3), 241, 1983.

59. Loftsson, T., Somogyi, G., and Bodor, N., Effect of choline esters and oleic acid on thepenetration of acyclovir, estradiol, hydrocortisone, nitroglycerin, retinoic acid and trifluorothy-midine across hairless mouse skin in vitro, Acta Pharm Nord, 1(5), 279, 1989.

60. Tenjarla, S.N., Kasina, R., Puranajoti, P., Omar, M.S., and Harris, W.T., Synthesis and evaluation ofN-acetylprolinate esters — novel skin penetration enhancers, Int J Pharm, 192(2), 147, 1999.

61. Kobayashi, D., Matsuzawa, T., Sugibayashi, K., Morimoto, Y., and Kimura, M., Analysis of thecombined effect of 1-menthol and ethanol as skin permeation enhancers based on a two-layerskin model, Pharm Res, 11(1), 96, 1994.




62. Bhatia, K.S. and Singh, J., Effect of linolenic acid/ethanol or limonene/ethanol and iontophor-esis on the in vitro percutaneous absorption of LHRH and ultrastructure of human epidermis,Int J Pharm, 180(2), 235, 1999.

63. Dayan, N. and Touitou, E., Carriers for skin delivery of trihexyphenidyl HCl: ethosomes vs.liposomes, Biomaterials, 21(18), 1879, 2000.

64. Verma, D.D. and Fahr, A., Synergistic penetration enhancement effect of ethanol and phos-pholipids on the topical delivery of cyclosporin A, J Control Release, 97(1), 55, 2004.

65. MacDonald, R.C., MacDonald, R.I., Menco, B.P., Takeshita, K., Subbarao, N.K., and Hu, L.R.,Small-volume extrusion apparatus for preparation of large, unilamellar vesicles, BiochimBiophys Acta, 1061(2), 297, 1991.

66. Berner, B., Mazzenga, G.C., Otte, J.H., Steffens, R.J., Juang, R.H., and Ebert, C.D., Ethanol:water mutually enhanced transdermal therapeutic system II: skin permeation of ethanol andnitroglycerin, J Pharm Sci, 78(5), 402, 1989.

67. Kadir, R., Stempler, D., Liron, Z., and Cohen, S., Delivery of theophylline into excised humanskin from alkanoic acid solutions: a ‘push–pull’ mechanism, J Pharm Sci, 76, 774, 1987.

68. Panchagnula, R., Salve, P.S., Thomas, N.S., Jain, A.K., and Ramarao, P., Transdermal delivery ofnaloxone: effect of water, propylene glycol, ethanol and their binary combinations on perme-ation through rat skin, Int J Pharm, 219(1–2), 95, 2001.

69. Verma, D.D., Verma, S., McElwee, K.J., Freyschmidt-Paul, P., Hoffmann, R., and Fahr, A.,Treatment of alopecia areata in the DEBR model using cyclosporin A lipid vesicles, EurJ Dermatol, 14(5), 1, 2004.

70. Lieb, L.M., Ramachandran, C., Egbaria, K., and Weiner, N., Topical delivery enhancement withmultilamellar liposomes into pilosebaceous units: I. in vitro evaluation using fluorescenttechniques with the hamster ear model, J Invest Dermatol, 99(1), 108, 1992.

71. Niemiec, S.M., Ramachandran, C., and Weiner, N., Influence of nonionic liposomal compos-ition on topical delivery of peptide drugs into pilosebaceous units: an in vivo study using thehamster ear model, Pharm Res, 12(8), 1184, 1995.

72. Bohm, M. and Luger, T.A., The pilosebaceous unit is part of the skin immune system,Dermatology, 196(1), 75, 1998.

73. Agarwal, R., Katare, O.P., and Vyas, S.P., The pilosebaceous unit: a pivotal route for topicaldrug delivery, Methods Find Exp Clin Pharmacol, 22(2), 129, 2000.




THE RETARDATION

OF PERCUTANEOUS

PENETRATION

V



Chapter 24

Fundamentals of RetardingPenetration

Jonathan Hadgraft and Barrie C. Finnin

CONTENTS

Introduction .................................................................................................................................... 361Nature and Structure of Skin.......................................................................................................... 362Physicochemical Determinants ...................................................................................................... 363Reducing Skin Absorption — Possibilities .................................................................................... 364Effects of Chemical Structure ......................................................................................................... 364Formulation Effects......................................................................................................................... 367Conclusions .................................................................................................................................... 368References....................................................................................................................................... 369

Introduction

The skin is constantly subjected to a barrage of chemicals and generally it acts as anextremely effective barrier. It prevents the ingress of xenobiotics and also stops us fromlosing excessive water. The reasons for this excellent barrier will be reviewed since ifthese are understood it is possible to identify how the permeability of the barrier can bealtered. A consideration of the extensive literature on the subject reveals that mostresearch has been conducted on ways of reducing the barrier properties of the stratumcorneum. This is to permit better access of therapeutic agents both for local (topical)effect and for systemic (transdermal) effect. Formulation approaches are often employedin which penetration enhancers are used. Publications on retarding penetration are farfrom common. Perhaps this is an oversight since there are a number of reasons why it isdesirable to retard permeation.

There are materials that have their effect on the skin surface and deeper penetrationcan cause problems. Examples of these include insect repellents and UV filters that areincorporated into sunscreens. Both of these products are used repeatedly over large body


361


surface areas and, in principle, this could give rise to an unwanted systemic burden.There have been reports in the press about some UV filters mimicking steroid hormones[1] and also publications which have shown that the insect repellent diethyl m-toluamide(DEET) has neurotoxicty issues in young children [2].

Potent insecticides are formulated into concentrates before use in, for example, cropspraying. These are handled and if the concentrate inadvertently comes into contact withthe skin, the active can be absorbed. It would be sensible to produce formulations thatare effective when diluted but have reduced dermal penetration potential both in theirconcentrated and diluted forms. There have been concerns about the use of organo-phosphates used in environmental situations such as spraying [3] and in animal dips [4].

There are household products that contain antiseptics, cleaning agents, etc. It wouldbe beneficial if these could retain their activity but have minimum dermal absorptionpotential. With the array of these products that are available it is perhaps surprising thatlittle has been written about the methodologies of retarding skin permeation. Thepurpose of this chapter is to examine the different possibilities that are available from afundamental context.

Nature and Structure of Skin

Before reviewing the different possibilities it is instructive to consider the nature andstructure of the skin. This is a very brief resume of the salient features that are relevant tounderstanding the mechanisms of permeation retardation.

The skin comprises three major layers, the outermost layer, the stratum corneumforms the major permeability barrier. The structure of this very thin membrane (~15 mm)is therefore the most important. The viable tissue is immediately underneath the stratumcorneum. It is metabolically active and can be a barrier for molecules that are extremelylipophilic in nature (log P [octanol water partition coefficient]>~4). This is because it islargely aqueous in nature and can be considered, in diffusional terms, as being similar toan aqueous protein gel. It is approximately 200 mm thick. The dermis is found under-neath the viable epidermis. It is approximately 2 mm in thickness and contains blood andlymph vessels. Once a molecule has reached this region, it is rapidly taken up by thecapillary network and is therefore systemically available.

The principal barrier to penetration is the stratum corneum and the reasons for itsexceptionally efficient barrier properties have been the subject of extensive research. Thelayer comprises dense overlapping dead cells, the corneocytes. The largest component ofthese cells is keratin. They are held together by ‘‘rivets’’ the corneodesmosomes and anintercellular matrix which contains a complex mixture of lipids. Unlike most biologicalmembranes, there are no phospholipids and the predominant lipid class is the ceramides.These structure themselves into bilayer arrays, the molecular dimensions of which can beseen by x-ray scattering [5] and, more recently, cryomicroscopy [6]. Although the stratumcorneum is interspersed with appendages, hair follicles (and associated sebaceousglands) and eccrine glands these are thought to play a fairly minor, if any, role inpercutaneous penetration [7].

The reasons for the barrier function of the skin are therefore:

1. A small area for diffusion, the intercellular channels, is available. If a permeant issolvent deposited and crystallizes in the center of a corneocyte it could beunavailable for penetration [8].




2. The intercellular route is tortuous. It has been estimated that the path length fordiffusion is 300 to 500 mm rather than the straight thickness of the stratumcorneum of 15 mm [9, 10].

3. The permeant has to cross, sequentially a number of structured bilayers, ittherefore has to cross both hydrophilic and lipophilic domains [7].

4. The lipid headgroups are closely packed and the methylene groups immediatelyadjacent to the headgroups are rather rigid providing a high microviscosity region[7, 11].

Given the above it is not surprising that it is difficult for molecules to penetratethe skin and most research has concentrated on improving penetration rather thanretarding it.

The bioavailability from many topical products is very low and it is not unusual to findexperimental bioavailabilities in the region of a few percent [12].

Physicochemical Determinants

It is also clear from the above section that the physicochemical properties of thepermeant and the formulation into which it is incorporated will have a significantinfluence on the rate of penetration. The membrane is complex and diffusion involvestransport through a heterogeneous environment. Despite this, it is possible to conductdiffusion experiments and the results appear to conform to Fick’s laws of diffusion. Thesimplest form of this is the first law which is used for steady-state diffusion. The steady-state flux ( J ) is given by

J ¼ DA @c=@X (24:1)

where D is the diffusion coefficient, A the area, and @c/@x the concentration gradient.If a solution of the drug is placed on the skin surface (of concentration, capp)

the concentration in the outer layers of the stratum corneum will be dictated by thepartition coefficient (K ) of the permeant between the stratum corneum and the appliedvehicle and

J ¼ DAK(capp cinner)=h (24:2)

where cinner is the concentration of the permeant in the lower layer of the stratumcorneum. In general, cinner < capp and Equation (24.2) reduces to

J ¼ DAK capp=h (24:3)

or

J ¼ A kp capp (24:4)

kp is the permeability coefficient which is a heterogeneous rate constant having unitsoften expressed in cm/h.

The maximum flux will be when a saturated solution is applied and capp is thesolubility limit of the permeant.


Retarding Penetration & 363


The above mathematical description shows that there are several physicochemicaldeterminants that are important in skin permeation, partition behavior, diffusional char-acteristics, and solubility.

The expression in Equation (24.1) is an approximation and more correctly the drivingforce for diffusion is the chemical potential (m). This means that concentration should bereplaced by activity (a). This can have significant effects particularly at higher concen-trations. The activity of a saturated solution is, by definition, 1. If saturated solutions areplaced on the skin and the solvents in which they are dissolved have no effect on thebarrier properties of the skin the fluxes should be the same [13]. This has been demon-strated for a number of systems and it is possible to choose a series of solvents in whichthe permeant has solubilities which vary by orders of magnitude but the flux is the same[14]. The choice of formulation can clearly have a profound impact on flux which will beseen in later discussions.

Reducing Skin Absorption — Possibilities

Perhaps the most obvious way of reducing skin absorption is to prevent the substancefrom actually reaching the skin. The use of barrier creams to protect the skin againstendogenous toxins is a long established practice. However, some recent studies [15, 16]have found no difference between the protective effects of conventional ‘‘barrier’’ creamsand their bases. More recently creams containing perfluorinated compounds have beenshown to decrease the flux of sulfur mustard across human skin in vitro [17] and rabbitskin in vivo [18]. The occupational use of these skin-protection creams has been reviewedrecently by Kresken and Klotz [19].

Since the skin is stratified it may be possible to reduce the absorption in differentregions and there may be alternative avenues available for achieving this. For example, inthe case of insect repellents it is important for the active to remain on the surface and notdiffuse too far into the stratum corneum. This would involve the choice of active and orformulation that left the material surface associated and less liable to permeate deeper.

There are actives, such as antifungals, which have their site of action in the stratumcorneum. It may be possible to select compounds which build up in the outer layer of theskin and because of their properties do not permeate significantly into the viable layers.

When the permeant partitions into the viable tissue it diffuses relatively rapidlycompared with its transit through the stratum corneum. It is not clear whether or notthere are other transport mechanisms that take place in this region such as fluid flow,perhaps as a result of lymph drainage. Additionally it should be remembered that thisregion is metabolically active and therefore materials can be broken down either to activeor inactive metabolites [20–23]. At the basal layer of the viable epidermis the permeantencounters the network of vessels transporting the blood. Rapid uptake is possible andthe material will become systemically active. Vasodilators are unlikely to affect the speedof uptake but vasoconstrictors can reduce it. This has obvious implications with regard tosystemic load and the concentration found in the region of the blood vessel network [24].

Effects of Chemical Structure

The chemical structure will affect simple factors such as the partition and solubilitycharacteristics. It is easier to identify those compounds which will permeate the skin




the best. These will be comparatively small in size, that is, they will diffuse relativelyrapidly. They will have a log P value in the region of 2 to 3 [7]. This will ensure that theywill be able to partition reasonably well between the hydrophilic and lipophilic domains.Finally they should have reasonable solubility in both oils and water [25]. Again this is areflection of the heterogeneous nature of the skin.

Extremely hydrophilic compounds will be unable to partition into the outer layers ofthe stratum corneum, and will therefore be poor permeants, their permeation will beretarded into the skin and hence through it. On the other hand extremely lipophiliccompounds will be able to partition into the skin lipids. They will experience problemsin crossing the aqueous headgroup regions of the skin lipids and they will certainly haveextreme difficulty partitioning out of the stratum corneum into the underlying viableepidermis. Their penetration into the stratum corneum will be moderate, depending ontheir solubility properties, but deeper penetration into the viable epidermis, dermis, andalso systemic circulation will be limited.

There is often confusion in the literature on the subject of partition effects. A simplealgorithm [26] shows how the permeability coefficient (from an aqueous solution) can bepredicted from log P and molecular mass (MW)

log kp(cm=h) ¼ 2:7þ 0:71 log p 0:0061 MW (24:5)

It is clear that the permeability coefficient increases with an increase in partitioncoefficient. However the flux, which is important in evaluating permeation rate, isdependent on both the permeability coefficient and the concentration in the appliedsolution (Equation [24.4]). The maximum flux is when the solution is saturated and forcompounds with high log P the aqueous solubility is very low. There is therefore abalance between a high log P leading to high permeability and high log P causing lowwater solubility. This is probably the reason that compounds with a log P ~ 2 have thebest permeation characteristics (but not the highest permeability coefficients).

The balance between log P and solubility is an important issue. The best permeantsare those with moderate log P and with good solubility in both oils and water. Thesolubility is also related, through intermolecular forces, to the melting point [27]. Com-pounds with low melting points tend to permeate better. Two of the best skin permeantsare nicotine and nitroglycerin, these two appear to possess optimum log P and solubilityproperties and both are at low melting points being oils at room temperature.

Therefore compounds with properties that have low permeation characteristicsare likely to have extreme values of log P, high melting point and be large. Highmelting point materials tend to be salts and many salts do have poor permeationcharacteristics. But again there are some misconceptions about permeation of ionizedcompounds through the skin. When an ionized compound is put on the skin there willbe a mixture of the ionized and unionized species depending on the pH of the localenvironment on the skin surface and the pKa of the permeant. The overall perme-ation rate will depend on the concentration of the two species and their permeabilitycoefficients

J total ¼ A(Kp(union)c(union) þ Kp(ion)c(ion)) (24:6)

Considering an aqueous solution and partition characteristics coupled with solubilityproperties the following observations can be made. When ionization is suppressedthe permeability coefficient will be high and the solubility low. When ionization is




high, permeability coefficient will be low but solubility will be elevated. Few studies havebeen conducted where saturated solutions of the same compound at different ionizationstates have been examined. In the case of ibuprofen it appears that there is a better fluxfrom a saturated solution of the fully ionized species than for a saturated solution of thefree acid [28]. It is true that at equal concentrations the free acid or free base will have ahigher flux than the ionized conjugate acid or base.

The issue of ionization effects is further complicated by ion-pair effects whereenhanced permeation can occur as a result of ion-pairing [29]. In addition it is possiblefor higher intermolecular forces to occur between molecules of the free base comparedwith its salt form. If this occurs it is possible for the free base to have a lower solubilitythan the salt form in a lipophilic solvent which will influence the permeation rate.

Structure–activity relationships (SAR) have been examined for skin permeation anda number of interesting factors can be identified that demonstrate which structuralfeatures can retard permeation. For example, the number of hydrogen bonding groupson a permeant affects its ability to diffuse [30, 31]. Addition of 1 to 3 hydrogen bondinggroups appears to decrease the diffusion coefficient by an order of magnitude pergroup. Thereafter (from 3 to 6) the effect appears to have reached a plateau. Therefore,addition of hydrogen bonding groups to a parent molecule appears to retard permea-tion. More subtle effects can also be seen and the degree of retardation isCOOH>CH2OH>phenolic OH>CH2–CO–CH2>CH2–O–CH2. It is thought thatthese effects are because of the interaction of the permeant with the polar headgroupsof the stratum corneum lipids.

Penetration enhancers have been identified with different structural properties [32],those that disrupt the packing of the structured lipids of the skin tend to possess polarhead groups and long alkyl chains. These intercalate into the ordered lipids which is howthey produce their effect. As they interact with the lipids it can also be anticipated thatthey will diffuse relatively slowly and be less likely to partition out of the stratumcorneum into the underlying tissue. This type of molecule would also be expected tohave retarded permeation.

There has been some discussion in the literature about skin binding but whether ornot there is genuine chemical binding in the stratum corneum is a matter of debate. Someexperiments that try to confirm binding involve the use of powdered keratin or pow-dered stratum corneum [33]. There is undoubtedly binding potential for many com-pounds with powdered keratin but a molecule diffusing through the skin does notencounter keratin in this form. There is a covalently bonded lipid envelope around thecorneocytes which would shield the permeant from direct contact with keratin. It ispossible that the permeant does ‘‘see’’ keratin on the skin surface and surface associationcan occur. If specific binding on the skin surface does happen, retarded permeationwould result.

