Liposomes - Hindawi Publishing Corporationdownloads.hindawi.com/journals/focusissues/820732.pdf ·...

Liposomes

Journal of Drug Delivery

Liposomes

Journal of Drug Delivery

Liposomes

Copyright © 2011 Hindawi Publishing Corporation. All rights reserved.

This is a focus issue published in volume 2011 of “Journal of Drug Delivery.” All articles are open access articles distributed under theCreative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided theoriginal work is properly cited.

Editorial Board

Sophia Antimisiaris, GreeceAbdul Basit, UKE. Batrakova, USAHeather Benson, AustraliaA. Bernkop-Schnurch, AustriaGuru V. Betageri, USAMarıa J. Blanco-Prieto, SpainG. Buckton, UKYılmaz Capan, TurkeyCarla Caramella, ItalyRoberta Cavalli, ItalyNevin Celeby, TurkeyRita Cortesi, ItalyAlekha K. Dash, USAMartin J. D’Souza, USAJeanetta du Plessis, South AfricaN. D. Eddington, USAA. Fadda, ItalyJia You Fang, TaiwanSven Frøkjær, DenmarkSanjay Garg, New ZealandAndrea Gazzaniga, Italy

Richard A. Gemeinhart, USALisbeth Illum, UKJuan M. Irache, SpainBhaskara R. Jasti, USAHans E. Junginger, ThailandDae-Duk Kim, Republic of KoreaYellela S. R. Krishnaiah, USAVinod Labhasetwar, USAClaus S. Larsen, DenmarkKang Choon Lee, USALee-Yong Lim, AustraliaRam I. Mahato, USAPhilippe Maincent, FranceEdith Mathiowitz, USAReza Mehvar, USABozena Michniak-Kohn, USATamara Minko, USAAmbikanandan Misra, IndiaA. K. Mitra, USAS. M. Moghimi, DenmarkA. Mullertz, DenmarkSteven H. Neau, USA

Ali Nokhodchi, UKAbdelwahab Omri, CanadaR. Pignatello, ItalyViness Pillay, South AfricaMorteza Rafiee-Tehrani, IranMichael Roberts, AustraliaPatrick J. Sinko, USAJohn Smart, UKQuentin R. Smith, USAHartwig Steckel, GermanySnow Stolnik-Trenkic, UKK. Takayama, JapanHirofumi Takeuchi, JapanIstvan Toth, AustraliaHasan Uludag, CanadaClaudia Valenta, AustriaJaleh Varshosaz, IranSubbu S. Venkatraman, SingaporeS. P. Vyas, IndiaChi H. Wang, SingaporeAdrian Williams, UKP. York, UK

Contents

A Liposomal Formulation Able to Incorporate a High Content of Paclitaxel and Exert PromisingAnticancer Effect, Pei Kan, Chih-Wan Tsao, Ae-June Wang, Wu-Chou Su, and Hsiang-Fa LiangVolume 2011, Article ID 629234, 9 pages

Targeted Liposomal Drug Delivery to Monocytes and Macrophages, Ciara Kelly, Caroline Jefferies,and Sally-Ann CryanVolume 2011, Article ID 727241, 11 pages

Characterization and In Vitro Skin Permeation of Meloxicam-Loaded Liposomes versus Transfersomes,Sureewan Duangjit, Praneet Opanasopit, Theerasak Rojanarata, and Tanasait NgawhirunpatVolume 2011, Article ID 418316, 9 pages

Liposome Technology for Industrial Purposes, Andreas Wagner and Karola Vorauer-UhlVolume 2011, Article ID 591325, 9 pages

Cellular Injury of Cardiomyocytes during Hepatocyte Growth Factor Gene Transfection withUltrasound-Triggered Bubble Liposome Destruction, Kazuo Komamura, Rie Tatsumi,Yuko Tsujita-Kuroda, Takatoshi Onoe, Kunio Matsumoto, Toshikazu Nakamura, Jun-ichi Miyazaki,Takeshi Horio, and Masaru SugimachiVolume 2011, Article ID 453619, 8 pages

In Vitro Gene Delivery Mediated by Asialofetuin-Appended Cationic Liposomes Associated withγ-Cyclodextrin into Hepatocytes, Keiichi Motoyama, Yoshihiro Nakashima, Yukihiko Aramaki,Fumitoshi Hirayama, Kaneto Uekama, and Hidetoshi ArimaVolume 2011, Article ID 476137, 13 pages

Effects of Polyethylene Glycol Spacer Length and Ligand Density on Folate Receptor Targeting ofLiposomal Doxorubicin In Vitro, Kumi Kawano and Yoshie MaitaniVolume 2011, Article ID 160967, 6 pages

Liposomes for Use in Gene Delivery, Daniel A. Balazs and WT. GodbeyVolume 2011, Article ID 326497, 12 pages

Liposome Model Systems to Study the Endosomal Escape of Cell-Penetrating Peptides: Transport acrossPhospholipid Membranes Induced by a Proton Gradient, Fatemeh Madani, Alex Peralvarez-Marın,and Astrid GraslundVolume 2011, Article ID 897592, 7 pages

Liposomal Tumor Targeting in Drug Delivery Utilizing MMP-2- and MMP-9-Binding Ligands,Oula Penate Medina, Merja Haikola, Marja Tahtinen, Ilkka Simpura, Sami Kaukinen, Heli Valtanen,Ying Zhu, Sari Kuosmanen, Wei Cao, Justus Reunanen, Tuula Nurminen, Per E. J. Saris, Peter Smith-Jones,Michelle Bradbury, Steven Larson, and Kalevi KairemoVolume 2011, Article ID 160515, 9 pages

Antibody-Hapten Recognition at the Surface of Functionalized Liposomes Studied by SPR: StericHindrance of Pegylated Phospholipids in Stealth Liposomes Prepared for Targeted RadionuclideDelivery, Eliot. P. Botosoa, Mike Maillasson, Marie Mougin-Degraef, Patricia Remaud-Le Saec,Jean-Francois Gestin, Yannick Jacques, Jacques Barbet, and Alain Faivre-ChauvetVolume 2011, Article ID 368535, 9 pages

A Review on Composite Liposomal Technologies for Specialized Drug Delivery, Maluta S. Mufamadi,Viness Pillay, Yahya E. Choonara, Lisa C. Du Toit, Girish Modi, Dinesh Naidoo, and Valence M. K. NdesendoVolume 2011, Article ID 939851, 19 pages

Recent Applications of Liposomes in Ophthalmic Drug Delivery, Gyan P. Mishra, Mahuya Bagui,Viral Tamboli, and Ashim K. MitraVolume 2011, Article ID 863734, 14 pages

Preparation and Characterization of Stealth Archaeosomes Based on a Synthetic PEGylated ArchaealTetraether Lipid, Julie Barbeau, Sandrine Cammas-Marion, Pierrick Auvray, and Thierry BenvegnuVolume 2011, Article ID 396068, 11 pages

Hindawi Publishing CorporationJournal of Drug DeliveryVolume 2011, Article ID 629234, 9 pagesdoi:10.1155/2011/629234

Research Article

A Liposomal Formulation Able to Incorporate a High Content ofPaclitaxel and Exert Promising Anticancer Effect

Pei Kan,1 Chih-Wan Tsao,1 Ae-June Wang,1 Wu-Chou Su,2 and Hsiang-Fa Liang1

1 Drug Delivery Lab, Biomedical Engineering Research Laboratories, Industrial Technology Research Institute, Hsinchu 31040, Taiwan2 Medical College and Hospital, National Cheng Kung University, Tainan 701, Taiwan

Correspondence should be addressed to Hsiang-Fa Liang, [email protected]

Received 24 June 2010; Revised 14 September 2010; Accepted 17 September 2010

Academic Editor: Kozo Takayama

Copyright © 2011 Pei Kan et al. This is an open access article distributed under the Creative Commons Attribution License, whichpermits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

A liposome formulation for paclitaxel was developed in this study. The liposomes, composed of naturally unsaturated andhydrogenated phosphatidylcholines, with significant phase transition temperature difference, were prepared and characterized.The liposomes exhibited a high content of paclitaxel, which was incorporated within the segregated microdomains coexisting onphospholipid bilayer of liposomes. As much as 15% paclitaxel to phospholipid molar ratio were attained without precipitatesobserved during preparation. In addition, the liposomes remained stable in liquid form at 4◦C for at least 6 months. The specialcomposition of liposomal membrane which could reduce paclitaxel aggregation could account for such a capacity and stability.The cytotoxicity of prepared paclitaxel liposomes on the colon cancer C-26 cell culture was comparable to Taxol. Acute toxicitytest revealed that LD50 for intravenous bolus injection in mice exceeded by 40 mg/kg. In antitumor efficacy study, the preparedliposomal paclitaxel demonstrated the increase in the efficacy against human cancer in animal model. Taken together, the novelformulated liposomes can incorporate high content of paclitaxel, remaining stable for long-term storage. These animal data alsodemonstrate that the liposomal paclitaxel is promising for further clinical use.

1. Introduction

Paclitaxel, an effective anticancer agent, has been appliedas the first-line drug against breast and ovarian cancers.However, more extensive clinical use is limited owing tothe drug’s low water solubility and the highly inflammatoryresponse to the current excipient, cremophore EL [1]. Thus,much effort has been made in eliminating the side effectsduring administration. A variety of formulations have beendeveloped to replace cremophore EL [2–12]. Among thoseformulations, liposome is regarded as one of the mostpromising drug carrier. It has many advantages over otherformulations, such as being the most biocompatible andbest able to reduce drug toxicity without changing drugefficacy against tumor cells. However, limited drug loadingand insufficient shelf stability remain prohibitive obstacles topractical application [1, 2, 12].

Conventional paclitaxel liposomes were prepared at aconfined paclitaxel/lipid molar ratio of approximately 3%,regardless of whether the liposomes were made of a mixture

of phosphatidyl glycerol (PG) [13, 14] or DOTAP (1,2-dioleoyl-3-trimethylammonium propane) [15] and phos-phatidyl choline (PC), or unsaturated [16, 17] or partiallyunsaturated PC [18]. At a drug-to-lipid molar ratio of 4%,the paclitaxel-liposomes are stable for only 2 days. Dur-ing preparation of paclitaxel liposomes, needle-like crystalprecipitates appear at a drug/lipid molar ratio up to 8%[19]. Incorporation of the hydrophilic polymer conjugatedphospholipid (methoxy polyethylene glycol-phosphatidylethanolamine), known to be able to stabilize liposomes andextend its circulation time in the bloodstream, that wasattempted [20]. But the PEGylated liposomes with a maximalpaclitaxel/lipid molar ratio of 3% quickly become unstable inone week of storage at 4◦C. On the other hand, the liposomalformulations of paclitaxel consisting of a special negativelycharged phospholipid, cardiolipid, and phosphatidyl cholinehave been described [21]. Increasing the paclitaxel/lipidmolar ratio to 9% causes the liposomes to be stable for onlyone month when stored in liquid form at 4◦C [21]. For the3% drug-to-lipid ratio of liposomal paclitaxel, it needs many

2 Journal of Drug Delivery

lipids to formulate, thus increases the cost of lipids and thevolume for injection in clinical, and would be significantlymore expensive than the commercial product (Taxol).

Korlach et al. reported the presence of a phaseseparation in giant unilamellar vesicles composed ofDPPC/DLPC/cholesterol was visualized [22]. It was specu-lated that there are many segregated microdomains coexist-ing on the membrane of liposomes constituted by two dif-ferent kinds of lipids. We hypothesized these microdomainsmight prevent the aggregation of hydrophobic drug to formcrystal precipitates.

The aim of the study was to develop a novel liposomalformulation, composed of naturally unsaturated and hydro-genated PC with significant phase transition temperaturedifference, capable of incorporating high paclitaxel content.The influences of the feeding ratio of hydrogenated PC tototal PC and the drug-to-lipid ratio on the particles size,drug incorporation efficiency, phase transition temperature,and the storage stability were evaluated. Additionally, in vitrocytotoxicity of prepared paclitaxel-loaded liposomes on C-26 colon cancer cell line was estimated. Moreover, plasmaexposure and acute toxicity of the paclitaxel liposomeswere studied in vivo. Finally, the antitumor efficacy of thepaclitaxel liposomes in PC14PE6/AS2 bearing nude mice wasalso examined.

2. Materials and Methods

2.1. Materials. Egg phosphatidylcholine (EPC, Lipoid E100),and hydrogenated egg phosphatidylcholine (HEPC, LipoidE PC-3) were obtained from Lipoid GmbH. Hydrogenatedsoy phosphatidylcholine (HSPC, Epikuron 200 SH) wereobtained from Lucas Meyer GmbH. Paclitaxel was purchasedfrom Hauser Chemical Res, Inc. Methoxy polyethyleneglycol 2000-disteary phosphatidyl ethanolamine (MPEG)was purchased from Shearwater Polymers, Inc. The otherchemicals were purchased from Sigma or Merck.

2.2. Preparation of Liposomes. Paclitaxel was added to thealcoholic admixture of EPC, HEPC, cholesterol (Chol), andMPEG with a given drug-to-lipid molar ratio as indicatedin the context. The solution was evaporated under vacuumto remove the solvent and formed a lipid film on thewall of the round-bottom flask at which time; aliquots of10% (w/v) sucrose were added to the flask for hydration.Large multilamellar liposomes were suspended, and thensonicated (XL2020, Misonix Inc., Farmingdale, NY, USA) for10 minutes to yield small unilamellar liposomes. Paclitaxel-containing liposomes underwent filtration through a 0.2 μmcellulose acetate membrane (Orange Scientific Co., Braine L’Alleud, Belgium) to remove possible paclitaxel precipitatesand achieve sterilization. Drug incorporation efficiency, rep-resenting the retention of paclitaxel in the filtered liposomeswith respect to the originally added drug, was determined byHPLC analysis. Laser particle size analyzer (N4 Plus, CoulterElectronics Inc., Hialeah, FL, USA) was used to measure theaverage particle size. The liposomes were sealed in the vialunder nitrogen and stored at 4◦C for further shelf stabilitytest.

2.3. HPLC Assay. High performance liquid chromatography(HPLC) was performed using an autosampler, controller,and dual wavelength absorbance detector with wavelengthset to 229 nm, all of which were obtained from Waters Co. A125 mm × 4 mm Lichrosphere 100 RP-18 column, obtainedfrom Merck, was employed to identify and quantify theconcentration of paclitaxel. The mobile phase was composedof 50% acetonitrile and 50% D.I. water eluted isocraticallythroughout the measurement. A sample was dissolved inmethanol before injection into a 20 μL sample loop. Theretention time of paclitaxel is 12 minute while the flow ratewas kept at 0.5 mL/min.

2.4. Differential Scanning Calorimetry (DSC) Studies. DSCmeasurements were performed using a differential scanningcalorimeter (Mettler-Toledo DSC 822e, Switzerland). Theliposome suspensions (total lipid concentration: 20 mg/mL)were heated at a programmed constant heating rate of 5◦Cper minute. Empty hermetically sealed aluminum pans wereused as reference.

2.5. Shelf Stability Analysis. Shelf stability of the paclitaxelliposomes at 4◦C was monitored at the predeterminedinterval time. Particle size was analyzed before filtration ofthe sample to remove the aggregated liposomes and paclitaxelprecipitates. The sample filtered through 0.2 μm celluloseacetate membrane then was prepared for measurement ofpaclitaxel concentration by HPLC.

2.6. Cytotoxicity Assay. C-26, a syngeneic colon tumor cellline, was inoculated at 5 × 103 cells per well in 96-wellmicrotiter plate. The cells were maintained with RPMI-1640 medium comprising 10% heat-inactivated fetal calfserum, 100 U/mL penicillin, and 100 mg/mL streptomycinat 37◦C in a 5% CO2 humidified incubator. The drug-containing solutions were added and incubated with cells for72 hours before the MTT assay [23]. Liposome vehicle at thecomparable lipid concentration was used as the control. Theoptical density readings were determined by an ELISA readerat 540 nm. Cell survival rate was calculated by internalizationof the optical density readings.

2.7. Pharmacokinetic Studies. All animal studies, includingpharmacokinetic study, acute toxicity, and efficacies ofprepared liposomes, were performed in compliance withthe “Guide for the Care and Use of Laboratory Animals”prepared by the Institute of Laboratory Animal Resources,National Research Council, USA and published by theNational Academy Press, revised in 1996.

For the pharmacokinetic studies, six to seven weeks oldfemale SD rats were purchased from the National Labora-tories of Animal Breeding and Research Center (NLABRC,Nangarng, Taipei, Taiwan). Rats were bred at least oneweek after received to obtain a stable habitable conditionbefore any experiment. The jugular vein was cannulatedand the cannula was exteriorized in the back of the neck.Taxol or liposomal paclitaxel was administrated throughthe jugular vein at the paclitaxel dose of 5 mg/kg rats.

Journal of Drug Delivery 3

Table 1: Characteristics and shelf stability of liposomes mainly composed of either natural EPC or HEPC alone.

Liposomecomposition

[Lipid] (mM)Drug/PL(mole%)

[Paclitaxel](mg/mL)a

Mean particlesize (nm)

I.E.b (%)Remaining contentc (%)

14 days 30 days

EPC/Chol/MPEG 20 3 0.45 142.9 88.4 89.3 77.9

(20/8/1) 20 7 0.52 174.1 42.1 67.8 35.4

HEPC/Chol/MPEG(20/8/1)

20 3 0.32 93.2 68.1 76.7 63.6

aConcentration of paclitaxel at day 0.bIncorporation efficiency = paclitaxel incorporated in liposomes/paclitaxel added.cRemaining content = [paclitaxel] at day N/[paclitaxel] at day 0.

Serial blood samples were withdrawn through the venouscatheter after the rats were awakened from anesthesia.Drug concentrations in plasma were analyzed by HPLC.The pharmacokinetic parameters of each formulation werecalculated using the WinNonLin pharmacokinetic software(Version 3.1, Pharsight Co., Mountainview, CA, USA).

2.8. Acute Toxicity Test. Six to eight weeks old male ICR micewere divided into four different groups (treated with Taxol20 mg/kg, Taxol 40 mg/kg, liposomal paclitaxel 20 mg/kg,or liposomal paclitaxel 40 mg/kg), consisting of 5 mice ineach group. Mice for each group were injected throughtail vein to examine the acute intravenous toxicity. Afterliposome administration, the mice were observed for 14 days.During the observation period, mice were observed daily formortality and clinical signs. The survival rate over 14 dayswas obtained.

2.9. Efficacy Test. Athymic BALB/c nude mice were obtainedfrom NLABRC and weighted 20–22 g at the start of theexperiments. The mice were housed in sterilized filter-toppedcages and maintained in sterile conditions. The humanlung adenocarcinoma cell line PC14PE6/AS2, a derivativeof PC14PE6, which was obtained from Dr. Wu-Chou Su(National Chung Kung University medical college, Taiwan).The PC14PE6/AS2 cells express higher VEGF proteins,microvessel density, and vascular permeability in tumors[24]. It was suggested that the enhanced permeability andretention (EPR) effect within the tumor site made colloidalsystems more effective on the treatment of cancer [25].On the day of implantation, 106 cells were inoculatedsubcutaneously into lower back for each mouse. Tumorvolume was determined by measuring orthogonal diametersof the tumor and calculated as 0.4× (a2×b), where “a” is thetumor width and “b” is its length in mm. Tumor formationmeasuring at least 250 mm3 was considered a positive take(day 0), at which time 4 groups, each contained 6 animals,were established. They were (1) normal saline control group,(2) Taxol 20 mg/kg treatment group, (3) liposomal paclitaxel20 mg/kg treatment group, and (4) liposomal paclitaxel40 mg/kg treatment group. The drugs and controls weregiven as a bolus into tail veins for 4 doses totally on day 1,3, 6, and 9. Animal mortality was checked daily, and tumorvolume and body weight was checked every other day. Miceshowing more than 20% body weight loss or tumors largerthan 10-fold of original size (∼2,500 mm3) were sacrificed.

3. Results

3.1. Liposomes Made of Single PC. The liposomes composedof either HEPC or EPC alone were prepared according tothe procedure described in Experimental Section. MPEG wasused in the formulation to stabilize liposomes. The MPEGto phospholipid molar ratio was limited to less than 5% toavoid misinterpretation with the combinative formulation[26]. The cholesterol compositions were optimized to obtainthe small liposome size and high drug incorporation. Theresults in Table 1 showed that the liposomes made mainlyof EPC incorporated up to 88% paclitaxel when the drugto phospholipid molar ratio was kept at 3%. However,drug incorporation efficiency fell to 42% when the drug tophospholipid molar ratio rose to 7%. The liposomes madeof HEPC incorporated below 70% paclitaxel when drug tophospholipid molar ratio was kept at 3%. The liposomeswould yield apparent white precipitates while paclitaxel tophospholipid molar ratio was elevated above 3%. The higherdrug incorporation efficiency for EPC liposomes resultedfrom the lower transition temperature of EPC (<0◦C) thatis flexible enough to entrap relatively more hydrophobicmolecules rather than rigid HPEC [27]. Table 1 also presentsthe shelf stability of the liposomes composed principally ofeither HEPC or EPC. These liposomes were monitored at 4◦Conly for one month because of the early appearance of whiteprecipitates. A decrease in drug incorporation efficiency inthe liposomes was confirmed by HPLC. The EPC liposomesexhibited an obvious decline in drug incorporation as thedrug to phospholipid molar ratio was increased to 7%. TheHEPC liposomes were considerably unstable too.

3.2. Combinative Formulation of Hydrogenated PC and Nat-ural PC. Two distinct phosphatidylcholines with significantphase transition temperature difference were used in thisstudy. HEPC is referred to as a phospholipid with highphase transition temperature, anticipated to be about 50–60◦C. Natural EPC containing high content of unsaturatedfatty acid chains is considered to have low-phase transitiontemperature below 0◦C. A series of combinations of HEPCand EPC were investigated in an attempt to develop astable liposome formulation for paclitaxel. Besides, 5 mol%MPEG and 10 mol% cholesterol to phospholipid molarratio were added to costabilize the liposomes. Their effectsof MPEG and cholesterol on paclitaxel incorporation andparticle size were minimized by the constant molar ratio.


Table 2: Paclitaxel incorporation efficiency and particle sizeof the liposomes made of EPC and HEPC. Liposomes wereprepared in accordance to the formulation (paclitaxel/totalPC/cholesterol/MPEG = 0.3/10/1/0.5).

Molar ratio of HEPC/totalPC (%)

Mean particle size(nm)

I.E. (%)

25 113.3 69.2

43 120.8 63.8

62 128.4 73.6

81 202.6 37.6

Table 3: Effects of increasing paclitaxel to lipid molar ratioon physical properties of liposomes. Liposome formulation iscomposed of PCs, cholesterol, and MPEG at the optimal molar ratio(EPC/HEPC/Chol/MPEG = 15/5/2/1).

PC(mM)

Drug/PLa

(mole%)[Paclitaxel](mg/mL)

Mean particlesize (nm)

I.E. (%)

20 7 1.0 114.3 84.5

40 7 2.0 115.8 82.4

20 10 1.3 116.2 78.8

20 15 2.1 125.4 81.0

20 20 2.9 134.9 85.1

20 25 2.3 146.3 54.6aPL represents total phospholipids including EPC, HEPC and MPEG.

The characteristics of the formulated liposomes are given inTable 2. When increasing HEPC molar ratio, this led to anincrease in the average diameter of the liposomes. Mean-while, the incorporation efficiency of paclitaxel graduallydecreased as the quantity of HEPC increased. The particlesize could be reduced but drug incorporation efficiency didnot change significantly when HEPC molar ratio decreasedbelow 62%. Therefore, 25% molar ratio of HEPC, with thesmallest particle size among the tested compositions, wasselected to test the drug loading capacity of the liposomeformulation.

3.3. Increasing Drug/Phospholipid Ratio. Drug loadingcapacity of the formulated liposomes described above wasinvestigated by further increasing paclitaxel to phospholipidmolar ratio from 7% to 25%. Phospholipids representHEPC, EPC, and MPEG. Drug to phospholipid molar ratiorepresents the originally added drug content. The effects ofincreasing drug to phospholipid ratio were examined onthe physical properties of drug incorporation and particlesize. Table 3 lists the drug incorporation efficiency andparticle size of the resultant liposomes. It was noteworthythat high paclitaxel content was effectively incorporated inthe liposomes. The incorporation efficiency was maintainedabove 80% even though the drug to phospholipid molarratio was increased to 20%. No precipitate was observedthroughout preparation.

The drug loading capacity of the liposomes was foundto be paclitaxel concentration dependent. Attempts to

Temperature (◦C)

Exo

ther

mic

14% paclitaxel

7% paclitaxel

0% paclitaxel

38 40 42 44 46

Figure 1: The DSC thermographs for the EPC/HEPC (4 : 1)liposomes without the drug as well as with 7 and 14 mol%paclitaxel.

incorporate extremely high paclitaxel tended to destabilizeliposomes. The drug incorporation efficiency dropped to55% when the drug to phospholipid molar ratio wasincreased to 25%. Such a high paclitaxel loading acceler-ated destabilization of liposomes. White precipitates andaggregated liposomes appeared shortly after sonication.Needle-like precipitates and many floccules can be seen byoptical microscope. The poor drug incorporation and lipidaggregation reflects instability of the liposomes with such ahigh drug to phospholipid ratio. Therefore, the maximumdrug loading in the stable liposomes during preparationwas anticipated to 20 mole%. Besides, liposome solutionswith various lipid concentrations were also prepared andexamined. It is evident that doubling lipid content (40 mM)with the same liposome composition affected neither drugincorporation nor particle size.

3.4. Phase Transition Temperature of Prepared Liposomes.To determine the influence of paclitaxel on phospholipidbilayer phase transitions, the DSC analysis was employed.The DSC thermographs for the EPC/HEPC (4 : 1) liposomeswithout the drug as well as with 7 and 14 mol% paclitaxelare shown in Figure 1. For the EPC/HEPC liposome, a lowmiscibility was observed that leads to a phase separation inthe temperature range of 39–44◦C. It can be observed fromFigure 1 that with increasing the paclitaxel concentrationin liposomes, the main transition temperature was shiftedslightly to a lower temperature from 41◦C to 39.5◦C.

3.5. Shelf Stability. Liposomal paclitaxel were stored at 4◦Cimmediately after preparation and sterilization. Particle size(Figure 2(a)) and paclitaxel concentration (Figure 2(b)) weremeasured periodically. The results in Figure 2 indicate thatthese two measurements were stable for most of the formu-lations over six months. The implication of shelf stabilityof the liposomes with paclitaxel to lipid ratio was revealed.At 25% of drug to phospholipid molar ratio the liposomes


0 1 2 3 4 5 60

20

40

60

80

100

120

140

160

180

200Pa

rtic

lesi

ze(n

m)

Time (months)

7%10%15%

20%25%

(a)

7%10%15%

20%25%

0 1 2 3 4 5 6

0.8

1.2

1.6

2.0

2.4

2.8

3.2

Pacl

itax

elco

nce

ntr

atio

n(m

g/m

L)

Time (months)

(b)

Figure 2: Shelf stability of the liposomes with increasing drug/phospholipid molar ratio at 4◦C. Legends mean the paclitaxel to phospholipidmolar ratio of the liposomes. The composition is tabulated in Table 2.

was rather unstable in both terms of drug incorporationand particle size. Particle size rose and drug incorporationfell apparently. Although the 20% liposomes possessed morethan 80% of incorporation efficiency at the beginning ofstorage, drug incorporation declined much more and fasterthan those of the other with a lower drug-to-lipid ratio.After six-month storage, the retention of paclitaxel in theliposomes dropped to 67% of the originally incorporatedamount. The particle size increased from 146 to 168 nmwithin the first two months. When the drug to phospholipidmolar ratio was maintained equal or below to 15%, all theformulated liposomes remained stable for at least 6 months.The particle size varied by 10 nm in maximum and almostunchanged drug incorporation occurred. A maximum drugloading capacity of the liposomes, which could be stablein long-term storage, thus was anticipated to be 15%–20%drug-to-lipid molar ratio.

3.6. Cytotoxicity. The liposome formula with 15% pacli-taxel was preceded with the cytotoxicity, acute toxicity,and pharmacokinetic tests. The paclitaxel concentration ofthe liposomes for 50% inhibition of C-26 cells (IC50) isapproximately 162 nM which is slightly higher than that ofTaxol (IC50 = 105 nM), as shown in Figure 3. The liposomevehicles without paclitaxel showed no cytotoxicity against C-26 tumor cells over the tested range.