It is also possible for material to be deposited in the center of the corneocytes on theskin surface. This would occur for example from solvent deposition. If the intercellularchannels are the predominant route material that crystallizes in the center of the deadcells, they would have a problem diffusing to the ‘‘active’’ intercellular space and wouldessentially be unavailable for deeper penetration.

Large chemical entities have difficulty diffusing through the skin and it is also possibleto retard the permeation of a specific agent, such as a UV filter by chemical modification.If the filter is bound to a polymer, which cannot diffuse through the stratum corneum, itsuptake into the deeper tissues and blood stream will be retarded. It is also possible toassociate the UV filter with nanoparticles or encapsulate it [34, 35].




Formulation Effects

Often the active cannot be selected on the basis of its chemical structure and may notlend itself to simple chemical modification. In this case the only strategy that can beconsidered is the formulation. It is well established that formulation can have a significanteffect on skin penetration and there are simple strategies that can be adopted to reduceskin permeation.

The driving force for diffusion is the chemical potential of the diffusant. If this canbe reduced permeation will also be attenuated. A simple example of this effect is thetreatment for washing phenol from the skin surface. If phenol comes into contact with theskin it should not be washed off with water. Instead glycerol should be used. The reasonfor this is that phenol has a much greater affinity for glycerol than water and therefore alower chemical potential in glycerol. From water the phenol will be ‘‘encouraged’’ todiffuse into the skin, whereas from glycerol the reverse is true.

In physicochemical terms, Equation (24.3) should more correctly be written

J ¼ DA a=g h (24:7)

where ‘‘a’’ is the thermodynamic activity of the diffusant in the formulation and g is theeffective activity coefficient of the diffusant in the stratum corneum. It should be remem-bered that a saturated solution (capable of delivering maximum flux) has an activity of1. If the solvent has no effect on the skin, all saturated solutions, irrespective of thesolubility, should all provide the same flux [13].

In the phenol case above, the reduced flux J from glycerol is a result of ‘a’ beingsmaller. It is also possible for formulation excipients to enter into the stratum corneumand alter g. The term a/g is essentially the same as Kc, the product of the partitioncoefficient and the applied concentration. The flux will be lower for a reduced parti-tion coefficient and at lower applied concentrations.

Thermodynamic effects have been examined in the past but in general this is forenhancing permeation and particular attention has been paid to supersaturated stateswhere the driving force is very high [36–40]. However, these formulations are inherentlyunstable and often contain antinucleant polymers to help stability [41, 42]. Similarly if adrug exists with a number of polymorphic forms, the one with the least energy can give areduced flux from a saturated solution [43]. Stability issues need also to be consideredin the choice of polymorphic forms. However, the most stable form is likely to have thelowest permeability.

Excipients and the active can diffuse into the skin at the same time. The presence ofsolvent can influence the activity state of the drug in the stratum corneum lipids, basicallyaltering the solubility properties of the skin. Attempts have been made to relate theingress of solvent to the solubility parameter and enhanced permeation is thought tooccur when the solubility parameter of the drug is similar to the solubility parameter ofthe skin lipids [44] (plus any influence from the solvent). However, this has been difficultto verify. The solubility parameter of the skin is estimated to be about 10 (cal/cm3)1/2.This could be shifted in the direction of the drug if an appropriate solvent is used, thiswould enhance absorption and alternatively if a solvent moves the solubility parameter ofthe skin lipids away from that of the active, permeation would be retarded.

Excipient molecules that have a polar head group and a long hydrophobic chain (orchains) can penetrate into the intercellular lipids and intercalate with the endogenousceramides. Usually this creates disruption of the packing of the ceramides and diffusion of




the active is increased. Molecules such as Azone1 act in this way [45]. Oleic acid alsointercalates but its phase separates and enhanced permeation is a result of transfer throughthe more fluid pools of oleic acid or at the interfacial defects [46]. Analogs of Azone havealso been identified as permeation enhancers but N-0915, whose structure is provided inFigure 24.1 where it is compared with Azone has been shown to decrease the permeationof metronidazole and the insect repellent, DEET [47]. Physicochemical studies coupledwith molecular graphics suggest that N-0915 intercalates into the skin lipids and ‘‘pulls’’ thepolar headgroups together. This makes the adjacent methylene groups pack closer to-gether and the microviscosity of this region is increased, the diffusion of any permeant istherefore reduced. The partial charges on the opposite oxygen atoms in N-0915 areimplicated in the headgroup attractions and it may be possible to design specific agentsthat will retard permeation by increasing the viscosity of the endogenous lipids.

Synergistic effects can also be anticipated and have been observed in the case ofpenetration enhancement. For example if an enhancer can be shown to increase D andalso increase the concentration of the diffusant in the outer layers of the skin the effectswill be multiplicative (Equation [24.3]). Similarly a judicious formulation approach couldreduce permeation by decreasing D and reducing K, the effects also being multiplicative.

Conclusions

This is an important subject that has received little attention. However, it is clear that itshould be possible to make significant steps in the reduction of dermal absorption if thisis required for safety reasons. These include:

Figure 24.1 The structures of Azone and N-0915 shown in two- and three-dimensions.




1. The specific design of actives so that they have physicochemical properties whichdecrease permeation

2. Chemical modification of the permeant, for example, bonding to a polymer to stopdermal uptake

3. Modification of the formulation to:(a) Reduce the thermodynamic activity of the permeant(b) Reduce the solubility of the permeant in the skin lipids(c) Increase the microviscosity of the skin lipids

With the increased awareness of the problems of systemic exposure, these strategieswill become commonplace in the future design pharmaceuticals, cosmetics, and agro-chemical formulations.

References1. M. Schlumpf, H. Jarry, W. Wuttke, R. Ma, and W. Lichtensteiger. Estrogenic activity and

estrogen receptor [beta] binding of the UV filter 3-benzylidene camphor: comparison with4-methylbenzylidene camphor. Toxicology 199: 109–120 (2004).

2. G. Briassoulis, M. Narlioglou, and T. Hatzis. Toxic encephalopathy associated with use ofDEET insect repellents: a case analysis of its toxicity in children. Human and ExperimentalToxicology 20: 8–14 (2001).

3. E.L. Baker, M. Zack, J.W. Miles, L. Alderman, M. Warren, R.D. Dobbin, S. Miller, and W.R.Teeters. Epidemic malathion poisoning in Pakistan malaria workers. The Lancet 311: 31–34(1978).

4. A.E. Watterson. Regulating pesticides in the UK: a case study of risk management problemsrelating to the organophosphate diazinon. Toxicology Letters 107: 241–248 (1999).

5. J.A. Bouwstra, G.S. Gooris, D. Downing, and W. Bras. The sructure of pig stratum-corneum —a wide and small-angle x-ray study. Biophysical Journal 64: A295–A295 (1993).

6. A. Al-Amoudi, L.P.O. Norlen, and J. Dubochet. Cryo-electron microscopy of vitreous sectionsof native biological cells and tissues. Journal of Structural Biology. In press.

7. J. Hadgraft. Skin, the final frontier. International Journal of Pharmaceutics 224: 1–18 (2001).8. K.D. McCarley and A.L. Bunge. Absorption into silicone rubber membranes from powders and

aqueous solutions. International Journal of Pharmaceutics 250: 169–180 (2003).9. W.J. Albery and J. Hadgraft. Percutaneous absorption: in vivo experiments. Journal of Phar-

macy and Pharmacology 31: 140–147 (1979).10. R.O. Potts and M.L. Francoeur. The influence of stratum-corneum morphology on water

permeability. Journal of Investigative Dermatology 96: 495–499 (1991).11. C.L. Gay, T.M. Murphy, J. Hadgraft, I.W. Kellaway, J.C. Evans, and C.C. Rowlands. An electron

spin resonance study of skin penetration enhancers. International Journal of Pharmaceutics49: 39–45 (1989).

12. R. J. Feldman and H. I. Maibach. Regional variation in percutaneous penetration of 14C cortisolin man. Journal of Investigative Dermatology 48: 181–183 (1967).

13. T. Higuchi. Physical chemical analysis of percutaneous absorption process from creams andointments. Journal of the Society of Cosmetic Chemists 11: 85–97 (1960).

14. J. N. Twist and J. L. Zatz. Influence of solvents on paraben permeation through idealized skinmodel membranes. Journal of the Society of Cosmetic Chemists 37: 429–444 (1986).

15. U. Berndt, W. Wigger-Alberti, B. Gabard, and P. Elsner. Efficacy of a barrier cream and itsvehicle as protective measures against occupational irritant contact dermatitis. Contact Derma-titis 42: 77–80 (2000).

16. D. Perrenoud, D. Gallezot, and G. van Melle. The efficacy of a protective cream in a real-worldapprentice hairdresser environment. Contact Dermatitis 45: 134–138 (2001).




17. R. P. Chilcott, J. Jenner, S. A. Hotchkiss, and P. Rice. Evaluation of barrier creams against sulfurmustard. I. In vitro studies using human skin. Skin Pharmacology and Applied Skin Physiology15: 225–235 (2002).

18. D. K. Liu, R. W. Wannemacher, T.H. Snider, and T. L. Hayes. Efficacy of the topical skinprotectant in advanced development. Journal of Applied Toxicology 19: S41–S45 (1999).

19. J. Kresken and A. Klotz. Occupational skin-protection products — a review. InternationalArchives of Occupational and Environmental Health 76: 355–358 (2003).

20. I. Steinstrasser and H. P. Merkle. Dermal metabolism of topically applied drugs: pathways andmodels reconsidered. Pharmaceutica Acta Helvetiae 70: 3–24 (1995).

21. S. Selim, J. Hartnagel, E. Ralph, T. G. Osimitz, K. L. Gabriel, and G. P. Schoenig. Absorption,metabolism, and excretion of N,N-diethyl-m-toluamide following dermal application to humanvolunteers. Fundamental and Applied Toxicology 25: 95–100 (1995).

22. R. U. Pendlington, D. L. Williams, J. T. Naik, and R. K. Sharma. Distribution of xenobioticmetabolizing enzymes in skin. Toxicology in Vitro 8: 525–527 (1994).

23. R. J. Martin, S. P. Denyer, and J. Hadgraft. Skin metabolism of topically applied compounds.International Journal of Pharmaceutics 39: 23–32 (1987).

24. K. D. McCarley and A. L. Bunge. Pharmacokinetic models of dermal absorption. Journal ofPharmaceutical Sciences 90: 1699–1719 (2001).

25. J. W. Hadgraft and G. F. Somers. Percutaneous absorption. Journal of Pharmacy and Pharma-cology 8: 625–634 (1956).

26. R. O. Potts and R. H. Guy. Predicting skin permeability. Pharmaceutical Research 9: 663–669(1992).

27. S. H. Yalkowsky and S. C. Valvani. Solubility and partitioning I: solubility of nonelectrolytes inwater. Journal of Pharmaceutical Sciences 69(8): 912–922 (1980).

28. J. Hadgraft and C. Valenta. pH, pK(a), and dermal delivery. International Journal of Pharma-ceutics 200: 243–247 (2000).

29. C. Valenta, U. Siman, M. Kratzel, and J. Hadgraft. The dermal delivery of lignocaine: influenceof ion pairing. International Journal of Pharmaceutics 197: 77–85 (2000).

30. W. J. Pugh, M. S. Roberts, and J. Hadgraft. Epidermal permeability — penetrant struc-ture relationships. 3. The effect of hydrogen bonding interactions and molecular size ondiffusion across the stratum corneum. International Journal of Pharmaceutics 138: 149–165(1996).

31. M. S. Roberts, W. J. Pugh, and J. Hadgraft. Epidermal permeability: penetrant structurerelationships. 2. The effect of H-bonding groups in penetrants on their diffusion through thestratum corneum. International Journal of Pharmaceutics 132: 23–32 (1996).

32. B. C. Finnin and T. M. Morgan. Transdermal penetration enhancers: applications, limitations,and potential. Journal of Pharmaceutical Sciences 88: 955–958 (1999).

33. R. C. Wester, M. Mobayen, and H. I. Maibach. In vivo and in vitro absorption and binding topowdered stratum corneum as methods to evaluate skin absorption of environmental chem-ical contaminants from ground and surface water. Journal of Toxicology and EnvironmentalHealth 21: 367–374 (1987).

34. P. Perugini, S. Simeoni, S. Scalia, I. Genta, T. Modena, B. Conti, and F. Pavanetto. Effectof nanoparticle encapsulation on the photostability of the sunscreen agent, 2-ethylhexyl-p-methoxycinnamate. International Journal of Pharmaceutics 246: 37–45 (2002).

35. R. Alvarez-Roman, G. Barre, R. H. Guy, and H. Fessi. Biodegradable polymer nanocapsulescontaining a sunscreen agent: preparation and photoprotection. European Journal of Pharma-ceutics and Biopharmaceutics 52: 191–195 (2001).

36. A. F. Davis and J. Hadgraft. Supersaturated solutions as topical drug delivery systems. In K. A.Walters and J. Hadgraft (eds), Pharmaceutical Skin Penetration Enhancement, Marcel Dekker,New York, 1993, pp. 243–267.

37. M. A. Pellett, A. F. Davis, and J. Hadgraft. Effect of supersaturation on membrane transport 2.Piroxicam. International Journal of Pharmaceutics 111: 1–6 (1994).




38. N. A. Megrab, A. C. Williams, and B. W. Barry. Oestradiol permeation across human skin,silastic and snake skin membranes — the effects of ethanol water co-solvent systems. Inter-national Journal of Pharmaceutics 116: 101–112 (1995).

39. K. Moser, K. Kriwet, C. Froehlich, Y. N. Kalia, and R. H. Guy. Supersaturation: enhancement ofskin penetration and permeation of a lipophilic drug. Pharmaceutical Research 18: 1006–1011(2001).

40. K. Moser, K. Kriwet, Y. N. Kalia, and R. H. Guy. Stabilization of supersaturated solutions of alipophilic drug for dermal delivery. International Journal of Pharmaceutics 224: 169–176(2001).

41. S. L. Raghavan, A. Trividic, A. F. Davis, and J. Hadgraft. Crystallization of hydrocortisoneacetate: influence of polymers. International Journal of Pharmaceutics 212: 213–221 (2001).

42. S. L. Raghavan, B. Kiepfer, A. F. Davis, S. G. Kazarian, and J. Hadgraft. Membrane transportof hydrocortisone acetate from supersaturated solutions; the role of polymers. Inter-national Journal of Pharmaceutics 221: 95–105 (2001).

43. B. W. Barry. Dermatological formulations: percutaneous absorption. In J. Swarbrick (ed.),Drugs and the Pharmaceutical Sciences, Vol. 18, Marcel Dekker, New York and Basel, 1983,p. 480

44. Z. Liron and S. Cohen. Percutaneous absorption of alkanoic acids II: application of regularsolution theory. Journal of Pharmaceutical Sciences 73: 538–542 (1984).

45. J. C. Beastall, J. Hadgraft, and C. Washington. Mechanism of action of azone as a percutan-eous penetration enhancer: lipid bilayer fluidity and transition temperature effects. Inter-national Journal of Pharmaceutics 43: 207–213 (1988).

46. B. Ongpipattanakul, R. R. Burnette, R. O. Potts, and M. L. Francoeur. Evidence that oleic acidexists in a separate phase within stratum corneum lipids. Pharmaceutical Research 8: 350–354(1991).

47. J. Hadgraft, J. Peck, D. G. Williams, W. J. Pugh, and G. Allan. Mechanisms of action of skinpenetration enhancers retarders: Azone and analogues. International Journal of Pharmaceut-ics 141: 17–25 (1996).




Chapter 25

Retardation Strategiesfor Sunscreen Agents

Carryn H. Purdon, Eric W. Smith, and Christian Surber

CONTENTS

Introduction .................................................................................................................................... 373Cyclodextrins and Photostability ................................................................................................... 374Transcutol1 .................................................................................................................................... 375Encapsulation Structures ................................................................................................................ 375Physical Properties of Organic Particulate UV-Absorbers ............................................................ 377Inorganic Materials ......................................................................................................................... 377Penetration Retarders ..................................................................................................................... 378Vehicle Effects................................................................................................................................. 379Conclusions .................................................................................................................................... 379References....................................................................................................................................... 380

Introduction

There is overwhelming evidence indicating that human skin is damaged in different waysby exposure to sunlight. Of the solar radiation reaching the earth’s surface, the ultraviolet(UV) component (290 to 400 nm) is a major factor leading to skin pathologies that rangein severity from inflammatory responses, cutaneous photoageing, dendritic keratitis toskin cancer [1–3]. The expanding knowledge of the deleterious effects of sunlight haspromoted the widespread use of topical sunscreen preparations [4, 5], which containchemicals that absorb, reflect, or scatter UV radiation [6] and are thereby highly effectiveskin protectants. Organic sunscreen agents are compounds that decrease the intensity ofUV light reaching the epidermal strata by absorbing the radiation (typical electronpromotion from a lower- to a higher-energy molecular orbital). The activated sunscreenmolecule dissipates the excess energy in the form of heat, by fluorescence, phosphor-escence, interaction with neighboring molecules or by undergoing photochemical


373


modifications [7]. Particulate sunscreens present a physical barrier between the incidentradiation and the epidermis, scattering or reflecting the radiation. However, to be effect-ive, these agents must remain on or in the outermost layers of the skin, the stratumcorneum (SC). One major drawback of current sunscreen formulations is that they areconstantly lost from the skin surface by abrasion with clothing, sweating, or swimming;requiring frequent reapplication for continued effectiveness. Moreover, several of thechemical sunscreens currently on the market exhibit intense irritancy and sensitizationreactions after absorption into the dermal strata in predisposed individuals, often causingsevere immunological problems [8–10]. Even if toxic manifestations are not evident inusers, this penetration of the sunscreen represents a loss from the surface with asubsequent reduction in sun protection effectiveness. A significant improvement insunscreen technology would be the development of a system that retards the penetrationof the chemical into the skin and binds the agent in the SC so that minimal loss occurs bydiffusion, abrasion, or moisture. The degree of sunscreen penetration depends stronglyon the physicochemical properties of the active compound and of the nature of thevehicle in which the sunscreen is applied, that is, polarity of the solvent, particle size,type of vehicle [11]. Therefore, the development of suitable products that preventpenetration of the sunscreen into the skin is a challenge for manufacturers. Some ofthe vehicular penetration retardation strategies being researched for sunscreens arereviewed below.

Cyclodextrins and Photostability

Cyclodextrins are cyclic, toroidal-shaped oligosaccharides with a hydrophilic externalsurface and a hydrophobic central core. They are capable of incorporating appropriatelysized, nonpolar compounds or some lipophilic moiety of a molecule into their apolarcavities, forming noncovalent inclusion complexes [12, 13]. This type of molecular en-capsulation can lead to changes in some of the physical and chemical properties of theincluded substance, such as the enhancement of stability to air and light and apparentaqueous solubility [12–14]. Moreover, cyclodextrin complexation can affect the topicalavailability of applied drugs, either increasing or decreasing their permeability into andthrough the skin [12, 15].

Butyl-methoxydibenzoylmethane (BM-DBM) is a widely used filter that provides pro-tection against UVA radiation in the 320 to 400 nm range. However, BM-DBM experiencesmarked photodegradation [16–21], forming highly reactive photolytic products that areexposed to the living tissues of the epidermis and dermis following percutaneous perme-ation. Scalia et al. [17, 21] have demonstrated that the degree of decomposition andfree radical formation upon exposure of BM-DBM to simulated sunlight were reducedby complexation with hydroxypropyl-b-cyclodextrin (HP-b-CD). The effects of HP-b-CDand sulfobutylether-b-cyclodextrin (SBE7-b-CD) on in vitro human skin penetration andretention of the sunscreen agent BM-DBM were investigated by Simeoni et al. [22].They report that approximately 14 to 16% of the applied dose of BM-DBM penetratedinto the skin tissue; however, no sunscreen was detected in the dermis and in the receiverphase. The greater proportion (84.6 to 95.5%) of the absorbed UV filter was localized in theSC with no significant differences between uncomplexed and complexed BM-DBM.Notable levels (2.3% of the applied dose) of the sunscreen agent accumulated in theepidermis from the preparation containing free BM-DBM. The epidermal concentration




of the UV filter was markedly reduced (0.7% of the applied dose) by complexationwith SBE7-b-CD, whereas HP-b-CD had no effect. The results demonstrated thatcomplexation of BM-DBM with SBE7-b-CD attained high sunscreen levels at the skinsurface where its action is most desirable, and produced lower concentrations of the activein the epidermis.

Transcutolh

Transcutol1 CG (diethylene glycol monoethyl ether) is a hydgroscopic liquid that isfreely miscible with both polar and nonpolar solvents. Transcutol has been recognizedas a potential transdermal permeation enhancer due to its nontoxicity, biocompatibilitywith skin, and excellent solubilizing properties [23]. However, Transcutol has also beenreported to increase the skin accumulation of topically applied compounds without aconcomitant increase in transdermal permeation [24, 25]. It is theorized that thisdepot effect is created by a swelling of SC intercellular lipids, without alteration of theirmultiple bilayer structure. The expanded lipid domain is then able to retain drugs(especially lipophilic compounds) to form the depot, with a simultaneous decrease intransdermal permeation.

Godwin et al. [23] studied the influence of Transcutol CG concentrations in sunscreenformulations on the transdermal permeation and skin accumulation of the UV abs-orbers 2-hydroxy-4-methoxybenzophenone (oxybenzone) and 2-octyl-4-methoxycin-namate (cinnamate).

When formulated alone, both these lipophilic sunscreens have been shown to per-meate through the skin and enter the systemic circulation [26]. In their study, theconcentration of the UV absorber was held constant at 6% (w/w) for all vehicle systemswhile the concentration of Transcutol CG was varied from 0 to 50% (w/w). The datademonstrated that both UV absorbers exhibited an increase in skin accumulation withincreasing concentrations of Transcutol CG. Skin accumulation of oxybenzone wassignificantly (P < 0.05) greater than that of cinnamate for all formulations investigated.However, no significant differences were found in the transdermal permeation of oxy-benzone or cinnamate for any of the formulations tested. The results of this studydemonstrate that the inclusion of Transcutol CG in sunscreen formulations appears toincrease the skin accumulation of the UV absorbers oxybenzone and cinnamate, withouta concomitant increase in transdermal permeation. Their data support the theory of theformation of an intracutaneous depot for both oxybenzone and cinnamate whenformulated with Transcutol CG; however, our group has not been able to confirm thistheory in vivo.

Encapsulation Structures

Colloidal drug carriers, including submicron emulsions, nanospheres, nanocapsules,liposomes, and lipid complexes, have been attracting increasing interest in recent yearsas drug delivery vehicles. These encapsulation systems have been evaluated for theintravenous administration of lipophilic drugs, as improved parenteral formulations,and as systems for site-specific drug delivery [27]. In general, two techniques havebeen used for the preparation of nanocapsules based on biodegradable polymers: the


Retardation Strategies for Sunscreen Agents & 375


emulsification–diffusion technique [28] and the solvent displacement procedure[29, 30]. The ideal medium in which an active ingredient is incorporated must providenot only the necessary solubility, but also maintain contact between the active ingredientand the skin. The nature of the colloidal carrier, and the effects of size and surface charge,influence drug penetration and permeation of UV filters into the skin [31–33].

Alvarez-Roman et al. [34] investigated the optimization of a solvent displacementmethod for poly(e-caprolactone) nanocapsules, using the lipophilic drug, octyl meth-oxycinnamate (OMC) as the oil core. In addition, these researchers evaluated the influenceof polysorbate 85 and poloxamer 188 as stabilizing agents, the OMC loading capacity, andthe photoprotective potential of the formulations. The OMC-nanocapsule-gel preparationresulted in a significantly better (P < 0.05) protection against UV-induced erythema than asimple OMC-gel. Sunscreen effectiveness implies that the sunscreens adhere to the skinmore efficiently as a protective film. These results suggest that the nanoparticles are able tocover the skin surface due to their high specific surface area. Sunscreen nanocapsules,therefore, show good potential as improved skin retention vehicles.

Liposomes and emulsions have been formulated from biocompatible excipients andcan easily be produced on a large scale. Compared to liposomes and emulsions, solidparticles afford protection of incorporated active compounds against chemical degrad-ation and allow more flexibility in modulating the release of the compound. The advan-tages of solid particles, emulsions, and liposomes were, therefore, combined by thedevelopment of solid lipid nanoparticles (SLNs) [35], produced by simply exchangingthe liquid lipid (oil) of the emulsions by a solid lipid.

Wissing and muller. [36] compared an SLN and a conventional o/w emulsion carriersystem for the sunscreen oxybenzone, by studying the in vitro rate of release with amembrane-free model and static Franz diffusion cells. It is reported that the release ratecould be decreased by up to 50% with the SLN formulation. Penetration of oxybenzoneinto SC on the forearm in vivo was also investigated by a tape stripping method. Incongruity with the in vitro data, it was shown that the active release rate could bedecreased by 30 to 60% with SLN formulations. In all test models, oxybenzone penetratedinto the skin more quickly and to a greater extent than from conventional emulsions. Theauthors concluded that using SLN as a carrier system offers two main advantages. SLNs actas physical sunscreens on their own — therefore the concentration of molecular sun-screen agents can be decreased while maintaining the formulation sun protection factor.Moreover, SLNs are able to provide a sustained release carrier system, enabling thesunscreen to remain longer at its site of action on the surface of the skin.

Similar results were obtained by Wissing et al. [37] when they compared the efficacyof a conventional o/w emulsion and crystalline lipid nanoparticles (CLN) incorporatingthe sunscreen benzophenone-3. This in vitro study based on the Transporee

ˆtest

[38] showed that the amount of molecular sunscreen can be decreased by up to 50%while maintaining the UV protection efficacy, simply because of the particulate nature ofthe CLN structures.

Nanocapsules (NC) have been introduced as a new generation of carriers for cos-metics and UV blockers for use on human skin and hair. Jimenez et al. [39] compared theporcine skin permeation of a lipophilic sunscreen, OMC, from different emulsions andencapsulated sunscreen-poly(e-caprolactone) nanocapsules. Their results showed thatthe use of NC-emulsions decreases the permeation of OMC through pig skin whencompared with equivalent w/o and o/w emulsions. NC-emulsions are, therefore, novelvehicle-type dispersion systems and can be used advantageously as sunscreen carriers tolower permeation of the active through the skin.




Physical Properties of Organic Particulate UV-Absorbers

Most of the UV filters in use are oil-soluble and, consequently, are incorporated into theoil phase of sunscreen emulsions, however, even solubility in the oil may be problematic.UV absorbers that are poorly soluble in oils and relatively insoluble in water may bemicronized to form aqueous dispersions of ultra-small particle size. The protectiveperformance of these particles depends on size, as both absorption and scattering playa role in the attenuation of UV light. To this end, it is desirable to achieve particle sizes inthe submicrometer range.

Herzog et al. [40] generated microparticles of a benzotriazole derivative in the 0.16 to40 mm range by milling particles in the presence of a dispersing agent. The UVabsorptionincreases with a decrease in particle size, while the light scattering shows a maximum ata certain particle size. These researchers investigated the UV-attenuating properties ofparticulate organic absorbers as a function of particle size, with special emphasis on thedifferentiation between absorption and scattering functionalities of the particles [41]. Theefficiency of the UV extinction of the dispersion increases with decreasing particle sizedown to a maximum extinction at a particle size of 80 nm, and the UV extinctiondecreases for particles smaller than 80 nm indicating an optimum at 80 nm. It wasfound from reflection spectroscopic measurements that scattering accounts for about10%, and absorption 90%, of the UV-attenuating effect of the particles.

Inorganic Materials

Micronized TiO2 particles with a diameter of about 15 nm are used in sunscreens asphysical UV filters. These particles are suspected to be absorbed through the SC into theepidermis or dermis via intercellular channels, hair follicles, and sweat glands. Thispermeation is undesirable because of the risk of damage to DNA and RNA by the photo-catalytic effects of the TiO2 after absorption of UV light [42]. Furthermore, the particles canactivate the immune system and accumulations of these particles in the skin can decreasethe threshold for allergies [43]. The function of the SC as a barrier against dermal uptakeof ultrafine particles was the subject of several investigations, which came to differentconclusions concerning the penetration depth of the particles [44, 45].

Researchers have investigated a number of other ultra fine, inorganic particles,including ceria (CeO2) [46] and zinc oxide, for efficacy as UV-protectants. Most of theseagents are ideal for cosmetic applications because they are relatively transparent tovisible light, but have excellent ultraviolet radiation absorption properties, and appeartransparent on the skin. However, many of these chemicals exhibit high photocatalyticpotential after UV activation [47]. This reactivity can be mediated by coating of theparticles (with amorphous silica for example) or by doping with a metal ion possessinglower valence and larger ionic size.

Menzel et al. [48] investigated the percutaneous penetration of TiO2 through pig skinand observed a penetration of particles through the SC into the underlying stratumgranulosum via the intercellular spaces, but the TiO2 particle concentration in the stratumspinosum was negligible. Hair follicles did not seem to be major penetration pathways asTiO2 was not detected inside these appendages. These findings show the importance ofcoating the TiO2 particles as a mechanism to reduce genetic damage in the skin.

Alternatively, any formulation mechanism that would retard the penetration of theparticles below the outermost SC layers would be highly advantageous. One possible




mechanism to achieve this may be the chemical sequestration of the organic active.To minimize absorption of sunscreen agents through the skin, organic materialsmay be incorporated in the nanospaces of inorganic materials, thereby avoiding directexposure to the SC biochemical environment. Layered double hydroxides (LDHs) consistof hydrotalcite-like layers and exchangeable interlayer anions [49–52]. The uniqueanion exchange capability of LDH enables the encapsulation of UV-absorbents witha negative charge. However, organic UV-absorbents are often de-intercalated by ananion exchange reaction with carbonate ions, since the selectivity of ion exchange ofcarbonate is higher. This problem may also be minimized by silica coating of the entirecomposite [53].

Penetration Retarders

Transdermal penetration enhancers have been synthesized and tested for their abilityto increase the amount of co-administered drug in a topical formulation that can bedelivered through the skin. The uniform, parallel arrangement of molecules in the lipidbilayer is disrupted by the presence of the enhancer, causing convolution of the bilayermolecules and, subsequently, decreasing the ability of these structures to act as a barrierto the passive diffusion of chemicals applied to the skin surface. The net result is areversible reduction in the barrier properties of the skin, and higher concentrationsof drug reaching the dermal circulation. Conversely, it should be possible to chemicallybind the molecules of the lipid bilayers more closely and rigidly together, therebyincreasing the barrier potential of this intercellular domain. This has been the premiseof the (relatively limited) research that has been carried out in the field of penetrationretarders.

Hadgraft et al. [54] reported the existence of compounds of a similar structure toAzone which acted as drug retardants rather than enhancers. Such agents would haveapplications in formulations that contain sunscreens, pesticides, or drugs with specificlocal-skin targeting. These authors tested Azone and five of its analogs using in vitrodiffusion-cell methodology and human cadaver skin. In addition, the compounds weretested for their ability to reduce the phase-transition temperature of dipalmitoyl phos-phatidylcholine (DPPC). The two drugs used for the evaluation were metronidazole anddiethyl m-toluamide (DEET), and all experiments were performed for 40 h. The com-pounds were placed in the donor chambers of Franz diffusion cells in a 1% ethanolicsolution and left for 2 h. This pretreatment was followed by the application of a finitedose of metronidazole or DEET in the ethanolic solution (5 mmol/ml). Based on thereduction of the phase-transition temperature of DPPC, the agents were ranked: N-0539~Azone>N-0721>N-0253¼N-0131>N-0915. Therefore, N-0915 increased the phase-transition temperature, suggesting that it enters the bilayer of the liposome and increasesthe carbon side-chain rigidity. This may imply that N-0539 would be a retarder rather thanan enhancer. All compounds, except for N-0915, showed some degree of enhancementactivity. In contrast, N-0915 produced an enhancement ratio at 40 h of only 0.2, comparedwith the control; a significant retardation rather than an enhancement. Experiments withDEET showed a similar trend in activity with all six compounds tested. To date this hasbeen one of the few investigations published that specifically identifies a penetrationretarder moiety — a species that theoretically has widespread applicability in the sun-screen industry.




Vehicle Effects

Similarly, there are relatively few studies examining the effect of vehicle viscosity oncutaneous penetration following the application of finite or small ‘‘in use’’ doses oftopical drug formulations. Cross et al. [55] compared the effect of viscosity on thein vitro penetration of benzophenone from four different types of emulsion formulations.The researchers maintained the same thermodynamic activity in all test modes, usingboth epidermal and high density polyethylene (HDPE) membranes, and allowed forcontrol of any possible vehicle–skin interactions. In addition, the change in percutaneouspenetration and retention kinetics of benzophenone from the emulsions followinginfinite and finite dose application was determined in an attempt to define the effectsof viscosity on actual ‘‘in use’’ conditions. In the latter situation, factors such as formula-tion evaporation (estimated from the rate of vehicle water loss) would be expected tomake a significant contribution to release kinetics (metamorphosis of the vehicle [56]).The results from the human epidermal penetration flux data indicate that while thepenetration flux decreased with formulation viscosity for the very thick (infinite dose)formulation, the penetration flux was increased over the control formulation with in-creasing viscosity for the very thin formulation (finite or ‘‘in use’’ dose). The epidermalmembrane retention also decreases with viscosity for the infinite dose. In contrast, theepidermal membrane retention for the finite dose appears to be unaffected by theviscosity of the formulation used. The penetration and retention profiles with viscosityare similar for HDPE membranes to that observed for human epidermal membranes.However, even the concepts of ‘‘finite’’ and ‘‘infinite’’ dose have been open to wideinterpretation by researchers [57].

The discrepancy in the infinite and finite dosing results is likely to arise from thediffering diffusion of benzophenone in the formulations, and skin hydration arising inthe two situations. Slower water evaporation is likely to result in a higher water contentin the residual film and an increase in skin penetration due to a higher diffusivity in amore hydrated membrane [58]. It is unlikely that the formulations have affected partition-ing into the skin as epidermal retention for the four vehicles was similar. Hence, the fluxof benzophenone-3 through both human epidermal and HDPE membranes decreaseswith increasing formulation viscosity. The clinical implications from the study is thatcaution should be exercised in assuming that more viscous formulations applied to theskin may retard the penetration of topically applied sunscreens. Viscous formulationsimpede the skin penetration of benzophenone under infinite dosing conditions, butappear to cause faster skin penetration using thin, ‘‘in use’’ formulations. The interestingaspect from a formulation viewpoint is that it may be possible to modulate absorptionby simple changes in the vehicle matrix, a conclusion that corroborates early findings ofHaigh et al. [59].