3.7. Pharmacokinetic Studies. Figure 4 and Table 4 showthe plasma concentration profile of paclitaxel and theirpharmacokinetic parameters, respectively, after i.v. injectionof liposomes and Taxol in rats. The AUC value of paclitaxelliposomes was slightly higher than that of Taxol. However,the liposomal paclitaxel in plasma declined quicker thanTaxol. It seems that incorporation of MPEG in the prepared

0.01 0.1 1 100

20

40

60

80

100

120

Cel

lsu

rviv

alra

te(%

)

Paclitaxel concentration (μM)

Liposome vehicleTaxolLiposomal paclitaxel (n = 5)

1E − 3

Figure 3: Survival rates of C-26 tumor cells exposing to theliposomes with or without paclitaxel and Taxol. The amount of theliposomes corresponding to the paclitaxel liposome was added asthe control.

liposome formulations did not prolong their circulationtime.

3.8. Acute Toxicity. Escalated dose of paclitaxel liposomeswas tested in ICR mice to determine the acute toxicity. Micewere divided into two groups treated with Taxol or liposomalpaclitaxel at doses of 20 and 40 mg/kg, respectively. Table 5shows the survival rate in all the groups over 14 days. It was


0 60 120 180 240 300 360

0.01

0.1

1

10

100

Pacl

itax

elle

veli

npl

asm

a(μ

g/m

L)

Time (min)

TaxolLiposomal paclitaxel (n = 4)

Figure 4: Plasma concentration profiles of paclitaxel after i.v.injection of Taxol or paclitaxel liposomes in rats (5 mg/kg aspaclitaxel). Each data represents the mean of 4 rats.

Table 4: Mean pharmacokinetic parameters of paclitaxel after i.v.injections of Taxol or paclitaxel liposomes at a dose of 5 mg/kg inrats.

T1/2 (hour) AUC0→∞ (μg h/mL)

Taxol 2.4± 0.4 8.5± 1.7

Liposomal paclitaxel 2.1± 1.0 12.7± 5.8

Table 5: Survival rate of mice received i.v. injections of Taxol orpaclitaxel liposomes at doses of 20 and 40 mg/kg.

Dose (mg/kg) Survival Rate

Taxol20 4/5

40 1/5

Liposomal paclitaxel 40 8/8

found that 3 of 5 mice with 40 mg/kg of Taxol group died onthe same day of injection. Afterwards, one of the other micedied on the third day. In contrast, all the mice in liposomalpaclitaxel group survived over the test period of 14 days.

3.9. Efficacy Test. The antitumor efficacy of distinct paclitaxelformulations was studied in AS-2 lung cancer bearing nudemice. In the case of the paclitaxel liposomes, weight loss wasobserved at a dose of 40 mg/kg. One mouse died on day 7and one on day 11. It was estimated the toxicity due to therepeated doses of the paclitaxel liposomes. In the case ofTaxol at a dose of 20 mg/kg, an evidence that one mouse diedon day 6 also indicated the repeated dose toxicity. BecauseTaxol at a dose of 40 mg/kg had caused a high mortality innude mice (>50%) in a preliminary study, we excluded thedose level in the current study. Figure 5(a) shows the progressof the tumor growth observed for 28 days. It was found

that the tumor size of the normal saline group increasedsignificantly with time. In contrast, the groups injectedwith distinct paclitaxel formulations significantly delayedthe tumor growth as compared to the normal saline group(P < .05). At the same dose of 20 mg/kg, liposomal paclitaxelseemed to delay the tumor growth more effectively thanTaxol. Once increasing the dose to 40 mg/kg, the liposomalpaclitaxel significantly inhibit the tumor growth for morethan 42 days as compared with other treated groups (P <.05). Although two mice died during dosing treatment forthe high dose of liposomal paclitaxel, the liposomal paclitaxel(20 and 40 mg/kg) significantly enhanced the mouse survivaltime to more than 30 days as compared with saline group(Figure 5(b), P < .05). The median survival time for micetreated with normal saline was 12.3 days, and treatment withTaxol slightly increased this survival to 19.7 days. Thus, theprepared liposomal paclitaxel provide benefits on reducingtumor volume, which correlated with a substantial increasein animal survival.

4. Discussion

Balasubramanian et al. reported that paclitaxel has atendency to undergo concentration-dependent aggregationin hydrophobic or relatively low polarity environments,forming intermolecular hydrogen bonds [28]. Restated, as alarge amount of paclitaxel is embedded in the hydrophobicdomain within the bilayer membrane, it is thermodynam-ically prone to self-aggregating, and thereby destabilizingthe liposomes [29]. The results imply the limited drugloading and the poor shelf stability of the current lipo-some formulations for paclitaxel. Much research [12–15,19] also supported the fact that the optimal paclitaxel tolipid molar ratio in the previous liposome formulationsis from 3% to 4%, and the liposomes is shelf stable onlywhen the drug-to-lipid molar ratio is kept equal to orbelow 3%. A higher drug-to-lipid molar ratio would leadto the occurrence of needle-like crystal precipitate duringpreparation.

To improve instability and poor drug payload of theconventional paclitaxel liposomes, we developed a formu-lation combining two sorts of PCs into liposomes, whichhave significant differences between their phase transitiontemperatures. Based on the material information providedby the manufacture, HEPC is referred to a phospholipid withlong hydrocarbon chain length and high phase transitiontemperature of 50–55◦C; on the contrary, the other (natu-rally occurring EPC) containing high content of unsaturatedfatty acid chains is considered to have a lower phasetransition temperature of −8◦C. The difference of phasetransition temperatures between the two PCs is estimatedto 60◦C at least. It could be speculated that the separatedphases, a gel phase and fluid (liquid-crystal) phase, on bilayermembrane at a given temperature were formed, like thegiant unilamellar liposomes made of DPPC and DLPC andvisualized by confocal microscope [22, 30]. Moreover, gel-gel[31, 32] and fluid-fluid [33] demixing of the binary phos-pholipid system have also been observed, especially whentheir hydrocarbon chain lengths are mismatched. Therefore,


0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30

0

2

4

6

8

10

12

14

16

SalineTaxol (20 mg/kg)

Liposomal paclitaxel (20 mg/kg)Liposomal paclitaxel (40 mg/kg)

Rat

ioof

tum

orsi

zech

ange

d

Time (days after start of treatment)

(a)

0 4 8 12 16 20 24 28 32 36 40 440

10

20

30

40

50

60

70

80

90

100

110

SalineTaxol (20 mg/kg)

Liposomal paclitaxel (20 mg/kg)Liposomal paclitaxel (40 mg/kg)

Time (days after start of treatment)

Surv

ival

rate

(%)

(b)

Figure 5: (a) The ratio of tumor size changed and (b) survival rate of different paclitaxel formulations on human lung adenocarcinoma(AS-2) bearing nude mice.

a combination of two phospholipids including HEPC andnatural EPC is reasonably expected to produce liposomeswith the many segregated microdomains coexisting on themembrane.

Accordingly, formation of the phase boundary wasspeculated to restrict the lateral diffusion across segregateddomains, hindering the self-aggregation of hydrophobicmolecules. A stable liposome formulation able to incorporatea high content of paclitaxel, therefore, can be made. Thecoexistence of lateral separate phospholipid regions pro-motes the incorporation of a large amount of hydrophobicpaclitaxel into the phospholipid bilayer. The hypothesis mayaccount for why the liposomes formulated in this study canincorporate more paclitaxel and remain more stable in long-term storage. The drug to phospholipid molar ratio canbe increased to 15%, which was significantly upgraded byapproximately sixfolds in comparison to the other liposomeformulations reported [12–15, 19]. The liposomes consistingof a combination of two phospholipids showed improveddrug loading capacity and shelf stability over those of theformulations with single phospholipid alone. The featureseven are superior to the previous liposome formulations withnegatively charged phospholipids [11–14, 21]. Furthermore,the liposomes still alleviate acute toxicity without changingits cytotoxicity against tumor cells, resembling the otherliposome formulations [2, 14, 21, 34, 35]. Pharmacokineticdata also exhibits a higher AUC in rats than Taxol. Despite,the liposomes were not able to circulate in blood as longas those composed of MPEG on the surface. This resultmay be attributed to the presence of reticuloendothelial(RES) system. Nanoparticles will usually be taken up by theliver, spleen, and other parts of the RES depending on theirsurface characteristics, especially for particles with morehydrophobic surfaces [36, 37]. However, the RES uptake of

liposomal paclitaxel may limit the systemic exposure of non-RES tissues, such as the bone marrow, to paclitaxel [37]. Dueto the alternant biodistribution of paclitaxel by liposomes,it may exert not only a direct effect on reduced toxicitybut also may underlie the preservation or enhancement ofantitumor efficacy following administration of liposomalpaclitaxel [37]. Regardless of the similar profile of the bloodexposure to Taxol, the prepared liposomal paclitaxel diddemonstrate the reduced toxicity and increase the efficacyagainst human cancer in animal model.

5. Conclusion

This study presents a novel liposomal formulation capable ofincorporating a high paclitaxel content, and remaining stablein long-term storage as well. Liposomes remained stable inliquid form at 4◦C for at least 6 months when the drug-to-lipid molar ratio was below 15%. In aspects of in vitroand in vivo efficacy studies, the paclitaxel liposomes exhibit acomparable cytotoxicity against colon cancer and enhancedefficacy against human lung tumor as compared withTaxol. As expected, the liposomes have lower acute toxicitysignificantly in mice than the current cremophore/alcoholformulation dose. These results demonstrate that the liposo-mal paclitaxel is promising as an anticancer treatment. Thenovel formulation has a potential to incorporate the highcontent of hydrophobic drug stably.

Acknowledgments

The authors would like to thank the Ministry of EconomicAffairs of the Republic of China for financially supportingthis research under Contract no 893WB4100. Free lipids werekindly gifts from Lipoid and Lucas Meyer.


References

[1] J. M. Terwogt, B. Nuijen, W. W. Huinink, and J. H. Beijnen,“Alternative formulations of paclitaxel,” Cancer TreatmentReviews, vol. 23, no. 2, pp. 87–95, 1997.

[2] M.-F. Shieh, I.-M. Chu, C.-J. Lee, P. Kan, D.-M. Hau, andJ.-J. Shieh, “Liposomal delivery system for taxol,” Journal ofFermentation and Bioengineering, vol. 83, no. 1, pp. 87–90,1997.

[3] A. O. Nornoo and D. S.-L. Chow, “Cremophor-free intra-venous microemulsions for paclitaxel. II. Stability, in vitrorelease and pharmacokinetics,” International Journal of Phar-maceutics, vol. 349, no. 1-2, pp. 117–123, 2008.

[4] P. Kan, Z.-B. Chen, C.-J. Lee, and I.-M. Chu, “Development ofnonionic surfactant/phospholipid o/w emulsion as a paclitaxeldelivery system,” Journal of Controlled Release, vol. 58, no. 3,pp. 271–278, 1999.

[5] E. Bilensoy, O. Gurkaynak, M. Ertan, M. Sen, and A. A. Hincal,“Development of nonsurfactant cyclodextrin nanoparticlesloaded with anticancer drug paclitaxel,” Journal of Pharmaceu-tical Sciences, vol. 97, no. 4, pp. 1519–1529, 2008.

[6] J. Liu, D. Meisner, E. Kwong, X. Y. Wu, and M. R. Johnston,“A novel trans-lymphatic drug delivery system: implantablegelatin sponge impregnated with PLGA-paclitaxel micro-spheres,” Biomaterials, vol. 28, no. 21, pp. 3236–3244, 2007.

[7] E. K. Park, S. B. Lee, and Y. M. Lee, “Preparation andcharacterization of methoxy poly(ethylene glycol)/poly(ε-caprolactone) amphiphilic block copolymeric nanospheres fortumor-specific folate-mediated targeting of anticancer drugs,”Biomaterials, vol. 26, no. 9, pp. 1053–1061, 2005.

[8] J. Pan and S.-S. Feng, “Targeted delivery of paclitaxel usingfolate-decorated poly(lactide)-vitamin E TPGS nanoparti-cles,” Biomaterials, vol. 29, no. 17, pp. 2663–2672, 2008.

[9] B. B. Lundberg, V. Risovic, M. Ramaswamy, and K. M. Wasan,“A lipophilic paclitaxel derivative incorporated in a lipidemulsion for parenteral administration,” Journal of ControlledRelease, vol. 86, no. 1, pp. 93–100, 2003.

[10] M. S. Tartis, J. McCallan, A. F. H. Lum et al., “Therapeuticeffects of paclitaxel-containing ultrasound contrast agents,”Ultrasound in Medicine and Biology, vol. 32, no. 11, pp. 1771–1780, 2006.

[11] M.-H. Bartoli, M. Boitard, H. Fessi et al., “In vitro and in vivoantitumoral activity of free, and encapsulated taxol,” Journal ofMicroencapsulation, vol. 7, no. 2, pp. 191–197, 1990.

[12] R. M. Straubinger, A. Sharma, M. Murray, and E. Mayhew,Taxol formulation, US patent 5415869, 1995.

[13] F. Sampedro, J. Partika, P. Santalo, A. M. Molins-Pujol, J.Bonal, and R. Perez-Soler, “Liposomes as carriers of differentnew lipophilic antitumour drugs: a preliminary report,”Journal of Microencapsulation, vol. 11, no. 3, pp. 309–318,1994.

[14] A. Sharma and R. M. Straubinger, “Novel taxol formula-tions: preparation and characterization of taxol-containingliposomes,” Pharmaceutical Research, vol. 11, no. 6, pp. 889–896, 1994.

[15] K. S. Warner, S. Kevin Li, and W. I. Higuchi, “Influence ofcationic lipids on the stability and membrane properties ofpaclitaxel-containing liposomes,” Journal of PharmaceuticalSciences, vol. 90, no. 8, pp. 1091–1105, 2001.

[16] R. Reszka, M. Brandl, I. Fichtner, and G. Warnke, “Liposome-encapsulated Taxol, its preparation and its use,” US patent6090955, 2000.

[17] D. Sharma, T. P. Chelvi, J. Kaur, and R. Ralhan, “Thermosensi-tive liposomal taxol formulation: heat-mediated targeted drug

delivery in murine melanoma,” Melanoma Research, vol. 8, no.3, pp. 240–244, 1998.

[18] L. Boni and J. Portnoff, “Taxane-containing phosphatidyl-choline liposomes,” US patent 5683715, 1997.

[19] C. Bernsdorff, R. Reszka, and R. Winter, “Interaction of theanticancer agent Taxol(TM) (paclitaxel) with phospholipidbilayers,” Journal of Biomedical Materials Research, vol. 46, no.2, pp. 141–149, 1999.

[20] P. Crosasso, M. Ceruti, P. Brusa, S. Arpicco, F. Dosio, andL. Cattel, “Preparation, characterization and properties ofsterically stabilized paclitaxel-containing liposomes,” Journalof Controlled Release, vol. 63, no. 1-2, pp. 19–30, 2000.

[21] A. Cabanes, K. E. Briggs, P. C. Gokhale, J. A. Treat, andA. Rahman, “Comparative in vivo studies with paclitaxeland liposome-encapsulated paclitaxel,” International Journalof Oncology, vol. 12, no. 5, pp. 1035–1040, 1998.

[22] J. Korlach, P. Schwille, W. W. Webb, and G. W. Feigen-son, “Characterization of lipid bilayer phases by confocalmicroscopy and fluorescence correlation spectroscopy,” Pro-ceedings of the National Academy of Sciences of the United Statesof America, vol. 96, no. 15, pp. 8461–8466, 1999.

[23] M. C. Alley, D. A. Scudiero, A. Monks et al., “Feasibility ofdrug screening with panels of human tumor cell lines using amicroculture tetrazolium assay,” Cancer Research, vol. 48, no.3, pp. 589–601, 1988.

[24] H.-H. Yeh, W.-W. Lai, H. H. W. Chen, H.-S. Liu, and W.-C. Su,“Autocrine IL-6-induced Stat3 activation contributes to thepathogenesis of lung adenocarcinoma and malignant pleuraleffusion,” Oncogene, vol. 25, no. 31, pp. 4300–4309, 2006.

[25] J. M. Koziara, T. R. Whisman, M. T. Tseng, and R. J. Mumper,“In-vivo efficacy of novel paclitaxel nanoparticles in paclitaxel-resistant human colorectal tumors,” Journal of ControlledRelease, vol. 112, no. 3, pp. 312–319, 2006.

[26] A. L. Klibanov, K. Maruyama, V. P. Torchilin, and L. Huang,“Amphipathic polyethyleneglycols effectively prolong the cir-culation time of liposomes,” FEBS Letters, vol. 268, no. 1, pp.235–237, 1990.

[27] J. A. Zhang, G. Anyarambhatla, L. Ma et al., “Develop-ment and characterization of a novel Cremophor� EL freeliposome-based paclitaxel (LEP-ETU) formulation,” EuropeanJournal of Pharmaceutics and Biopharmaceutics, vol. 59, no. 1,pp. 177–187, 2005.

[28] S. V. Balasubramanian, J. L. Alderfer, and R. M. Straubinger,“Solvent- and concentration-dependent molecular interac-tions of taxol (Paclitaxel),” Journal of Pharmaceutical Sciences,vol. 83, no. 10, pp. 1470–1476, 1994.

[29] M. R. Wenk, A. Fahr, R. Reszka, and J. Seelig, “Paclitaxelpartitioning into lipid bilayers,” Journal of PharmaceuticalSciences, vol. 85, no. 2, pp. 228–231, 1996.

[30] L. A. Bagatolli and E. Gratton, “Two photon fluorescencemicroscopy of coexisting lipid domains in giant unilamellarvesicles of binary phospholipid mixtures,” Biophysical Journal,vol. 78, no. 1, pp. 290–305, 2000.

[31] C. Gliss, H. Clausen-Schaumann, R. Gunther, S. Odenbach,O. Randl, and T. M. Bayerl, “Direct detection of domainsin phospholipid bilayers by grazing incidence diffraction ofneutrons and atomic force microscopy,” Biophysical Journal,vol. 74, no. 5, pp. 2443–2450, 1998.

[32] I. P. Sugar, T. E. Thompson, and R. L. Biltonen, “MonteCarlo simulation of two-component bilayers: DMPC/DSPCmixtures,” Biophysical Journal, vol. 76, no. 4, pp. 2099–2110,1999.

[33] J. Y. A. Lehtonen, J. M. Holopainen, and P. K. J. Kinnunen,“Evidence for the formation of microdomains in liquid


crystalline large unilamellar vesicles caused by hydrophobicmismatch of the constituent phospholipids,” Biophysical Jour-nal, vol. 70, no. 4, pp. 1753–1760, 1996.

[34] A. Sharma, E. Mayhew, L. Bolcsak et al., “Activity of pacli-taxel liposome formulations against human ovarian tumorxenografts,” International Journal of Cancer, vol. 71, no. 1, pp.103–107, 1997.

[35] J. Treat, N. Damjanov, C. Huang, S. Zrada, and A. Rahman,“Liposomal-encapsulated chemotherapy: preliminary resultsof a phase I study of a novel liposomal paclitaxel,” ONCOL-OGY, vol. 15, no. 5, pp. 44–48, 2001.

[36] L. Brannon-Peppas and J. O. Blanchette, “Nanoparticle andtargeted systems for cancer therapy,” Advanced Drug DeliveryReviews, vol. 56, no. 11, pp. 1649–1659, 2004.

[37] G. J. Fetterly and R. M. Straubinger, “Pharmacokinetics ofpaclitaxel-containing liposomes in rats,” AAPS PharmSci, vol.5, no. 4, article 32, 2003.


Review Article

Targeted Liposomal Drug Delivery toMonocytes and Macrophages

Ciara Kelly,1, 2 Caroline Jefferies,2 and Sally-Ann Cryan1

1 School of Pharmacy, Royal College of Surgeons in Ireland, Dublin 2, Ireland2 Department of Molecular & Cellular Therapeutics, Royal College of Surgeons in Ireland, Dublin 2, Ireland

Correspondence should be addressed to Ciara Kelly, [email protected]

Received 30 July 2010; Accepted 27 September 2010

Academic Editor: Juan M. Irache

Copyright © 2011 Ciara Kelly et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

As the role of monocytes and macrophages in a range of diseases including infectious disease, inflammatory diseases, cancer,and atherosclerosis is better understood, strategies to target these cell types are of growing importance both scientifically andtherapeutically. As particulate carriers, liposomes naturally target cells of the mononuclear phagocytic system (MPS), particularlymacrophages. Loading drugs into liposomes can therefore offer an efficient means of drug targeting to MPS cells. Physicochemicalproperties including size, charge, and lipid composition can have a very significant effect on the efficiency with which liposomestarget MPS cells. Small, negatively charged liposomes appear to target macrophages most efficiently by interaction with scavengerreceptors on the macrophage cell surface. MPS cells express a range of receptors including scavenger receptors, integrins, mannosereceptors, and Fc-receptors that can be targeted by the addition of ligands to liposome surfaces. These ligands include peptides,antibodies, and lectins and have the advantages of increasing target specificity and avoiding the need for cationic lipids to triggerintracellular delivery. The goal for targeting monocytes/macrophages using liposomes includes not only drug delivery but alsopotentially a role in cell ablation and cell activation for the treatment of conditions including cancer, atherosclerosis, HIV, andchronic inflammation.

1. Introduction

Mononuclear phagocytes such as monocytes, macrophages,and dendritic cells are intrinsically involved in innateimmunity. As the designation denotes, the chief role of thesecells is phagocytosis whereby cells will engulf and destroyapoptotic cells, pathogens, and other targets. This occurseither through employing opsonin receptor-dependentmechanisms via complement- and Fc-receptors, or opsoninreceptor-independent mechanisms via lectin-receptors, scav-enger receptors, stearylamine receptors or CD14 [1].

Due to its pivotal role in inflammation, the mononuclearphagocytic system (MPS) is an important target for drugdelivery to treat disease. For certain diseases such as chronicobstructive pulmonary disease (COPD), asthma, atheroscle-rosis, and cancer [2–4] and for pathogenic infections includ-ing tuberculosis [5], human immunodeficiency virus (HIV),and Leishmaniasis [6], the inflammatory process is a keydriver of both disease progression as well as pathogenesis.

Thus strategies aimed at targeting the MPS are highlyattractive. In general however these cells are reputed to bedifficult targets [7], particularly where intracellular deliveryof the active is required such as for gene delivery [8].Therefore the development of delivery systems that can targetmonocytes/macrophages intracellularly is crucial and couldpotentially open up new treatment paradigms for a range ofdiseases.

Liposomes are the most widely investigated delivery sys-tem for phagocyte-targeted therapies providing advantagessuch as low immunogenicity, biocompatibility, cell specificityand drug protection. However, there are also shortcomingssuch as poor scale-up, cost, short shelf life, and in somecases toxicity and off target effects. Parenterally administeredliposomes are naturally cleared by the MPS. Liposomaldelivery systems targeting other cell types outside the MPSare modified to evade phagocytosis; for example, “stealthliposomes” include poly-ethylene-glycol (PEG) into theirformulations to shield the liposomes from the MPS and


increase their circulatory lifespan [9]. Consequently, numer-ous studies have been carried out to develop formulationsthat avoid monocyte/macrophage clearance, the corollaryof which is that there is now greater knowledge of themechanisms of binding and uptake that can be harnessed fordrug targeting to monocyte/macrophage cells.

2. Monocytes and Macrophages

Cell origin, lineage, and function in the MPS are complexand remain under considerable investigation. In essence,monocytes differentiate from hematopoietic stem cells,specifically granulocyte/macrophage progenitors in the bonemarrow and enter the periphery as circulating monocytes.Various microenvironmental cues determine monocyte fatewhich can lead to differentiation into macrophage anddendritic cells [10]. However monocytes are not simplymacrophage and dendritic cell precursors but are alsoimmune effector cells [11].

Under inflammatory conditions, circulating monocytescan be recruited to the site of infection or injury, and oncethere, differentiate. However under steady state conditions,local proliferation maintains resident macrophages in sitessuch as the lungs and liver. Macrophages (M∅s) are centralplayers in the development, progression, and resolution ofinflammation [12]. They are polarized following activationinto classic (or M1) and alternative (or M2) macrophages[13–15]. M1 macrophages are activated in response to micro-bial products such as lipopolysaccharide (LPS) or cytokineslike interferon-γ (IFN-γ) and tumour necrosis factor α(TNFα) and are characterized by a strong propensity topresent antigen. In a polarized response, M1 cells are thoughtto kill intracellular microorganisms and produce abundantproinflammatory cytokines such as TNF-α, interleukin (IL)-12, IL-23, and proinflammatory mediators like nitric oxide(NO) and reactive oxygen intermediates (ROI).

On the other hand, M2 macrophages are promotedby various signals such as IL-4, IL-13, glucocorticoids,IL-10, immune complexes and some pathogen-associatedmolecular patterns (PAMPs) that elicit different M2 forms(M2a, b and c). They function in inflammation resolutionand tissue remodelling. Pathogen Recognition Receptors(PRRs) have evolved to recognise conserved molecular-associated molecular patterns (PAMPS) from pathogens,such as lipopolysaccharide or bacterial DNA motifs. TheToll-like receptors (TLRs) are one such family whose ligandshave generated much excitement over the last decade asimmunostimulatory adjuvants in vaccine development [16].Engagement of TLRs by their cognate ligands will activateantigen presenting cells, stimulate cytokine secretion thatregulates the adaptive immune response, and promote upregulation of costimulatory molecules in order to improveantigen presentation to T cells. Thus incorporation of TLRligands or immunomodulatory moieties into liposomes hasbeen a strategy for improving efficacy of both vaccinedevelopment and drug targeting [17]. For example, asTLR ligands have been shown to activate macrophages anddendritic cells and enhance antigen-specific T cell responses,then enhanced uptake of PAMP-coated liposomes into these

cells would be expected. However, whilst TLR ligands andPAMPs in general can increase liposome uptake, their abilityto stimulate and activate macrophages and enhance antigen-specific T cell activation and immune reactivity wouldsuggest that their potential inflammatory properties may bean issue for general use in targeting strategies [18]. In thisrespect other target receptors such as the scavenger receptorsand mannose receptors may prove more appropriate.

In addition Tumour-Associated Macrophages (TAMs)are an M2-like macrophage population that promote tumourgrowth via angiogenesis and metastasis, at least in part, by therelease of proangiogenic factors including vascular endothe-lial growth factor (VEGF) and matrix metalloproteinases[19]. Thus targeting strategies aimed at discriminatingagainst M1 and M2 macrophages may be very attractive forcancer chemotherapy in the future [20]. With respect tocancer therapeutics, dendritic cells are major antigen pre-senting cells that play important roles in cancer detection andelimination through the activation of T cells, and interest liesin targeting these cells for cancer immunotherapies [21].

3. Liposomal Drug Targeting

Liposome drug delivery systems harness the physiologicalrole of these cells to provide specific targeting and enhancedrug efficacy. Mononuclear phagocytes play major roles inmetabolism such as cholesterol and bilirubin metabolismand pathogen clearance [12]. Hence, cell surface receptorsare expressed, for example, scavenger receptors that allow theidentification and uptake of materials which can be targetedfor drug delivery. Targeting of liposomes to monocytes andmacrophages can be achieved by modifying lipid composi-tion to control physicochemical properties such as size andcharge and by the inclusion of surface ligands includingproteins, peptides, antibodies, polysaccharides, glycolipids,glycoproteins, and lectins (Figure 1 and Table 1).

3.1. Physicochemical Properties. Specific liposome propertieshave been shown to facilitate uptake into monocytes andmacrophages and are a simple and effective means oftargeting these cells.

3.1.1. Liposome Size. Recently, a detailed study by Epstein-Barash et al. compared the effect of liposome size and chargeon the bioactivity of liposomal bisphosphonates in a widerange of cell types in vitro including monocyte/macrophagecell lines (THP-1, J774, and RAW 264 cells) and primarycells (neutrophils, monocytes, kupffer cells, endothelial cells,and smooth muscle cells) and in vivo [24]. Liposomesranged in size from 50 to 800 nm in diameter and werecomposed of lipids with neutral, positive, or negativecharge. It was concluded that small (85 nm) negativelycharged liposomes composed of neutral 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC), anionic distearoyl-phophatidylglycerol (DSPG), and cholesterol at a molar ratio3 : 1 : 2 were optimum for internalisation by MPS cells whilelarge and positively charged liposomes induced cytokineactivation and toxicity [24, 38].