Conclusions

A number of strategies have been individually evaluated to limit the absorption ofsunscreens after topical application. It would be interesting to judiciously combinethese scientific concepts that have been investigated individually, into an integratedtopical delivery system of chemical composition, formulation microstructure and specificpenetration retarder such that the delivered sunscreen chemical is held in a bound




reservoir in the layers of the SC, preventing penetration into the lower strata and dermis,and minimizing surface loss. In this manner, the sun protecting agent would be held atits optimal site of action — the surface of the skin — for a prolonged period, and wouldinduce minimal sensitizing or irritancy activity because of exposure to the systemiccirculation. Perhaps the addition of a specific keratin binding agent may extend thelongevity of the sunscreen even further. Clearly sunscreen optimization technology isstill in its infancy in terms of maintaining the active protectant at its site of action.Hopefully, we will be able to report the development of optimized sun-induced skincancer protection vehicles in the next edition of this series.

References1. Pathak, M.A. (1991) Ultraviolet radiation and the development of non-melanoma and

melanoma skin cancer: clinical and experimental evidence. Skin Pharmacology: TheOfficial Journal of the Skin Pharmacology Society, 4, 85–94.

2. Ziegler, A., Jonason, A.S., Leffell, D.J., Simon, J.A., Sharma, H.W., Kimmelman, J., Remington,L., Jacks, T., and Brash, D.E. (1994) Sunburn and p53 in the onset of skin cancer. Nature, 372,773–776.

3. Hochberg, M. and Enk, C.D. (1999) Partial protection against epidermal IL-10 transcription andLangerhans cell depletion by sunscreens after exposure of human skin to UVB. Photochemistryand Photobiology, 70, 766–772.

4. Hayden, C.G., Roberts, M.S., and Benson, H.A. (1998) Sunscreens: are Australians getting thegood oil? Australian and New Zealand Journal of Medicine, 28, 639–646.

5. Green, A., Williams, G., Neale, R., Hart, V., Leslie, D., Parsons, P., Marks, G.C., Gaffney, P.,Battistutta, D., and Frost, C. (1999) Daily sunscreen application and betacarotene supplemen-tation in prevention of basal-cell and squamous-cell carcinomas of the skin: a randomisedcontrolled trial. The Lancet, 354, 723–729.

6. Patel, N.P., Highton, A., and Moy, R.L. (1992) Properties of topical sunscreen formulations.A review. The Journal of Dermatologic Surgery and Oncology, 18, 316–320.

7. Broadbent, J.K., Martincigh, B.S., Raynor, M.W., Salter, L.F., Moulder, R., Sjoberg, P., andMarkides, K.E. (1996) Capillary supercritical fluid chromatography combined with atmo-spheric pressure chemical ionisation mass spectrometry for the investigation of photoproductformation in the sunscreen absorber 2-ethylhexyl-p-methoxycinnamate. Journal of Chroma-tography A, 732, 101–110.

8. Deflandre, A. and Lang, G. (1988) Photostability assessment of sunscreens. Benzylidenecamphor and dibenzoylmethane derivatives. International Journal of Cosmetic Science, 10,53–62.

9. Dromgoole, S.H. and Maibach, H.I. (1990) Sunscreening agent intolerance: contact andphotocontact sensitization and contact urticaria. Journal of American Academy of Dermatol-ogy, 22, 1068–1078.

10. Mariani, E., Neuhoff, C., Bargagna, A., Bonina, F., Giacchi, M., De Guidi, G., and Velardita, A.(1998) Synthesis, in vitro percutaneous absorption and phototoxicity of new benzylidenederivatives of 1,3,3-trimethyl-2-oxabicyclo[2.2.2]octan-6-one as potential UV sunscreens. Inter-national Journal of Pharmaceutics, 161, 65–73.

11. Benech-Kieffer, F., Wegrich, P, Schwarzenbach, R., Klecak, G., Weber, T., Leclaire, J., andSchaefer, H. (2000) Percutaneous absorption of sunscreens in vitro: interspecies comparison,skin models and reproducibility aspects. Skin Pharmacology and Applied Skin Physiology, 13,324–35.

12. Rajewski, R.A. and Stella, V.J. (1996) Pharmaceutical applications of cyclodextrins. 2. In vivodrug delivery. Journal of Pharmaceutical Sciences, 85, 1142–1169.




13. Loftsson, T. and Brewster, M.E. (1996) Pharmaceutical applications of cyclodextrins. 1. Drugsolubilization and stabilization. Journal of Pharmaceutical Sciences, 85, 1017–1025.

14. Uekama, K., Hirayama, F., and Irie, T. Cyclodextrin drug carrier systems. Chemical Reviews, 98,2045–2076.

15. Loftsson, T. and Masson, M. (2001) Cyclodextrins in topical drug formulations: theory andpractice. International Journal of Pharmaceutics, 225, 15–30.

16. Schwack, W. and Rudolph, T. (1995) Photochemistry of dibenzoyl methane UVA filters Part 1.Journal of Photochemistry and Photobiology B: Biology, 28, 229–234.

17. Scalia, S., Villani, S., Scatturin, A., Vandelli, M.A., and Forni, F. (1998) Complexation of thesunscreen agent, butyl-methoxydibenzoylmethane, with hydroxypropyl-[beta]-cyclodextrin.International Journal of Pharmaceutics, 175, 205–213.

18. Tarras-Wahlberg, N., Stenhagen, G., Larko, O., Rosen, A., Wennberg A.-M., and Wennerstrom,O. (1999) Changes in ultraviolet absorption of sunscreens after ultraviolet irradiation. Journalof Investigative Dermatology, 113, 547–553.

19. Chatelain, E. and Gabard, B. (2001) Photostabilization of butyl methoxydibenzoylmethane(Avobenzone) and ethylhexyl methoxycinnamate by bis-ethylhexyloxyphenol methoxyphe-nyl triazine (Tinosorb S), a new UV broadband filter. Photochemistry and Photobiology, 74,401–406.

20. Damiani, E., Greci, L., Parsons, R., and Knowland, J. (1999) Nitroxide radicals protect DNAfrom damage when illuminated in vitro in the presence of dibenzoylmethane and a commonsunscreen ingredient. Free Radical Biology and Medicine, 26, 809–816.

21. Scalia, S., Casolari, A., Iaconinoto, A., and Simeoni, S. (2002) Comparative studies of theinfluence of cyclodextrins on the stability of the sunscreen agent, 2-ethylhexyl-p-methoxycin-namate. Journal of Pharmaceutical and Biomedical Analysis, 30, 1181–1189.

22. Simeoni, S., Scalia, S., and Benson, H.A.E. (2004) Influence of cyclodextrins on in vitro humanskin absorption of the sunscreen, butyl-methoxydibenzoylmethane. International Journal ofPharmaceutics, 280, 163–171.

23. Godwin, D.A., Kim, N.-H., and Felton, L.A. (2002) Influence of transcutol CG on the skinacccumulation and transdermal permeation of ultraviolet absorbers. European Journal ofPharmaceutics and Biopharmaceutics, 53, 23–27.

24. Ritschel, W.A., Panchagnula, R., Stemmer, K., and Ashraf, M. (1991) Development of anintracutaneous depot for drugs. Binding, drug accumulation and retention studies, and mech-anism of depot. Skin Pharmacology, 4, 235–245.

25. Panchagnula, R. and Ritschel, W.A. (1991) Development and evaluation of an intracutaneousdepot formulation of corticosteroids using Transcutol as a cosolvent: in-vitro, ex-vivo, and in-vivo rat studies. Journal of Pharmacy and Pharmacology, 43, 609–614.

26. Treffel, P. and Gabard, B. (1996) Skin penetration and sun protection factor of ultra-violetfilters from two vehicles. Pharm Res, 13, 770–4.

27. Allemann, E., Gurny, R., and Doelker, E. (1993) Drug-loaded nanoparticles — preparationmethods and drug targeting issues. European Journal of Pharmaceutics and Biopharmaceu-tics, 39, 173–191.

28. Quintanar-Guerrero, D., Fessi, H., Allemann, E., and Doelker, E. (1996) Influence of stabilizingagents and preparative variables on the formation of poly(-lactic acid) nanoparticles by anemulsification–diffusion technique. International Journal of Pharmaceutics, 143, 133–141.

29. Fessi, H., Devissaguet, J.-P., Puisieux, F., and Thies, C. (1988) Procede de preparation desystemes colloIdaux dispersibles d’une substance, sous forme de nanoparticules. FrenchPat, 2, 608–988.

30. Al Khouri Fallouh, N., Roblot-Treupel, L., Fessi, H., Devissaguet, J.P., and Puisieux, F. (1986)Development of a new process for the manufacture of polyisobutylcyanoacrylate nanocap-sules. International Journal of Pharmaceutics, 28, 125–132.

31. Zeevi, A., Klang, S., Alard, V., Brossard, F., and Benita, S. (1994) The design and character-ization of a positively charged submicron emulsion containing a sunscreen agent. Inter-national Journal of Pharmaceutics, 108, 57–68.




32. Treffel, P. and Gabard, B. (1996) Skin penetration and sun protection factor of ultra-violetfilters from two vehicles. Pharmaceutical Research, 13, 770–774.

33. Gupta, V.K., Zatz, J.L., and Rerek, M. (1999) Percutaneous absorption of sunscreens throughmicro-Yucatan pig skin in vitro. Pharmaceutical Research, 16, 1602–1607.

34. Alvarez-Roman, R., Barre, G., Guy, R.H., and Fessi, H. (2001) Biodegradable polymer nano-capsules containing a sunscreen agent: preparation and photoprotection. European Journal ofPharmaceutics and Biopharmaceutics, 52, 191–195.

35. Muller, R.H., Radtke, M., and Wissing, S.A. (2002) Solid lipid nanoparticles (SLN) and nano-structured lipid carriers (NLC) in cosmetic and dermatological preparations. Advanced DrugDelivery Reviews, 54, S131–S155.

36. Wissing, S.A. and Muller, R.H. (2002) Solid lipid nanoparticles as carrier for sunscreens: in vitrorelease and in vivo skin penetration. Journal of Controlled Release, 81, 225–233.

37. Wissing, S.A. and Muller, R.H. (2002) The development of an improved carrier system forsunscreen formulations based on crystalline lipid nanoparticles. International Journal ofPharmaceutics, 242, 373–375.

38. Diffey, B.L. and Farr, P.M. (1991) Sunscreen protection against UVB, UVA and blue light: anin vivo and in vitro comparison. The British Journal of Dermatology, 124, 258–263.

39. Jimenez, M.M., Pelletier, J., Bobin, M.F., and Martini, M.C. (2004) Influence of encapsulationon the in vitro percutaneous absorption of octyl methoxycinnamate. International Journal ofPharmaceutics, 272, 45–55.

40. Herzog, B., Katzenstein, A., Quass, K., Stehlin, A., and Luther, H. (2004) Physical propertiesof organic particulate UV-absorbers used in sunscreens: I. Determination of particle sizewith fiber-optic quasi-elastic light scattering (FOQELS), disc centrifugation, and laser diffrac-tometry. Journal of Colloid and Interface Science, 271, 136–144.

41. Herzog, B., Quass, K., Schmidt, E., Muller, S., and Luther, H. (2004) Physical properties oforganic particulate UV absorbers used in sunscreens: II. UV-attenuating efficiency as functionof particle size. Journal of Colloid and Interface Science, 276, 354–363.

42. Dunford, R., Salinaro, A., Cai, L., Serpone, N., Horikoshi, S., Hidaka, H., and Knowland,J. (1997) Chemical oxidation and DNA damage catalysed by inorganic sunscreen ingredients.FEBS Letters, 418, 87–90.

43. Granum, B., Gaarder, P.I., Groeng, E.-C., Leikvold, R.-B., Namork, E., and Lovik, M. (2001) Fineparticles of widely different composition have an adjuvant effect on the production of allergen-specific antibodies. Toxicology Letters, 118, 171–181.

44. Pflucker, F., Wendel, V., Hohenberg, H., Gartner, E., Will, T., Pfeiffer, S., Wepf, R., and Gers-Barlag, H. (2001) The human stratum corneum layer: an effective barrier against dermal uptakeof different forms of topically applied micronised titanium dioxide. Skin Pharmacology andApplied Skin Physiology, 14 (suppl 1), 92–97.

45. Tan, M.H., Commens, C.A., Burnett, L., and Snitch, P.J. (1996) A pilot study on the percutan-eous absorption of microfine titanium dioxide from sunscreens. The Australasian Journal ofDermatology, 37, 185–187.

46. Yabe, S. and Sato, T. (2003) Cerium oxide for sunscreen cosmetics. Journal of Solid StateChemistry, 171, 7–11.

47. Cai, R., Hashimoto, K., Itoh, K., Kubota, Y., and Fujishima, A. (1991) Photokilling of malignantcells with ultrafine TiO2 powder. Bulletin of Chemical Society of Japan, 64, 1268–1273.

48. Menzel, F., Reinert, T., Vogt, J., and Butz, T. (2004) Investigations of percutaneous uptakeof ultrafine TiO2 particles at the high energy ion nanoprobe LIPSION. Nuclear Instrumentsand Methods in Physics Research Section B: Beam Interactions with Materials and Atoms,219–220, 82–86.

49. Taylor, R.M. (1984) The rapid formation of crystalline double hydroxy salts and other com-pounds by controlled hydrolysis. Clay Minerals, 19, 591–603.

50. Sato, T., Onai, S., Yoshioka, T., and Okuwaki, A. (1993) Causticization of sodium carbonatewith rock-salt type magnesium aluminium oxide formed by the thermal decomposition of




hydrotalcite-like layered double hydroxide. Journal of Chemical Technology and Biotechnol-ogy, 57, 137–140.

51. Cavani, F., Trifiro, F., and Vaccari, A. (1991) Hydrotalcite-type anionic clays: preparation,properties and applications. Catalysis Today, 11, 173–301.

52. Lakraimi, M., Legrouri, A., Barroug, A., De Roy, A., and Besse, J. (2000) Preparation of a newstable hybrid material by chloride-2,4-dichlorophenoxyacetate ion exchange into the zinc–aluminium-chloride layered double hydroxide. Journal of Materials Chemistry, 10, 1007–1011.

53. El-Toni, A.M., Yin, S., and Sato, T. (2005) Silica coating of Zn2Al/4-hydroxy-3-methoxybenzoicacid nanocomposites via seeded polymerization technique. Materials Chemistry and Physics,89, 154–158.

54. Hadgraft, J., Peck, J., Williams, D.G., Pugh, W.J., and Allan, G. (1996) Mechanisms of actionof skin penetration enhancers/retarders: Azone and analogues. International Journal ofPharmaceutics, 141, 17–25.

55. Cross, S.E., Jiang, R., Benson, H.A.E., and Roberts, M.S. (2001) Can increasing the viscosityof formulations be used to reduce the human skin penetration of the sunscreen oxybenzone?The Journal of Investigative Dermatology, 117, 147–150.

56. Surber, C. and Smith, E.W. (2005) The mystical effects of dermatological vehicles. Dermatol-ogy, 210, 157–168.

57. Surber, C. and Davis, A.F. (2002) Bioavailability and bioequivalence of dermatological formu-lations. In Walters, K.A. (ed.), Dermatological and Transdermal Formulations. Marcel Dekker,Inc., New York, Basel, pp. 401–498.

58. Roberts, M. and Walker, M. (1993) Water — the most natural skin penetration enhancer. InWalters, K. and Hadgraft, J. (eds), Skin Penetration Enhancement. Marcel Dekker, New York,pp. 1–30.

59. Haigh, J.M., Smith, E.W., Meyer, E., and Fassihi, R. (1992) Influence of the oil phase dispersionin a cream base on the in vivo release of betamethasone 17-valerate. S.T.P. PharmaceuticalSciences, 2, 259–264.




Chapter 26

Military Perspectivesin Chemical PenetrationRetardation

Ernest H. Braue, Jr., Bryce F. Doxzon, Horace L. Lumpkin, Kelly A. Hanssen,Robert S. Stevenson, Robin R. Deckert, and John S. Graham

CONTENTS

Introduction .................................................................................................................................... 385

Active Barrier Creams..................................................................................................................... 388Efficacy Evaluation Methods ...................................................................................................... 389Results and Conclusions............................................................................................................. 391

Transdermal Chemical Inhibitors................................................................................................... 393Conclusions .................................................................................................................................... 394References....................................................................................................................................... 395

Introduction

The Joint Forces of the U.S. military must operate across the continuum of globalcontingency operations. These forces have an immediate need to safely operate, survive,and sustain operations in chemical, biological, radiological, nuclear (CBRN), toxic indus-trial material (TIM), toxic industrial chemical (TIC), and new threat agent (NTA) hazard-ous environments. To accomplish this mission, service members use a combinationof protective equipment including a chemical protective suit and barrier skin creams. Ifthe systems fail to protect the service member from the effects of toxic agents, pre- andpostexposure treatment therapies and decontamination systems are available to limit theharmful consequences of exposure. This chapter will focus on the use of barrier skincreams to retard the percutaneous absorption of chemical warfare agents (CWAs).


385


Passive Barrier Creams ................................................................................................................... 387

CWAs represent a real and growing threat to both U.S. Armed Forces and civilians.The use of CWAs in the last three decades includes that by the Soviets in Cambodia(yellow rain, tricothecene mycotoxins) [1], the Iraqis against Iran (sulfur mustard andtabun) [2], and the Iraqis against its own Kurdish population at Halabja (sulfur mustard,hydrogen cyanide) [3]. In World War I, almost one-third of Allied casualties were hospi-talized as a result of CWA injuries [4]. In 1995, a Japanese religious cult terrorized thecivilian population by releasing sarin in a Tokyo subway [5]. This attack resulted in over1000 casualties and 12 deaths. Most recently, a plan by Al Qaeda terrorists to use sarin onthe European Parliament Building in Strasbourg was prevented by German police [6].These latter examples demonstrate that the civilian population is no longer immune tothe threat of CWAs.

The U.S. army classifies CWAs into seven categories [7], cyanides, nerve agents, lungtoxicants, vesicants, incapacitating agents, tear gases, and vomiting gases. The twocategories of concern for percutaneous exposure are the vesicants and nerve agents.Vesicants cause irritation and vesication (blistering) of the skin and mucous membranesespecially the respiratory tract and lung. Sulfur mustard (HD, bis[2-chloroethyl]sulfide)and Lewisite (L, 2-chlorovinyl dichloroarsine) are the two most prevalent vesicatingagents. HD exposure to the skin is insidious, causing no immediate discernible effectswithin the first several hours. Erythema, edema, and finally blister formation occur 12 to24 h after exposure depending on the site and dose. While HD exposure is generallylethal in only massive exposures, its vesicating properties are incapacitating both phys-ically and psychologically, and lesions may require up to 4 months for complete healing.Lewisite exposure causes instant pain and lesions generally heal within several weeks.

Nerve agents are highly toxic organophosphorous (OP) compounds that are chem-ically related to some insecticides (parathion, malathion) [8]. The traditional nerve agentsinclude tabun (GA, ethyl N,N-dimethyl-phosphoramidocyanidate), sarin (GB, isopropyl-methylphosphonofluoridate), soman (GD, 1,2,2-trimethylpropyl methylphosphonofluor-idate), and VX (o-ethyl S-[2-(diisopropylamino)ethyl] methylphosphonothiolate). Theseagents inhibit acetylcholinesterase, an enzyme responsible for hydrolyzing the neuro-transmitter acetylcholine in the nervous system and throughout the whole body. Thesigns of poisoning include miosis (constriction of the pupils), increased tracheobronchialsecretions, bronchial constriction, increased sweating, urinary and fecal incontinence,muscle fasciculations, tremors, convulsions or seizures, respiratory failure, and ultimatelydeath. The relative prominence and severity of a given sign is highly dependent on theroute and degree of exposure.

The primary barrier to skin absorption of CWAs, as well as other chemicals, isprovided by the stratum corneum, the layer of dead epithelial cells that forms theoutermost portion of the epidermis. The thickness of the epidermis, which is reflectedin the number of cell layers within the stratum corneum, varies between species andbody location. In humans, the thickness of the epidermis varies between 15 and 150 mmover most of the body, except for the palms and soles which may be as thick as 400 to 600mm [9]. Barrier effectiveness increases with the thickness of the stratum corneum. Skinhydration and temperature also affect the barrier properties of skin [10]. Warm, moistareas with thin skin, such as the perineum, external genitalia, axillae, antecubital fossae,and neck, are much more sensitive to exposure. Hair follicles may also provide a moredirect entry route for agent into the dermis, since the epithelial tissue that surrounds thehair shaft and comprises the follicle may not be as thick as the surface epithelium [11].

The diffusion coefficients for permeation of CWAs through the stratum corneum aresufficiently high to require effective decontamination to be conducted within the first few




minutes of exposure. The current protection scheme against CWAs for U.S. servicemembers consists of a chemically resistant outer layer of clothing known as battledressovergarment (BDO), protective mask (M40), butyl rubber gloves, and rubber overboots[12]. This ensemble allows continued operation in a chemically contaminated area, butresults in decreased performance and increased heat retention. Barrier creams are used inconjunction with protective clothing to increase efficacy at critical junctures of the BDO,such as where the gloves meet the sleeves.

Passive Barrier Creams

Applying a topical protectant to vulnerable skin surfaces prior to entry into a chemicalcombat arena was proposed as a protective measure against percutaneous CWA toxicitysoon after the use of HD by Germany at Ypres, Belgium in 1917 [13]. In the summer of1917, the U.S. army began examining various soaps and ointments for protective cap-abilities. Although several simple formulations were found to be effective in reducing‘‘skin redness’’ produced by agents such as hydrogen sulfide, no product was availablebefore the end of World War I [14]. Research in the area of protective ointments continuedafter the war, but this effort did not produce a fielded product before the beginning ofWorld War II. During World War II, a concentrated effort to develop ointments forprotection against HD took place at the Chemical Warfare Service, Edgewood Arsenal,Maryland. The Army produced the M-5 protective ointment, which was manufactured in1943 and 1944. However, because of limited effectiveness, odor, and other cosmeticcharacteristics, the M-5 ointment was no longer issued to soldiers by the mid 1950s [15].

Between 1950 and the early 1980s, the focus on research shifted to medical counter-measures rather than protective creams. Beginning in the early 1980s, a limited researcheffort produced two nonactive barrier skin cream formulations based on a blend ofperfluorinated polymers. The two formulations, developed at the U.S. Army MedicalResearch Institute of Chemical Defense (USAMRICD), Aberdeen Proving Ground, Mary-land, were transferred to advanced development in October 1990 [16]. The best formu-lation was selected and progressed through development with an Investigational NewDrug (IND) filed with the Food and Drug Administration (FDA) in 1994 and approval of aNew Drug Application (NDA) in 2000. This new product was called Skin ExposureReduction Paste Against Chemical Warfare Agents (SERPACWA). SERPACWA consistedof fine particles of polytetrafluoroethylene (PTFE) solid dispersed in a fluorinated poly-ether. The excellent barrier properties of this polymer blend were related to the lowsolubility of most materials in it. Only highly fluorinated solvents like Freon1 wereobserved to show appreciable solubility. SERPACWA is now a standard issue item toU.S. forces when there is a threat of CWA use. Operationally, SERPACWA is designed to beused on the skin at the BDO closures and on other vulnerable skin areas to enhanceprotection.

Efficacy testing showed that ICD2289 (an early formulation of SERPACWA), spread asa thin layer about 0.1 mm thick, formed an effective barrier on the skin of animals andreduced the toxicity of CWA [17]. Data in a clipped rabbit model showed that ICD2289was effective against the percutaneous penetration of HD by reducing the size of skinlesions by 81%. In these studies, animals were exposed to 1 ml of neat HD for 4 h beforedecontamination with 0.5% bleach. Lesions were evaluated 24 h after exposure.

T-2 mycotoxin, a trichothecene produced by various species of fungi, is the onlybiological warfare agent that is skin active [7]. In a clipped rabbit model, ICD2289


Military Perspectives in Chemical Penetration Retardation & 387


provided complete protection from the effects of T-2 mycotoxin exposure including theformation of erythema, edema, dermatitis, folliculitis, and necrosis. In these studies,animals were exposed to 2 ml of a methanol solution of T-2 mycotoxin (12.5 mg ml1)and decontaminated with swabs containing Dyna-Hex (a preparation containing 4%chlorohexidine gluconate) at 1, 2, 4, or 6 h after exposure. Lesion was evaluated at 24,32, and 48 h postexposure.

ICD2289 also demonstrated effective protection against percutaneous exposure to thenerve agents VX and thickened (5% methyl methacrylamide) soman (TGD). In a clippedrabbit model, ICD2289 reduced 24-h lethality by 92% compared with unprotected ani-mals challenged for 4 h with 0.5 mg kg1 of neat VX (a 10 LD50 dose). In addition, theICD2289 protected animals challenged with VX retained sufficient red blood cell (RBC)acetylcholinesterase (AChE) activity (50% of baseline) to sustain life. In the same animalmodel, ICD2289 protected animals challenged with TGD (3.35 mg kg1, 1 LD50) yieldedRBC AChE activity values of 57%, while unprotected animals had values of only 19% ofbaseline. Both protected and unprotected animals had sufficient AChE activity to sustainlife, and only a very small percentage of the animals died within 24 h of exposure. Therewas no difference in the lethality rate between ICD2289-protected and unprotectedanimals in these experiments, most likely because the dose selected produced only asmall number of deaths (2 of 24 animals) even though the historical LD50 challenge dosewas used.

The effect on ICD2289’s efficacy was evaluated when used in conjunction with thestandard army insect repellent cream containing N,N-diethyl-m-toluamide (DEET). WhenICD2289 was applied to the skin of rabbits followed by application of insect repellentand challenged with HD, ICD2289 no longer protected the site. The protective benefitof ICD2289 was also lost when insect repellent was applied to rabbit skin before applyingICD2289 and challenged with topical HD. If the insect repellent was applied to the skinand wiped off with a dry cloth (but not a wet cloth), ICD2289 regained some of itsprotective properties.

In contrast to the HD data, ICD2289 remained effective against percutaneous VXin the presence of the insect repellent. In the clipped rabbit model, animals challengedwith 10 LD50s of VX (0.5 mg kg1) were significantly protected by ICD2289 whetherinsect repellent was applied before or after ICD2289 application. DEET-containinginsect repellent decreased the protection offered by ICD2289 against VX, but did notnegate it.

Active Barrier Creams

SERPACWA extended the protection afforded by the current protective garments andallowed a longer window for decontamination, but it did not neutralize CWAs into lesstoxic products. Furthermore, although SERPACWA provided excellent protection againstliquid challenges of GD, VX, and HD, its protection against HD and GD vapor was lessthan optimal.

To overcome the limitations of SERPACWA, the USAMRICD began development of animproved SERPACWA that would act as both a protective barrier and an active destructive

molecules that potentially could be used to neutralize or detoxify CWAs were knownfor a long time. These compounds fell into three general classes: oxidizers, reducers, andnucleophiles. An important limitation, however, was that the final formulation must not




matrix to detoxify CWAs. The concept is demonstrated in Figure 26.1. The types of

irritate the skin. This restriction eliminated many of the most reactive species. The aproticnonpolar environment of SERPACWA provided a unique but challenging medium foractive moieties to neutralize CWA. Reaction mechanisms that did not involve chargedtransition states were favored in this medium. The improved SERPACWA containing areactive matrix became known as active topical skin protectant (aTSP). aTSP providedincreased protection without degrading a warfighter’s performance. Four criteria wereestablished for aTSP. First, the aTSP must neutralize CWAs including HD, GD, and VX.Second, the barrier properties of SERPACWA must be maintained or increased. Third, theprotection against HD and GD vapor must increase. And fourth, the cosmetic character-istics (e.g., odor, texture) of the TSP must be maintained [18].

Using the two components of SERPACWA, perfluorinated-polyether oil and polytetra-fluoroethylene solid, as a base cream, USAMRICD scientists evaluated over 150 differentactive components. Classes of compounds tested included organic polymers, enzymes,hybrid organic–inorganic materials, polyoxometalates, inorganic composites, inorganicoxides, metal alloys, and small organic molecules. These compounds were incorporatedinto the base cream to produce over 500 candidate formulations [19, 20].

Efficacy Evaluation Methods

A Decision Tree Network (DTN) was used to evaluate for efficacy and down select to thebest formulations18. The DTN was divided into two pathways: one for vesicants and theother for nerve agents. Within these pathways, there were three blocks each with adecision point. The first block consisted of a series of three in vitro modules used todetermine the initial efficacy of candidate formulations and to eliminate noneffectivecandidates before animal testing. The second block consisted of in vivo modules, and thethird block consisted of an advanced animal module to determine the influence of time,water, and interactions with other products [21, 22]. The critical DTN tests that were usedin the down-selection model are summarized in the next few paragraphs.

Figure 26.1 Concept for an active topical skin protectant. Pretreatment skin cream formulationprovided physical barrier to prevent absorption by the skin and active compounds to neutralizeblister agents (sulfur mustard) and nerve agents (i.e., GA, GB, GD, and VX).




The pig HDV (sulfur mustard vapor) test evaluated the efficacy of aTSPs in a weanlingpig model challenged with saturated HD vapor. The aTSPs were spread in a thin layer,0.15 mm thick, on the depilated dorsa. The liquid HD was saturated onto filter paperfitted into the top of the cap (1.1 cm2) so that liquid could not run down. The standardsaturated vapor cup was used in a 30-min challenge. The effectiveness of the aTSPwas determined by measuring the degree of erythema that developed on the skin 24h following exposure. Erythema was measured objectively using a reflectance colorim-eter [18].

The rabbit HDL (sulfur mustard liquid) test evaluated the efficacy of aTSPs in a clippedrabbit model challenged with HD liquid. In this test, a 0.15-mm layer of aTSP appliedto the clipped dorsa was challenged with 1.0 ml of liquid HD spread by a 12-mm disk.After a 4-h exposure the sites were decontaminated. The effectiveness of the aTSP wasdetermined by measuring the lesion areas of protected and nonprotected sites 24 hpostexposure.

The rabbit GDV (soman vapor) test evaluated the efficacy of aTSPs in a clipped rabbitmodel challenged with saturated GD vapor. In this test, a 0.15-mm layer of aTSP spreadon the clipped dorsa was challenged with two vapor cups (7 cm2) each containing 28 mgGD per kg of body weight. The liquid GD was saturated onto filter paper fitted into thetop of the cap so that liquid could not run down. The caps were left in place for 4 h. Afterthe exposure period, the caps were removed and the exposure sites decontaminated.This exposure dose was lethal to all animals not protected with aTSP. The effectiveness ofthe aTSP was determined by 24-h lethality.

The GP GDL (soman liquid) test evaluated the efficacy of aTSPs in a clipped guineapig model challenged with liquid GD. In this test, a 0.15-mm layer of aTSP applied to theclipped dorsa was challenged with neat GD applied at a rate of 100 mg per kg of bodyweight (3.3 LD50s) spread by a 12-mm disk. After a 2-h exposure the sites were decon-taminated. The effectiveness of the aTSP was determined by 24-h lethality.

The rabbit VXL (VX liquid) test evaluated the efficacy of aTSPs in a clipped rabbitmodel challenged with liquid VX. In this test, a 0.15-mm layer of aTSP applied to theclipped dorsa was challenged with neat VX applied at the rate of 0.50 mg per kg of bodyweight (12.8 LD50s) spread by a 12-mm disk. After a 4-h exposure the sites weredecontaminated. The effectiveness of the aTSP was determined by 24-h lethality.

The solid phase microextraction (SPME) tests were used to demonstrate that aTSPformulations actually neutralized CWAs into less toxic materials. This test used a head-space SPME technique for the collection of CWAs. Samples collected on the extractionfilament were analyzed by gas chromatography or mass spectroscopy. In a small vial, 100mg of a TSP formulation were challenged with 0.1 ml of neat CWA (HD, GD, or VX). Theheadspace above the mixture was sampled 30 min after challenge to determine theamount of CWA that remained in the vial. Efficacy was determined by calculating thepercent loss of CWA.

The wash test was used to estimate how well the aTSP would remain on the skin inthe presence of water and/or sweat. In a weanling pig model, a 0.15-mm layer of aTSPapplied to the depilated dorsa (1.1 cm2) was washed five times with 1.7 ml of standardsaline solution (0.9%) for a total volume of 8.5 ml. The percentage of the aTSP washed offthe site was visually estimated.

The wipe test was used to estimate how well the aTSP would resist removal by normalabrasion from clothing. In a weanling pig model, a 0.15-mm layer of aTSP applied to thedepilated dorsa (1.1 cm2) was wiped with a foam swab. The percentage of the aTSPwiped off the site was visually estimated.




Results and Conclusions

A total of 17 active moieties, out of the 150 tested, demonstrated significantly (p¼ 0.05)improved protection compared with SERPACWA against HD in at least one DTN test. Thesuccessful active moieties are listed in Table 26.1. Against a GD challenge, 15 activemoieties demonstrated significantly ( p¼ 0.05) improved protection compared with SER-

A weighted and unweighted criteria model was established for the down-selectionprocess. Using this model, two candidate formulations were selected for transition toadvanced development. The lead formulation was a mixture of organic polymers, sur-factants, and the base cream of perfluorinated-polyether oil and polytetrafluoroethylenesolid. The backup formulation contained S-330 and the base cream.

formulation and the backup aTSP formulation. Percent toxicity is a normalized efficacyscale where 100 is no protection and 0 is complete protection. The ‘‘*’’ symbol indicatesresults that represent a significant ( p¼ 0.05) improvement compared with SERPACWA.The lead formulation containing organic polymers provided significantly improved

Table 26.1 List of Active Moieties Demonstrating Significantly (p < 0.05)Improved Protection Compared with SERPACWA against Sulfur Mustard

Active Moieties Source Reference

S-330 Sigma-Aldrich cat # S706485; CAS # 19103-02-7 [18b,1]iodobenzene diacetate Sigma-Aldrich cat # 17,872-1, CAS # 3240-34-4 [18e]Nanoreactors Army Research Lab, APG, MD,

proprietary compoundsAmbergard, XE-5551 resin Rohm and Haas, Philadelphia, PA [18e]Polysilsesquioxanes Dr Kenneth Shea, University California at Irvine [18h]Polyoxometalates Eltron Research Corp, Boulder, CO and

Dr Craig Hill, Emory University, Atlanta, GA[18c,18d]

Titanium or manganesecoated metal alloy

Mainstream Engineering Corp., Rockledge, FL [18f]

Gold or copper catalysts Dr Craig Hill, Emory University, Atlanta, GA [18d]Magnesium oxide reactive

nanoparticlesNanoscale Materials, Inc., Manhattan, KS [18g]

Silicon dioxide Dr Kenneth Shea, University California at Irvine [18h]Ethanolamine matrix Army Research Lab, APG, MD,

proprietary compoundsCerium or copper or

titanium dioxideDr Craig Hill, Emory University,

Atlanta, GA, proprietary compoundsNanophase catalysts Biopraxis, Inc., proprietary compoundsPolyoxometalates on titanium

dioxide reactive nanoparticlesDr Craig Hill, Emory University, Atlanta, GA

and Nanoscale Materials, Inc., Manhattan, KS[18i]

Polyoxometalates on magnesiumoxide reactive nanoparticles

Dr Craig Hill, Emory University, Atlanta, GAand Nanoscale Materials, Inc., Manhattan, KS

[18i]

Silver catalysts Dr Craig Hill, Emory University, Atlanta, GA [18d]Organic polymers USAMRICD, APG, MD and TDA Research,

Wheat Ridge, CO, proprietary compounds

CAS #¼Chemical Abstract Service Registry Numbers.