Table 1: Examples of therapeutic applications using monocyte/macrophage-targeted liposomes.

Ligand Active Disease Reference

Anionic lipids

Dexamethasone Atherosclerosis [22]

SLPI Inflammatory lung disease [23]

Bisphosphonates Restnosis [24]

Rifampicin Tuberculosis [25]

Dideoxycytidine-5′-triphosphate HIV [26]

Clarithromycin Mycobacterium avium infection [27]

Peptides

Muramyl tripeptide (MTP) MTP-phosphotidylethanolamine Osteosarcoma [28]

Arg-Gly-Asp (RGD) Diclofenac sodium (model drug) Cerebrovascular disease [29]

Antibodies

Anti-VCAM-1 Prostaglandins Atherosclerosis [30]

Anti-CC52 — Colon Cancer [31]

Anti-CC531 — Colon Adenocarcinoma [32]

Anti-CD11c/DEC-205 tumour antigen (OVA) Cancer [21]

Lectins

Mann-C4-Chol Dexamethasone palmitate Inflammatory lung disease [33]

Man2DOG — — [34]

Aminophenyl-α-D-mannopyranoside Doxorubicin Experimental visceral leishmaniasis [6]

Ciprofloxacin Respiratory infection [5]

Man3-DPPE OVA [35]

— Gastric cancer [36]

Other Ligands

Maleylated bovine serum albumin (MBSA) [25]

O-steroly amylopectin (O-SAP) [25]

Fibronectin [37]

Galactosyl [37]

While greater uptake of small liposomes (<100 nm) byMPS cells has been reported in the literature [37], manyother studies have shown liposome uptake by MPS cellsto be improved with increased size [39–41]. Optimal sizetherefore is likely to be dependent on multiple factorsincluding the target cell and specific properties of theliposome formulation, for example, receptor mediated ornonreceptor mediated uptake. Additionally in vitro resultsoften differ from in vivo findings [24, 40]. Particularly whenadministered parentally, liposomes will interact with variouscirculatory components and are then cleared by hepatocytesin vivo [40, 42].

3.1.2. Liposome Charge. Cationic liposomes are associatedwith efficient cellular delivery of drug cargoes and routinelyapplied for in vitro gene delivery [43]. Electrostatic interac-tions between positively charged liposomes and the nega-tively charged cell membranes and cell surface proteoglycans[44] facilitate cell uptake. Unfortunately, cationic liposomescan cause cytotoxicity limiting their safety for clinicaluse [45]. In RAW264.7 macrophages cationic liposomescontaining stearylamine (SA) have previously been shown toinduce apoptosis through mitochondrial pathways generat-ing reactive oxygen species (ROS), releasing cytochrome c,

caspase-3 and -8 and more recently activating protein kinaseC (PKC) δ possibly by cell surface proteoglycan interaction[38, 46–48]. Consequently interest for drug delivery hasturned to neutral and anionic liposomes.

Negatively charged lipids such as phosphatidylserine (PS)and phosphatidylglycerol (PG) are preferentially recognisedby macrophages [37]. Studies comparing phosphotidyl-choline (PC; neutral) and PS-composed liposomes haveestablished negative liposome formulations to have enhancedmacrophage internalisation [49]. Additionally, studies by usto quantify this difference have found a 5.3-fold increase inthe association of negatively charged 1,2-dioleoyl-sn-glycero-3-phospho-L-serine (DOPS):Cholesterol liposomes with amacrophage cell model, differentiated THP-1 cells, com-pared to neutral 1,2-dioleoyl-sn-glycero-3-phosphocholine(DOPC):Cholesterol liposomes (Figure 2) an effect whichwas also seen in vivo [50]. Negative charge can also beachieved by the incorporation of dicetylphosphate (DCP)[25, 40]. Vyas et al. showed a 3.4-fold increase in rifampicinlung retention in rats when rifampicin was encapsulatedin negatively charged DCP, PC, and cholesterol-composedliposomes and a 1.3-fold increase when encapsulated in thecorresponding neutral liposomes compared to free drug afteraerosol administration [25].


Pathogen

Immunoliposome

Anionic liposome

Macrophage

Scavenger receptors

Lectin receptors

Fc receptors

Integrins

Mannosylated liposome

Peptide coated liposome

“Trojan liposomes”

Figure 1: Summary of liposomal targeting strategies to macrophages.

The composition of the inner membrane leaflet ofeukaryotic cells [1] consists of PS and phosphatidyletha-nolamine (PE) with an outer layer of PC and sphingomyelin(SM) [51, 52]. In an apoptotic or necrotic event, PS will beexposed on the outer cell surface, and monocytic phago-cytosis is induced. It is believed that PS targets scavengerreceptors (SRs) on macrophages (Figure 1) but there mayalso be receptors specific for PS recognition. Moreover PS canactivate complement and associate with plasma apolipopro-teins such as ApoE promoting phagocytosis by macrophages[53]. There are six classes of SRs with A, B, and D as the mostlikely participants in liposome recognition [53]. However,not all phagocytes have the same affinity for these anioniclipids. According to Foged et al., PS and PG liposomes werefound to have minimal association with human monocyte-and bone marrow-derived dendritic cells [54].

In addition PS is a non-bilayer lipid (along withphosphatidylethanolamine; PE) which is frequently used in

the development of pH-sensitive and fusogenic liposomespromoting intracellular drug delivery [51]. For instance,liposomes composed of DOPE and PS have been assessed aspH-sensitive carriers of plasmid DNA to RAW 264.7 alveolarmacrophages [55]. Recently Andreakos et al. developed anovel amphoteric liposome for the delivery of antisenseoligonucleotides to sites of inflammation in experimentalarthritis [56]. The novel formulation known as Nov038 iscationic at low pH and anionic at neutral pH, facilitatingcomplexation to nucleic acids and avoiding nonspecificblood interactions, respectively. The group reported targeteddelivery to sites of inflammation as well as blood, liver,spleen, and inguinal lymph node mononuclear cells. Inaddition, Nov038 administration was well tolerated withefficient antisense oligonucleotide delivery in vivo.

3.2. Ligands. In addition to controlling the physicochemicalproperties of liposomes to enhance targeting, ligands can


0

200

400

600

800

1000

1200

1400

1600

1800

2000

Untreated DOPC DOPS

Flu

ores

cen

ce/m

gpr

otei

n

∗∗

Figure 2: Uptake of neutral (DOPC : Chol 7 : 3) and anionic(DOPS : Chol 7 : 3) liposomes by differentiated THP-1 cells after 2hours (n = 6± SEM) ∗P < .05; ∗∗P < .001.

be incorporated into liposome formulations to specificallytarget monocytes, macrophages, and dendritic cells. Usinga ligand targeting strategy for liposome drug delivery hasthe advantages of potentially increasing target specificity andavoiding the need for cationic lipids to trigger intracellulardelivery. A multitude of ligands are currently being assessedincluding peptides, antibodies, proteins, polysaccharides,glycolipids, glycoproteins, and lectins which make use ofmononuclear phagocytes characteristic receptor expressionand phagocytic innate processes (Figure 1 and Table 1). Herewe will briefly look at three of the most commonly studiedsystems peptide, antibody, and lectin directed delivery.

3.2.1. Peptides. Cell targeting peptides (CTPs) and cellpenetrating peptides (CPPs) have been conjugated to lipo-somes to improve cell-specific targeting and cell uptake,respectively, to a range of cell types [57]. Peptide sequencessuch as GGPNLTGRW (GGP-peptide) have been shownto selectively associate with neutrophils and monocytes[58, 59]. GGP-peptide-coated liposomes, with 500 externalligands per liposome, show 30.9 times greater associationto monocytes than uncoated liposomes [58]. Arg-Gly-Asp(RGD) peptide has also been incorporated into liposomeformulations to target integrin receptors expressed bymonocytes [29, 60, 61] (Figure 1). Magnetic RGD-coatedliposomes achieved an increase of approximately 15% drugrecovery from monocytes and neutrophils compared touncoated magnetic liposomes [29].

3.2.2. Antibodies. Immunoliposomes are liposomes coupledwith antibodies which can be used to target cell-specificantigens. In the case of phagocyte targeting, the use ofnonspecific and monoclonal antibodies can lead to liposomeopsonisation and uptake by macrophages. In vivo liposomesinteract with a wide variety of serum proteins includingimmunoglobulins, apolipoproteins, and complement pro-teins [42, 53] and may also activate complement leading toenhanced uptake by the MPS. However, protein interaction,complement activation, and opsonisation depend greatly onthe physicochemical properties of the liposomes such as size,surface charge, cholesterol content, and lipid composition

[42, 53]. For example, some studies have reported comple-ment activation to be greater with increasing liposome size[53] although observed activation has not always been ofsignificance [24].

Immunoglobulins (Igs) are recognised by Fc receptorson the surface of phagocytic cells which are involved inphagocytosis as well as antigen presentation [21] (Figure 1).Interest has focused on the FcγRI receptor as a target whichrecognises IgG and is expressed by monocytes, macrophages,activated neutrophils, and DCs [21]. Opsonisation is gen-erally Fc-receptor mediated and has previously been shownto significantly enhance liposome uptake by monocytes andmacrophages [32]. Opsonisation of non-immunoliposomesby immunoglobulins, for example, IgM and IgG, can alsooccur in vivo leading to enhanced uptake by macrophages[53].

Antibodies have been coupled to the surface of liposomesor distally via their Fc-region to liposome-attached PEG[31, 32]. Koning et al. showed increased Kupffer cell uptakewith greater antibody surface density [31, 32]. Dendritic cellshave been targeted with histidine-tagged antibody fragmentsattached to a novel chelator lipid, 3(nitrilotriacetic acid)-ditetradecylamine (NTA3-DTDA), incorporated into stealthliposomes via the DC receptors DEC-205 and CD11c [21].

3.2.3. Lectins. Immune cells including alveolar macrophages,peritoneal macrophages, monocyte-derived dendritic cells,and Kupffer cells constitutively express high levels of themannose receptor (MR). Macrophages and DCs can there-fore be targeted via mannosylated nanoparticles (Figure 1).The MR is a C-type lectin 175-kD type I transmembraneprotein [62, 63] whose ligands possess a terminal nonreduc-ing sugar such as mannose, glucose, N-acetylglucosamine,and fucose [64, 65]. These receptors play numerous roles inimmune function including antigenic recognition, endocy-tosis, and antigen presentation, and are critically involvedin homeostatic maintenance, inflammation and immuneresponses [66, 67]. Hence MR can identify and engulfpathogens such as Mycobacterium tuberculosis and Leishma-nia donovani via surface sugar antigens.

It should be noted that there are a wide variety of lectinswith mannose affinity including MR, dendritic cell-specificintercellular adhesion molecule-3 (DC-SIGN) and Endo 180,and many mannose receptor expressing cells but expressionand recognition profiles differ between cell types [66]. This isparticularly evident during inflammation where expressionof MR is altered in DCs [68]. Here we will focus on liposomesdesigned specifically for macrophage MR recognition (areceptor that is not expressed by circulating monocytes).

Mannosylated liposomes have repeatedly been shownto preferentially target macrophages and DCs attainingenhanced cellular uptake both in vitro and in vivo with betterin vitro/in vivo correlation than for nonligand containingliposomes [5, 6, 33–36, 41, 49, 66, 69–76]. Mannosylation hasbeen achieved by the incorporation of ligands such as alkylmannosides [70], Cholesten-5-yloxy-N-(4-((1-imino-2-α-thioglycosylethyl)amino)butyl)formamide (Mann-C4-Chol)[33, 74, 75, 77], Mann-His-C4-Chol [77], Man2DOG[34], 4-aminophenyl-a-D-mannopyranoside [5, 69], and


manntriose (Man3)-DPPE [35, 36, 71] into the liposomeformulations or by liposome coating with p-aminophenyl-α-D-mannopyranoside [6]. We have prepared a range ofmannosylated liposome, and quantified the increase in cellassociation with a macrophage-like cell model, differentiatedTHP-1 cells. Mannosylated liposomes significantly increasedliposome association with the macrophages compared touncoated controls (Figure 3) [78].

Over the past decade Hasida and colleagues have ledthe way in the development of mannosylated liposomestargeted to macrophages and DCs for the delivery of anti-inflammatory agents dexamethasone palmitate [33] andNuclear factor κ-B (NFκB) decoy and anticancer agents CpGoligonucleotides and DNA [79]. Intratracheally administeredMan-C4-Chol liposomes were shown to be preferentiallytaken up by alveolar macrophages which was mediated viaMR endocytosis as revealed by inhibition studies. Manno-sylation and the extent of this mannosylation significantlyimproved liposome internalisation by macrophages [72].The ability of these liposomes to efficiently deliver theirload has been the focus of a more recent study in whichthe use of bubble liposomes and ultrasound in combinationwith mannosylated liposomes to deliver plasmid DNA tomacrophages and dendritic cells was assessed [73]. Signifi-cant enhancement of transfection efficiencies was reportedusing these formulations in comparison to plasmid DNAalone and unmodified liposomes.

4. Liposome Drug Delivery forthe Treatment of Disease

4.1. Infection. A major role of mononuclear phagocytes isthe capture and presentation of pathogenic antigens. Certainpathogens are capable of surviving macrophage phagocytosissuch as Brucella species [80], HIV [81, 82], and mycobacteria[83]. As a result viruses and bacteria can be harbouredand proliferate within these cells. Macrophages can betterwithstand the cytopathic effects of HIV than T cells [81, 82],while some pathogens such as certain brucella species impairthe apoptotic ability of macrophages and monocytes [80],and subsequently survival time of the pathogen-infected cellis extended. As these cells can cross tissue barriers such as theblood brain barrier (BBB), the virus can spread unrestricted[81].

The ability of these pathogens to infect, evade the host’sphagocytic mechanisms, and replicate creating pathogenreservoirs that can disseminate throughout the body stressesthe importance of the development of targeted therapeuticsto macrophages and other phagocytic cells. Liposome deliv-ery to these pathogen reservoirs has received some attention[84, 85]. Targeting strategies studied to-date include the useof negatively charged liposomes containing PG [26, 27],sterically stabilized immunoliposomes incorporating surfaceanti-HLA-DR antibodies [86], tuftsin [87], galactosylated[88], and mannosylated [89] liposomes (Table 1). Overall inthese studies, the liposome encapsulation of anti-infectiveswas generally found to decrease cellular toxicity, modifypharmacokinetics, and improve targeting thereby enhancingthe overall efficacy of the anti-infective agents.

0

200

400

600

800

1000

1200

1400

1600

1800

Flu

ores

cen

ce/m

gpr

otei

n

Untreated Uncoated liposome Mannosylatedliposome

∗∗

Figure 3: Uptake of uncoated and mannosylated liposomes bymacrophage like differentiated THP-1 cells after 2 hours [78]. (n =6± SEM) ∗P < .05; ∗∗P < .001.

4.2. Inflammation and Cancer. Mononuclear phagocytes arerecruited to sites of injury and cancer, and these sites becomeareas with a high macrophage presence. As inflammatorycells, macrophages release proinflammatory cytokines suchas TNFα further increasing inflammation. This process canbe utilized in two ways for drug targeting. Firstly, cells canbe targeted and activated to bestow tumour suppressiveproperties for cancer therapy [7]. Secondly, for inflammatorydisease, the inflammatory response can be reduced usinganti-inflammatory drugs or cell killing to deplete mono-cyte/macrophage cell populations.

Activation of macrophages is a means of augmentingantitumor immune responses [4] by the induction ofproinflammatory mediators such as TNFα, IL-8, and nitricoxide (NO) [28]. For instance liposomal delivery of hexade-cylphosphocholine [2], JBT3002, a synthetic lipopeptide [3],the tetrapeptide (Thr-Lys-Pro-Arg) tuftsin, and muramyltripeptide phosphatidylethanolamine (MTP-PE) [28] hasbeen investigated. MTP-PE is a synthetic glycopeptide thatcan activate monocytes and macrophages promoting tumourregression [28]. A liposomal MTP-PE formulation (L-MTP-PE; mifamurtide) is currently in clinical trials for high riskosteosarcoma.

Bisphosphonates, for example, clodronate and alen-dronate, are extensively used in the treatment of osteo-porosis but have also shown the ability to induce apop-tosis in monocytes and macrophages. Interest lies in theirtherapeutic potential for inflammatory disorders. To datea range of potential therapies for inflammatory relatedconditions including nerve injury-associated hyperalgesia[90], endometriosis [91], lung cancer cell metastasis [92],arthritis [93], restinosis [24, 94], and hyperlipidemia [95]have been assessed using liposome-mediated bisphosphonatedelivery. Other inducers of macrophage apoptosis have beeninvestigated such as propamidine [96] and locally admin-istered inhibitors such as cycloheximide for atherosclerosistreatment [95].

4.3. Cardiovascular Disease. The role of monocytes/macro-phages in the development of atherosclerosis is undisputed[97, 98]. Following endothelial cell damage, monocytes


are recruited to the site via the release of chemokines.Following extravasation to the intima, recruited and residentmacrophages play a critical role in the development of theatherosclerotic plaque via the scavenging of oxidised LDLand the ultimate differentiation into foam cells which formthe atheroscelotic plaque core. The glycoprotein CD36 is cen-tral to this process. CD36 is a member of the scavenger recep-tor class B which is expressed on macrophages/monocytes,platelets, and endothelial cells. Its importance in atheroscle-rosis has clearly been established through studies in theApoE-deficient mice, demonstrating that inactivation ofCD36 results in substantially reduced lesion size. Thereforetargeting of CD36-expressing macrophages in atheroscleroticlesions using a ligand, for example, the growth peptideHexarelin, can be envisaged to have a dual effect—thedelivery of therapeutic agents to the lesion and the neu-tralisation of LDL uptake. Hexarelin, a member of thehexapeptide growth hormone-releasing peptides (GHRPs),binds to CD36 receptors [99].

Investigations into liposome targeting to atheroscleroticlesions have looked at their potential for delivery of con-trast agents for diagnostic imaging [100, 101] and anti-inflammatory drugs for therapy development. For instance,Chono and colleagues have investigated liposomal deliveryto macrophages as a therapeutic approach to atherosclerosisin several studies [22, 40, 102] using anionic liposomesconsisting of egg yolk phosphotidylcholine (PC), cholesterol,and DCP at a molar ratio 7 : 2 : 1 and sized to 70, 200 and500 nm. In vitro uptake by macrophages and foam cellswas improved with increasing particle size [22, 40, 102];however, in vivo, optimal aortic delivery in atherogenicmice was achieved using 200 nm liposomes. In addition,various studies have shown significant antiatheroscleroticeffects in vivo by liposomal delivery of dexamethasone,cyclopentenone prostaglandins, and serum amyloid A (SAA)peptide fragments [22, 30, 103].

4.4. Cerebral Ischemia and Stroke. The role of the innateimmune system and infiltrating macrophages and residentmicroglia in cerebral ischemia is currently an area ofintense investigation. Inflammation, be it sterile or infection-induced, plays an important part in cerebral ischemicinjury. Interestingly CD36 is upregulated in a number ofinflammatory and pathological conditions, such as cerebralischemia and stroke. Both CD36 and TLR2 are upregulatedon microglia and infiltrating macrophages under ischemicconditions and triggering either will induce a potent inflam-matory response [104, 105]. One study investigated the useof infiltrating macrophages to deliver a systemically admin-istered gene therapy in stroke [106]. Plasmids expressingenhanced green fluorescent protein (EGFP) and fibroblastgrowth factor-2 (FGF-2) were complexed with cationicliposomes, administered into the femoral vein resulting inexpression of EGFP and FGF-2 in infiltrating macrophagesand in the cerebral infarction.

4.5. Other. There has also been some attention paid to“Trojan monocytes” for drug delivery to the brain [107] asa means of delivering drugs to inaccessible sites (Figure 1).

Delivery of drugs to the brain is greatly hampered by theextremely selective permeability of the blood brain barrier(BBB). However, immune cells such as phagocytes can crossthis barrier. Therefore by targeting circulating mononuclearcells with drug-loaded liposomes, this natural BBB uptakeprocess can be harnessed for drug delivery.

Previous studies have used RGD-liposomes [29, 60, 61]as well as magnetic liposome formulations [29, 108] fordelivery to the brain via monocytes and neutrophils. Aferganet al. prepared PG-composed liposomes for the deliveryof the neurotransmitter serotonin [109]. In vivo studiesshowed localisation to the brain to be improved by liposomeencapsulation and that the delivered liposomes were intact.FACS analysis of rabbit blood 4 hours posttreatment showedhigher uptake of liposomes by monocytes over granulocytes.Uptake was also observed by monocytes and neutrophils invivo and in vitro but it was shown that monocytes were theneurodelivery cells by an alendronate monocyte depletionstudy [109]. More recently Saiyed et al. developed azi-dothymidine 5′-triphosphate (AZTTP) containing magneticliposomes as a therapeutic for neuroAIDS [108]. Magneticnanoparticles (Fe3O4, magnetite) were encapsulated withAZTTP in neutral liposomes, and transmigration of the lipo-somes in monocytes was monitored across an in vitro BBBmodel in the presence of a magnet. By magnetic liposomeendocytosis, monocytes become magnetic and respondedto magnetic fields [29]. The transmigration of magneticmonocytes was significantly increased in the presence of amagnet in comparison to nonmagneticlinebreak monocytes.

A study by Matsui et al. examined the potential ofperipheral blood monocytes (PBMCs) and human peri-toneal macrophages as drug carriers in gastric cancer [36].Oligomannose-coated liposomes were successfully targetedto monocytes and macrophages showing significantly higheruptake than bare liposomes. These liposome-loaded humanmonocytes and macrophages were found to accumulate atthe disease target site micrometastases and milky spots of theomentum in mice and ex vivo in resected human omentum.

5. Conclusion

As the role of monocytes and macrophages in a range ofdiseases including infectious disease, inflammatory diseases,cancer, and atherosclerosis is better understood, strategiesto target these cell types are of growing importance bothscientifically and therapeutically. Efficient methods of tar-geting these cells can facilitate efficient drug delivery butalso potentially facilitate cell activation and ablation. Theproperties of liposomes mean they naturally target cells ofthe MPS, particularly macrophages. This natural targetingcapacity can be harnessed for drug delivery. By controllingthe liposome physicochemical properties including size,charge, and lipid composition, natural targeting can befurther enhanced. A range of ligand-mediated strategies forliposome targeting to MPS cells have been explored includingpeptide-, antibody-, and lectin-coating to specifically targetdrug-loaded liposomes to some of the many receptor typesexpressed on macrophage and monocyte cells.


Acknowledgment

The authors would like to acknowledge the support receivedfrom the Irish Health Research Board (HRB) under Grantno. PHD/2007/11.

References

[1] K. S. Ravichandran and U. Lorenz, “Engulfment of apoptoticcells: signals for a good meal,” Nature Reviews Immunology,vol. 7, no. 12, pp. 964–974, 2007.

[2] I. Eue, “Growth inhibition of human mammary carci-noma by liposomal hexadecylphosphocholine: participationof activated macrophages in the antitumor mechanism,”International Journal of Cancer, vol. 92, no. 3, pp. 426–433,2001.

[3] I. Eue, R. Kumar, Z. Dong, J. J. Killion, and I. J. Fidler, “Induc-tion of nitric oxide production and tumoricidal properties inmurine macrophages by a new synthetic lipopeptide JBT3002encapsulated in liposomes,” Journal of Immunotherapy, vol.21, no. 5, pp. 340–351, 1998.

[4] J. J. Killion and I. J. Fidler, “Therapy of cancer metasta-sis by tumoricidal activation of tissue macrophages usingliposome-encapsulated immunomodulators,” Pharmacologyand Therapeutics, vol. 78, no. 3, pp. 141–154, 1998.

[5] S. Chono, T. Tanino, T. Seki, and K. Morimoto, “Efficientdrug targeting to rat alveolar macrophages by pulmonaryadministration of ciprofloxacin incorporated into manno-sylated liposomes for treatment of respiratory intracellularparasitic infections,” Journal of Controlled Release, vol. 127,no. 1, pp. 50–58, 2008.

[6] L. Kole, L. Das, and P. K. Das, “Synergistic effect ofinterferon-γ and mannosylated liposome- incorporated dox-orubicin in the therapy of experimental visceral leishmania-sis,” Journal of Infectious Diseases, vol. 180, no. 3, pp. 811–820,1999.

[7] N. van Rooijen and A. Sanders, “The macrophage astarget or obstacle in liposome-based targeting strategies,”International Journal of Pharmaceutics, vol. 162, no. 1-2, pp.45–50, 1998.

[8] B. Burke, S. Sumner, N. Maitland, and C. E. Lewis,“Macrophages in gene therapy: cellular delivery vehicles andin vivo targets,” Journal of Leukocyte Biology, vol. 72, no. 3,pp. 417–428, 2002.

[9] M. L. Immordino, F. Dosio, and L. Cattel, “Stealth lipo-somes: review of the basic science, rationale, and clinicalapplications, existing and potential,” International Jurnal ofNanomedicine, vol. 1, no. 3, pp. 297–315, 2006.

[10] C. Varol, S. Yona, and S. Jung, “Origins and tissue-context-dependent fates of blood monocytes,” Immunology and CellBiology, vol. 87, no. 1, pp. 30–38, 2009.

[11] C. S. Robbins and F. K. Swirski, “The multiple roles ofmonocyte subsets in steady state and inflammation,” Cellularand Molecular Life Sciences, vol. 67, no. 16, pp. 2685–2693,2010.

[12] M. Naito, “Macrophage differentiation and function inhealth and disease,” Pathology International, vol. 58, no. 3, pp.143–155, 2008.

[13] M. Benoit, B. Desnues, and J.-L. Mege, “Macrophage polar-ization in bacterial infections,” Journal of Immunology, vol.181, no. 6, pp. 3733–3739, 2008.

[14] A. Mantovani, A. Sica, and M. Locati, “Macrophage polar-ization comes of age,” Immunity, vol. 23, no. 4, pp. 344–346,2005.

[15] F. O. Martinez, A. Sica, A. Mantovani, and M. Locati,“Macrophage activation and polarization,” Frontiers in Bio-science, vol. 13, no. 2, pp. 453–461, 2008.

[16] A. Makkouk and A. M. Abdelnoor, “The potential use oftoll-like receptor (TLR) agonists and antagonists as prophy-lactic and/or therapeutic agents,” Immunopharmacology andImmunotoxicology, vol. 31, no. 3, pp. 331–338, 2009.

[17] D. Christensen, E. M. Agger, L. V. Andreasen, D. Kirby, P.Andersen, and Y. Perrie, “Liposome-based cationic adjuvantformulations (CAF): past, present, and future,” Journal ofLiposome Research, vol. 19, no. 1, pp. 2–11, 2009.

[18] M. Diwan, P. Elamanchili, H. Lane, A. Gainer, and J. Samuel,“Biodegradable nanoparticle mediated antigen delivery tohuman cord blood derived dendritic cells for induction ofprimary T cell responses,” Journal of Drug Targeting, vol. 11,no. 8-10, pp. 495–507, 2003.

[19] S. B. Coffelt, R. Hughes, and C. E. Lewis, “Tumor-associatedmacrophages: effectors of angiogenesis and tumor progres-sion,” Biochimica et Biophysica Acta, vol. 1796, no. 1, pp. 11–18, 2009.

[20] H.-P. Gerber, E. Olazoglu, and I. S. Grewal, “Targetinginflammatory cells to improve anti-VEGF therapies in oncol-ogy,” Recent Results in Cancer Research, vol. 180, pp. 185–200,2010.

[21] C. L. Van Broekhoven, C. R. Parish, C. Demangel, W. J.Britton, and J. G. Altin, “Targeting dendritic cells withantigen-containing liposomes: a highly effective procedurefor induction of antitumor immunity and for tumorimmunotherapy,” Cancer Research, vol. 64, no. 12, pp. 4357–4365, 2004.