PACWA. These active moieties are listed in Table 26.2.

Figure 26.2 shows the in vivo data comparing SERPACWA efficacy with the lead aTSP

protection in every model. The backup formulation, containing S-330, provided signifi-cantly improved protection for HD but only equivalent protection for VX and GD.

PACWA efficacy with the lead aTSP formulation and the backup aTSP formulation. Thepercentage of toxicity is a normalized efficacy scale where 100 means no destruction ofagent in the SPME test and easy removal in the wash and wipe test. The ‘‘*’’ symbolindicates significant ( p¼ 0.05) improvement compared with SERPACWA. The lead for-mulation containing organic polymers provided significantly improved protection inevery model except for the wash test. The backup formulation, containing S-330, pro-vided improved protection for HD and VX and equivalent protection for GD in theneutralization tests. It also provides improved protection in the wipe test and equivalentprotection in the wash test.

The lead and backup formulations were transitioned to advanced development.Depending on the FDA approval process and funding, it will likely take 7 to 10 additionalyears to field this improved SERPACWA. When fielded, aTSP will significantly increase theprotection provided to our warfighters and civilians from the effects of CWAs.

Table 26.2 List of Active Moieties Demonstrating Significantly (p 5 0.05)Improved Protection Compared with SERPACWA against Soman

Active Moieties Source Reference

Nanoreactors Army Research Lab, APG, MD,proprietary compounds

Organophosphorus acid anhydridehydrolase (OPAA)

Altus Biologics, Inc., Cambridge, MA [18a]

Sodium hypochlorite (HTH) Many commercial distributors,CAS # 7681-52-9

Gold or copper catalysts Dr Craig Hill, Emory University,Atlanta, GA

[18d]

Iron or copper or lanthanum catalysts Dr Craig Hill, Emory University,Atlanta, GA, proprietary compounds

Zinc oxide reactive nanoparticle Nanoscale Materials, Inc., Manhattan, KS [18g]Magnesium oxide reactive nanoparticles Nanoscale Materials, Inc., Manhattan, KS [18g]Titanium dioxide reactive nanoparticles Nanoscale Materials, Inc., Manhattan, KS [18g]Diethanolamine matrix Army Research Lab, APG, MD,

proprietary compoundsCerium or copper on titanium dioxide

reactive nanoparticleDr Craig Hill, Emory University, Atlanta,

GA, proprietary compoundsNanophase catalysts Biopraxis, Inc., proprietary compoundsPolyoxometalates on titanium

dioxide reactive nanoparticlesDr Craig Hill, Emory University, Atlanta,

GA and Nanoscale Materials, Inc.,Manhattan, KS

[18i]

Polyoxometalates on magnesiumoxide reactive nanoparticles

Dr Craig Hill, Emory University, Atlanta,GA and Nanoscale Materials, Inc.,Manhattan, KS

[18i]

Calcium oxide reactive nanoparticle Nanoscale Materials, Inc., Manhattan, KS [18g]Organic polymers USAMRICD, APG, MD and

TDA Research, Wheat Ridge,CO, proprietary compounds




Figure 26.3 shows the results from the critical DTN in vitro tests comparing SER-

CAS #¼Chemical Abstract Service Registry Numbers.

Transdermal Chemical Inhibitors

The emphasis for CWA chemical penetration retardation for the U.S. military has been onprotective suits and barrier skin creams. There was, however, a small effort to identifymaterials that would change the skin penetration rates to make the skin more resistant.

0

20

40

60

80

100

120

SERPAWCA Lead Backup

% T

oxi

city

Pig HDV Rabbit HDL Rabbit VXL Rabbit GDV GP GDL

* * * * * * *

Figure 26.2 Critical DTN in vivo tests comparing SERPACWA efficacy with the lead aTSPformulation and the backup aTSP formulation. Percentage of toxicity is a normalized efficacyscale where 100 is no protection and 0 is complete protection. * indicates significant (p < 0.05)improvement compared with SERPACWA. ‘‘L’’ with agent name indicates liquid, ‘‘V’’ indicatesvapor. The lead formulation containing organic polymers provided significantly improved pro-tection in every model. The backup formulation, containing S-330, provided improved protec-tion for HD, and equivalent protection for VX and GD.

0

20

40

60

80

100

120

SERPAWCA Lead Backup

Per

cen

tag

e o

f to

xici

ty

SPME-HDL SPME-GDL SPME-VXL WASH WIPE

* * ** *

Figure 26.3 Critical DTN in vitro tests comparing SERPACWA efficacy with the lead aTSPformulation and the backup aTSP formulation. Percentage of toxicity is a normalized efficacyscale where 100 means no destruction of agent in the SPME tests and easy removal in the washand wipe tests. * indicates significant (p < 0.05) improvement compared with SERPACWA. ‘‘L’’with agent name indicates liquid, ‘‘V’’ indicates vapor. The lead formulation containing organicpolymers provided significantly improved protection in every model except for the wash test.The backup formulation, containing S-330, provided improved protection for HD and VX andequivalent protection for GD in the neutralization tests. It also provided improved protection inthe wipe test and equivalent protection in the wash test.




A prototype compound, N-dodecanoyl-2-oxazolidone [23], known as ICD2837 wasevaluated in the weanling pig HD vapor test [18]. Because the compound was believed tooperate not as a physical barrier, but rather by actually modifying the penetrationcharacteristics of the stratum corneum, it was applied to the skin as a 2% (w/v) ethanolsolution. The application rate was 25 ml per 0.5 cm2. Experimental sites were challengedwith saturated HD vapor for 15 min instead of the standard 30-min challenge used in theDTN module. Figure 26.4 summarizes the test results. Compound ICD2837 providedsignificant protection when compared with the nonprotected control sites. The qualitycontrol standard in this test, ICD2289 (an early formulation of SERPACWA), providedbetter protection when spread at a thickness of 0.2 mm. However, when spread at 0.1mm, the current application rate for warfighters, it provided no protection (E.H. Braue, Jr.,Unpublished data, 1999). Thus, ICD2837 provided better protection than ICD2289.

When ICD2837 was used in combination with ICD2289 as a multilayer TSP, the resultswere disappointing. Compound ICD2837 was applied as a 2% solution in the weanlingpig model as described above. After a drying time of 2 h, ICD2289 was spread at either a

the results. Applying ICD2837 in combination with ICD2289, spread as a 0.2-mm layer,did not improve the efficacy afforded by ICD2289 alone. In fact, the protection ICD2837provided when applied alone was lost when combined with ICD2289 spread at 0.1 mm.The multilayer approach has possibilities as a concept, but has not yet been successfullyimplemented using the penetration modifying compound ICD2837.

Conclusions

Chemical Penetration Retardation is important to the U.S. military to reduce the threatfrom CWAs. Recent events demonstrate that the civilian population may also be apotential target. The most effective means for reducing or eliminating the percutaneous

0

1

2

3

4

5

6

7

8

9

Control STD 2289 2837

Rel

ativ

e er

yth

ema

*

*

Figure 26.4 Evaluation of ICD2837 challenged for 15 min with saturated HD vapor in theweanling pig test. Relative erythema is the mean (+SEM) difference between before and afterexposure a* reflectance values. Controls are positive control sites with no aTSP. STD 2289 is aquality control standard (0.2 mm thick) of ICD2289, an early SERPACWA formulation. 2837 isa 2% (w/v) ethanol solution of N-dodecanoyl-2-axazolidone applied at the rate of 25 ml per 0.5cm2. ICD2837 provided significantly (p < 0.05, n 5 6) better protection than the positive controlsites, which had no aTSP. The standard, ICD2289, provided significantly (p < 0.05, n 5 6) betterprotection than ICD2837.




0.1- or 0.2-mm layer and challenged with HD vapor for 15 min. Figure 26.5 summarizes

exposure to toxic materials including CWAs is the chemical protective suit. SERPACWA isa fielded passive barrier skin cream that increases the efficacy of the protective suit. Anew aTSP that incorporates a reactive moiety into the SERPACWA base cream wastransitioned to advanced development and should be ready for fielding in 7 to 10years. This new active product dramatically improves the protection from CWAs andmay change or reduce the need for a full protective suit in the future.

References1. Bartley, R.L. and Kucewicz, W.P., Foreign Affairs, 61, 805–826, 1993.2. UN Security Council, Report of the specialists appointed by the Secretary–General to investi-

gate allegations by the Islamic Republic of Iran concerning the use of chemical weapons,S/16433, 26 March 1984, 11–12.

3. (a) Spiers, E.M., Over 10,000 casualties were reported., Chemical and Biological Weapons:A Study in Proliferation, St Martin’s Press, New York, 1994, 18; (b) Kirkham, N., Cyanidebombers lay waste a town, The Daily Telegraph, 22 March 1988, 1.

4. Heller, C.E., Leavenworth Papers. Chemical Warfare in World War I: The American Experience,Combat Studies Institute, Fort Leavenworth, Kensas, 1984, 91–92.

5. Woodall, J., Lancet, 350, 296, 1997.6. Bamber, D., Hastings, C., and Syal, R., Bin Laden British cell planned gas attack on European

cessed January 2002).7. Takafuji, E.T. and Kok, A.B., in Textbook of Military Medicine, Medical Aspects of Chemical

and Biological Warfare, Sidell, F.R., Takafuji, E.T., and Franz, D.R. (Eds), Office of the SurgeonGeneral at TMM Publications, Washington, D.C., 1997, 118–119 and 609–610.

0

2

4

6

8

10

12

Contol STD 2289 2837/2289 2837/2289

Rel

ativ

e er

yth

ema

0.2 mm 0.2 mm

0.1 mm

* *

Figure 26.5 Evaluation of ICD2837 and ICD2289 in combination challenged for 15 min withsaturated HD vapor in the weanling pig test. Relative erythema is the mean (+SEM) differencebetween before and after exposure a* reflectance values. Controls are positive control sites withno aTSP. STD 2289 is a quality control standard (0.2 mm thick) of ICD2289, an early SERPACWAformulation. 2837/2289 is a combination of a 2% (w/v) ethanol solution of N-dodecanoyl-2-axazolidone (ICD2837) applied at the rate of 25 ml per 0.5 cm2 and ICD2289 applied as either a0.1- or 0.2-mm thick layer. Applying ICD2837 in combination with ICD2289, spread as a 0.2-mmlayer, did not improve the efficacy afforded by ICD2289 alone. In fact, the protection of ICD2837alone was lost when combined with ICD2289 spread at 0.1 mm. This was observed becauseICD2289 spread alone at 0.1 mm does not provide protection compared with control (Braue, Jr.E.H., Unpublished data, 1999).




Parliament. London Sunday Telegraph, 16 September 2001, http://www.telegraph.co.uk (ac-

http://www.telegraph.co.uk

8. Romano, J.A. Jr. et al. [Authors are Romano, McDonough, Sheridan, and Sidell], Health effectsof low–level exposure to nerve agents, in Chemical Warfare Agents: Toxicity at Low Levels,Somani, S.M. and Romano, J.A. Jr. (Eds), CRC Press, Boca Raton, FL, 2001, Chapter 1.

9. Odland, F.F., in Biochemistry and Physiology of the Skin, 1, Goldsmith, L.A. (Ed.), OxfordUniversity Press, New York, 1983.

10. O’Hern, M.R., Dashiell, T.R., and Tracey, M.F., in Textbook of Military Medicine, Part I: MedicalAspects of Chemical and Biological Warfare, Sidell, F.R., Takafuji, E.T., and Franz, D.R. (Eds),Office of the Surgeon General at TMM Publications, Washington, D.C., 1997, 144 and 205–208.

11. Axelrod, D.J. and Hamilton, J.G., Am. J. Pathol., 23, 389–411, 1947.12. O’Hern, M.R., Dashiell, T.R., and Tracey, M.F., in Textbook of Military Medicine, Part I: Medical

Aspects of Chemical and Biological Warfare, Sidell, F.R., Takafuji, E.T., and Franz, D.R. (Eds),Office of the Surgeon General at TMM Publications, Washington, D.C., 1997, 371–372.

13. Papirmeister B. et al. [Authors are Papirmeister, Feister, Robinson, and Ford], Medical Defenseagainst Mustard Gas: Toxic Mechanisms and Pharmacological Implications, CRC Press,Boca Raton, FL, 1991, 2.

14. Papirmeister B. et al. [Authors are Papirmeister, Feister, Robinson, and Ford], Medical Defenseagainst Mustard Gas: Toxic Mechanisms and Pharmacological Implications, CRC Press,Boca Raton, FL, 1991, 3.

15. Romano, J.R., United States Army Medical Research Institute of Chemical Defense, AberdeenProving Ground, MD, personal communication, 2001.

16. McCreery, M.J., U.S. Patent 5,607,979, 4 March 1997.17. Liu, D.K. et al. [Authors are Liu, Wannemacher, Snider, and Hayes], J. Appl. Toxicol., 19, S41–

S45, 1999.18. Braue, E.H. Jr., Development of a reactive topical skin protectant, J. Appl. Toxicol., 19, S47–

S53, 1999.19. Hobson, S.T., Lehnert, E.K., and Braue, E.H. Jr., The U.S. Army reactive topical skin protectant

(rTSP): challenges and successes, MRS Symposium Series CC: Hybrid Organic InorganicMaterials [Online], 628, CC10.8.1–CC10.8.8, 2000.

20. Patents covering this work: (a) Braue, E.H. Jr. et al. [Authors are Braue, Hobson, Govardhan,and Khalaf], Active Topical Skin Protectants Containing OPAA Enzymes and CLECs, U.S. Patent6,410,603, 25 June 2002; (b) Braue, E.H. Jr., et al. [Authors are Braue, E.H., Mershon, Braue,C.R., and Way], Active Topical Skin Protectants Containing S-330, U.S. Patent 6,472,438, 29October 2002; (c) Braue, E.H. Jr. et al. [Authors are Braue, Hobson, White, and Bley], ActiveTopical Skin Protectants Using Polyoxometalates, U.S. Patent 6,420,434, 16 July 2002; (d)Braue, E.H. Jr. et al. [Authors are Braue, Hobson, Hill, Boring, and Rhule], Active TopicalSkin Protectants Using Polyoxometalates and/or Coinage Metal Complexes, U.S. Patent6,414,039, 2 July 2002; (e) Braue, E.H. Jr., Hobson, S.T., and Lehnert, E.K., Active TopicalSkin Protectants, U.S. Patent 6,472,437, 27 October 2002; (f) Hobson, S.T., Braue, E.H. Jr., andBack, D., Active Topical Skin Protectants Using Polymer Coated Metal Alloys, U.S. Patent6,437,005, 20 August 2002; (g) Hobson, S.T. et al. [Authors are Hobson, Braue, Lehnert,Klabunde, Koper, and Decker], Active Topical Skin Protectants Using Reactive Nanoparticles,U.S. Patent 6,403,653, 11 June 2002; (h) Hobson, S.T., Braue, E.H. Jr., and Shea, K., ActiveTopical Skin Protectants Using Organic Inorganic Polysilsesquioxane Materials, U.S. Patent6,417,236, 9 July 2002; (i) Hobson, S.T. et al. [Authors are Hobson, Braue, Lehnert, Klabunde,Decker, Hill, Rhule, Boring, and Koper], Active Topical Skin Protectants Using Combinations ofReactive Nanoparticles and Polyoxometalates or Metal Salts, U.S. Patent 6,410,603, 25 June2002; and (j) Hill, C.L. et al. [Authors are Hill, Xu, Rhule, Boring, Hobson, and Braue],Polyoxometalate Materials, Metal-Containing Materials, and Methods of Use Thereof, U.S.Patent 6,723,349, 20 April 2004.

21. Hobson, S.T and Braue, E.H. Jr., Development of multifunctional perfluorinated polymerblends as an active barrier cream against chemical warfare agents, Polymeric Materials: Scienceand Engineering, 84, 80, 2001.




22. Snider, T.H., Matthews, M.C., and Braue, E.H. Jr., A model for assessing efficacy of topical skinprotectants against sulfur mustard vapor using hairless guinea pigs, Toxicol., 19, S55–S58,1999.

23. Peck, J.V., Minaskanian, G., and Hadgraft, J., U.S. Patent 6,086,905, 2000. Contact James E.




Sheldon, email to [email protected] for additional information.

mailto:[email protected]

COMMERCIAL

APPLICATIONS

OF PENETRATION

ENHANCERS

VI



Chapter 27

Preclinical and ClinicalDevelopment of aPenetration EnhancerSEPA 0009

Thomas C. K. Chan

CONTENTS

Introduction .................................................................................................................................... 402SEPA ................................................................................................................................................ 402Scientific Rationale for Designing SEPA-Type Compounds.......................................................... 402

Experimental Data ...................................................................................................................... 403Physical–Chemical Characteristics of SEPA................................................................................ 404How Formulation Affects Skin Permeation Performance ......................................................... 405

Early Successes with SEPA ............................................................................................................. 406Proof of Concept ........................................................................................................................ 406

In Vitro Absorption of Indomethacin .................................................................................... 406In Vivo Percutaneous Absorption .......................................................................................... 406Scalp Hair Growth in Balding Stumptail Macaque................................................................ 407

First Clinical Use of SEPA ........................................................................................................... 407Nonclinical Evaluations of SEPA .................................................................................................... 409Clinical Safety ................................................................................................................................. 412

Description of Selected Key Clinical Safety and Efficacy Studies............................................. 412Clinical Programs............................................................................................................................ 412

Opterone1 .............................................................................................................................. 412EcoNail ........................................................................................................................................ 414Topiglan1 ............................................................................................................................... 414



401


Introduction

Skin penetration enhancers have a long history of development, but none to date havebeen routinely incorporated into topical formulations. Limitations to their use haveincluded incompatibility with the drugs they are coupled with and general safety orlocal irritation issues. Numerous compounds have been evaluated for skin penetrationenhancing activity, including sulfoxides (such as dimethylsulfoxide, DMSO), Azone1

(e.g., laurocapram), pyrrolidones (for example, 2-pyrrolidone, 2P), alcohols and alkanols(ethanol and decanol), glycols (for example, propylene glycol, PG, a common excipientin topically applied dosage forms), surfactants (also common in dosage forms), andterpenes.

Penetration enhancers ideally are compatible with all drugs with which they areformulated with, pharmacologically inactive and without any potential local or systemicsafety concerns, and predictable in their activity to enhance penetration of drugs throughthe skin. Soft Enhancement of Percutaneous Absorption (SEPA1) is a series of suchenhancers. The ‘‘soft’’ in the name refers to the rapid breakdown of the enhancers,hence their reversible (nondamaging) effects on skin, more specifically on the stratumcorneum (SC).

SEPA

SEPA 0009 (Figure 27.1), 2-n-nonyl-1,3-dioxolane, belongs to a group of alkyl-substitutedacetals and cycloacetals (1,3 dioxolanes). Many members from these classes of agents areused as food additives for human consumption and served as synthetic flavoring sub-stances and adjuvants.

SEPA 0009 (hereafter refer to as SEPA) is synthesized by the condensation of ethyleneglycol and decyl aldehyde (decanal).

Scientific Rationale for Designing SEPA-Type Compounds

For effective transdermal drug delivery, topical vehicles have to overcome the naturalprotective function of the skin to allow transit of both large and small molecules withoutany adverse or permanent effects. SEPA enhances skin penetration by altering the fluidity

Figure 27.1 SEPA 0009.




of the lipid layers in the stratum corneum and temporarily alters the alignment of thoselipids, thereby reducing the diffusional barrier to allow drug molecules to penetratethrough the skin barrier.

Experimental Data

The interactions between SEPA and components of human SC were characterized usingFourier-transform infrared (FTIR) spectroscopy, differential scanning calorimetry (DSC),and scanning electron microscopy (SEM). Human abdominal and breast skin specimenswere obtained from healthy patients via a plastic surgery department in a hospital. The SCwas isolated from the whole skin after immersion in warm water.

FTIR spectra of SC samples, both SEPA-treated and untreated, were recorded for atleast five different skin samples over a range of temperatures. Table 27.1 shows theassignment of the bands present in the FTIR spectrum of human SC. The FTIR of SEPAhad several bands partially overlapping human SC. Therefore, to see possible changesinduced by SEPA on the SC lipid component, the SEPA spectra were subtracted from

treatment at different temperatures. FTIR at temperatures between 18 and 1208C showeda significant influence of SEPA treatment on human SC samples in the frequency andintensity of several absorption bands. This is attributed to an increase in mobility of thehydrocarbon chain of skin lipids, or an increased disorder of the lipid layer, coupled witha modification of protein hydrophobic interactions.

The DSC thermogram of human SC containing 30 w/w% of absorbed water has fourendothermic transitions at 35, 70, 85, and 1008C. Transitions at 40 and 708C are attributedto the melting of sebaceous and intercellular lipids, respectively. Transitions at 80 and1008C are attributed to lipid–protein interactions and keratin denaturation, respectively.Human SC samples treated with SEPA for 5 min produced thermograms in which the lipidtransitions almost completely disappeared, but displayed two new endotherms at lowertemperatures. When SC samples were treated with SEPA for 120 min, all peaks almostcompletely disappeared.

The ability of SEPA to affect SC lipid structure was strongly supported by the demon-strated influence of SEPA treatment on the intensity and position of SC lipid thermaltransitions shown by DSC analysis.

Table 27.1 Band Assignment in the FTIR Human SC

Frequency (cm1) Assignment

3500 to 3400 O–H stretching3400 to 3300 N–H stretching2959 CH3 asymmetrical stretching (lipids)2917 to 2928 CH2 asymmetrical stretching (lipids)2872 CH3 symmetrical stretching (lipids)2856 CH2 symmetrical stretching (lipids)1740 to 1730 Carbonyl ester C¼O stretching (proteins)1700 to 1600 (amide I) Peptide C¼O stretching (proteins)1600 to 1480 (amide II) Peptide N–H bending (proteins)1480 to 1330 CH2 and CH3 bending (lipids and proteins)1330 to 1190 (amide III) Peptide C–N stretching and N–H bending (proteins)


Preclinical and Clinical Development of SEPA 0009 & 403

those samples treated with SEPA. Figure 27.2 illustrates the fluidizing of SC after SEPA


These lipid changes were confirmed by SEM showing a significant loosening of the SCcorneocyte packing. Fragments of analyzed SC samples were dried under vacuum toconstant weight prior to DSC analysis. The vacuum treatment apparently removed theSEPA from the SC and the lipid structure was restored implying the reversibility of anyeffects of SEPA on human SC.

Physical–Chemical Characteristics of SEPA

Development of SEPA was targeted towards using the FDA designation of Generally

SEPA was synthesized by condensing ethylene glycol and decyl aldehyde (decanal).These components are rapidly metabolized after penetrating the SC into a fatty acidand a glycol.

3000 2900 28002950 2850

18C115C

Wavenumber (cm−1)

Abs

orba

nce

(a.u

.)

1800 1500 1200 900

18C115C

Wavenumber (cm−1)

Abs

orba

nce

(a.u

.)

Figure 27.2 FTIR spectra of human SC treated with SEPA after subtracting the SEPA spectrumrecorded at the same temperature.



Recognized As Safe (GRAS) (Figure 27.3), chemicals in its construction, and in fact


For maximum skin penetration, the number of carbon atoms in the chain was foundto be critical; 7 to 12 carbons were determined to be an optimal chain length. In orderto have a consistent approach to all potential formulations with SEPA, a nine carbonmember of the SEPA family was selected for commercial development as the bestcompromise that can effectively enhance the transdermal delivery of a wide variety ofactive pharmaceutical ingredients. SEPA is 2-n-nonyl-1,3-dioxolane, a clear, colorless oil.It freezes at 08C and has a boiling point of 89 to 908C. Its molecular weight is 200.31.

How Formulation Affects Skin Permeation Performance

Skin penetration enhancement is very much dependent on the overall formulationcarrying the drug. Small lipid-soluble molecules can partition into the stratum corneumand then diffuse across the lipid bilayers in membranes. However, water soluble mol-ecules cannot penetrate significantly, other than through aqueous pathways such asthose in sweat gland ducts and hair follicles. These pathways likely provide major routesfor iontophoresis (application of low voltage across skin to buffers to move ions andmolecules across the skin). Iontophoresis, coupled with a skin penetration enhancer, canresult in much increased transdermal flux.

Optimally formulated topical drugs can result in controlled release into the blood-stream through intact skin while avoiding the effects of first-pass metabolism in the liverand gastrointestinal tract following oral dosing. Such formulations provide increasedpatient convenience and compliance.

The mechanism of percutaneous penetration can be described by assuming that theskin is a semipermeable membrane (permeability coefficient¼ 1/resistance, R). Resist-ance can be due to the active ingredient, the vehicle in the formulation, characteristics ofthe SC, and other epidermal tissues and the dermis.

The ultimate rate of absorption is affected by molecular size and whether or not theactive ingredient is hydrophilic or lipophilic.

The final formulation containing the active ingredient is critical to successful percu-taneous penetration. Important formulation factors include a high concentration gradientof the active ingredient; viscosity of the vehicle; and activity of the penetration enhancer.The condition of the skin also affects absorption. Relevant factors can include skinthickness, hydration, temperature, and vascular perfusion.

A large number of drugs ranging in molecular weights up to about 1200 D have beenevaluated in vitro using SEPA. Substantially larger molecules appear unlikely to penetratehuman skin in sufficient quantity to be clinically useful. However, animal skin tends to be

H3C

H3C

O

O

OH

OH

OHC+

SEPA®

GRAS compound

Figure 27.3 Development of SEPA was targeted toward using GRAS precursors.




much more permeable than human skin — for example, cow skin is 400 times morepermeable — and so it is possible that absorption of larger molecules might be usefullyenhanced in veterinary medicine.

Early Successes with SEPA

Proof of Concept

In vivo and in vitro test systems have shown SEPA to be an effective transdermalpermeation enhancer. When formulated with other excipients, SEPA has increasedin vitro transdermal flux of agents as diverse as prostaglandins, nonsteroidal antiinflam-matory agents, steroids, hormones, vasodilators and others, through hairless mouse,porcine, or human skin. The mechanism of enhanced permeation appears to be areversible modification (liquefaction) of the lipid layers of the SC, allowing diffusion ofactive agents into the epidermis and further into the skin.

In Vitro Absorption of Indomethacin

An early study directly comparing trandermal delivery of the NSAID indomethacin bySEPA (2%) and by Azone (5%) was conducted using excised hairless rat skin undercontrolled conditions. Indomethacin was applied to the skin after it was dissolved inthe vehicle of ethanol and propylene glycol and varying amounts of the test enhancers.With no enhancer, the absorption of indomethacin through the skin was very low. Theaddition of SEPA to the solution increased absorption several fold over the control andthe solution containing Azone2. These results are summarized in Table 27.2.

In Vivo Percutaneous Absorption

Groups of hairless female rats were tested in a 1-week bioavailability study using theFeldmann and Maibach method.3 The test solutions were applied to the dorsal skin whichwas then covered with a nonocclusive device to prevent oral contamination. An intraper-itoneal injection of indomethacin served as a control to the test solutions which were: 100ml/5 cm2 of indomethacin; 100 ml/5 cm2 of indomethacin þ SEPA 5, 10, or 20%. The totalurinary excretion of indomethacin was 11.3, 13.5, 22.0, and 21.3%, corresponding to thesolution containing 0, 5, 10, and 20% v/v of SEPA, respectively. After i.p. injection, 87% ofthe injected dose was recovered in urine. The absorption enhancing effect for SEPAappeared to reach a plateau at 10%.

Table 27.2 In Vitro Percutaneous Absorption ofIndomethacin through Hairless Rat Skin SEPA vs Azone

Total amounts of applied indomethacin absorbed through rat skin (%)(mean + SD, n ¼ 6)

Control 7.7 + 3.7Azone (5%) 19.0 + 4.6SEPA (2%) 42.5 + 6.6




Scalp Hair Growth in Balding Stumptail Macaque4

This study to determine if SEPA would augment scalp hair growth effects of topicalminoxidil was conducted in groups of female monkeys. Four groups of five monkeyseach were treated once or twice a day with minoxidil or its vehicle; similar groups weretreated with SEPA vehicle or SEPA þ minoxidil for 16 weeks. The SEPA þ minoxidiltreatments resulted in significantly greater hair growth (as measured by hair weight)compared to the minoxidil-alone treatments. Steady-state urinary excretion of minoxidilwas greater from the SEPA þ minoxidil treated animals than from the minoxidil treatedanimals. The data suggest that SEPA influences the topical delivery of minoxidil in thisanimal model. It is not known if the increased hair weight is a result of improved skinpenetration of minoxidil or if the hair follicle itself is targeted by the SEPA þ minoxidilformulations.

First Clinical Use of SEPA

In view of the primate hair growth data, one of the first major clinical programs was aformulation of SEPA with minoxidil. This was at the time when topical minoxidil hadbeen shown to induce hair growth. The goal of the program was to maximize this activityusing SEPA. A major concern for any new excipient is that no new toxicity is introduceddue to interactions between it and the active, and that contact sensitization and irritationat the site of application do not occur.

An initial study was conducted to determine the potential of SEPA 5% þ Rogaine1

(minoxidil) to induce delayed contact hypersensitivity in human subjects. Under theconditions of a repeated insult (semiocclusive) patch test procedure, the SEPA or minox-idil formulation was very mildly to moderately irritating to approximately 42% (22/53) ofthe test panelists. There was no evidence of induced allergic contact dermatitis in thehuman subjects.

A study was conducted to determine the contact sensitization potential of topical 2.0%minoxidil in 2 and 5% SEPA vehicle. Panel A evaluated a formulation that contained 2%topical minoxidil in a 2% SEPA vehicle; Panel B evaluated a formulation that contained2% topical minoxidil in 5% SEPA vehicle. A total of 25 volunteers (healthy adults of bothsexes ranging in age from 19 to 49) were enrolled in each panel. No contact sensitizingpotential was demonstrated for any of the concentrations tested.

The skin sensitization potential of 2.5% nonaqueous minoxidil solution with 5, 10, or15% SEPA was determined using the Magnusson and Kligman Maximization Test designin 163 healthy volunteers.5 This was a double blind, evaluator-blinded, repeat insultpatch test. Both the 2.5% minoxidil solutions with 5 and 10% SEPA, respectively, wereconsidered to possess ‘‘weak’’ contact sensitizing potential in normal human skin, whilethe 2.5% minoxidil plus 15% SEPA solution was rated ‘‘weak to mild.’’ These results couldhave been the result of increased penetration of minoxidil, since minoxidil is thought tobe a sensitizer and previous studies with SEPA by itself have not shown sensitizationpotential.

A series of 10-day primary irritation studies in healthy volunteers using the techniqueoutlined by Kligman and Wooding6 have been conducted to determine cumulativeirritancy potential of minoxidil plus SEPA solutions. In the first, the primary irritationpotential of 2.5% nonaqueous minoxidil solution with either 2 or 5% SEPA was mild.Three other nonaqueous formulations were also tested. Cumulative irritation score for 5%minoxidil ranged from 5.0 to 12.0, with a mean of 8.4 (theoretical range: 0 to 40). Peeling




or desquamation was observed in three subjects. The desquamation was attributed tothe propylene glycol in the minoxidil vehicle, which is known to enhance exfoliationespecially under occlusive dressings. In addition, since previous studies have shown thatSEPA is essentially nonirritating to human skin, some of the irritation was attributed tominoxidil. A final set of formulations containing 2.5% minoxidil and 5, 10, or 15% SEPAwere also tested. The minoxidil formulation with up to 15% SEPA, and the correspondingvehicle solutions with SEPA, but without minoxidil, were found to have a mild degree ofprimary irritancy which were not significantly greater than a marketed facial moisturizer(negative control). In nearly all cases, the mean cumulative scores of the vehicle withSEPA were lower than those of either the corresponding formulations with minoxidil orthe marketed cosmetic. Since SEPA alone has been shown in other human studies toproduce only minimal erythema in cumulative irritation studies, the irritation observed inthis study with the SEPA vehicle may be due to the propylene glycol, which is known tobe a skin irritant.

The photocontact allergenic potential of 2.5% nonaqueous minoxidil solution with 5,10, or 15% SEPA was determined in 156 healthy Caucasian adult volunteers. The studyinvolved a period of exposure to the test solutions followed by a later challenge usingUVA light. There were no adverse reactions or unexpected side effects of any kind,except for mild to moderate erythema, scaling and tanning, which developed duringthe induction phase, and which are expected responses following repeated exposures toultraviolet radiation. Following the challenge, there were erythematous and pruriticreactions in those groups with higher concentrations of SEPA that were very suggestiveof delayed contact hypersensitivity. All symptoms resolved uneventfully within 7 to 8days. These findings seem to indicate that concomitant UV light exposure may enhanceor amplify the contact sensitizing capacity of 2.5% minoxidil in 10 or 15% SEPA. In bothcases, however, the sensitization rates are low and both treatments would be classified ashaving a ‘‘mild’’ potential for sensitization.

The potential for phototoxicity with 2.5% minoxidil in a nonaqueous formulationcontaining either 5, 10, or 15% SEPA was measured by occlusive patch testing in 30healthy adults (11 males and 19 females). After 6 h of exposure, half the patches wereremoved and exposed to 20 J/cm2 long wave ultraviolet (UVA 320 to 400 nm) while theother half of the patches served as nonirradiated controls. No phototoxicity was observedin any of the test groups. No skin reactions or abnormal responses were recorded at anytimepoint after irradiation with UVA. These results showed no indication for potentialphototoxicity in any of the SEPA formulations with minoxidil.

Nine healthy volunteers with male pattern baldness completed a pharmacokineticevaluation of SEPA–minoxidil formulations. Subjects were randomized into three groups,and received twice-daily applications of 1 ml of the formulation applied to 100 cm2 of thescalp for 3 weeks. The three nonaqueous formulations contained SEPA at 5, 10, or 15% inthe vehicle (equivalent to 100, 200, or 300 mg/d of SEPA), respectively.

Blood samples taken prior to treatment on day 21 revealed that serum concentrationsof SEPA go up in a dose-related manner (mean values were 3.48, 4.18 and 9.41 ng/mlfor the groups receiving 5, 10, and 15% SEPA in the vehicle, respectively). After dosing,SEPA concentrations had increased at 1 h postdose, and remained in the same generalrange for up to 4 h. At 8 h postdose, mean concentrations were declining. SEPAconcentrations were below or approaching the limits of quantification (2.5 ng/ml) by24 h in the subjects receiving 5% SEPA, by 48 h in the subjects receiving 10% SEPA, and by96 h in subjects receiving 15% SEPA. Mean AUC(0-inf) increased in a generally dose-proportional manner. Half-lives ranged from 6 to 12 h for the 5% SEPA group to 21 to




37 h for the 15% SEPA group. This is considerably less than that noted in nonclinicalstudies using labeled material, and the nonclinical values may reflect species differencesas well as the longer elimination of SEPA metabolites. An additional significant note is thatthese values obtained with clinical formulations are vastly lower than those seen in thenonclinical toxicology studies, suggesting a wide margin of safety for SEPA.