[22] S. Chono, Y. Tauchi, Y. Deguchi, and K. Morimoto,“Efficient drug delivery to atherosclerotic lesions and theantiatherosclerotic effect by dexamethasone incorporatedinto liposomes in atherogenic mice,” Journal of Drug Target-ing, vol. 13, no. 4, pp. 267–276, 2005.

[23] A. M. Gibbons, N. G. McElvaney, C. C. Taggart, and S.-A.Cryan, “Delivery of rSLPI in a liposomal carrier for inhala-tion provides protection against cathepsin L degradation,”Journal of Microencapsulation, vol. 26, no. 6, pp. 513–522,2009.

[24] H. Epstein-Barash, D. Gutman, E. Markovsky et al., “Physic-ochemical parameters affecting liposomal bisphosphonatesbioactivity for restenosis therapy: internalization, cell inhi-bition, activation of cytokines and complement, and mecha-nism of cell death,” Journal of Controlled Release, vol. 146, no.2, pp. 182–195, 2010.

[25] S. P. Vyas, M. E. Kannan, S. Jain, V. Mishra, and P. Singh,“Design of liposomal aerosols for improved delivery ofrifampicin to alveolar macrophages,” International Journal ofPharmaceutics, vol. 269, no. 1, pp. 37–49, 2004.

[26] C. Oussoren, M. Magnani, A. Fraternale et al., “Liposomesas carriers of the antiretroviral agent dideoxycytidine-5’-triphosphate,” International Journal of Pharmaceutics, vol.180, no. 2, pp. 261–270, 1999.

[27] I. I. Salem and N. Duzgunes, “Efficacies of cyclodextrin-complexed and liposome-encapsulated clarithromycinagainst Mycobacterium avium complex infection in humanmacrophages,” International Journal of Pharmaceutics, vol.250, no. 2, pp. 403–414, 2003.

[28] A. Nardin, M.-L. Lefebvre, K. Labroquere, O. Faure,and J.-P. Abastado, “Liposomal muramyl tripeptide phos-phatidylethanolamine: targeting and activating macrophagesfor adjuvant treatment of osteosarcoma,” Current CancerDrug Targets, vol. 6, no. 2, pp. 123–133, 2006.


[29] S. Jain, V. Mishra, P. Singh, P. K. Dubey, D. K. Saraf,and S. P. Vyas, “RGD-anchored magnetic liposomes formonocytes/neutrophils-mediated brain targeting,” Interna-tional Journal of Pharmaceutics, vol. 261, no. 1-2, pp. 43–55,2003.

[30] P. I. Homem de Bittencourt Jr., D. J. Lagranha, A.Maslinkiewicz et al., “LipoCardium: endothelium-directedcyclopentenone prostaglandin-based liposome formulationthat completely reverses atherosclerotic lesions,” Atheroscle-rosis, vol. 193, no. 2, pp. 245–258, 2007.

[31] G. A. Koning, H. W. M. Morselt, A. Gorter et al., “Interactionof differently designed immunoliposomes with colon cancercells and Kupffer cells. An in vitro comparison,” Pharmaceu-tical Research, vol. 20, no. 8, pp. 1249–1257, 2003.

[32] G. A. Koning, J. A. A. M. Kamps, and G. L. Scherphof,“Interference of macrophages with immunotargeting ofliposomes,” Journal of Liposome Research, vol. 12, no. 1-2, pp.107–119, 2002.

[33] W. Wijagkanalan, Y. Higuchi, S. Kawakami, M. Teshima, H.Sasaki, and M. Hashida, “Enhanced anti-inflammation ofinhaled dexamethasone palmitate using mannosylated lipo-somes in an endotoxin-induced lung inflammation model,”Molecular Pharmacology, vol. 74, no. 5, pp. 1183–1192, 2008.

[34] S. Espuelas, C. Thumann, B. Heurtault, F. Schuber, and B.Frisch, “Influence of ligand valency on the targeting of imma-ture human dendritic cells by mannosylated liposomes,”Bioconjugate Chemistry, vol. 19, no. 12, pp. 2385–2393, 2008.

[35] M. Ishii and N. Kojima, “Mucosal adjuvant activity ofoligomannose-coated liposomes for nasal immunization,”Glycoconjugate Journal, vol. 27, no. 1, pp. 1–9, 2009.

[36] M. Matsui, Y. Shimizu, Y. Kodera, E. Kondo, Y. Ikehara, andH. Nakanishi, “Targeted delivery of oligomannose-coatedliposome to the omental micrometastasis by peritonealmacrophages from patients with gastric cancer,” CancerScience, vol. 101, no. 7, pp. 1670–1677, 2010.

[37] F. Ahsan, I. P. Rivas, M. A. Khan, and A. I. Torres Suarez,“Targeting to macrophages: role of physicochemical proper-ties of particulate carriers—Liposomes and microspheres—on the phagocytosis by macrophages,” Journal of ControlledRelease, vol. 79, no. 1–3, pp. 29–40, 2002.

[38] S. Takano, Y. Aramaki, and S. Tsuchiya, “Physicochem-ical properties of liposomes affecting apoptosis inducedby cationic liposomes in macrophages,” PharmaceuticalResearch, vol. 20, no. 7, pp. 962–968, 2003.

[39] S. Chono, T. Tanino, T. Seki, and K. Morimoto, “Influenceof particle size on drug delivery to rat alveolar macrophagesfollowing pulmonary administration of ciprofloxacin incor-porated into liposomes,” Journal of Drug Targeting, vol. 14,no. 8, pp. 557–566, 2006.

[40] S. Chono, Y. Tauchi, and K. Morimoto, “Influence of particlesize on the distributions of liposomes to atheroscleroticlesions in mice,” Drug Development and Industrial Pharmacy,vol. 32, no. 1, pp. 125–135, 2006.

[41] S. Chono, T. Tanino, T. Seki, and K. Morimoto, “Uptakecharacteristics of liposomes by rat alveolar macrophages:influence of particle size and surface mannose modification,”Journal of Pharmacy and Pharmacology, vol. 59, no. 1, pp. 75–80, 2007.

[42] X. Yan, G. L. Scherphof, and J. A. A. M. Kamps, “Liposomeopsonization,” Journal of Liposome Research, vol. 15, no. 1-2,pp. 109–139, 2005.

[43] I. S. Zuhorn, J. B. F. N. Engberts, and D. Hoekstra,“Gene delivery by cationic lipid vectors: overcoming cellular

barriers,” European Biophysics Journal, vol. 36, no. 4-5, pp.349–362, 2007.

[44] C. M. Wiethoff, J. G. Smith, G. S. Koe, and C. R. Middaugh,“The potential role of proteoglycans in cationic lipid-mediated gene delivery: studies of the interaction of cationiclipid-DNA complexes with model glycosaminoglycans,” TheJournal of Biological Chemistry, vol. 276, no. 35, pp. 32806–32813, 2001.

[45] H. Lv, S. Zhang, B. Wang, S. Cui, and J. Yan, “Toxicityof cationic lipids and cationic polymers in gene delivery,”Journal of Controlled Release, vol. 114, no. 1, pp. 100–109,2006.

[46] S. Iwaoka, T. Nakamura, S. Takano, S. Tsuchiya, and Y.Aramaki, “Cationic liposomes induce apoptosis throughp38 MAP kinase-caspase-8-Bid pathway in macrophage-likeRAW264.7 cells,” Journal of Leukocyte Biology, vol. 79, no. 1,pp. 184–191, 2006.

[47] Y. Aramaki, S. Takano, and S. Tsuchiya, “Cationic liposomesinduce macrophage apoptosis through mitochondrial path-way,” Archives of Biochemistry and Biophysics, vol. 392, no. 2,pp. 245–250, 2001.

[48] M. Arisaka et al., “Involvement of protein kinase Cdelta ininduction of apoptosis by cationic liposomes in macrophage-like RAW264.7 cells,” FEBS Letters, vol. 584, no. 5, pp. 1016–1020, 2010.

[49] I. J. Fidler, A. Raz, and W. E. Fogler, “Design of liposomesto improve delivery of macrophage-augmenting agents toalveolar macrophages,” Cancer Research, vol. 40, no. 12, pp.4460–4466, 1980.

[50] A. Gibbons et al., “Liposome Encapsulation InfluencesPharmacokinetics and Activity of recombinant SecretoryLeukocyte Protease Inhibitor (rSLPI) after Local Deliveryto the Lungs: an In Vitro and In Vivo Assessment,” inProceedings of the American Association of PharmaceuticalScientists Annual Meeting and Exposition, Atlanta, Ga, USA,2008.

[51] I. M. Hafez and P. R. Cullis, “Roles of lipid polymorphism inintracellular delivery,” Advanced Drug Delivery Reviews, vol.47, no. 2-3, pp. 139–148, 2001.

[52] G. L. Scherphof and J. A. A. M. Kamps, “Receptor versus non-receptor mediated clearance of liposomes,” Advanced DrugDelivery Reviews, vol. 32, no. 1-2, pp. 81–97, 1998.

[53] S. M. Moghimi and A. C. Hunter, “Recognition bymacrophages and liver cells of opsonized phospholipid vesi-cles and phospholipid headgroups,” Pharmaceutical Research,vol. 18, no. 1, pp. 1–8, 2001.

[54] C. Foged, C. Arigita, A. Sundblad, W. Jiskoot, G. Storm,and S. Frokjaer, “Interaction of dendritic cells with antigen-containing liposomes: effect of bilayer composition,” Vaccine,vol. 22, no. 15-16, pp. 1903–1913, 2004.

[55] P. Sun, M. Zhong, X. Shi, and Z. Li, “Anionic LPDcomplexes for gene delivery to macrophage: preparation,characterization and transfection in vitro,” Journal of DrugTargeting, vol. 16, no. 9, pp. 668–678, 2008.

[56] E. Andreakos, U. Rauchhaus, A. Stavropoulos et al., “Ampho-teric liposomes enable systemic antigen-presenting cell-directed delivery of CD40 antisense and are therapeuticallyeffective in experimental arthritis,” Arthritis and Rheumatism,vol. 60, no. 4, pp. 994–1005, 2009.

[57] R. L. Juliano, R. Alam, V. Dixit, and H. M. Kang, “Cell-targeting and cell-penetrating peptides for delivery of ther-apeutic and imaging agents,” Wiley Interdisciplinary Reviews.Nanomedicine and Nanobiotechnology, vol. 1, no. 3, pp. 324–335, 2009.


[58] E. Karathanasis, C. M. Geigerman, C. A. Parkos, L. Chan,R. V. Bellamkonda, and D. L. Jaye, “Selective targetingof nanocarriers to neutrophils and monocytes,” Annals ofBiomedical Engineering, vol. 37, no. 10, pp. 1984–1992, 2009.

[59] L. Mazzucchelli, J. B. Burritt, A. J. Jesaitis et al., “Cell-specificpeptide binding by human neutrophils,” Blood, vol. 93, no. 5,pp. 1738–1748, 1999.

[60] J. Qin, D. Chen, H. Hu, M. Qiao, X. Zhao, and B. Chen, “Bodydistributioin of RGD-mediated liposome in brain-targetingdrug delivery,” Yakugaku Zasshi, vol. 127, no. 9, pp. 1497–1501, 2007.

[61] J. Qin, D. Chen, H. Hu, Q. Cui, M. Qiao, and B. Chen,“Surface modification of RGD-liposomes for selective drugdelivery to monocytes/neutrophils in brain,” Chemical andPharmaceutical Bulletin, vol. 55, no. 8, pp. 1192–1197, 2007.

[62] T. E. Wileman, M. R. Lennartz, and P. D. Stahl, “Identifi-cation of the macrophage mannose receptor as a 175-kDamembrane protein,” Proceedings of the National Academy ofSciences of the United States of America, vol. 83, no. 8, pp.2501–2505, 1986.

[63] E. P. McGreal, L. Martinez-Pomares, and S. Gordon, “Diver-gent roles for C-type lectins expressed by cells of the innateimmune system,” Molecular Immunology, vol. 41, no. 11, pp.1109–1121, 2004.

[64] V. L. Shepherd, Y. C. Lee, P. H. Schlesinger, and P. D.Stahl, “L-Fucose-terminated glycoconjugates are recognizedby pinocytosis receptors on macrophages,” Proceedings of theNational Academy of Sciences of the United States of America,vol. 78, no. 2, pp. 1019–1022, 1981.

[65] P. D. Stahl, J. S. Rodman, M. J. Miller, and P. H. Schlesinger,“Evidence for receptor-mediated binding of glycoproteins,glycoconjugates, and lysosomal glycosidases by alveolarmacrophages,” Proceedings of the National Academy of Sci-ences of the United States of America, vol. 75, no. 3, pp. 1399–1403, 1978.

[66] J. M. Irache, H. H. Salman, C. Gamazo, and S. Espuelas,“Mannose-targeted systems for the delivery of therapeutics,”Expert Opinion on Drug Delivery, vol. 5, no. 6, pp. 703–724,2008.

[67] U. Gazi and L. Martinez-Pomares, “Influence of the mannosereceptor in host immune responses,” Immunobiology, vol.214, no. 7, pp. 554–561, 2009.

[68] E. J. McKenzie, P. R. Taylor, R. J. Stillion et al., “Mannosereceptor expression and function define a new population ofmurine dendritic cells,” Journal of Immunology, vol. 178, no.8, pp. 4975–4983, 2007.

[69] S. Chono, K. Kaneko, E. Yamamoto, K. Togami, and K.Morimoto, “Effect of surface-mannose modification onaerosolized liposomal delivery to alveolar macrophages,”Drug Development and Industrial Pharmacy, vol. 36, no. 1,pp. 102–107, 2010.

[70] A. Engel, S. K. Chatterjee, A. Al-arifi, D. Riemann, J. Langner,and P. Nuhn, “Influence of spacer length on interactionof mannosylated liposomes with human phagocytic cells,”Pharmaceutical Research, vol. 20, no. 1, pp. 51–57, 2003.

[71] H. Takagi, M. Numazaki, T. Kajiwara et al., “Cooperation ofspecific ICAM-3 grabbing nonintegrin-related 1 (SIGNR1)and complement receptor type 3 (CR3) in the uptake ofoligomannose-coated liposomes by macrophages,” Glycobi-ology, vol. 19, no. 3, pp. 258–266, 2009.

[72] W. Wijagkanalan, S. Kawakami, M. Takenaga, R. Igarashi, F.Yamashita, and M. Hashida, “Efficient targeting to alveolar

macrophages by intratracheal administration of mannosy-lated liposomes in rats,” Journal of Controlled Release, vol.125, no. 2, pp. 121–130, 2008.

[73] K. Un, S. Kawakami, R. Suzuki, K. Maruyama, F. Yamashita,and M. Hashida, “Enhanced transfection efficiency intomacrophages and dendritic cells by a combination methodusing mannosylated lipoplexes and bubble liposomes withultrasound exposure,” Human Gene Therapy, vol. 21, no. 1,pp. 65–74, 2010.

[74] S. Kawakami, A. Sato, M. Nishikawa, F. Yamashita, and M.Hashida, “Mannose receptor-mediated gene transfer intomacrophages using novel mannosylated cationic liposomes,”Gene Therapy, vol. 7, no. 4, pp. 292–299, 2000.

[75] S. Kawakami, J. Wong, A. Sato, Y. Hattori, F. Yamashita,and M. Hashida, “Biodistribution characteristics of manno-sylated, fucosylated, and galactosylated liposomes in mice,”Biochimica et Biophysica Acta, vol. 1524, no. 2-3, pp. 258–265,2000.

[76] S. Chono, K. Kaneko, E. Yamamoto, K. Togami, and K.Morimoto, “Effect of surface-mannose modification onaerosolized liposomal delivery to alveolar macrophages,”Drug Development and Industrial Pharmacy, vol. 36, no. 1,pp. 102–107, 2010.

[77] K. Nakamura, Y. Kuramoto, H. Mukai, S. Kawakami, Y.Higuchi, and M. Hashida, “Enhanced gene transfection inmacrophages by histidine-conjugated mannosylated cationicliposomes,” Biological and Pharmaceutical Bulletin, vol. 32,no. 9, pp. 1628–1631, 2009.

[78] C. Kelly et al., “Mannosylated liposomes for targeted deliveryof siRNA to alveolar macrophages,” in Proceedings of the 7thBritish Society for Gene Therapy Annual Conference, London,UK, 2010.

[79] Y. Kuramoto, S. Kawakami, S. Zhou, K. Fukuda, F. Yamashita,and M. Hashida, “Use of mannosylated cationic lipo-somes/immunostimulatory CpG DNA complex for effectiveinhibition of peritoneal dissemination in mice,” Journal ofGene Medicine, vol. 10, no. 4, pp. 392–399, 2008.

[80] J. Dornand, A. Gross, V. Lafont, J. Liautard, J. Oliaro, and J.-P. Liautard, “The innate immune response against Brucella inhumans,” Veterinary Microbiology, vol. 90, no. 1–4, pp. 383–394, 2002.

[81] C. A. Carter and L. S. Ehrlich, “Cell biology of HIV-1infection of macrophages,” Annual Review of Microbiology,vol. 62, pp. 425–443, 2008.

[82] E. Cassol, M. Alfano, P. Biswas, and G. Poli, “Monocyte-derived macrophages and myeloid cell lines as targets of HIV-1 replication and persistence,” Journal of Leukocyte Biology,vol. 80, no. 5, pp. 1018–1030, 2006.

[83] J. Pieters and J. Gatfield, “Hijacking the host: survival ofpathogenic mycobacteria inside macrophages,” Trends inMicrobiology, vol. 10, no. 3, pp. 142–146, 2002.

[84] S. Gunaseelan, K. Gunaseelan, M. Deshmukh, X. Zhang,and P. J. Sinko, “Surface modifications of nanocarriers foreffective intracellular delivery of anti-HIV drugs,” AdvancedDrug Delivery Reviews, vol. 62, no. 4-5, pp. 518–531, 2010.

[85] E. Ojewole, I. Mackraj, P. Naidoo, and T. Govender, “Explor-ing the use of novel drug delivery systems for antiretroviraldrugs,” European Journal of Pharmaceutics and Biopharma-ceutics, vol. 70, no. 3, pp. 697–710, 2008.

[86] J. Bestman-Smith, P. Gourde, A. Desormeaux, M. J. Trem-blay, and M. G. Bergeron, “Sterically stabilized liposomesbearing anti-HLA-DR antibodies for targeting the primarycellular reservoirs of HIV-1,” Biochimica et Biophysica Acta,vol. 1468, no. 1-2, pp. 161–174, 2000.


[87] A. K. Agrawal and C. M. Gupta, “Tuftsin-bearing liposomesin treatment of macrophage-based infections,” AdvancedDrug Delivery Reviews, vol. 41, no. 2, pp. 135–146, 2000.

[88] M. Garg and N. K. Jain, “Reduced hematopoietic toxicity,enhanced cellular uptake and altered pharmacokinetics ofazidothymidine loaded galactosylated liposomes,” Journal ofDrug Targeting, vol. 14, no. 1, pp. 1–11, 2006.

[89] M. Garg, A. Asthana, H. B. Agashe, G. P. Agrawal, and N.K. Jain, “Stavudine-loaded mannosylated liposomes: in-vitroanti-HIV-1 activity, tissue distribution and pharmacokinet-ics,” Journal of Pharmacy and Pharmacology, vol. 58, no. 5,pp. 605–616, 2006.

[90] T. Liu, N. Van Rooijen, and D. J. Tracey, “Depletion ofmacrophages reduces axonal degeneration and hyperalgesiafollowing nerve injury,” Pain, vol. 86, no. 1-2, pp. 25–32,2000.

[91] E. Haber, H. D. Danenberg, N. Koroukhov, R. Ron-El, G.Golomb, and M. Schachter, “Peritoneal macrophage deple-tion by liposomal bisphosphonate attenuates endometriosisin the rat model,” Human Reproduction, vol. 24, no. 2, pp.398–407, 2009.

[92] K. Hiraoka, M. Zenmyo, K. Watari et al., “Inhibition of boneand muscle metastases of lung cancer cells by a decrease in thenumber of monocytes/macrophages,” Cancer Science, vol. 99,no. 8, pp. 1595–1602, 2008.

[93] P. J. Richards, B. D. Williams, and A. S. Williams, “Suppres-sion of chronic streptococcal cell wall-induced arthritis inLewis rats by liposomal clodronate,” Rheumatology, vol. 40,no. 9, pp. 978–987, 2001.

[94] H. D. Danenberg, G. Golomb, A. Groothuis et al., “Lipo-somal Alendronate Inhibits Systemic Innate Immunity andReduces In-Stent Neointimal Hyperplasia in Rabbits,” Circu-lation, vol. 108, no. 22, pp. 2798–2804, 2003.

[95] M. V. Calin, I. Manduteanu, E. Dragomir et al., “Effectof depletion of monocytes/macrophages on early aorticvalve lesion in experimental hyperlipidemia,” Cell and TissueResearch, vol. 336, no. 2, pp. 237–248, 2009.

[96] N. Van Rooijen, “Manipulation of Kupffer cells by liposomeencapsulated clodronate and propamidine—synergistic andantagonistic effects of liposomal phospholipids and drugs,”International Journal of Pharmaceutics, vol. 162, no. 1-2, pp.51–58, 1998.

[97] E. Galkina and K. Ley, “Immune and inflammatory mecha-nisms of atherosclerosis,” Annual Review of Immunology, vol.27, pp. 165–197, 2009.

[98] K. J. Woollard and F. Geissmann, “Monocytes in atheroscle-rosis: subsets and functions,” Nature Reviews Cardiology, vol.7, no. 2, pp. 77–86, 2010.

[99] A. Demers, N. McNicoll, M. Febbraio et al., “Identification ofthe growth hormone-releasing peptide binding site in CD36:a photoaffinity cross-linking study,” Biochemical Journal, vol.382, no. 2, pp. 417–424, 2004.

[100] K. C. Briley-Saebo, W. J. M. Mulder, V. Mani et al., “Magneticresonance imaging of vulnerable atherosclerotic plaques:current imaging strategies and molecular imaging probes,”Journal of Magnetic Resonance Imaging, vol. 26, no. 3, pp.460–479, 2007.

[101] A. Maiseyeu, G. Mihai, T. Kampfrath et al., “Gadolinium-containing phosphatidylserine liposomes for molecularimaging of atherosclerosis,” Journal of Lipid Research, vol. 50,no. 11, pp. 2157–2163, 2009.

[102] S. Chono, Y. Tauchi, and K. Morimoto, “Pharmacokineticanalysis of the uptake of liposomes by macrophages and foamcells in vitro and their distribution to atherosclerotic lesions

in mice,” Drug Metabolism and Pharmacokinetics, vol. 21, no.1, pp. 37–44, 2006.

[103] S. P. Tam, J. B. Ancsin, R. Tan, and R. Kisilevsky, “Peptidesderived from serum amyloid A prevent, and reverse, aorticlipid lesions in apoE-/- mice,” Journal of Lipid Research, vol.46, no. 10, pp. 2091–2101, 2005.

[104] E. Kim, A. T. Tolhurst, Y. Q. Lu, X.-Y. Chen, M. Febbraio, andS. Cho, “CD36/fatty acid translocase, an inflammatory medi-ator, is involved in hyperlipidemia-induced exacerbation inischemic brain injury,” Journal of Neuroscience, vol. 28, no.18, pp. 4661–4670, 2008.

[105] G. Ziegler, D. Harhausen, C. Schepers et al., “TLR2 has adetrimental role in mouse transient focal cerebral ischemia,”Biochemical and Biophysical Research Communications, vol.359, no. 3, pp. 574–579, 2007.

[106] S. Tanaka, K. Kitagawa, S. Sugiura et al., “Infiltratingmacrophages as in vivo targets for intravenous gene deliveryin cerebral infarction,” Stroke, vol. 35, no. 8, pp. 1968–1973,2004.

[107] K. Park, “Trojan monocytes for improved drug delivery tothe brain,” Journal of Controlled Release, vol. 132, no. 2, p. 75,2008.

[108] Z. M. Saiyed, N. H. Gandhi, and M. P.N. Nair, “Magneticnanoformulation of azidothymidine 5′-triphosphate for tar-geted delivery across the blood-brain barrier,” InternationalJournal of Nanomedicine, vol. 5, no. 1, pp. 157–166, 2010.

[109] E. Afergan, H. Epstein, R. Dahan et al., “Delivery ofserotonin to the brain by monocytes following phagocytosisof liposomes,” Journal of Controlled Release, vol. 132, no. 2,pp. 84–90, 2008.


Research Article

Characterization and In Vitro Skin Permeation ofMeloxicam-Loaded Liposomes versus Transfersomes

Sureewan Duangjit, Praneet Opanasopit, Theerasak Rojanarata, and Tanasait Ngawhirunpat

Faculty of Pharmacy, Silpakorn University, Sanamchan Palace Campus, Nakhon Pathom 73000, Thailand

Correspondence should be addressed to Tanasait Ngawhirunpat, [email protected]

Received 16 May 2010; Revised 11 September 2010; Accepted 18 October 2010

Academic Editor: Jia You Fang

Copyright © 2011 Sureewan Duangjit et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

The goal of this study was to develop and evaluate the potential use of liposome and transfersome vesicles in the transdermaldrug delivery of meloxicam (MX). MX-loaded vesicles were prepared and evaluated for particle size, zeta potential, entrapmentefficiency (%EE), loading efficiency, stability, and in vitro skin permeation. The vesicles were spherical in structure, 90 to 140 nmin size, and negatively charged (−23 to −43 mV). The %EE of MX in the vesicles ranged from 40 to 70%. Transfersomes provideda significantly higher skin permeation of MX compared to liposomes. Fourier Transform Infrared Spectroscopy (FT-IR) andDifferential Scanning Calorimetry (DSC) analysis indicated that the application of transfersomes significantly disrupted thestratum corneum lipid. Our research suggests that MX-loaded transfersomes can be potentially used as a transdermal drug deliverysystem.

1. Introduction

Transdermal drug delivery systems (TDDs) offer a numberof potential advantages over conventional methods suchas injectable and oral delivery [1]. However, the majorlimitation of TDDs is the permeability of the skin; it is per-meable to small molecules and lipophilic drugs and highlyimpermeable to macromolecules and hydrophilic drugs. Themain barrier and rate-limiting step for diffusion of drugsacross the skin is provided by the outermost layer of the skin,the stratum corneum (SC) [2]. Several strategies have beendeveloped to overcome the skin’s resistance, including the useof prodrugs, ion pairs, liposomes, microneedles, ultrasound,and iontophoresis [3–6].

Various types of liposomes (LPs) exist, such as traditionalliposomes, niosomes, ethosomes, and transfersomes [3, 7–12]. Various LPs have been extensively investigated forimproving skin permeation enhancement. Liposomes arepromising carriers for enhancing skin permeation becausethey have high membrane fluidity. Previous reports indicatethat liposomes can deliver a large quantity of hydrophilicdrugs (e.g., sodium fluorescein [13], carboxyfluorescein[14]), lipophilic drugs (e.g., retinoic acid [11], tretinoin

[12]), proteins, and macromolecules through the skin. Manyfactors influence the percutaneous penetration behavior ofLPs, including particle size, surface charge, lipid composi-tion, bilayer elasticity, lamellarity, and type of LP [7, 12].

Cevc’s group introduced Transfersomes, which are thefirst generation of elastic vesicles. Transfersomes are pre-pared from phospholipids and edge activators. An edgeactivator is often a single-chain surfactant with a highradius of curvature that destabilizes the lipid bilayers ofthe vesicles and increases the deformability of the bilayers.Sodium cholate, sodium deoxycholate, Span 60, Span 65,Span 80, Tween 20, Tween 60, Tween 80, and dipotassiumglycyrrhizinate were employed as edge activators. Comparedwith subcutaneous administration, transfersomes improvedin vitro skin permeation of various drugs, penetrated intactskin in vivo, and efficiently transferred therapeutic amountsof drugs [9, 15, 16]. However, the mechanism by which LPsand their analogs deliver drugs through the skin is not fullyunderstood [14].

Meloxicam (Figure 1) has low aqueous solubility, andit is a highly potent, nonsteroidal anti-inflammatory drug(NSAID) that is used for treatment of rheumatoid arthritisand osteoarthritis [6, 17–19]. MX shows similar efficacy


S

HN

N

N

OH O

S

MXO O

(a)

EPC

N+ OP

OO

O

O

O

OO−

(b)

Chol

HO

H

H

H

(c)

NaO O

O−Na+

(d)

NaChol OH

OHHOH

O

O−Na+

(e)

DCP

O

O

P O

OH

(f)

Figure 1: The chemical structure of meloxicam and the lipid compositions of the liposomes.

for reducing pain and inflammatory symptoms, but it haslower toxicity than other NSAIDs. Although MX is relativelypotent and safe, its limitations include low solubility, lowincorporation in formulations, and low skin permeation[6, 18–25]. In this study, vesicles were used as a novel MXtransdermal drug delivery system. The system was developedand evaluated for its physicochemical characteristics, such asparticle size, surface charge, entrapment efficiency, loadingefficiency, stability, and in vitro skin permeation. The typeof vesicles (liposomes and transfersomes), the compositionof lipid in the liposomes (cholesterol), and transfersomes(cholesterol and surfactants) were evaluated. Three surfac-tants that differ in length of carbon chains were used forthe preparation of transfersomes: sodium oleate (NaO, C18),sodium cholate (NaChol, C24), and dicetylphosphate (DCP,C32). Characterization of skin permeation was performedusing FTIR and DSC. Figure 1 shows the chemical structureof meloxicam and the lipid compositions of the liposomes.


2.1. Materials. Phosphatidylcholine (PC) from eggs waspurchased from GmbH. Cholesterol (Chol) was purchasedfrom Carlo Erba Reagenti. Sodium cholate (NaChol) was

purchased from Acros Organics. Sodium oleate (NaO)and dicetylphosphate (DCP) were purchased from Sigma-Aldrich. Meloxicam (MX) was supplied from Fluka.

2.2. Preparation of Meloxicam-Loaded Liposomes, Transfer-somes, and Suspensions. Liposomes containing a controlledamount of PC and various amounts of MX were formulated.The MX concentration was varied from 2.5 to 70.0 wt. % ofthe PC. The sonication method was used to prepare differentformulations; they were composed of bilayer-forming PCand either Chol, NaO, NaChol, or DCP in a molar ratioof 10 : 2. The PC, Chol, NaO, NaChol, DCP, and MX wereeach briefly dissolved in chloroform:methanol (2 : 1 v/v).In preparing MX-loaded liposomes and transfersomes, thematerials were deposited in a test tube, and the solvent wasevaporated with nitrogen gas. The lipid film was placed ina desiccator connected to a vacuum pump for a minimumof 6 h to remove the remaining organic solvent. The driedlipid film was hydrated with Tris buffer. Following hydration,the dispersion was sonicated in a bath for 30 min andthen probe-sonicated for 2 cycles of 30 min. The lipidcompositions of the different formulations utilized in thisstudy are listed in Table 1.


Table 1: The lipid compositions of the different formulations used in study.

Name (molar ratio)Composition (%W/V)

MX PC Chol NaO NaChol DCP PBS ph 7.4

MX/PC (2 : 10) 0.07 0.77 — — — — 100 mL

MX/PC/Chol (2 : 10 : 2) 0.07 0.77 0.07 — — — 100 mL

MX/PC/NaO (2 : 10 : 2) 0.07 0.77 — 0.06 — — 100 mL

MX/PC/NaO/Chol (2 : 10 : 2 : 2) 0.07 0.77 0.07 0.06 — — 100 mL

MX/PC/NaChol (2 : 10 : 2) 0.07 0.77 — — 0.08 — 100 mL

MX/PC/NaChol/Chol (2 : 10 : 2 : 2) 0.07 0.77 0.07 — 0.08 — 100 mL

MX/PC/DCP (2 : 10 : 2) 0.07 0.77 — — — 0.11 100 mL

MX/PC/DCP/Chol (2 : 10 : 2 : 2) 0.07 0.77 0.07 — — 0.11 100 mL

0.2μm

(a)

0.1μm

(b)

0.1μm

(c)

0.2μm

(d)

0.1μm

(e)

0.1μm

(f)

Figure 2: Transmission electron microscopy of MX loaded in vesicles. (a) visualization of MX loaded in liposomes (PC) (10,000x), (b)visualization of MX loaded in liposomes (PC) (30,000x), (c) visualization of MX loaded in liposomes (PC) (50,000x), (d) Visualizationof MX loaded in transfersomes (PC/NaChol) (10,000x), (e) visualization of MX loaded in transfersomes (PC/NaChol) (30,000x), and (f)visualization of MX loaded in transfersomes (PC/NaChol) (50,000x).

For the preparation of MX suspensions, the saturatedsolubility of MX in water was determined to ensure excessdrug in MX suspension. The solubility of MX was deter-mined by adding excess amount of MX to 5 mL of waterin a glass vial and stirring by a magnetic stirrer for 24 h.The sample was filtered through 0.45 μm membrane filter inorder to remove undissolved drugs in the saturated solution.The concentration of MX was analyzed by HPLC. The MXsuspension was prepared by adding MX to distilled water ata concentration 2 times higher than the solubility of MX andstirring for 24 h to ensure constant thermodynamic activitythroughout the course of the permeation experiment. Theparticle size of MX suspension was determined, and theMX suspension was used in the skin permeation experi-ment.

2.3. Characterization of Liposomes and Transfersomes

2.3.1. Particle Size and Surface Charge. The droplet sizeand zeta potential of the liposomes and transfersomes weredetermined by a Laser Scattering Particle Size DistributionAnalyzer and Zeta Potential Analyzer at room temperature.One mL of the liposome and transfersome suspensions werediluted with 14 mL and 2 mL deionized water, respectively.

2.3.2. Transmission Electron Microscopy. Transmission Elec-tron Microscopy (TEM) was used to visualize the liposomaland transfersomal vesicles. The vesicles were dried on acopper grid and adsorbed with filter paper. After drying,the sample was viewed under the microscope at 10–100 kmagnification at an accelerating voltage of 100 kV.


0

20

40

60

80

100

En

trap

men

teffi

cien

cy(%

)

0

20

40

60

80

100

Load

ing

effici

ency

(mg/

g)

2.5 5 10 20 30 50 70

Initial amount of meloxicam (% to PC)

(a)

0

20

40

60

80

100

En

trap

men

teffi

cien

cy(%

)

0

20

40

60

80

100

Load

ing

effici

ency

(mg/

g)

MX

/PC

MX

/PC

/Ch

ol

MX

/PC

/NaO

MX

/PC

/NaC

hol

MX

/PC

/NaO

/Ch

ol

MX

/PC

/NaC

hol

/Ch

ol

MX

/PC

/DC

P

MX

/PC

/DC

P/C

hol

∗∗

∗ ∗

(b)

Figure 3: (a) The effect of initial amount of meloxicam (2.5, 5, 10,20, 30, 50, and 70%) added in liposomes on percentage entrapmentefficiency (white bar) and loading efficiency (fill square) of meloxi-cam loaded in liposomes composed of PC. Each value represents themean±SD (n = 3) (b) The percentage entrapment efficiency (whitebar) and loading efficiency (fill square) of meloxicam loaded in dif-ferent formulations: (shaded square) liposomes and (white square)transfersomes. Each value represents the mean ± SD (n = 6).

2.3.3. Entrapment Efficiency (%EE) and Loading Efficiency.The concentration of MX in the formulation was determinedby HPLC analysis after disruption of the vesicles (liposomesand transfersomes) with Triton X-100 (0.1% w/v) at a1 : 1 volume ratio and appropriate dilution with PBS (pH7.4). The vesicle/Triton X-100 solution was centrifugedat 10,000 rpm at 4◦C for 10 min. The supernatant wasfiltered with a 0.45 μm nylon syringe filter. The entrapmentefficiencies and the loading efficiencies of the MX-loadedformulation were calculated by (1) and (2), respectively.

% entrapment efficiency =(CL

Ci

)× 100, (1)

where CL is the concentration of MX loaded in the formula-tion as described in the above methods, and Ci is the initialconcentration of MX added into the formulation

loading efficiency = Dt

Lt, (2)

where Dt is the total amount of MX in the formulation andLt is the total amount of PC added into the formulation.

2.3.4. Stability Evaluation of Liposomes and Transfersomes.Liposomes and transfersomes were stored at 4 ± 1◦C and22 ± 1◦C (room temperature, RT) for 30 days. Both thephysical and the chemical stability of MX were evaluated.The physical stability was assessed by visual observation forsedimentation and particle size determination. The chemicalstability was determined by measuring the MX content byHPLC on days 0, 1, 7, 14, and 30.

2.4. In Vitro Skin Permeation Study. Shed snake skin fromthe Siamese cobra (Naja kaouthia) was used as a modelmembrane for the skin permeation study because of itssimilarity to human skin in lipid content and permeability.The skin samples were mounted between the two half-cellsof a side-by-side diffusion chamber with a 37◦C water jacketto control the temperature. The dorsal surface of the skin wasplaced in contact with the donor chamber, which was filledwith the liposome formulation. The receptor chamber wasfilled with 0.1 M PBS (pH 7.4) and stirred with a star-headTeflon magnetic bar driven by a synchronous motor. At timeintervals of 0.5, 1, 2, 4, 8 and, 24 h, a 1 mL aliquot of receptorwas withdrawn, and the same volume of fresh medium wasadded back into the chamber. The concentration of MX inthe samples was analyzed by HPLC. The concentration ofpermeants in the samples was analyzed by HPLC, and thecumulative amount was plotted against time. The steady-state flux was determined as the slope of linear portion of theplot. Lag time was also obtained by extrapolating the linearportion of the penetration profile to the abscissa.

2.5. HPLC Analysis. The MX concentration was analyzed byHPLC [28] using an Eclipse XDB-C18 column. The mobilephase was a mixture of potassium dihydrogen phosphate pH4.4, methanol, and acetonitrile at a ratio of 45 : 45 : 10 (v/v/v).A 20 μL injection volume was used with a flow rate of1.0 mL/min, and UV detection was viewed at 364 nm. Thequantitative determination of MX in the tested samplewas obtained from the calibration curve, which gave goodlinearity at the range of 0.1–50 μg/mL.

2.6. Characterization of Snake Skin after Skin Permeation

2.6.1. FT-IR Analysis of Shed Snake Skin. Following theskin permeation study, the skin was washed with waterand blotted dry by keeping in the desiccator for 24 h. Thespectrum of the snake skin was recorded in the range of4000–500 cm−1 using an FT-IR spectrophotometer. The FT-IR spectrum of the untreated skin was also recorded and usedas a control.


0

20

40

60

80

100M

Xre

mai

nin

g(%

)

Day 1 Day 7 Day 14 Day 30

(a)

0

20

40

60

80

100

MX

rem

ain

ing

(%)

Day 1 Day 7 Day 14 Day 30

(b)

Figure 4: The percentage of meloxicam remaining in vesicles composed of different compositions: (solid diamond) PC, (white diamond)PC/Chol, (solid triangle) PC/NaO, (white triangle) PC/NaO/Chol, (solid circle) PC/NaChol, (white circle) PC/NaChol/Chol, (solid square)PC/DCP, and (white square) PC/DCP/Chol following storage at (a) 4◦C and (b) RT for 30 days. Each value represents the mean ± SD(n = 3).

0

1

2

3

4

5

Cu

mu

lati

vesk

inpe

rmea

tion

per

area

(μg/

cm2)

0 5 10 15 20 25

Time (h)

(a)

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35Fl

ux

(μg/

cm2/h

)

0

MX

susp

ensi

on

MX

/PC

MX

/PC

/Ch

ol

MX

/PC

/NaO

MX

/PC

/NaC

hol

MX

/PC

/NaO

/Ch

ol

MX

/PC

/NaC

hol

/Ch

ol

MX

/PC

/DC

P

MX

/PC

/DC

P/C

hol

∗∗

∗, ∗∗

∗

∗

(b)

Figure 5: (a) The skin permeation profile of meloxicam from (solid circle) MX suspensions (control) and (solid square) MX/PC/NaChol.(b) The fluxes of meloxicam through shed snake skin from different formulations: (solid square) control, (shaded square) liposomes, and(white square) transfersomes. Different values ∗ were statistically significant (P < .05) compared with MX suspensions (control). Differentvalues ∗∗ were statistically significant (P < .05) compared with liposomes. Each value represents the mean ± SD (n = 3–6).

2.6.2. Differential Scanning Calorimetry (DSC) Analysis ofShed Snake Skin. Thermal analysis of the skin after thepermeation study prepared with the same method as FTIRwas performed with a Sapphire DSC. The skin sample (2 mg)was weighed into an aluminum crimp pan. The samples wereheated from −30 to 320◦C at a heating rate of 10◦C/min.All DSC measurements were collected under a nitrogenatmosphere with a flow rate of 100 mL/min.

2.7. Data Analysis. Data are expressed as the means ±standard deviation (SD) of the mean, and statistical analysiswas carried out employing the one-way analysis of variance

(ANOVA) followed by an LSD post hoc test. A value of P < .05was considered statistically significant.

3. Results and Discussion

3.1. Physicochemical Characteristics of Liposomes and Trans-fersomes. The particle size range for all formulations, exceptthe MX suspensions, was less than 200 nm (89 to 137 nm)with a narrow size distribution. The particle size range ofthe MX suspensions was significantly larger than that ofthe liposomes (Table 2). The vesicles containing cholesterolhad a slightly lower particle size than without cholesterol.


(a)

(b)(c)

(d)

(e)(f)(g)

Abs

orba

nce

3000 2950 2900 2850 2800

Wavenumber (cm−1)

(a)

(a)

(b)

(c)

(d)

(e)

(f)

(g)

En

doth

erm

ic

80 120 160 200 240 280

Temperature (◦C)

(b)

Figure 6: (a) FT-IR spectra profile of shed snake skin after 24 h transfersomes skin permeation. (a) Untreated skin, (b) PC/NaO, (c)PC/NaO/Chol, (d) PC/NaChol, (e) PC/NaChol/Chol, (f) PC/DCP, and (g) PC/DCP/Chol and (b) DSC thermogram of shed snake skin after24 h MX suspensions (control) and transfersomes skin permeation. (a) MX suspensions, (b) PC/NaO, (c) PC/NaO/Chol, (d) PC/NaChol,(e) PC/NaChol/Chol, (f) PC/DCP, and (g) PC/DCP/Chol.

Table 2: Particle size and zeta potential in various formulations.

Name Particle size (nm) Zeta potential (mV)

MX suspensions 2411± 84.2 −19.3± 0.7

MX/PC 107.0± 5.0 −35.0± 0.5

MX/PC/Chol 100.3± 0.6 −23.5± 0.2

MX/PC/NaO 107.4± 0.5 −43.4± 0.1

MX/PC/NaO/Chol 100.5± 0.6 −23.1± 0.0

MX/PC/NaChol 93.0± 1.0 −32.7± 0.7

MX/PC/NaChol/Chol 88.6± 0.7 −28.9± 0.5

MX/PC/DCP 137.2± 6.1 −35.2± 0.6

MX/PC/DCP/Chol 126.5± 1.6 −29.3± 0.5

Each value represents the mean± SD (n = 3).

These results might be attributed to cholesterol causing thebilayer to be more compact [10, 26, 29–31]. The particlesize of the transfersomes with different types of surfactantdid not show a significant difference. These results indicatedthat the particle size of the vesicles was not affected by lipidcomposition (cholesterol) and surfactant.

The zeta potential of all vesicle formulations werenegative (−23 to −43 mV) due to the net charge of thelipid composition in the formulations. PC is a zwitterioniccompound with an isoelectric point (pI) between 6 and7 [32]. Under experimental conditions (pH 7.4), wherethe pH was higher than its pI, PC carried a net negativecharge. The surfactants used were anionic surfactants, andthe anion form of MX was also the predominant form atpH 7.4 [25]. Therefore, a negative charge in all formulationswas observed. Because the negatively charged liposomeformulations strongly improved skin permeation of drugs intransdermal delivery [12], these formulations were chosen tobe tested for MX permeation in our study.

The morphology of the two-dimensional vesicles wasfurther evaluated by TEM, justifying the vesicular charac-teristics. MX loaded in liposomes prepared from PC andPC/NaChol was spherical in shape (Figures 2(a), 2(b), and2(c)) and spherical with unilamellar vesicles (Figures 2(d),2(e), and 2(f)), respectively.

3.2. Entrapment Efficiency and Loading Efficiency. Theentrapment efficiencies and loading efficiencies of the MX-loaded formulations are presented in Figure 3(a). The 2.5%MX-LP formulation had the highest entrapment efficiencybut the lowest loading efficiency, while the 70% MX-LPformulation showed the highest loading efficiency but thelowest entrapment efficiency. Therefore, there should be anoptimum ratio between PC and MX for developing MX-loaded vesicles as carriers for transdermal drug delivery. Theoptimum ratio, which offered high entrapment efficiencyand high loading efficiency, was 10% MX-LP. This ratio wasused to prepare the vesicles.

The entrapment efficiency and loading efficiency oftransfersome formulations were significantly higher thanthe liposome formulations (Figure 3(b)). The entrapmentefficiency of MX in the vesicles ranged from 38% to 71%.The entrapment of MX in liposomes was lower than trans-fersomes except in formulations with DCP. This result mightbe attributed to interactions between the surfactants (NaOand NaChol) and MX when the complex was inserted intothe transfersomes bilayer. Fang et al. reported that addingsurfactant (sodium stearate) to phosphatidylethanolaminevesicles significantly increased the entrapment efficiency of5-aminolevulinic acid [26]. The results indicated that thetype of carrier systems and lipid composition affected theentrapment efficiency and loading efficiency of MX in thevesicle formulations.


The entrapment efficiency of the vesicles with andwithout cholesterol did not show a significant differ-ence. However, the entrapment efficiencies of the trans-fersome formulations changed depending on the typeof surfactant used and ranked PC/NaO(C18)>PC/NaChol(C24)>PC/DCP(C32). The lower the carbon chain length ofthe surfactants in the formulation, the higher the entrapmentefficiency. The increase in the carbon chain length of thesurfactant increased the lipophilicity and the solubility oflipophilic drug in the bilayer [10, 27]. This characteristicmay explain the increase in entrapment efficiency of MXin the bilayer of the vesicles. Surfactant may also competewith MX when arranging in the bilayer and therefore excludethe drug as it assembles into the bilayer of the vesicles. Thedata indicated that the entrapment efficiency and loadingefficiency are independent of cholesterol but dependent onthe surfactant in the formulations.

3.3. Stability Evaluation of Liposomes and Transfersomes.Liposomes and transfersomes were stored at 4◦C or RT for 30days. The physical (particle size determination) and chemical(percent MX remaining in the formulation) stability of thevesicles are presented in Table 3 and Figure 4, respectively.No sedimentation was found in any vesicle formulation afterfresh preparation. After storage at 4◦C for 30 days, there wasno sedimentation, but the average size of the vesicles in allformulations slightly increased. Nevertheless, the average sizeremained under 200 nm (Table 3). After storage at RT for7 days, no sedimentation was present in any formulation(data not shown). When evaluating the chemical stabilityof the vesicles, the percentage of MX remaining at 4◦C for30 days was in the range of 93% to 99% (Figure 4(a)),but it was 4% to 33% for the samples at RT (Figure 4(b)).The degradation rate of the MX-loaded vesicles stored at4◦C was not significantly different than those that werefreshly prepared. This reveals that the degradation of MXis independent of lipid composition but dependent on thestorage temperature and age.

3.4. In Vitro Skin Permeation Study. Figure 5(a) illustratesthe permeation profiles of MX suspensions (control) andMX-loaded transfersomes with NaChol. The cumulativeamount of drug increased linearly with time after a short lagtime (0.5–0.8 h). This linear accumulation was also observedfor other formulations (data not shown). Figure 5(b) showsthe flux (F) of MX through the snake skin calculatedfrom the permeation profiles. The F of MX permeatedthrough the skin in all vesicle formulations was significantlyhigher than the MX suspensions. The vesicle systems wereable to promote skin permeation of an active drug by avariety of mechanisms: (a) the free drug mechanism, (b) thepenetration-enhancing process of the liposome components,(c) vesicle adsorption to and/or fusion with the SC, and (d)intact vesicle penetration into and through the intact skinand the localization at the site of action [33–35]. Moreover,the similar predominance to the lipid bilayer of biologicalmembranes [36] and the nanometer size range of the vesiclesmay be also influenced [7, 26, 30]. These results indicated

that the vesicle system can overcome the barrier functionof the stratum corneum by various mechanisms and theirphysicochemical properties.

The F of MX permeated through the skin in transfer-somes was significantly higher than in liposomes. Transfer-somes have shown to be successful in the delivery of drugsinto the skin, including diclofenac, triamcinolone acetonide,hydrocortisone, and estradiol. Because transfersomes arecomposed of PC and surfactants, they can squeeze throughthe pores in the SC, which are smaller than one-tenth theirdiameter [3]. They can also adsorb onto or fuse with theSC, and the intact vesicle can penetrate into and through theintact skin.

The F of MX in the vesicles composed of cholesterol wasslightly lower than vesicles without cholesterol. An increasein cholesterol could lead to increased stability and rigidityand decrease the permeability of the lipid bilayer, whichmay cause lower release of MX and lower permeation ofMX through the skin [31]. The F of MX permeated fromtransfersomes with different compositions of surfactants areranked as follows: NaO (C18)∼NaChol (C24)>DCP (C32).The lower the carbon chain length of the surfactant inthe formulation, the higher the skin permeation of MX.The particle size and %EE of the vesicles composed ofNaO and NaChol were smaller and higher than vesiclescontaining DCP, respectively. These results indicated thatthe barrier function of stratum corneum can be overcomeby several factors, including physicochemical properties ofvesicle systems (size, charge, and %EE), lipid composition(cholesterol, surfactant), and type of vesicle system (lipo-somes, transfersomes).

The research results indicated that the skin permeabilityof MX-loaded transfersomes and liposomes were greaterthan that of MX suspensions and that both PC and surfactantwere key factors. Surfactants are enhancers that solubilizethe lipophilic compound; they also have the potential tosolubilize the lipid within the SC. Surfactants swell theSC, interact with the intercellular keratin, and fluidizethe SC lipid to create channels that allow increased drugdelivery.

3.5. Characterization of the Skin. The FT-IR spectrum of thesnake skin as a model for the SC provided a measure offluidity of the SC lipid. The comparison of the spectral profileof the untreated skin and treated skin with transfersomes,with and without cholesterol, resulted in shifts to higher fre-quencies. There was an absorbance broadening for both theC–H (CH2) asymmetric stretching peak near 2920 cm−1 andthe C–H (CH2) symmetric stretching peak near 2850 cm−1

(Figure 6(a)) [37]. The data indicated that flexibility of theSC lipid upon application of transfersomes occurred. Thus,it can be hypothesized that transfersomes permeated throughthe skin by disruption of the SC lipid structure.

The disruption of the SC lipid by the application oftransfersomes was further evaluated by DSC (Figure 6(b)).The SC lipid of the snake skin exist as a solid gel attemperature of 244◦C. In the DSC study, when the skinwas treated with transfersomes, which exists as liquid


Table 3: Particle size of formulations composed of different formulations following storage at 4◦C for 30 days.

NamePracticle size (nm)

Day 0 Day 1 Day 7 Day 14 Day 30

MX/PC 107.0± 5.0 113.4± 4.3 114.0± 1.1 114.5± 3.7 126.9± 16.0

MX/PC/Chol 100.3± 0.6 130.3± 15.5 159.0± 1.2 163.1± 2.5 182.6± 4.5

MX/PC/NaO 107.4± 0.5 93.8± 2.3 91.7± 0.9 93.8± 6.9 97.4± 2.0

MX/PC/Nao/Chol 100.5± 0.6 99.9± 1.1 96.1± 1.2 100.5± 5.5 110.6± 25.7

MX/PC/NaChol 93.0± 1.0 93.0± 1.0 93.6± 2.0 94.5± 1.6 92.1± 2.1

MX/PC/NaChol/Chol 88.6± 0.7 74.0± 2.5 87.4± 7.8 85.4± 4.3 85.1± 2.0

MX/PC/DCP 137.2± 6.1 144.5± 6.8 152.4± 1.2 162.3± 2.9 162.0± 4.9

MX/PC/DCP/Chol 126.5± 1.6 131.6± 3.9 139.5± 2.8 166.3± 12.9 184.9± 3.0

Each value represents the mean± SD (n = 3).

state vesicles, their thermal properties shifted (meltingpoint; Tm) as follows: PC/NaChol, 198◦C; PC/NaO, 207◦C;PC/DCP, 218◦C; PC/NaChol/Chol, 207◦C; PC/NaO/Chol,222◦; PC/DCP/Chol, 221◦C. The data indicated that the Tm

of skin treated with transfersomes was significantly lowerthan that of the untreated skin. The change into lower tran-sition temperature suggests an increase in the gross fluidityof the SC lipids. This is consistent with the general view thatthe mechanism of action of the surfactant is attributed to thealteration of the lipid organization and an increase in lipidlamellae disorder in the SC. Moreover, the Tm of the skintreated with transfersomes with cholesterol was significantlyhigher than those without cholesterol. If cholesterol could becomplexed with phospholipids in the skin, it could add morestructure to the bilayer. These results were in accordance withskin permeation data showing that transfersomes increasedthe skin permeation of MX, and the addition of cholesterol inthe transfersomes also led to a decrease in skin permeation ofMX when compared with transfersomes without cholesterol.Transfersomes may be used as alternative carriers for trans-dermal drug delivery potential because they interact withsolid gel phase SC lipids and thus leading to disruption andfluidization of the SC lipid.

4. Conclusion

In the present study, MX-loaded transfersomes were success-fully prepared by a sonication method. The use of surfactantscontaining medium length carbon chains, including NaO(C18) and NaChol (C24), in the transfersomes resulted in ahigh entrapment efficiency. Transfersomes provide greaterMX skin permeation than liposome and MX suspensions.The mechanism of this increase in MX permeation maybe through transfersomes’ disruption of the SC lipid. Thedata indicate that the barrier function of SC was affectedby several factors, including physicochemical propertiesof vesicle systems (size, charge, %EE), lipid composition(cholesterol, surfactant), and type of vesicle system (lipo-somes, transfersomes). Our research suggests that utilizingMX-loaded transfersomes as a transdermal therapeutic agentshows potential.

Acknowledgments

The authors wish to thank the Thailand Research Fundsthrough the Golden Jubilee Ph.D. Program (Grant no.PHD/0141/2550), the Thailand Research Funds (Grant no.RSA 5280001) for financial support.

References

[1] M. M. Badran, J. Kuntsche, and A. Fahr, “Skin penetrationenhancement by a microneedle device (Dermaroller�) invitro: dependency on needle size and applied formulation,”European Journal of Pharmaceutical Sciences, vol. 36, no. 4-5,pp. 511–523, 2009.

[2] H. Trommer and R. H. H. Neubert, “Overcoming the stratumcorneum: the modulation of skin penetration. A review,” SkinPharmacology and Physiology, vol. 19, no. 2, pp. 106–121, 2006.

[3] B. W. Barry, “Novel mechanisms and devices to enablesuccessful transdermal drug delivery,” European Journal ofPharmaceutical Sciences, vol. 14, no. 2, pp. 101–114, 2001.

[4] E. W. Smith and H. I. Maibach, Percutaceous PenetrationEnhancement, Taylor & Francis, Boca Raton, Fla, USA, 2ndedition, 2006.

[5] A. C. Williams and B. W. Barry, “Penetration enhancers,”Advanced Drug Delivery Reviews, vol. 56, no. 5, pp. 603–618,2004.

[6] Y.-C. Ah, J.-K. Choi, Y.-K. Choi, H.-M. Ki, and J.-H. Bae,“A novel transdermal patch incorporating meloxicam: invitro and in vivo characterization,” International Journal ofPharmaceutics, vol. 385, pp. 12–19, 2010.

[7] P. Karande and S. Mitragotri, “Enhancement of transdermaldrug delivery via synergistic action of chemicals,” Biochimicaet Biophysica Actas, vol. 1788, no. 11, pp. 2362–2373, 2009.

[8] J. Montanari, A. P. Perez, F. Di Salvo et al., “Photodynamicultradeformable liposomes: design and characterization,”International Journal of Pharmaceutics, vol. 330, no. 1-2, pp.183–194, 2007.

[9] G. Cevc, D. Gebauer, J. Stieber, A. Schatzlein, and G. Blume,“Ultraflexible vesicles, transfersomes, have an extremelylow pore penetration resistance and transport therapeuticamounts of insulin across the intact mammalian skin,”Biochimica et Biophysica Acta, vol. 1368, no. 2, pp. 201–215,1998.