Finally, the Upjohn Company in collaboration with MacroChem conducted three large32-week efficacy (for hair growth) and safety studies that exposed over 1900 male andfemale patients with androgenous alopecia to minoxidil or SEPA formulations. Theproject was stopped because the primary endpoint, an increase in follicular count, wasnot met. However, the hair that did grow in the SEPA or minoxidil group was thicker byweight. There is no indication that there were any safety concerns.

Nonclinical Evaluations of SEPA

SEPA has undergone extensive testing similar to that of a new chemical entity. As such, ithas been evaluated in nonclinical studies that have included pharmacological activityscreens, pharmacokinetic, biodistribution and metabolism studies, acute and repeat dosetoxicity studies, and mutagenicity, carcinogenicity and reproductive toxicity studies.

The outcome of this extensive testing programpermits the following broad conclusions:

1. SEPA is essentially pharmacologically inert.2. SEPA increases the absorption of topically administered drugs.3. SEPA has been evaluated in clinical and nonclinical studies, all of which have

demonstrated that topically applied SEPA, while clearly absorbed into the systemiccirculation of both man and experimental animals, has not produced systemictoxicity.

4. The results of the toxicity studies have shown that formulations containing up to10% SEPA (w/v) possess no systemic toxicological effects. However, some formu-lations and concentrations of SEPA can be irritating to the skin of certain species,especially in fur bearing mammals.

5. SEPA is nonmutagenic, nonteratogenic.6. In a 2-year rat study, SEPA was shown to be noncarcinogenic. In a 26-week

oncogenicity study in Tg.AC transgenic mouse model, animals exhibited moderateto marked desquamation (large flakes to denudement) after being treated forseveral weeks at 250 and 1500 mg/kg/d. Papilloma development was observedfollowing dermal desquamation in these mice. Drug safety experts have examinedthe safety data package on SEPA and concluded that this response is related to thefrank skin injury and chronic inflammation seen at these high doses and not toSEPA. At lower doses up to 50 mg/kg/d, no desquamation and no subsequentpapilloma development were noted.

7. The chemistry of SEPA suggests that the most likely route of metabolism is openingof the dioxolane ring at the labile acetal carbon, with resultant formation ofethylene glycol and decanal. Literature reports on the metabolism of similar com-pounds, including long chain cyclic acetals of glycerol, budesonide (a topicalglucocorticoid), 1,3 benzodioxole, and doxophylline support this hypothesis. De-canal is a common, naturally occurring fatty acid aldehyde, approved by FDA as aGRAS substance used in foods. Because of its low toxicity and rapid metabolism to



These studies are outlined in Table 27.3.


Table 27.3 Summary of SEPA Testing Program (Data on File, MacroChem Corporation)

Test Species (N/Dose) RouteSEPA Dose andFormulation Findings

Pharmacology Mice>80 activity screens

PO; IP;In vitro

1000; 100 mg/kg of100% SEPA

MTD > 1000 mg/kg PO; NOEL ¼ 30 mg/kg IPNo significant pharmacological activity

PK (metabolism) Rats IP 14C-label Major metabolite in urine identified asethylene glycol and decanoic acid

PK (metabolites) Rats IP 2-[14C]-label 48 h postdose: 62% of administeredSEPA-dose recovered in urine asethylene glycol. Other urinemetabolites dioxolane-ring-intact

Toxicology (acute) Rats (5) IP 1250 to 5000 mg/kg Mild to noted lethargy for 1.25,2.5 g/kg at 24 h; recovery by 72 h.2.5 g/kg < LD50 < 5 g/kg

Toxicology (acute)Intact and abraded skin

Rabbits (10, 3) Topical 2000 mg/kg of100% SEPA

No systemic toxicityLD50 > 2 g/kg; dermal irritation(slight redness at 24 h, pustules anddry skin at 5 d)

Primary dermal irritation Rabbits (3) Topical 2000 mg/kg of100% SEPA

Not a primary irritant(score of 0.08 out of 8)

Skin sensitization Guinea pigs (20) Topical 100% SEPA NonsensitizingMutagenicity (+S9) TA98, TA100, TA1535,

TA15375, andE. coli WP2 uvrA

In vitro Up to 5 mg/plate Nonmutagenic

Mutagenicity (+S9) TA97a, TA100,TA102, TA1535

In vitro Up to 0.15 mg/plate Nonmutagenic

Micronucleus assay(daily 3d)

Mice (5M, 5F) PO Up to 5000 mg/kg LD50 ~ 5 g/kg/dNo increase in micronuclei

Smith

andMaib

ach/Percu

taneousPenetratio

nEnhan

cers

2ndedn

TF2152_c027

Final

Proof

page

410

8.9.2005

11:39pm

410

&Percu

taneousPenetratio

nEn

han

cers


Unscheduled DNA synthesis(UDS) assay

Rat hepatocytes In vitro Up to 0.3 mg/ml No increase in UDS

Developmental toxicology(embryo-fetal development)

Rabbits(18 time-mated F)

Topical 2.5 to 10% in IPA No developmental toxicity, no teratogenicity

Developmental toxicology(embryo-fetal development)

Rats (26 time-mated F) Topical 3 to 30% in IPA No teratogenicity. At 10 and 30% doserelated increase in incidence ofdevelopmental variation, rudimentary rib.May be secondary to F0 toxicity

Fertility and developmentaltoxicity (though implantation)

Rats (30M, 30F) Topical(alternatingdosing sites)

1 to 10% SEPAin IPA

Fewer implants and liver fetuses at 10% SEPA(considered secondary to stress assoc withdermal irritation)

No developmental toxicity No teratogenicityFertility and developmental

toxicity (pre- and postnataldevelopment)

Rats (25F/group) Topical 0 to 395 mg/kg/din IPA

NOEL for reduced gestation BW was 117 mg/kgNOEL for reduced BW gain was 395 mg/kgF1 offspring: no effects on developmental,behavioral or reproductive parameters.F1 NOEL set at 395 mg/kg

Oncogenicity (daily 26 weeks) Tg.AC Mice (20M, 20F) Topical 10 to 1500mg/kgin acetone

Nononcogenic at doses where no dermaleffects noted (up to 50mg/kg/d)

Moderate to marked desquamation(large flakes to denudement) at 250and 1500mg/kg/d

Papillomas associated with dermaldesquamation (In Tg. AC mice, abrasions,dermal irritation, wounding resultsin papillomas)

Toxicology and oncogenicity(daily 2 year)

Rat (50M, 50F) Topical 39 to 395 mg/kgin IPA

NoncarcinogenicNo systemic toxicity

(includes complete microscopic evaluation)

Smith

andMaib

ach/Percu

taneousPenetratio

nEnhan

cers

2ndedn

TF2152_c027

Final

Proof

page

411

8.9.2005

11:39pm

Preclin

icalan

dClin

icalDeve

lopmentofSEPA

0009&

411


decanoic acid, the small amount of decanal resulting from absorption of SEPA is notexpected to present any health risk to humans.

8. Pharmacokinetic studies with radiolabeled-SEPA have indicated SEPA is readilyabsorbed through the skin, rapidly distributed to tissues, and metabolized to CO2

and ethylene glycol. The Environmental Protection Agency sets 2 mg/kg/d as thepermissible exposure for ethylene glycol. This level is considerably higher thanlevels detected in our studies.

9. Metabolism studies in animals have suggested that the principal metabolites ofSEPA after systemic absorption are ethylene glycol and decanoic acid (a simple fattyacid). Long-term toxicity evaluation by the National Toxicology Program of theNational Institutes of Health (NIH) has shown ethylene glycol not to be a rodentcarcinogen.

Clinical Safety

Description of Selected Key Clinical Safety and Efficacy Studies

There have been over 4000 human subjects who were exposed to SEPA alone or incombination with a variety of known active drugs. There was no systemic toxicity that isattributed to SEPA in these human subjects.

Of most import, in standard tests for delayed contact sensitivity and primary irritation,exposure to SEPA alone showed SEPA to be free of these properties.

Briefly, the first study determined the potential of SEPA to induce delayed contacthypersensitivity in healthy volunteers. The SEPA formulation consisted of a 30% solutionin light mineral oil. A 0.2 ml volume of this SEPA formulation was applied to a Parke-Davis occlusive patch, and the patch applied to the back of each subject. Under theconditions of a repeated insult (occlusive) patch test procedure, SEPA (30% v/v in lightmineral oil) did not induce clinically significant irritation nor was there any evidence ofinduced allergic contact dermatitis in human subjects.

The second study determined the potential cumulative irritation associated withformulations of SEPA (2, 5, 15, and 30% in light mineral oil; a fifth formulation wasmineral oil only) under occlusive and semiocclusive conditions as well with no occlusion.Twenty-two healthy volunteers, 19 to 63 years of age, participated. Total cumulativeirritation scores showed that the four SEPA formulations did not have cumulative irritationscores higher than that obtained with the mineral oil alone. Thus, the four SEPA formu-lations would not be considered to have clinically significant cumulative irritation poten-tial under occlusive or semiocclusive conditions, and no irritation potential when appliedwith no occlusion.

Clinical Programs

Opterone1

Opterone is a SEPA-enhanced topical testosterone cream formulation developed to treatmale hypogonadism (testosterone deficiency). This is a condition that is often under-diagnosed clinically, but discovered by chance when a hormone workup is orderedor when a patient complains of erectile dysfunction (ED). Hypogonadism is diagnosedwhen a patient’s serum total testosterone concentration is lower than 300 ng/dl, together




with one or more of the following presenting symptoms: impotence and decreasedsexual desire; fatigue and low motivation, mood depression; regression of secondarysexual characteristics; and osteoporosis.

Testosterone is an example of a drug candidate that is ideally suited for transdermaldelivery. Oral testosterone delivery is not practical because of significant first-passmetabolism, and potential for liver injury during chronic oral administration. Deepintramuscular injections of testosterone tend to be painful and require a clinic visit foreach dose. Furthermore, the initial supra-physiological testosterone peak in the systemiccirculation is often associated with outbreak of acne, aggression, and unwanted hairgrowth, while the serum testosterone trough seen during the week before the next doseis associated with the undesirable symptoms mentioned above. Early attempts at trans-dermal testosterone delivery were in the form of skin patches, but patient acceptance waslow due to an unacceptably high level of skin irritation. The first topical testosterone gelcame to the market in 2001, and because of good patient acceptance, expanded thetopical testosterone therapy market to annual sales of more than US$300 million. Twoother topical testosterone products, one gel and one buccal adhesive tablet, have beenapproved for marketing in the US recently.

In vitro transdermal drug delivery experiments using human cadaver skin in staticdiffusion cells (Figure 27.4) showed that formulations containing SEPA deliver 200 to500% more testosterone per gram of applied dose over a 24-h period. An early clinicalpharmacokinetics study in hypogonadal males using a first-generation testosterone orSEPA hydroalcoholic gel showed that 2.5 g of the testosterone or SEPA gel deliveredequivalent amounts of testosterone systemically when compared to 5.0 g of the marketedgel. While the increased delivery of testosterone was desirable, serum levels of totaltestosterone returned to baseline by 12 h. A second generation testosterone cream wasdeveloped that has demonstrated comparable enhancement of drug delivery in vitro

Mean ± SE. (n = 24)

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

0 2 4 6 8 10 12 14 16 18

Time (h)

Tes

tost

ero

ne

flu

x (m

g/c

m2 /

h)

Androgel: 1%T Macro gel: 1%T, 5%SEP A Opterone: 1%T, 5%SEP A

Figure 27.4 Transdermal drug penetration data from human cadaver skin using a static Franzdiffusion cell.




but exhibits a more extended delivery profile. This new Opterone cream is currentlyundergoing clinical testing in hypogonadal patients.

EcoNail

EcoNail is a SEPA-based nail lacquer containing econazole, a well-known antifungaldrug, being developed to treat onychomycosis. The mechanism of action of SEPA inthis formulation is different from its general lipid fluidization activity in the stratumcorneum. In this formulation, SEPA promotes the release of econazole from the lacquerfilm. Econazole has been shown to penetrate human nail well.7 The presence of SEPA inthe lacquer softens the matrix to allow econazole to diffuse out of the lacquer to create alarge concentration gradient at the lacquer–nail interface. The durability of the lacquer,which provides an occlusive environment over the nail, combined with econazole’sability to penetrate the nail plate, resulted in delivery of very high concentrations ofeconazole into the ventral nail plate and into the nail bed.8 In the same experiments, wehave demonstrated that radiolabeled SEPA did not penetrate human nails. Currenttreatments of onychomycosis include orally administered antifungal drugs which havethe potential of injuring the liver, and a topical lacquer which has a combined myco-logical and clinical cure rate of approximately 7%. EcoNail is currently undergoingclinical testing in onychomycosis patients.

Topiglan1

Topiglan was originally formulated as a topical hydroalcoholic gel containing SEPA (5%)and alprostadil (1%) intended to treat male ED. Early studies suggested efficacy in asingle-dose, randomized, placebo-controlled office study among men with moderate-to-severe ED using subjective observation criteria in a clinical setting (Goldstein et al.). In alarger ‘‘at home’’ study in 541 patients with moderate-to-severe ED, Topiglan did notshow statistically significant clinical activity in the intent-to-treat patient population, butsuggested improvement of erectile function in a subset of protocol-conforming patients.The most commonly encountered side effects were associated with application sitediscomfort. As a consequence, the formulation was reformulated from a gel to a creamthat contained less alcohol with comparable transdermal drug delivery characteristics.This new cream elicited excellent erectile responses in experiments performed in ananimal model.9 A recently completed penile tumescence pharmacodynamic study inmild-to-moderate ED patients showed that Topiglan did not meet its clinical endpoint.There are many possible explanations for such disparate results between animals andhumans. One potential explanation is that penile circulation in animals may be anatom-ically different from human circulation. In humans, it is possible that alprostadil wasremoved by cutaneous blood vessels before it could diffuse into the corpora cavernosa toelicit an erectile response.

Conclusions

Inert, nontoxic chemical penetration enhancers can play a major role in delivering drugsthrough skin. The wide range of molecules that transdermal penetration enhancers suchas SEPA can deliver will improve the possibilities of topical delivery to local or systemic




targets for a growing range of active pharmaceutical ingredients. The clinical need foralternatives to solid-dosage oral delivery has never been greater, as the populations of themost industrialized nations become older and experience increasing difficulty swallow-ing tablets and capsules. Between 50 and 75% of America’s 3 million nursing homeresidents have some difficulty in swallowing. Pediatric patients also fall into this samecategory as do patients in chemo and radiotherapies or those patients recovering aftersurgery.

Regulatory positioning of chemical enhancers today remains somewhat of a ‘‘blackhole,’’ but as manufacturers accept the need to validate the safety of those products, andthe FDA becomes more comfortable with the data generated, we will see a wide range ofabsorption-enhanced products reaching the market in the near future. There are severalSEPA-based product candidates under consideration for clinical development to treatlocal and systemic diseases. Due to the very nature of transdermal drug delivery, appli-cation to a very large skin area is generally not well received by patients. Products thatrequire the application to a small skin area or those that are designed to act locally at theapplication sites are most suited for SEPA-based delivery. In addition to pharmaceuticals,the enhanced delivery of cosmetic ingredients and personal care products represents yetanother area underserved by current drug delivery technologies.

References1. Thermal and spectroscopic characterization of interactions between 2-nonyl-1,3-dioxolane and

stratum corneum components. Morganti, F et al. Journal of Bioactive and Compatible Polymers14: 162–177, 1999.

2. Enhancement of indomethacin percutaneous absorption effect of 2-n-nonyl-1,3-dioxolane.Doucet, O, Hagar, H, and Marty, JP. STP Pharmacetical Sciences 1(1): 89–93, 1991.

3. Percutaneous penetration of some pesticides and herbicides in man. Feldmann, RJ and Mai-bach, HI. Toxicology and Applied Pharmacology 28(1): 126–132, 1974.

4. The penetration enhancer SEPA augments stimulation of scalp hair growth by topical minoxidilin the balding stumptail macaque. Diani, AR, Shull, K, Zaya, MJ, and Brunden MN. SkinPharmacology 8: 221–228, 1995.

5. The identification of contact allergens by animal assay. The guinea pig maximization test.Magnusson, B and Kligman, AM. Journal of Investigative Dermatology 52(3): 268–276, 1969.

6. A method for the measurement and evaluation of irritants on human skin. Kligman, AM andWooding, WM. Journal of Investigative Dermatology 49: 78–94, 1967.

7. Nail penetration: focus on topical delivery of antifungal drugs for onychomycosis treatment.Sun, Y et al. in Topical Absorption of Dermatological Products, Bronaugh and Maibach, Eds,pp. 437–458, Marcel Dekker, Inc., New York, 2002.

8. Enhanced econazole penetration into human nail by 2-n-nonyl-1,3-dioxolane. Hui, X et al.Journal of Pharmaceutical Sciences 92: 142–148, 2003.

9. Feline penile erection induced by topical glans penis application of combination alprostadil andSEPA (Topiglan). Usta, MF et al. International Journal of Impotence Research 16: 73–77, 2004.




Percutaneous Penetration Enhancers

Documents

Transcript of Percutaneous Penetration Enhancers