[10] A. R. Mohammed, N. Weston, A. G. A. Coombes, M.Fitzgerald, and Y. Perrie, “Liposome formulation of poorly


water soluble drugs: optimisation of drug loading and ESEManalysis of stability,” International Journal of Pharmaceutics,vol. 285, no. 1-2, pp. 23–34, 2004.

[11] L. Montenegro, A. M. Panico, A. Ventimiglia, and F. P.Bonina, “In vitro retinoic acid release and skin permeationfrom different liposome formulations,” International Journalof Pharmaceutics, vol. 133, no. 1-2, pp. 89–96, 1996.

[12] C. Sinico, M. Manconi, M. Peppi, F. Lai, D. Valenti, andA. M. Fadda, “Liposomes as carriers for dermal delivery oftretinoin: in vitro evaluation of drug permeation and vesicle-skin interaction,” Journal of Controlled Release, vol. 103, no. 1,pp. 123–136, 2005.

[13] N. Perez-Cullell, L. Coderch, A. De La Maza, J. L. Parra, and J.Estelrich, “Influence of the fluidity of liposome compositionson percutaneous absorption,” Drug Delivery, vol. 7, no. 1, pp.7–13, 2000.

[14] D. D. Verma, S. Verma, G. Blume, and A. Fahr, “Liposomesincrease skin penetration of entrapped and non-entrappedhydrophilic substances into human skin: a skin penetrationand confocal laser scanning microscopy study,” EuropeanJournal of Pharmaceutics and Biopharmaceutics, vol. 55, no. 3,pp. 271–277, 2003.

[15] M. M. A. Elsayed, O. Y. Abdallah, V. F. Naggar, and N. M.Khalafallah, “Deformable liposomes and ethosomes: mech-anism of enhanced skin delivery,” International Journal ofPharmaceutics, vol. 322, no. 1-2, pp. 60–66, 2006.

[16] A. Viriyaroj, T. Ngawhirunpat, M. Sukma, P. Akkaramongkol-porn, U. Ruktanonchai, and P. Opanasopit, “Physicochemi-cal properties and antioxidant activity of gamma-oryzanol-loaded liposome formulations for topical use,” PharmaceuticalDevelopment and Technology, vol. 14, no. 6, pp. 665–671, 2009.

[17] R. Ambrus, P. Kocbek, J. Kristl, R. Sibanc, R. Rajko, andP. Szabo-Revesz, “Investigation of preparation parameters toimprove the dissolution of poorly water-soluble meloxicam,”International Journal of Pharmaceutics, vol. 381, no. 2, pp. 153–159, 2009.

[18] H.-K. Han and H.-K. Choi, “Improved absorption of meloxi-cam via salt formation with ethanolamines,” European Journalof Pharmaceutics and Biopharmaceutics, vol. 65, no. 1, pp. 99–103, 2007.

[19] J.-S. Chang, Y.-B. Huang, S.-S. Hou, R.-J. Wang, P.-C. Wu, andY.-H. Tsai, “Formulation optimization of meloxicam sodiumgel using response surface methodology,” International Journalof Pharmaceutics, vol. 338, no. 1-2, pp. 48–54, 2007.

[20] J.-W. Bae, M.-J. Kim, C.-G. Jang, and S.-Y. Lee, “Determina-tion of meloxicam in human plasma using a HPLC methodwith UV detection and its application to a pharmacokineticstudy,” Journal of Chromatography B: Analytical Technologiesin the Biomedical and Life Sciences, vol. 859, no. 1, pp. 69–73,2007.

[21] R. Jantharaprapap and G. Stagni, “Effects of penetrationenhancers on in vitro permeability of meloxicam gels,”International Journal of Pharmaceutics, vol. 343, no. 1-2, pp.26–33, 2007.

[22] R. Quintana, L. Kopcow, G. Marconi, E. Young, C. Yovanovich,and D. A. Paz, “Inhibition of cyclooxygenase-2 (COX-2) bymeloxicam decreases the incidence of ovarian hyperstimula-tion syndrome in a rat model,” Fertility and Sterility, vol. 90,no. 4, pp. 1511–1516, 2008.

[23] N. Seedher and S. Bhatia, “Mechanism of interaction ofthe non-steroidal antiinflammatory drugs meloxicam andnimesulide with serum albumin,” Journal of Pharmaceuticaland Biomedical Analysis, vol. 39, no. 1-2, pp. 257–262, 2005.

[24] Y. Yuan, S.-M. Li, F.-K. Mo, and D.-F. Zhong, “Investigationof microemulsion system for transdermal delivery of meloxi-cam,” International Journal of Pharmaceutics, vol. 321, no. 1-2,pp. 117–123, 2006.

[25] P. Luger, K. Daneck, W. Engel, G. Trummlitz, and K. Wagner,“Structure and physicochemical properties of meloxicam, anew NSAID,” European Journal of Pharmaceutical Sciences, vol.4, no. 3, pp. 175–187, 1996.

[26] Y.-P. Fang, Y.-H. Tsai, P.-C. Wu, and Y.-B. Huang, “Compar-ison of 5-aminolevulinic acid-encapsulated liposome versusethosome for skin delivery for photodynamic therapy,” Inter-national Journal of Pharmaceutics, vol. 356, no. 1-2, pp. 144–152, 2008.

[27] C. Bernsdorff, A. Wolf, R. Winter, and E. Gratton, “Effectof hydrostatic pressure on water penetration and rotationaldynamics in phospholipid-cholesterol bilayers,” BiophysicalJournal, vol. 72, no. 3, pp. 1264–1277, 1997.

[28] T. Ngawhirunpat, P. Opanasopit, T. Rojanarata, P. Akkara-mongkolporn, U. Ruktanonchai, and P. Supaphol, “Devel-opment of meloxicam-loaded electrospun polyvinyl alcoholmats as a transdermal therapeutic agent,” PharmaceuticalDevelopment and Technology, vol. 14, no. 1, pp. 70–79, 2009.

[29] S. Vemuri and C. T. Rhodes, “Preparation and characterizationof liposomes as therapeutic delivery systems: a review,”Pharmaceutica Acta Helvetiae, vol. 70, no. 2, pp. 95–111, 1995.

[30] D. D. Verma, S. Verma, G. Blume, and A. Fahr, “Particle size ofliposomes influences dermal delivery of substances into skin,”International Journal of Pharmaceutics, vol. 258, no. 1-2, pp.141–151, 2003.

[31] J.-Y. Fang, T.-L. Hwang, Y.-L. Huang, and C.-L. Fang,“Enhancement of the transdermal delivery of catechins byliposomes incorporating anionic surfactants and ethanol,”International Journal of Pharmaceutics, vol. 310, no. 1-2, pp.131–138, 2006.

[32] E. Chain and I. Kemp, “The isoelectric points of lecithin andsphingomyelin,” Biochemical Journal, vol. 28, no. 6, pp. 2052–2055, 1934.

[33] G. M. El Maghraby, B. W. Barry, and A. C. Williams, “Lipo-somes and skin: from drug delivery to model membranes,”European Journal of Pharmaceutical Sciences, vol. 34, no. 4-5,pp. 203–222, 2008.

[34] G. M. M. El Maghraby, A. C. Williams, and B. W. Barry,“Can drug-bearing liposomes penetrate intact skin?” Journalof Pharmacy and Pharmacology, vol. 58, no. 4, pp. 415–429,2006.

[35] G. M. M. El Maghraby, A. C. Williams, and B. W. Barry,“Skin delivery of oestradiol from deformable and traditionalliposomes: mechanistic studies,” Journal of Pharmacy andPharmacology, vol. 51, no. 10, pp. 1123–1134, 1999.

[36] J. Cladera, P. O’Shea, J. Hadgraft, and C. Valenta, “Influenceof molecular dipoles on human skin permeability: use of6-ketocholestanol to enhance the transdermal delivery ofbacitracin,” Journal of Pharmaceutical Sciences, vol. 92, no. 5,pp. 1018–1027, 2003.

[37] V. Dubey, D. Mishra, and N. K. Jain, “Melatonin loadedethanolic liposomes: physicochemical characterization andenhanced transdermal delivery,” European Journal of Pharma-ceutics and Biopharmaceutics, vol. 67, no. 2, pp. 398–405, 2007.


Review Article

Liposome Technology for Industrial Purposes

Andreas Wagner1 and Karola Vorauer-Uhl2

1 Polymun Scientific Immunbiologische Forschung GmbH, Nuβdorfer Lande 11, 1190 Vienna, Austria2 Department of Biotechnology, University of Natural Resources and Applied Life Sciences, Muthgasse 11, 1190 Vienna, Austria

Correspondence should be addressed to Andreas Wagner, [email protected]

Received 30 June 2010; Accepted 20 October 2010

Academic Editor: Adrian Williams

Copyright © 2011 A. Wagner and K. Vorauer-Uhl. This is an open access article distributed under the Creative CommonsAttribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work isproperly cited.

Liposomes, spherical vesicles consisting of one or more phospholipid bilayers, were first described in the mid 60s by Banghamand coworkers. Since then, liposomes have made their way to the market. Today, numerous lab scale but only a few large-scaletechniques are available. However, a lot of these methods have serious limitations in terms of entrapment of sensitive moleculesdue to their exposure to mechanical and/or chemical stress. This paper summarizes exclusively scalable techniques and focuseson strengths, respectively, limitations in respect to industrial applicability. An additional point of view was taken to regulatoryrequirements concerning liposomal drug formulations based on FDA and EMEA documents.

1. Introduction

1.1. History, Definition, and Classification of Liposomes. Thestory of success of liposomes was initiated by Bangham andhis colleagues in the early 1960s who observed that smearsof egg lecithin reacted with water to form quite intricatestructures. They were analyzed by electron microscopy show-ing that a multitude of vesicles were formed spontaneously.These more or less homogenous lipid vesicles were firstcalled smectic mesophases [1]. Later on, a colleague ofBangham termed them—more euphoniously—liposomes[2]. In the following years, liposomes were primarily used asartificial membrane models mimicking simple cell systemsfor the investigation of transport functions and mechanisms,permeation properties, as well as adhesion and fusionkinetics. Liposomes were very soon recognized as promisingcandidates for drug delivery systems [3, 4], and in this regardmore and more tailor-made formulations were investigatedfor certain purposes such as medical applications, cosmeticsbut also in food and agricultural industry, whereby themain activities were focused on pharmaceutical and inparticular biopharmaceutical applications. The first mostprominent products are Doxil (Sequus) and DaunoXome(Gilead, Nexstar). Both are indicated as anticancer drugs,which were successfully tested in clinical studies, followed by

the US Food and Drug Administration (FDA) approval in the1990s.

In general, liposomes are defined as spherical vesicleswith particle sizes ranging from 30 nm to several microme-ters. They consist of one or more lipid bilayers surroundingaqueous compartments, where the polar head groups areoriented towards the interior and exterior aqueous phases.However, self-aggregation of polar lipids is not restricted toconventional bilayer structures which depend on tempera-ture, molecular shape, and environmental and preparationconditions but may self-assemble into various kinds ofcolloidal particles [5, 6].

Due to this fact, the liposome family includes variouskinds of colloidal particles and structures which hampersystematic classification. However, they can be classifiedby structure, composition, and preparation, as shown inTable 1.

Technology and application are driven by two majorfacts. First, the transfer from academic bench to a highly reg-ulated, high technology industry was difficult for liposometechnology because of the lack of appropriate methods toproduce large quantities in a controlled and reproduciblemanner. Although several methods are suitable for large-scale production, their development, implementation, andquality control needed a certain time. Second, early clinical


Table 1: Classification of commonly known lipid vesicles according to their structures and/or preparation.

Identification Definition

ArcheosomesArcheosomes are vesicles consisting of archebacteria lipids which are chemically distinct from eukariotic andprokariotic species. They are less sensitive to oxidative stress, high temperature, and alkaline pH [7, 8].

Cochleates

Cochleates are derived from liposomes which are suspended in an aqueous two-phase polymer solution, allowingthe logic partitioning of polar molecule-based structures by phase separation. The liposome containingtwo-phase polymer solution treated with positively charged molecules such as Ca2+ or Zn2+ forms a cochleateprecipitate of a particle dimension less than 1 μm [9].

DendrosomesDendrosomes represent a family of novel, nontoxic, neutral, biodegradable, covalent or self-assembled,hyperbranched, dendritic, spheroidal nanoparticles which are easy to prepare, inexpensive, highly stable as wellas easy to handle and apply, compared with other existing synthetic vehicles for gene delivery [10].

Dried reconstitutedvesicles (DRV)

By this preparation technique, small, “empty” unilamellar vesicles, containing different lipids or mixtures ofthem, are prepared. After mixing those SUVs with the solubilised drug, dehydration is performed. By addition ofwater, rehydration leads to the formation of large quantities of rather inhomogeneous, multilamellar vesicleswhich need further processing [11].

Ethosomes

Ethosomal systems are much more efficient at delivering to the skin, in terms of quantity and depth, than eitherconventional liposomes or hydroalcoholic solutions. Ethosomal drug permeation through the skin wasdemonstrated in diffusion cell experiments. Ethosomal systems composed of soy phosphatidylcholine and about30% of ethanol were shown to contain multilamellar vesicles by electron microscopy [12].

ImmunoliposomesLiposomes modified with antibodies, Fab’s, or peptide structures on the bilayer surface were established for invitro and in vivo application [13, 14].

Immunosomes

Immunosomes are prepared by the anchorage of glycoprotein molecules to preformed liposomes. Under theelectron microscope, immunosomes look like homogenous spherical vesicles (50–60 nm) evenly covered withspikes. Immunosomes have structural and immunogen characteristics closer to those of purified and inactivatedviruses than any other form of glycoprotein lipids association [15].

Immune stimulatingcomplex (ISCOM)

ISCOMs are spherical, micellar assemblies of about 40 nm. They are made of the saponin mixture Quil A,cholesterol, and phospholipids. They contain amphiphilic antigens like membrane proteins. ISCOMs alreadyhave a built-in adjuvant, Quillaja saponin, which is a structural part of the vehicle [16].

LipoplexesCationic lipid-DNA complexes, named lipoplexes, are efficient carriers for cell transfection but have certaindrawbacks due to their toxicity. These toxic effects may result from either cationic lipids or nucleic acids [17, 18].

LUVETsLUVETs are large unilamellar vesicles prepared by extrusion technique, mainly performed with high-pressuresystems [19].

NiosomesNiosomes are small unilamellar vesicles made from nonionic surfactants also called Novasomes. Their chemicalstability is comparable to that of archeosomes [20].

pH-sensitive liposomes

Four basic classes of pH-sensitive liposomes have been described previously. The first class combinespolymorphic lipids, such as unsaturated phosphatidylethanolamines, with mild acidic amphiphiles that act asstabilizers at neutral pH. This class of pH-sensitive liposomes has been the most intensively studied. The secondclass includes liposomes composed of lipid derivatives resulting in increased permeability to encapsulatedsolutes. A third class of pH-sensitive liposomes utilizes pH-sensitive peptides or reconstituted fusion proteins todestabilize membranes at low pH. The final and most current class of pH-sensitive liposomes uses pH-titratablepolymers to destabilize membranes following change of the polymer conformation at low pH [21].

Polymerised liposomesPolymerized phosphatidyl choline vesicles (35–140 nm) have been synthesized from lipids bearing one or twomethacrylate groups per monomer. Compared to nonpolymeric analogues, these vesicles exhibited improvedstability and controllable time-release properties [22].

ProliposomesProliposomes are defined as dry, free-flowing particles that immediately form a liposomal dispersion on contactwith water [23, 24].

ProteosomesVesicles of bacterial origin were solubilised, followed by ammonium sulphate precipitation and dialysis againstdetergent buffer. Proteins and peptides are noncovalently complexed to the membrane, making them highlyimmunogenic [25].

Reverse-phaseevaporation vesicles(REV)

Vesicles are formed by evaporation of oil in water emulsions resulting in large unilamellar liposomes [26].

Stealth liposomes

In the early 1990s, this liposome engineering process culminated with the observation that coating of liposomeswith polyethylene glycol (PEG), a synthetic hydrophilic polymer, would improve their stability and lengthenstheir half-lives in circulation, rendering the use of glycolipids obsolete. PEG coating inhibits protein adsorptionand opsonization of liposomes, thereby avoiding or retarding liposome recognition by the reticuloendothelialsystem (RES). These PEG-coated liposomes are also referred to as sterically stabilized or stealth liposomes. ThePEG stabilizing effect results from local surface concentration of highly hydrated groups that sterically inhibitboth hydrophobic and electrostatic interactions of a variety of blood components at the liposome surface[27–33].


Table 1: Continued.

Identification Definition

Temperature-sensitiveliposomes

Temperature-sensitive liposomes are considered to be a promising tool to achieve site-specific delivery of drugs.Such liposomes have been prepared using lipids which undergo a gel-to-liquid crystalline phase transition a fewdegrees above physiological temperature. However, temperature sensitization of liposomes has been attemptedusing thermosensitive polymers. So far, functional liposomes have been developed according to this strategywhose content release behavior, surface properties, and affinity to cell surface can be controlled in atemperature-dependent manner [34, 35].

TransfersomesTransfersomes consist of phosphatidylcholine and cholate and are ultradeformable vesicles with enhancedskin-penetrating properties [36].

VirosomesVirosomes are small unilamellar vesicles containing influenza hemagglutinin, by which they became fusogenicwith endocytic membranes. Coincorporation of other membrane antigens induces enhanced immune responses[37].

trials were not as successful as expected because the stabilityof conventional liposomes was low, caused by inefficientpreparation, physical properties, and unfavorable choice oflipids. Furthermore, they were to a great extent cleared byliver and spleen very rapidly so that neither a prolongedbiological half-life nor specific targeting was achieved. Morestable conventional liposomes and second-generation for-mulations, such as the stealth technology, gave new impulsesto the industry as well as to clinicians with the developmentof industrial processes in the 1990s.

1.2. Liposome Technology and Regulatory Requirements. Inthe last decade, the European Agency of the Evaluation ofMedical Products (EMA) as well as the FDA has implementedthe subject of liposome into their guidelines.

Currently, EMA has not yet published any summarizingdocument or guideline which is dealing exclusively withnanoparticular structures. However, general aspects of lipo-somes are covered in several guidelines such as “Note ofGuidance on the Quality, Preclinical and Clinical Aspectsof gene transfer medicinal,” and “Guideline on adjuvant invaccines for human use”.

Regarding appropriate regulations, FDA published adraft version in 2001 entitled “Liposome Drug Products:chemistry, manufacturing, and controls, human pharma-cokinetics and bioavailability and labeling documentation.”This draft version includes recommendations explicitly forliposome drug products submitted in new drug applications(NDAs). In detail, recommendations concerning the submis-sion of a new liposomal product are given regarding phys-iochemical properties, description of manufacturing processand process controls, and control of excipients and drugproducts. Control of excipients includes all parameters whichare necessary to define lipid components, including descrip-tion, characterization, manufacture, and stability. Control ofdrug products is dealing with the specifications. In principal,the recommendations of the ICH (International Conferenceon Harmonization) guideline Q6A “Specifications, TestProcedures and Acceptance criteria for New Drug Sub-stances and New Drug Products: Chemical Substances” areappropriate, but additional testing is necessary. In particular,physicochemical parameters are critical for product qualityfor each batch. Furthermore, aspects are addressed such as

assaying encapsulated and nonencapsulated drug substance,lipid components, and degradation products, as well as invitro tests for drug release from liposomes. The second partof this document is dealing with human pharmacokineticsand bioavailability. In particular, requirements concerningthe quality and potency of bioanalytical methods are dis-cussed. Therefore, the recommendations are focused on thevalidation of these methods and the capability to distinguishbetween encapsulated and nonencapsulated drug substances.Similar recommendations are given for in vivo integrity andstability considerations, respectively. For safety assessment,validated in vitro assays are recommended to simulate theliposomal release and/or interaction with lipoproteins andother proteins in the blood. In an additional chapter, studiesfor pharmacokinetics and bioavailability are recommended,such as mass balance studies and pharmacokinetic studies.Finally, general recommendations concerning the labelingrequirements are given. This draft guidance does notprovide recommendations on clinical efficacy and safetystudies, nonclinical pharmacology and/or toxicology studies,bioequivalence studies or those to document the sameness,liposomal formulations of vaccine adjuvant or biologics, anddrug-lipid complexes. Unfortunately, during the intensivediscussion process no conclusion regarding the appropriateapproaches to access pharmacokinetics and bioavailabilitywas achieved. Hence, this document has only draft status tothis date.

In 2006, a reflection paper was published on nanotech-nology-based medicinal products for human use reflectingthe current thinking and the initiatives by EMA in view ofrecent developments in relation to this scope. As mentionedin this document, medicinal products containing nanoparti-cles, including liposomes, have already been authorized bothin EU and US under the existing regulatory frameworks.Nevertheless, the European Commission has developed anumber of initiatives with emphasis on safety and ethicalconsiderations but also to evaluate the appropriateness ofexisting methodologies to assess the potential risks associatedwith nanotechnology. In this context, it is mentioned thatthere is still insufficient knowledge and data concerningnanoparticles characterization, their detection and measure-ment, the persistence of nanoparticles in humans and theenvironment, and all aspects of toxicology related to theseparticles to allow satisfactory risk assessments. In order to


deal with this issue, the EMEA has created the InnovativeTask Force for the coordination of scientific and regulatorycompetence. Because novel applications of nanotechnologywill span the regulatory boundaries between medicinalproducts and medical devices, the mechanism of action willbe the key to decide whether a product should be regulatedas a medical product or a medical device. Furthermore, eval-uation of the quality, safety, efficacy, and risk managementmust be discussed in more detail. In conclusion, it is likelythat the evaluation of such new products will require specialconsiderations. Therefore, EMA will promote this processeither to develop specific guidelines or for the update ofexisting once.

2. Preparation Techniques

2.1. General Introduction into Techniques. Lipid moleculeshave to be introduced into an aqueous environment for thepreparation of liposomes independent of liposome size andstructure. A general overview representing the correlation ofthe way of lipid hydration, respectively, the way of primaryliposome formation with the resulting liposome structure,was originally developed by Lasic [38].

Several ways of treating the lipids are known to supportthe hydration of these molecules, as lipid molecules them-selves are poorly soluble in aqueous compartments. Theseprocedures can be categorized as shown in Table 2.

Additional methods have been developed such as freezethawing, freeze drying, and extrusion. However, they areall based on preformed vesicles. In the following sections,liposome preparation techniques are described with respectto the principle of lipid hydration/liposome formation aswell as process design and description. In addition, theadvantages and disadvantages of each technique are pointedout. Furthermore, focus is given on discussing the techniqueswith respect to their applicability regarding large-scaleproduction for clinical purposes and good manufacturingpractice (GMP) relevant issues.

3. Mechanical Methods

3.1. Preparation by Film Methods. Properties of lipid formu-lations can vary depending on the composition (cationic,anionic, and neutral lipid species). However, the samepreparation method can be used for all lipid vesiclesregardless of composition. The general steps of the procedureare preparation of the lipids for hydration, hydration withagitation, and sizing to a homogeneous distribution ofvesicles [40].

Since then, many different variations of this method havebeen developed differing in the organic solvents used for lipidsolubilization, the way of lipid drying, and the way of filmrehydration.

Despite the various modifications, all these methods havein common that heterogeneous populations of multilamellarliposomes are produced. However, vesicle size is influencedby the lipid charge. Charged lipids form smaller liposomeswith less lamellae. Other influencing parameters are the

nature of the aqueous phase as well as energy and powerinput of agitation.

The film method has several advantages. It can be usedfor all different kinds of lipid mixtures. In addition, themethod is easy to perform, and high encapsulation rates oflipid as well as aqueous soluble substances can be achievedbecause high lipid concentrations can be used.

One major drawback of this method is the difficulty ofscaling up to several tens of liters. Furthermore, the processbecomes more time and cost intensive because additionalprocessing is recommended for a defined liposome suspen-sion, whereby product losses are generated.

Several downsizing techniques have been established inorder to make the heterogeneous vesicles more uniform. Thefirst published downsizing method was sonication [41]. Avery high energy input based on cavitation is applied to theliposomal dispersion either directly with a tip or indirectly ina bath sonicator.

Other methods also aiming at breaking down the largeMLVs are homogenization techniques, either by shear orpressure forces. In this group, methods are included such asmicrofluidization, high-pressure homogenization, and shearforce-induced homogenization techniques.

The most defined method for downsizing is the extrusiontechnique whereby liposomes are forced through filters withwell defined pores.

3.2. Homogenization Techniques. Similar to the ultrasoundmethods, homogenization techniques have been used in biol-ogy and microbiology for breaking up the cells. Therefore,many scientists have used them for reducing the size andnumber of lamellae of multilamellar liposomes.

The French press [42] originally was established forbreaking up cells under milder and more appropriateconditions compared to the ultrasound techniques, becauselipids as well as proteins or other sensitive compoundsmight be degraded during the sonication procedure. Thissystem is normally used in the volume of 1 to 40 mL andtherefore is not suitable for large-scale production. However,a scale-up-based strategy on this technique was establishedas the microfluidization. This continuous and scaleablevariation of the French press technique enforces downsizingof Liposomes by collision of larger vesicles at high pressurein the interaction chamber of the microfluidizer.

Starting volumes from 50 mL upwards are applicable,and again high pressures are used for disruption of multi-lamellar systems. The system works in a pressure range of0–200 bar and is equipped with heating and cooling systemsto control sample temperature during processing [43]. Theliposome suspension passes the exchangeable orifices severaltimes (up to thousands of passes). Liposomes are formedin the size range from 50 to 100 nm by this process. Thistechnique is suitable for large-scale production and sterileliposome preparation.

In contrast to the microfluidizer, where the fluid streamis split and mixed by collision in a mixing chamber, homog-enizers work on a different principle. In a homogenizer, thefluid beam is pressed with high pressure through an orifice,


Table 2: Methods of liposome preparation and the resulting product. Partly from Lasic and Barenholz [39].

Method Vesicles

Mechanical methods

Vortex or hand shaking of phospholipid dispersions MLV

Extrusion through polycarbonate filters at low or medium pressure OLV, LUV

Extrusion through a French press cell “Microfluidizer” technique Mainly SUV

High-pressure homogenization Mainly SUV

Ultrasonic irritation SUV of minimal size

Bubbling of gas BSV

Methods based on replacement of organic solvent(s) by aqueous media

Removal of organic solvent(s) MLV, OLV, SUV

Use of water-immiscible solvents: ether and petroleum MLV, OLV, LUV

Ethanol injection method LUV

Ether infusion (solvent vaporization) LUV, OLV, MLV

Reverse-phase evaporation

Methods based on detergent removal

Gel exclusion chromatography SUV

“Slow” dialysis LUV, OLV, MLV

Fast dilution LUV, OLV

Other related techniques MLV, OLV, LUV, SUV

and this beam collides with a stainless steel wall. Theliposome suspension is continuously pumped through thehomogenizer system, where high pressures are generated todownsize lipid vesicles [44].

The most prominent scalable downsizing method isthe extrusion. Size reduction is managed under mild andmore reproducible conditions compared to those discussedabove. In this method, preformed vesicles are forced throughdefined membranes by a much lower pressure as described inthe French press method. Extrusion through polycarbonatefilters was first published by Olson et al. in 1979 [45].Mayer et al. [19] performed extensive studies on varyinglipid compositions and the influence on extrusion behaviorand membrane properties. Depending on the apparatus andscale, the diameters of these membranes range from 25 to142 mm. Lipex Biomembranes Inc., now Northern LipidsInc., invented a vessel system for extrusion which is marketedfrom the mL scale to several liters. As suggested for alldownsizing methods, liposomes should be extruded abovethe Tc of the lipid composition; this system can be tempered.The Lipex extruder system is available in a jacketed mode toallow extrusion at higher temperatures.

An alternative is the Maximator device, established bySchneider et al. [46]. It is a continuous extrusion deviceworking with a pumping system. The Maximator consistsof a thermostable glass supply vessel directly connectedto a pneumatic high pressure piston pump. The latter isdriven by either oxygen or nitrogen at pressures below0.5 MPa (5 bar or 75 psi). The propellant gas does not comeinto contact with the liposome suspension. The resultingoperating (extrusion) pressure—which can be adjusted viathe reduction valve in the control device for the propellantgas (3)—can be as high as 12 MPa (120 bar or 1800 psi) withthe current equipment.

All the presented extrusion methods have in commonthat the reproducibility of downsizing is extremely high.Systems with a heating device can either be used withsaturated and unsaturated lipids and are; therefore, allpurpose systems.

The main disadvantage of this method is the long-lastingpreparation starting with preformed liposomes, eventuallyan additional freeze-thaw step, and finally the extrusion.In these entire procedures, high product losses may begenerated, especially if clogging of the extrusion membranesoccurs, which may cause technical limitations with large-scale production of high-priced goods.

4. Methods Based on Replacement ofOrganic Solvents by Aqueous Media

The liposome preparation methods described in this sectionhave in common that organic solvents, either water miscibleor immiscible, are replaced by an aqueous solution. Thisreplacement is either performed by injection of the lipidcarrying organic solution into the aqueous phase—the injec-tion methods—or by stepwise addition of aqueous phase tothe organic phase, in particular ethanol—the proliposome-liposome method. In addition, the emulsification methods,namely, the reverse-phase evaporation method and the dou-ble emulsion technique, are based on the replacement of awater-immiscible solvent by an aqueous phase, thus formingliposomes with high encapsulation rates of hydrophilic aswell as lipid phase soluble substances.

4.1. The Ethanol Injection Method. This technique was firstreported in the early 1970s by Batzri and Korn [47] as oneof the first alternatives for the preparation of SUVs withoutsonication. By the immediate dilution of the ethanol in


the aqueous phase, the lipid molecules precipitate and formbilayer planar fragments [48] which themselves form intoliposomal systems, thereby encapsulating aqueous phase.Batzri and Korn performed their experiments with a very lowlipid concentration resulting in small liposomes and poorencapsulation efficiency.

The preparation parameters influencing liposome size,size distribution, and drug encapsulation efficiency wereinvestigated in more detail by Kremer et al. in 1977 [49].They determined the lipid concentration in ethanol as theonly liposome formation influencing parameter. Neitherstirring rate of the aqueous phase nor injection velocity hada significant influence on liposome size and size distribution.

Another modified ethanol injection method was devel-oped by Maitani et al. [50] which is more or less a com-bination of the ethanol injection method, the proliposomemethod, and the reverse-phase evaporation technique.

This method has many advantages as the technique isin principle easy to scale up, and ethanol is a very harmlesssolvent, accepted by the authorities also for injectables at amaximum of 0.1% [51]. Some other solvents might also beused, but one has to keep in mind the regulations for residualsolvents classified into different categories by the Europeanor US Pharmacopoeia. Stano et al. [52] emphasize theadvantage of preparing monomodal distributed liposomes inthe size range of about 100 nm and furthermore point outthe suitability of the entrapment of lipophilic substances. Anadditional advance is the improvement of product shelf lifedue to the absence of mechanical forces which might lead todrug and/or lipid degradation.

Therefore, further development was initiated by severalgroups. In 1995, Naeff [53] published the development ofa liposome production technique in industrial scale basedon the ethanol injection technique. Their production systemwas used for the liposomal encapsulation of econazole, animidazole derivative for the topical treatment of dermatomy-cosis, and combined the principles of the ethanol injectionsystem and high shear homogenization.

Additional production technology patents from severalcompanies were filed dealing with liposome productionsystems based on the ethanol injection technique (Optime,Liposome Comp. Martin, Tenzel) [54–57].

Wagner et al. have also extensively worked in this field,leading to the development of the cross-flow injectionsystem. Based on the ethanol injection technique, theydeveloped a scalable and sterile production technique leadingfrom the conventional batch process to a continuous proce-dure [58].

Herein, the principal item is the cross-flow injectionmodule [59], especially designed for this purpose. Thisspecially conceived unit has the benefit of defined andcharacterized injection streams and permits liposome man-ufacture regardless of production scale because scale isdetermined only by free disposable vessel volumes. By this,process development is performed in lab scale at a volumeof about 20 mL. Once the parameters are defined, an easyscale-up can be performed by changing the process vesselsonly. In addition, these process vessels can be sterilized, eitherby steam or autoclavation. All raw materials such as buffer

solutions, lipid ethanol solution, and even N2 for applyingthe injection pressure are transferred into the sanitized andsterilized system via 0.2 μm filters to guarantee an asepticproduction [60].

Liposome size can be controlled by the local lipidconcentration at the injection point which is defined by thelipid concentration in ethanol, the injection whole diameter,the injection pressure, and the flow rate of the aqueousphase. By varying these parameters, different liposome sizessuitable for the intended purpose can be prepared. Thesedefined process parameters are furthermore responsible forhighly reproducible results with respect to vesicle diametersand encapsulation rates [61]. Tangential flow filtration is thenext process step to remove ethanol as well as not entrappeddrug.

Another important advantage of this method is thesuitability of the entrapment of many different drug sub-stances [61] such as large hydrophilic proteins by passiveencapsulation, small amphiphilic drugs by a one-step remoteloading technique, or membrane association of antigens forvaccines [62].

4.2. Proliposome-Liposome Method. The proliposome-liposome method is based on the conversion of the initialproliposome preparation into a liposome dispersion bydilution with an aqueous phase [50]. This method issuitable for the encapsulation of a wide range of drugswith varying solubility in water and alcohol and hasextremely high encapsulation efficiencies when comparedwith other methods based on passive entrapment. Turanekand coworkers [63, 64] have developed a sterile liposomeproduction procedure based on this method. Reproducibleliposome preparation is feasible in a controlled mannerby exact controlling of the dilution rate and processtemperature. Additionally, the authors claim their methodas being easy to scale up, which makes this method analternative approach for the production of liposomes forclinical application.

4.3. Reverse-Phase Evaporation (REV). Similarly to the abovepresented injection methods, lipid is hydrated via solubi-lization in an organic phase followed by introduction intoan aqueous phase. The organic phase should be immisciblewith the aqueous phase, thus an oil/water emulsion iscreated, which is diluted with further aqueous phase forliposome formation [65]. The advantage of this very popularpreparation technique is a very high encapsulation rateup to 50%. One variation of the microemulsion tech-nique, the double emulsion technique, further improves theencapsulation rates and results in unilamellar liposomes[26]. A possible drawback of this efficient method is theremaining solvent or the proof of their absence especiallyfor using them for pharmaceutical purposes. The otherimportant issue is large-scale production which might befeasible if appropriate shear mixing devices for the creationof the microemulsion and pumps for the dilution step areavailable.


5. Methods Based on Detergent Removal

In this group of liposome preparation procedures, deter-gents, such as bile salts or alkylglycosides, are used for thesolubilization of lipids in micellar systems. In contrast tolipids, detergents are highly soluble in both aqueous andorganic media. There is equilibrium between the detergentmolecules in the aqueous phase and the lipid environmentof the micelle. The size and shape of the resulting vesiclesare depending on the chemical nature of the detergent,their concentration, and the lipids used. To date, the mostfrequently applied method for membrane protein reconstitu-tion involves the cosolubilization of membrane proteins andphospholipids [66–68]. Common procedures of detergentremoval from the mixed micelles are dilution [69], gel chro-matography [70], and dialysis through hollow fibers [71]or through membrane filters [72]. Additionally, detergentscan also be removed by adsorption to hydrophobic resinsor cyclodextrins [73]. Dialysis of mixed micelles against anaqueous medium was first described by Kagawa and Racker[74]. This method for vesicle formation is based on theretention of the micelle, whereas free detergent molecules areeliminated. Goldin [72] describe the use of pure cellulosefor this approach. In order to gain better control in theformation of proteoliposomes, Wagner et al. developed anew technique, where they combine an advanced ethanolinjection technique, the cross-flow injection technique [58],with detergent dilution within one operational step. Thereby,lipid vesicles are formed immediately after injection into amicellar protein solution. As described earlier, the multipleinjection technique [59], previously used for high yieldpassive encapsulation of water-soluble proteins, can beadapted for this one-step detergent dilution/vesicle formingprocess [62].

6. Final Remarks

Numerous studies for the pharmaceutical application ofliposomes have appeared during the past few decades.They have attracted great interest as models for biologi-cal membranes, diagnostics, nutrients, and other bioactiveagents. Nevertheless, the pharmaceutical application, as drugcarriers for specific targeting, controlled, and/or sustainedrelease, as well as for vaccination, was and still is the drivingforce for the development of innovative technologies. Fromthis expertise, one can derive that liposomes are versatilecarrier systems which need to be custom made in terms ofin vitro and in vivo properties. In the last decades, numerouspreparation techniques were established for this purpose,whereby most of them are in particular suitable for thelaboratory and less for industrial approach. However, largescale capacities are required for the preparation of clinicalmaterial as well as for marketed products providing sterile,well-characterized, and stable products. Unfortunately, theavailability of certain production methods as well as thequality aspects depend on the characteristics of the lipidsthemselves. This limits the choice of liposome types fromwhich one can select when optimizing liposome-based drugtherapy.

Though many preparation methods were investigatedin the 1980s and 1990s, little attention has been paid tothe transfer of technology to industry. Thus, presently theadvancement is primarily focused on large-scale manufac-turing. Stringent control of the product is required to ensurethe predictable therapeutic effect, whereas acceptance criteriahave to be defined for the quality as well as the process.Additionally, quality issues regarding unwanted by-products,such as residues of organic solvents and/or degradationproducts, are just as important as pyrogen-free and sterileconditions. In particular, the latter aspect still is a big issuefor industrial processes. Until now, no general acceptablemethod could be successfully established. Commonly usedprocesses to achieve sterility for pharmaceutical products aresterile filtration or autoclaving. Both methods are of eitherno or only limited suitability for liposomal drug products.In many cases, degradation and/or unacceptable productloss in combination with drug release and instability arethe consequences. Currently, many manufacturers try toimplement alternative strategies, such as lyophilization andproduction processes in closed containments equipped withsterile filter barriers, to solve this essential problem.

In conclusion, besides the development of new liposomaldrug formulations, researchers as well as manufacturers arerequired to establish processes which are state of the art in thepharmaceutical industry. The realization of maybe future,more complex liposome structures with advanced efficacywill to a great extent dependent on those achievements.

References

[1] A. D. Bangham, “Lipid bilayers and biomembranes,” AnnualReview of Biochemistry, vol. 41, pp. 753–776, 1972.

[2] G. Sessa and G. Weissmann, “Phospholipid spherules (lipo-somes) as a model for biological membranes,” Journal of LipidResearch, vol. 9, no. 3, pp. 310–318, 1968.

[3] G. Gregoriadis, P. D. Leathwood, and B. E. Ryman, “Enzymeentrapment in liposomes,” FEBS Letters, vol. 14, no. 2, pp. 95–99, 1971.

[4] Y. E. Rahman, M. W. Rosenthal, E. A. Cerny, and E. S. Moretti,“Preparation and prolonged tissue retention of liposomeencapsulated chelating agents,” Journal of Laboratory andClinical Medicine, vol. 83, no. 4, pp. 640–647, 1974.

[5] D. D. Lasic, “Kinetic and thermodynamic effects on thestructure and formation of phosphatidylcholine vesicles,”Hepatology, vol. 13, no. 5, pp. 1010–1012, 1991.

[6] A. S. Ulrich, “Biophysical aspects of using liposomes asdelivery vehicles,” Bioscience Reports, vol. 22, no. 2, pp. 129–150, 2002.

[7] L. Krishnan, C. J. Dicaire, G. B. Patel, and G. D. Sprott,“Archaeosome vaccine adjuvants induce strong humoral, cell-mediated, and memory responses: comparison to conven-tional liposomes and alum,” Infection and Immunity, vol. 68,no. 1, pp. 54–63, 2000.

[8] J. W. Conlan, L. Krishnan, G. E. Willick, G. B. Patel,and G. D. Sprott, “Immunization of mice with lipopeptideantigens encapsulated in novel liposomes prepared from thepolar lipids of various Archaeobacteria elicits rapid and pro-longed specific protective immunity against infection with thefacultative intracellular pathogen, Listeria monocytogenes,”Vaccine, vol. 19, no. 25-26, pp. 3509–3517, 2001.


[9] S. Gould-Fogerite, M. T. Kheiri, F. Zhang et al., “Targetingimmune response induction with cochleate and liposome-based vaccines,” Advanced Drug Delivery Reviews, vol. 32, no.3, pp. 273–287, 1998.

[10] M. N. Sarbolouki, M. Sadeghizadeh, M. M. Yaghoobi, A.Karami, and T. Lohrasbi, “Dendrosomes: a novel family ofvehicles for transfection and therapy,” Journal of ChemicalTechnology and Biotechnology, vol. 75, no. 10, pp. 919–922,2000.

[11] G. Gregoriadis, D. Davis, and A. Davies, “Liposomes asimmunological adjuvants: antigen incorporation studies,”Vaccine, vol. 5, no. 2, pp. 145–151, 1987.

[12] E. Touitou, N. Dayan, L. Bergelson, B. Godin, and M. Eliaz,“Ethosomes—novel vesicular carriers for enhanced delivery:characterization and skin penetration properties,” Journal ofControlled Release, vol. 65, no. 3, pp. 403–418, 2000.

[13] A. Huang, S. J. Kennel, and L. Huang, “Interactions ofimmunoliposomes with target cells,” Journal of BiologicalChemistry, vol. 258, no. 22, pp. 14034–14040, 1983.

[14] S. M. Sullivan, J. Connor, and L. Huang, “Immunoliposomes:preparation, properties, and applications,” Medicinal ResearchReviews, vol. 6, no. 2, pp. 171–195, 1986.

[15] P. Perrin, P. Sureau, and L. Thibodeau, “Structural andimmunogenic characteristics of rabies immunosomes,” Devel-opments in Biological Standardization, vol. 60, pp. 483–491,1985.

[16] G. F. A. Kersten and D. J. A. Crommelin, “Liposomes andISCOMs,” Vaccine, vol. 21, no. 9-10, pp. 915–920, 2003.

[17] I. A. Khalil, K. Kogure, H. Akita, and H. Harashima, “Uptakepathways and subsequent intracellular trafficking in nonviralgene delivery,” Pharmacological Reviews, vol. 58, no. 1, pp. 32–45, 2006.

[18] S. Audouy and D. Hoekstra, “Cationic lipid-mediated trans-fection in vitro and in vivo (review),” Molecular MembraneBiology, vol. 18, no. 2, pp. 129–143, 2001.

[19] L. D. Mayer, M. J. Hope, and P. R. Cullis, “Vesicles of variablesizes produced by a rapid extrusion procedure,” Biochimica etBiophysica Acta, vol. 858, no. 1, pp. 161–168, 1986.

[20] J. M. Brewer and J. Alexander, “Studies on the adjuvant activityof non-ionic surfactant vesicles: adjuvant-driven IgG2a pro-duction independent of MHC control,” Vaccine, vol. 12, no. 7,pp. 613–619, 1994.

[21] D. C. Drummond, M. Zignani, and J.-C. Leroux, “Currentstatus of pH-sensitive liposomes in drug delivery,” Progress inLipid Research, vol. 39, no. 5, pp. 409–460, 2000.

[22] S. L. Regen, A. Singh, G. Oehme, and M. Singh, “Polymerizedphosphatidyl choline vesicles. Stabilized and controllabletime-release carriers,” Biochemical and Biophysical ResearchCommunications, vol. 101, no. 1, pp. 131–136, 1981.

[23] B. H. Jung, S. J. Chung, and C. K. Shim, “Proliposomesas prolonged intranasal drug delivery systems,” STP PharmaSciences, vol. 12, no. 1, pp. 33–38, 2002.

[24] N. I. Payne, P. Timmins, and C. V. Ambrose, “Proliposomes:a novel solution to an old problem,” Journal of PharmaceuticalSciences, vol. 75, no. 4, pp. 325–329, 1986.

[25] G. H. Lowell, L. F. Smith, R. C. Seid, and W. D. Zollinger,“Peptides bound to proteosomes via hydrophobic feet becomehighly immunogenic without adjuvants,” Journal of Experi-mental Medicine, vol. 167, no. 2, pp. 658–663, 1988.

[26] F. Szoka Jr. and D. Papahadjopoulos, “Procedure for prepa-ration of liposomes with large internal aqueous space andhigh capture by reverse-phase evaporation,” Proceedings of theNational Academy of Sciences of the United States of America,vol. 75, no. 9, pp. 4194–4198, 1978.

[27] T. M. Allen, C. Hansen, F. Martin, C. Redemann, and A. F.Yau-Young, “Liposomes containing synthetic lipid derivativesof poly(ethylene glycol) show prolonged circulation half-livesin vivo,” Biochimica et Biophysica Acta, vol. 1066, no. 1, pp. 29–36, 1991.

[28] A. L. Klibanov, K. Maruyama, V. P. Torchilin, and L. Huang,“Amphipathic polyethyleneglycols effectively prolong the cir-culation time of liposomes,” FEBS Letters, vol. 268, no. 1, pp.235–237, 1990.

[29] J. Senior, C. Delgado, D. Fisher, C. Tilcock, and G. Gregoriadis,“Influence of surface hydrophilicity of liposomes on theirinteraction with plasma protein and clearance from thecirculation: studies with poly(ethylene glycol)-coated vesicles,”Biochimica et Biophysica Acta, vol. 1062, no. 1, pp. 77–82, 1991.

[30] G. Blume and G. Cevc, “Liposomes for the sustained drugrelease in vivo,” Biochimica et Biophysica Acta, vol. 1029, no.1, pp. 91–97, 1990.

[31] D. D. Lasic, F. J. Martin, A. Gabizon, S. K. Huang, and D.Papahadjopoulos, “Sterically stabilized liposomes: a hypothe-sis on the molecular origin of the extended circulation times,”Biochimica et Biophysica Acta, vol. 1070, no. 1, pp. 187–192,1991.

[32] M. C. Woodle and D. D. Lasic, “Sterically stabilized lipo-somes,” Biochimica et Biophysica Acta, vol. 1113, no. 2, pp.171–199, 1992.

[33] S. M. Moghimi and J. Szebeni, “Stealth liposomes and longcirculating nanoparticles: critical issues in pharmacokinetics,opsonization and protein-binding properties,” Progress inLipid Research, vol. 42, no. 6, pp. 463–478, 2003.

[34] K. Kono, “Thermosensitive polymer-modified liposomes,”Advanced Drug Delivery Reviews, vol. 53, no. 3, pp. 307–319,2001.

[35] D. Needham and M. W. Dewhirst, “The development andtesting of a new temperature-sensitive drug delivery systemfor the treatment of solid tumors,” Advanced Drug DeliveryReviews, vol. 53, no. 3, pp. 285–305, 2001.

[36] A. Paul and G. Cevc, “Noninvasive administration of proteinantigens: transdermal immunization with bovine serum albu-min in transfersomes,” Vaccine Research, vol. 4, no. 3, pp. 145–164, 1995.

[37] R. Gluck, “Adjuvant activity of immunopotentiating reconsti-tuted influenza virosomes (IRIVs),” Vaccine, vol. 17, no. 13-14,pp. 1782–1787, 1999.

[38] D. D. Lasic, Ed., Liposomes—From Physic to Application,Elsevier Science Publishe, 1993.

[39] D. Lasic and Y. Barenholz, Handbook of Nonmedical Applica-tions of Liposomes, vol. 3, chapter 3, CRC-Press, New York, NY,USA, 1996.

[40] A. D. Bangham, M. M. Standish, and J. C. Watkins, “Diffusionof univalent ions across the lamellae of swollen phospho-lipids,” Journal of Molecular Biology, vol. 13, no. 1, pp. 238–252,1965.

[41] C.-H. Huang, “Studies on phosphatidylcholine vesicles. For-mation and physical characteristics,” Biochemistry, vol. 8, no.1, pp. 344–352, 1969.

[42] Y. Barenholzt, S. Amselem, and L. D., “A new methodfor preparation of phospholipid vesicles (liposomes)—frenchpress,” FEBS Letters, vol. 99, no. 1, pp. 210–214, 1979.

[43] E. Mayhew, R. Lazo, W. J. Vail, J. King, and A. M. Green, “Char-acterization of liposomes prepared using a microemulsifier,”Biochimica et Biophysica Acta, vol. 775, no. 2, pp. 169–174,1984.

[44] M. Brandl, D. Bachmann, M. Drechsler, and K. H. Bauer,“Lipopsome preparation by a new high pressure homogenizer


Gaulin Micron LAB 40,” Drug Development and IndustrialPharmacy, vol. 16, no. 14, pp. 2167–2191, 1990.

[45] F. Olson, C. A. Hunt, and F. C. Szoka, “Preparation ofliposomes of defined size distribution by extrusion throughpolycarbonate membranes,” Biochimica et Biophysica Acta, vol.557, no. 1, pp. 9–23, 1979.

[46] T. Schneider, A. Sachse, G. Rossling, and M. Brandl, “Large-scale production of liposomes of defined size by a new contin-uous high pressure extrusion device,” Drug Development andIndustrial Pharmacy, vol. 20, no. 18, pp. 2787–2807, 1994.

[47] S. Batzri and E. D. Korn, “Single bilayer liposomes preparedwithout sonication,” Biochimica et Biophysica Acta, vol. 298,no. 4, pp. 1015–1019, 1973.

[48] D. D. Lasic, “Mechanisms of liposome formation,” Journal ofLiposome Research, vol. 5, no. 3, pp. 431–441, 1995.

[49] J. M. H. Kremer, M. W. J. V. D. Esker, C. Pathmamanoharan,and P. H. Wiersema, “Vesicles of variable diameter preparedby a modified injection method,” Biochemistry, vol. 16, no. 17,pp. 3932–3935, 1977.

[50] Y. Maitani, H. Soeda, W. Junping, and K. Takayama, “Mod-ified ethanol injection method for liposomes containing β-sitosterol β-d-glucoside,” Journal of Liposome Research, vol. 11,no. 1, pp. 115–125, 2001.

[51] (Eds), European Pharmakopeia, european treaty seriesnr.:no50, 1997.

[52] P. Stano, S. Bufali, C. Pisano et al., “Novel camptothecinanalogue (Gimatecan)-containing liposomes prepared by theethanol injection method,” Journal of Liposome Research, vol.14, no. 1-2, pp. 87–109, 2004.

[53] R. Naeff, “Feasibility of topical liposome drugs produced onan industrial scale,” Advanced Drug Delivery Reviews, vol. 18,no. 3, pp. 343–347, 1996.

[54] R. A. Tenzel and D. F. Aitcheson, “Preparation of Uniform-SizeLiposomes,” US patent office, US 5, 000, 887, 1991.

[55] F. Martin, “High-encapsulation liposome processing method,”US patent office, US 4, 752, 425, 1988.

[56] M. T. Baker and W. Heriot, “Method and Apparatus forLiposome Production,” PCT, WO 00/29103, 2000.

[57] C. Yiournas and D. F. H. Wallach, “Method and Apparatus forProducing Lipid Vesicles,” US patent office, US 4, 895, 452990.

[58] A. Wagner, K. Vorauer-Uhl, G. Kreismayr, and H. Katinger,“The crossflow injection technique: an improvement of theethanol injection method,” Journal of Liposome Research, vol.12, no. 3, pp. 259–270, 2002.

[59] A. Wagner, K. Vorauer-Uhl, G. Kreismayr, and H. Katinger,“Enhanced protein loading into liposomes by the multiplecrossflow injection technique,” Journal of Liposome Research,vol. 12, no. 3, pp. 271–283, 2002.

[60] A. Wagner, K. Vorauer-Uhl, and H. Katinger, “Liposomesproduced in a pilot scale: production, purification andefficiency aspects,” European Journal of Pharmaceutics andBiopharmaceutics, vol. 54, no. 2, pp. 213–219, 2002.

[61] A. Wagner, M. Platzgummer, G. Kreismayr et al., “GMPproduction of liposomes—a new industrial approach,” Journalof Liposome Research, vol. 16, no. 3, pp. 311–319, 2006.

[62] A. Wagner, G. Stiegler, K. Vorauer-Uhl et al., “One step mem-brane incorporation of viral antigens as a vaccine candidateagainst HIV,” Journal of Liposome Research, vol. 17, no. 3-4,pp. 139–154, 2007.

[63] D. Deamer and A. D. Bangham, “Large volume liposomes byan ether vaporization method,” Biochimica et Biophysica Acta,vol. 443, no. 3, pp. 629–634, 1976.

[64] J. Turanek, D. Zaluska, and J. Neca, “Linkup of a fast proteinliquid chromatography system with a stirred thermostated

cell for sterile preparation of liposomes by the proliposome-liposome method: application to encapsulation of antibiotics,synthetic peptide immunomodulators, and a photosensitizer,”Analytical Biochemistry, vol. 249, no. 2, pp. 131–139, 1997.

[65] J. Turanek, A. Kasna, D. Zaluska, and J. Neca, “Preparation ofsterile liposomes by proliposome-liposome method,” Methodsin Enzymology, vol. 367, pp. 111–125, 2003.

[66] S. Frokjaer, “Double emulsion vesicles,” in Liposomes. APractical Approach, R. R. C. New, Ed., IRLPress, Oxford, UK,1989.

[67] R. R. H. Anholt, “Solubilization and reassembly of themitochondrial benzodiazepine receptor,” Biochemistry, vol. 25,no. 8, pp. 2120–2125, 1986.

[68] M. L. Jackson and B. J. Litman, “Rhodopsin-phospholipidreconstitution by dialysis removal of octyl glucoside,” Bio-chemistry, vol. 21, no. 22, pp. 5601–5608, 1982.

[69] A. J. M. Driessen and W. Wickner, “Solubilization and func-tional reconstitution of the protein-translocation enzymesof Escherichia coli,” Proceedings of the National Academy ofSciences of the United States of America, vol. 87, no. 8, pp. 3107–3111, 1990.

[70] P. Schurtenberger, N. Mazer, S. Waldvogel, and W. Kanzig,“Preparation of monodisperse vesicles with variable sizeby dilution of mixed micellar solutions of bile salt andphosphatidylcholine,” Biochimica et Biophysica Acta, vol. 775,no. 1, pp. 111–114, 1984.

[71] J. Brunner, P. Skrabal, and H. Hauser, “Single bilayer vesiclesprepared without sonication: physico chemical properties,”Biochimica et Biophysica Acta, vol. 455, no. 2, pp. 322–331,1976.

[72] S. M. Goldin, “Formation of unilamellar lipid vesicles ofcontrollable dimensions by detergent dialysis,” Biochemistry,vol. 18, no. 19, pp. 4173–4176, 1979.

[73] M. H. W. Milsmann, R. A. Schwendener, and H. G. Weder,“The preparation of large single bilayer liposomes by a fast andcontrolled dialysis,” Biochimica et Biophysica Acta, vol. 512, no.1, pp. 147–155, 1978.

[74] Y. Kagawa and E. Racker, “Partial resolution of the enzymescatalysing oxidative phosphorylation. Reconstitution of vesi-cles catalysing adenosinetriphosphate exchange,” The Journalof Biological Chemistry, vol. 246, pp. 5477–5487, 1971.


Research Article

Cellular Injury of Cardiomyocytes during HepatocyteGrowth Factor Gene Transfection with Ultrasound-TriggeredBubble Liposome Destruction

Kazuo Komamura,1 Rie Tatsumi,1 Yuko Tsujita-Kuroda,1 Takatoshi Onoe,2

Kunio Matsumoto,3 Toshikazu Nakamura,3 Jun-ichi Miyazaki,4 Takeshi Horio,5

and Masaru Sugimachi1

1 Department of Cardiovascular Dynamics, Research Institute, National Cerebral and Cardiovascular Center, 5-7-1 Fujishiro-dai,Suita, Osaka 565-8565, Japan

2 Department of Nursing Science, Taisei Gakuin University, Sakai 587-8555, Japan3 Division of Molecular Regenerative Medicine, Department of Biochemistry and Molecular Biology,Osaka University Graduate School of Medicine, Suita 565-0871, Japan

4 Division of Stem Cell Regulation Research, G6, Osaka University School of Medicine, Suita 565-0871, Japan5 Division of Hypertension, National Cerebral and Cardiovascular Center, Suita 565-8565, Japan

Correspondence should be addressed to Kazuo Komamura, [email protected]

Received 6 August 2010; Accepted 31 October 2010

Academic Editor: Hasan Uludag

Copyright © 2011 Kazuo Komamura et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

We transfected naked HGF plasmid DNA into cultured cardiomyocytes using a sonoporation method consisting of ultrasound-triggered bubble liposome destruction. We examined the effects on transfection efficiency of three concentrations of bubbleliposome (1 × 106, 1 × 107, 1 × 108/mL), three concentrations of HGF DNA (60, 120, 180 μg/mL), two insonification times (30,60 sec), and three incubation times (15, 60, 120 min). We found that low concentrations of bubble liposome and low concentrationsof DNA provided the largest amount of the HGF protein expression by the sonoporated cardiomyocytes. Variation of insonificationand incubation times did not affect the amount of product. Following insonification, cardiomyocytes showed cellular injury, asdetermined by a dye exclusion test. The extent of injury was most severe with the highest concentration of bubble liposome. Inconclusion, there are some trade-offs between gene transfection efficiency and cellular injury using ultrasound-triggered bubbleliposome destruction as a method for gene transfection.

1. Introduction

Ultrasound-triggered bubble liposome destruction (sono-poration) has been proposed as a safe nonviral means ofgene therapy that can target many different cells or tissues.In the field of cardiovascular medicine, this method mayhave significant potential for the introduction of therapeuticgenes directly into vascular cells or cardiomyocytes [1, 2].Sonoporation can only be clinically effective if the dose-effect relationship between the amount of bubble liposomeand transfection efficiency is first established. However, few

reports have already examined this dose-effect relationshipand the safety of the procedure [3].

Transfection efficiency in sonoporation depends onvarious conditions including type of microbubble, mode ofultrasound, frequency of ultrasound, intensity of acousticpressure, concentration of microbubble, dose of DNA,duration of insonification, incubation time of cell withDNA, repeat count of insonification, type of targeted cell,and other physicochemical conditions like temperature andcarbon dioxide concentration, [3]. Greenleaf et al. reportedthat ultrasound acoustic pressure, DNA concentration, and


repeat count of insonification correlated with transfectionrate [4]. Teupe et al. demonstrated that duration of insoni-fication did not affect transfection rate [5]. Then, Chen etal. showed that transfection rate reached plateau when DNAconcentration was increased [6].

Greenleaf et al. also showed that transfection rate peakedand fell off according to the change in liposome concen-tration [4]. They thought it might be derived from cellulartoxicity of large amount of liposome. Li et al. reportedthat cell viability decreased along with the increase inmicrobubble concentration [1]. Guo et al. demonstrated thatcell viability decreased with the increase in duration ofinsonification [7]. Suzuki et al. and Li et al. showed that cellviability decreased with the increase in ultrasound acousticpressure [8, 9].

On the basis of those previous findings, we planned toexamine the effects of amount of plasmid DNA, liposomeconcentration, duration of insonification, repeat count ofinsonification, and time of incubation with liposome, cell,and DNA on transfection rate, which was measured by meansof HGF protein release into culture medium.


2.1. Cell Culture. Primary cultures of neonatal ventricularmyocytes were prepared as described previously [10]. Briefly,apical halves of cardiac ventricles from 1-day-old Wistarrats were separated, minced, and dispersed with 0.1% col-lagenase type II (Worthington Biochemical Corp., Freehold,NJ). Myocytes were segregated from nonmyocytes using adiscontinuous Percoll gradient (Sigma Chemical Co., Inc., St.Louis, MO). After centrifugation, the upper layer consisted ofa mixed population of nonmyocyte cell types and the lowerlayer consisted almost exclusively of cardiac myocytes. Afterthe myocytes had been incubated twice on uncoated 10-cm culture dishes for 30 minutes to remove any remainingnonmyocytes, the nonattached viable cells were plated ongelatin-coated 24-well culture plates or 10-cm culture dishesand then cultured in DMEM (Life Technologies, GrandIsland, NY) supplemented with 10% FCS (Life Technologies,Grand Island, NY). After 24-hour incubation in DMEMwith FCS, the culture medium was changed to serum-freeDMEM, and all experiments were performed 24 hours later.This purification procedure has well been established [11,12], and >95% of the cells obtained by this method werecardiomyocytes.

2.2. Plasmid DNA. Preparation of rat hepatocyte growth fac-tor (HGF) expression plasmid DNA was described previously[13]. Briefly, rat HGF cDNA cloned by polymerase chainreaction was inserted into the unique Xho I site betweenthe cytomegalovirus immediate-early enhancer-chicken β-actin hybrid promoter and rabbit β-globin poly A siteof the pCAGGS expression plasmid [14]. The resultingplasmid, pCAGGS-HGF, was grown in Escherichia coliDH5α (Figure 1(a)). The plasmid was purified with aQUIAGEN plasmid DNA kit and dissolved in TE buffer. Thepurified plasmid DNA was stored at −20◦C and diluted to

the required concentration with distilled water immediatelybefore use.

2.3. Bubble Liposome. Liposome microbubble, SHU 508A,consists of palmitic acid and galactose and providesechogenic micron-sized air bubbles when suspended inwater. The diameter of bubbles ranges from 2 to 8 μm, and97% are smaller than 6 μm [15]. These air bubbles arestabilized by palmitic acid, which forms a molecular filmthat lowers the surface tension of the aqueous vehicle.The SHU 508A bubbles are nontoxic, have a neutral pH,are biodegradable, and are made from a physiologicallyoccurring substance. The physiochemical properties of SHU508A bubbles are typical of a saccharide solution [15].

2.4. Experiment on Ultrasound Mode. Before performingthe experiments for dose-effect relationships using liposomesonoporation, we needed to determine the most appropri-ate ultrasound mode for the sonoporation procedure forefficient transfection. We tested four modes of ultrasound:pulsed wave Doppler, color flow Doppler, continuous waveDoppler, and harmonic power Doppler, which are availablewith standard echocardiographic machinery in a clinical lab-oratory. We performed a simple transfection experiment atthe same acoustic pressure of 0.5 W/cm2 for each ultrasoundmode, using a single condition with 60 μg HGF plasmidDNA, 1 × 107 particles/mL of SHU 508A liposome, 30 secinsonification, 15 min of DNA incubation, and 3 repetitionsof insonification.

Rat neonatal cardiomyocytes were inoculated and grownto confluence in DMEM+10% FCS. After confluence hadbeen reached in a 35 mm Petri dish, the medium was changedto fresh defined serum-free medium. Plasmid DNA wasdiluted with distilled water immediately before the transfec-tion. Each experiment was performed on 20 dishes. Cells oneach dish were treated with ultrasound (Figure 1(b)). Pulsedwave Doppler, color flow Doppler, and continuous waveDoppler were insonified from PSK-25AT acoustic transducerwith Toshiba SSA-380A (Toshiba Medical Systems), andharmonic power Doppler was insonified from S3 transducerwith Sonos 5500 (Phillips Medical Systems). The exper-imental results are shown in Figure 2. Continuous-waveDoppler ultrasound was the most efficacious and was usedfor subsequent experiments.

2.5. Experiments for Dose-Effect Relations. The mediumin 35 mm Petri dishes containing the cardiomyocyteswas changed to fresh defined serum-free medium fromDMEM+10% FCS. Rat HGF plasmid DNA was diluted withdistilled water, and a volume corresponding to 60, 120, or180 μg was added to each of the 20 Petri dishes per DNA dose.Cells on each dish were then treated with continuous-waveDoppler ultrasound (frequency of 2.5 MHz and acousticintensity of 0.5 W/cm2 from a PSK-25AT acoustic transducerwith Toshiba SSA-380A Ultrasound system) with SHU 508Aliposome (1 × 107 particles/mL) for acoustic exposure timeof 30 or 60 seconds at room temperature (Figure 1(b)).In a separate series of experiments, we tested four liposome


hCMV-IE enhancer

Chicken β-actinpromoter

pCAGGS-HGF6.8 kb

Rabbit β-globin

polyA

Rat HGF

Km

(a)

DMEM + FCS

Continuous wave ultrasound

Culture dish

Ultrasound transducer

Rat neonatalcardiomyocytes

(b)

Figure 1: (a) Structure of the expression plasmid pCAGGS-HGF. The expression cassette of pCAGGS-HGF contains chicken β-actinpromoter, rat HGF, and rabbit β-globin poly A. (b) Experimental setup. The transducer was attached to the bottom of the dish.

0

0.1

0.2

0.3

0.4

0.5

0.6

(ng/

mL

/mg)

DNA alone CFD PWD CWD Harmonic

∗

∗ ∗∗ ∗∗∗ †

∗ ∗∗∗

Figure 2: Comparison of four modes of ultrasound for sono-poration. Cells treated with continuous-wave Doppler ultrasoundyielded the largest amount of HGF protein indicating this to be themost effective ultrasound mode. CFD: color flow Doppler; PWD:pulsed wave Doppler; CWD: continuous wave Doppler; Harmonic:harmonic power Doppler. ∗P < .05 versus DNA alone; ∗∗P < .05versus CFD; ∗∗∗P < .05 versus PWD; †P < .05 versus Harmonic.

concentrations (0, 1 × 106, 1 × 107 or 1 × 108 particles/mL),three insonification repetitions (1 insonification only, 3 or 5insonifications for 30 seconds), and three DNA incubationtimes (15, 60 or 120 min). After the incubation, the culturemedium was changed to normal DMEM+10% FCS and

the cells were cultured for 72 hours. In a separate setof experiments, we examined the effect of culture periodon the amount of DNA product that is HGF protein bydiscontinuing culture at 24, 48, and 72 hours and measuringthe amount of rat HGF protein in the medium. The totalamount of protein content in the cultured cells was measuredand used to correct the HGF level in each dish. We measuredrat HGF protein using an EIA kit (Institute of ImmunologyCo., Ltd., Tokyo, Japan) [13] and protein content of culturedcells using a Modified Lowry Protein Assay Kit (PierceBiotechnology, Rockford).

2.6. Viability of Cultured Cells. To determine the safetyof sonoporation, in a separate experiment, cultured cellswere exposed to 0.1% trypan blue for 5 min just afterultrasound insonification. This allowed assessment of sar-colemmal membrane damage and was performed for eachconcentration of liposome, each insonification time, andeach number of repetitions of insonification. The numberof stained and unstained cells in the dishes was countedand used to calculate the percentage of intact cells [16].The degree of cellular injury caused by sonoporation wasdetermined by examining the insonified cells by scanningelectron microscopy (Hitachi S-4800). Immediately afterultrasound insonification in the presence of liposome, thecardiomyocytes were fixed with phosphate-buffered 2.5%glutaraldehyde for 4 hours, followed by postfixation with 1%


osmium tetroxide for 1 hour, and then were conventionallyprepared for scanning electron microscopy.

2.7. Statistical Analysis. Data were expressed as the mean ±SEM. Comparisons of parameters from experimental groupswere performed with unpaired t tests and resulting P-values were corrected according to the Bonferroni method.In analyses, P < .05 was considered to indicate statisticalsignificance.

3. Results

3.1. Effect of Culture Period on HGF Protein Production bySonoporated Cardiomyocytes. The concentration of HGFprotein in the culture medium increased as the culture periodafter ultrasonic transfection was extended. The transfectionconsisted of three 30-sec insonifications a 15-min incubationwith HGF DNA (60 μg) and liposome (1 × 107 particle/mL)(Figure 3(a)). After 72 hours of culture, HGF protein concen-tration in the culture medium was measured and correctedusing the protein content of the cultured cells.

3.2. Effect of the Amount of Plasmid DNA on HGF Protein Pro-duction by Sonoporated Cardiomyocytes. HGF protein con-centration in the culture medium was 0.54±0.049 ng/mL/mgand was highest when 60 μg of DNA was administered witha liposome concentration of 1 × 107 particles/mL, a15-minincubation, and three 30-sec insonification. Although thenominal mean values of HGF protein after transfection of120 and 180 μg DNA were lower than those after transfectionof 60 μg, the differences were not statistically significant(Figure 3(b)).

3.3. Effect of Incubation Period with Plasmid DNA and Lipo-some on HGF Protein Production by Sonoporated Cardiomy-ocytes. HGF protein concentration in the culture mediumwas 0.56 ± 0.053 ng/mL/mg and was highest when theincubation time was 15 min with a liposome concentrationof 1 × 107 particles/mL, 60 μg DNA, and three 30-secinsonification. Although the mean values of HGF proteinafter transfection for 60 and 120 min were lower than thoseafter 15 min incubation, the differences were not statisticallysignificant (Figure 3(c)).

3.4. Effect of Insonification Time on HGF Protein Productionby Sonoporated Cardiomyocytes. HGF protein concentrationin the culture medium was 0.59 ± 0.052 ng/mL/mg and washighest when the insonification period was 30 sec with 60 μgDNA, a liposome concentration of 1 × 107 particles/mL,and 15-min incubation. There was no significant differencein HGF production in cells insonified for 30 and 60 min(Figure 3(d)).

3.5. Effect of Liposome Concentration on HGF Protein Produc-tion by Sonoporated Cardiomyocytes. HGF protein concen-tration in the culture medium was 0.53 ± 0.053 ng/mL/mgand was nominally highest when the liposome concentrationwas 1 × 107 particles/mL and insonification consisted of

three 30-sec ultrasound exposures, though it was statisticallysimilar to that obtained with 1×106 particles/mL. At a higherliposome concentration of 1×108 particles/mL, HGF proteinconcentration decreased (Figure 3(e)).

3.6. Effect of Repetition of Insonification on HGF Protein Pro-duction by Sonoporated Cardiomyocytes. HGF protein con-centration in the culture medium was 0.54 ± 0.053 ng/mL/mg and was highest when three 30-sec insonifications weregiven, with a liposome concentration of 1× 107 particles/mLand 60 mg DNA. This protein production was statisti-cally higher than in cells given one or five insonifications(Figure 3(f)).

3.7. Effect of Insonification Time on Cell Viability. The per-centage of dead cells was 14.7 ± 0.9% and was higher in thecells given five 30-sec insonifications at a liposome concen-tration of 1 × 107 particles/mL (Figure 4(a)). There was nostatistical difference between 30- and 60-sec insonification.

3.8. Effect of Liposome Concentration on Cell Viability. Thepercentage of dead cells increased with increasing concen-trations of liposome (Figure 4(b)). The dead cell count was24.8 ± 2.9% and was highest when the liposome concentra-tion was 1×108 particles/mL and three 30-sec insonificationswere used.

3.9. Effect of Number of Insonification Repetitions on Cell Via-bility. The percentage of dead cells increased as the numberof insonification repetitions increased (Figure 4(c)). Thedead cell count was 14.7 ± 0.9% and was highest whenfive repetitions of the insonification step were given, witha liposome concentration of 1× 107 particles/mL.

3.10. Scanning Electron Microscopy Observations of Sonopo-rated Cardiomyocytes. No particular changes were evidenton the surfaces of untreated control cultured cardiomyocyteswhen viewed with the scanning electron microscope at lowand high magnification (Figures 5(a) and 5(b)). After sono-poration with a low concentration of liposome (Figure 5(c))and with a high concentration of liposome (Figure 5(d)),microdimples or pores were observed on the surfaces of thecultured cardiomyocytes.

4. Discussion

Considerable efforts have been made to develop methodsthat will allow effective and safe introduction of vectors intocells for gene therapy. However, we still need a breakthroughin the form of a novel vector that will transform cells athigh efficiency and with low risk of adverse effects. This isespecially true in cardiovascular medicine, where malignantcellular transformation is rare [17]. One of the promisingcandidates for safe and efficacious gene transfection is anaked plasmid vector that has been modified to have highaffinity for cardiovascular tissues but which has no built-in viral components [17, 18]. We have developed a methodfor electroporation of a cytokine gene for treatment of


0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

(ng/

mL

/mg)

Baseline 24 hrs 48 hrs 72 hrs

∗

∗ ∗∗

∗ ∗∗ ∗∗∗

(a)

0

0.1

0.2

0.3

0.4

0.5

0.6

(ng/

mL

/mg)

DNA alone 60μg 120μg 180μg

∗ ∗∗

(b)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

(ng/

mL

/mg)

DNA alone 15 min 60 min 120 min

∗∗

∗

(c)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

(ng/

mL

/mg)

DNA alone 30 s 60 s

∗ ∗∗

(d)

0

0.1

0.2

0.3

0.4

0.5

0.6(n

g/m

L/m

g)

DNAalone

DNA +US

1×106

1×107

1×108

∗ ∗∗∗ ∗∗ ∗∗∗

∗ ∗∗

(e)

0

0.1

0.2

0.3

0.4

0.5

0.6

(ng/

mL

/mg)

DNA alone 1 time 3 times 5 times

∗

∗ ∗∗ ∗∗∗

∗

(f)

Figure 3: (a) Effect of culture period after transfection of HGF DNA on HGF protein production using 60 μg of DNA and 1×107 particles/mLliposome with three 30-sec insonifications and 15-min incubation with DNA. Baseline was the concentration of rat HGF protein in theculture medium around rat cardiomyocytes without any intervention at the beginning of cell culture. ∗P < .05 versus baseline; ∗∗P <.05 versus 24 hours after the onset of culture; ∗∗∗P < .05 versus 48 hours after the onset of culture. (b) Effect of amount of plasmidDNA on HGF protein production using 1× 107 particles/mL liposome with three 30-sec insonifications and 15-min incubation with DNA.“DNA alone” indicates the concentration of rat HGF protein in the culture medium of cardiomyocytes treated with 60 μg DNA withoutinsonification. ∗P < .05 versus DNA alone. (c) Effect of incubation period of cardiomyocytes with plasmid DNA and liposome on HGFprotein production using 60 μg of DNA and 1 × 107 particles/mL liposome with three 30-sec insonifications. ∗P < .05 versus DNA alone.(d) Effect of insonification time on protein production using 60 μg of DNA, 1 × 107 particles/mL liposome, and 15-min incubation withDNA, and three 30- or 60-sec insonifications. ∗P < .05 versus DNA alone. (e) Effect of liposome concentration on HGF protein productionusing 60 μg of DNA with three 30-sec insonifications and 15-min incubation with DNA. ∗P < .05 versus DNA alone; ∗∗P < .05 versus 0particles/mL; ∗∗∗P < .05 versus 1×108 particles/mL. (f) Effect of repetition of insonification on HGF protein production using 6 μg of DNAand 1 × 107 particles/mL liposome with 15-min incubation with DNA. ∗P < .05 versus DNA alone; ∗∗P < .05 versus 1 time; ∗∗∗P < .05versus 5 times.

0

5

10

15

20

25

30

(%)

Baseline US alone 30 s 60 s

∗ ∗

(a)

0

5

10

15

20

25

30

(%)

Baseline USalone

1×106

1×107

1×108

∗∗

∗ ∗∗ ∗∗∗

(b)

0

5

10

15

20

25

30

(%)

Baseline USalone

30 s×1

30 s×3

30 s×5

∗∗

∗ ∗∗

(c)

Figure 4: (a) Effect of insonification time on cell viability using 60 μg of DNA, 1× 107 particles/mL liposome, and 15-min incubation withDNA, and three 30- or 60-sec insonifications. “US alone” represents the percentage of dead cells immediately after three 30-sec insonificationsin the absence of liposome and DNA. ∗P < .05 versus baseline. (b) Effect of liposome concentration on cell viability using 60 μg of DNA andthree 30-sec insonifications and 15-min incubation with DNA. ∗P < .05 versus baseline; ∗∗P < .05 versus 1 × 106 particles/mL; ∗∗∗P < .05versus 1×108 particles/mL. (c) Effect of repetitions of insonification on cell viability using 60 μg of DNA, 1×107 particles/mL liposome, and15-min incubation with DNA. ∗P < .05 versus baseline; ∗∗P < .05 versus 30 sec × 1.


K3 4 3 kV x 2 K 15μm

(a)

K3 5 3 kV x 10 K 3μm

(b)

K6 4 3 kV x 10 K 3μm

(c)

K2 3 3 kV x 10 K 3μm

(d)

Figure 5: (a) and (b) Scanning electron microscopic images of intact cell surfaces of cultured cardiomyocytes. Scale dots are indicated onthe images. (c) Image of a cell surface immediately after sonoporation using 1 × 106 particles/mL liposome. (d) Image of a cell surfaceimmediately after sonoporation using 1× 108 particles/mL liposome.

cardiomyopathy [13]. However, using electric shock fortransfection is not clinically practical. For this reason, we arepursuing the present sonoporation method as a protocol forgene transfection.

The HGF protein used in the present study is foundin a wide variety of cell types and has multiple biologicalproperties, including mitogenic, motogenic, morphogenicand antiapoptotic activities [19]. Several lines of evidenceindicate that this molecule has potential for therapeutic usefor treatment of heart failure, myocardial infarction, angina,and hypertension [20–22]. HGF may also have enormoustherapeutic potential for hepatic and renal disorders, inaddition to cardiovascular diseases [23–26].

In the present study, we showed variations in amount ofHGF plasmid DNA, liposome concentration, the duration ofinsonification, and incubation time of the cardiomyocyteswith liposome and DNA, and their dose relationships withthe final amount of HGF protein released from the culturedneonatal cardiomyocytes. We found that specific amountsof liposome and repetitions of insonification were neededfor effective protein production from cardiomyocytes. How-ever, high concentrations of bubble liposome and largenumbers of repeat insonifications resulted in decreased cellviability.

Plasma membrane sonoporation induced by ultrasoundand subsequent self-sealing has been reported in previousinvestigations [27–29]. However, the exact mechanism bywhich membrane sonoporation causes substance incorpora-tion into the cell is not yet understood. Some investigators

speculate that the membrane poration results in both trans-fection efficiency and cellular damage. In the present study,scanning microscopy images revealed some microdimplesor pores on the cell surface after sonoporation, which didnot exist on the surface of control cardiomyocytes. Thenumbers of dimples or pores tended to increase with higherconcentrations of liposome. Thus, we speculate that thesedimples or pores on the cell surface might be related totransfection efficiency and might be evidence of cellularinjury by sonoporation. Previous studies of sonoporation ofvascular walls revealed that microbubble destruction wouldcause rupture of microvessels and extravasation [30–33],which would cancel out some benefits of sonoporation.Thus, the poration and self-sealing mechanism needs to befully investigated and optimized.

A sonoporation technique targeting the cardiovascularsystem has now been developed for gene transfection tomyocardium, limb skeletal muscle, and arteries [34–37].For a variety of target tissues, a number of microbubbles,including liposomes, and a range of ultrasound modes havebeen developed. The optimal combination of the type ofmicrobubble, ultrasound mode, and target tissue still needsto be resolved [38–40]. However, the principal types ofultrasound used for sonoporation have included pulsed waveDoppler or continuous wave Doppler with acoustic pressureranging 0.5–5 W/cm2 [34–37]. In the present study, we foundthat continuous wave Doppler at a standard frequency forclinical use, that is, 2.5 MHz and the usual acoustic pressureof 0.5 W/cm2, was most effective with our cardiomyocytes.


The reason we used one of the standard ultrasound modeswith standard settings for clinical use is that we would like touse our sonoporation system eventually in a clinical setting.

The present study has several limitations. To avoid thecomplexity of numerous combinations of experimental con-ditions, such as amount of DNA, concentration of liposome,duration of insonification, repeat count of insonification,length of incubation time, and culture period after genetransfection, we only used several practical combinations foran in vitro experiment for cultured cardiomyocytes. Thus, wemight have missed other multimodal aspects of dose-effectrelationships among these conditions.

5. Conclusion

HGF DNA was successfully transferred to cultured car-diomyocytes using sonoporation with a defined liposomeconcentration and a mode of insonification. A number oftrade-offs between transfection efficiency and cellular injuryhave to be balanced to optimize this sonoporation method.

Acknowledgments

This study was supported by a Grant-in-Aid for ScientificResearch 14570709 from the Ministry of Education, Culture,Sport, Science, and Technology of Japan and by the Programfor Promotion of Fundamental Studies in Health Sciences ofthe Pharmaceuticals and Medical Devices Agency (PMDA).

References

[1] T. Li, K. Tachibana, M. Kuroki, and M. Kuroki, “Gene transferwith echo-enhanced contrast agents: comparison betweenAlbunex, Optison, and Levovist in mice—initial results,”Radiology, vol. 229, no. 2, pp. 423–428, 2003.

[2] S. Tsunoda, O. Mazda, Y. Oda et al., “Sonoporation usingmicrobubble BR14 promotes pDNA/siRNA transduction tomurine heart,” Biochemical and Biophysical Research Commu-nications, vol. 336, no. 1, pp. 118–127, 2005.

[3] L. B. Feril Jr. and T. Kondo, “Biological effects of low intensityultrasound: the mechanism involved, and its implications ontherapy and on biosafety of ultrasound,” Journal of RadiationResearch, vol. 45, no. 4, pp. 479–489, 2004.

[4] W. J. Greenleaf, M. E. Bolander, G. Sarkar, M. B. Goldring,and J. F. Greenleaf, “Artificial cavitation nuclei significantlyenhance acoustically induced cell transfection,” Ultrasound inMedicine and Biology, vol. 24, no. 4, pp. 587–595, 1998.

[5] C. Teupe, S. Richter, B. Fisslthaler et al., “Vascular genetransfer of phosphomimetic endothelial nitric oxide synthase(S1177D) using ultrasound-enhanced destruction of plasmid-loaded microbubbles improves vasoreactivity,” Circulation,vol. 105, no. 9, pp. 1104–1109, 2002.

[6] S. Chen, R. V. Shohet, R. Bekeredjian, P. Frenkel, and P. A.Grayburn, “Optimization of ultrasound parameters for car-diac gene delivery of adenoviral or plasmid deoxyribonucleicacid by ultrasound-targeted microbubble destruction,” Journalof the American College of Cardiology, vol. 42, no. 2, pp. 301–308, 2003.

[7] D.-P. Guo, X.-Y. Li, P. Sun et al., “Ultrasound-targetedmicrobubble destruction improves the low density lipoproteinreceptor gene expression in HepG2 cells,” Biochemical and

Biophysical Research Communications, vol. 343, no. 2, pp. 470–474, 2006.

[8] R. Suzuki, T. Takizawa, Y. Negishi, N. Utoguchi, and K.Maruyama, “Effective gene delivery with novel liposomalbubbles and ultrasonic destruction technology,” InternationalJournal of Pharmaceutics, vol. 354, no. 1-2, pp. 49–55, 2008.

[9] H. L. Li, X. Z. Zheng, H. P. Wang, F. Li, Y. Wu, and L. F.Du, “Ultrasound-targeted microbubble destruction enhancesAAV-mediated gene transfection in human RPE cells in vitroand rat retina in vivo,” Gene Therapy, vol. 16, no. 9, pp. 1146–1153, 2009.

[10] T. Horio, T. Nishikimi, F. Yoshihara, H. Matsuo, S. Takishita,and K. Kangawa, “Inhibitory regulation of hypertrophy byendogenous atrial natriuretic peptide in cultured cardiacmyocytes,” Hypertension, vol. 35, no. 1, part 1, pp. 19–24, 2000.

[11] J.-S. Zheng, M. O. Boluyt, L. O’Neill, M. T. Crow, and E.G. Lakatta, “Extracellular ATP induces immediate-early geneexpression but not cellular hypertrophy in neonatal cardiacmyocytes,” Circulation Research, vol. 74, no. 6, pp. 1034–1041,1994.

[12] M. Harada, H. Itoh, O. Nakagawa et al., “Significance ofventricular myocytes and nonmyocytes interaction during car-diocyte hypertrophy: evidence for endothelin-1 as a paracrinehypertrophic factor from cardiac nonmyocytes,” Circulation,vol. 96, no. 10, pp. 3737–3744, 1997.

[13] K. Komamura, R. Tatsumi, J.-I. Miyazaki et al., “Treatmentof dilated cardiomyopathy with electroporation of hepatocytegrowth factor gene into skeletal muscle,” Hypertension, vol. 44,no. 3, pp. 365–371, 2004.

[14] K. Matsumoto and T. Nakamura, “Hepatocyte growth factor(HGF) as a tissue organizer for organogenesis and regenera-tion,” Biochemical and Biophysical Research Communications,vol. 239, no. 3, pp. 639–644, 1997.

[15] B. B. Goldberg, JI. B. Liu, and F. Forsberg, “Ultrasoundcontrast agents: a review,” Ultrasound in Medicine and Biology,vol. 20, no. 4, pp. 319–333, 1994.

[16] S. Kato, G. Takemura, R. Maruyama et al., “Apoptosis, ratherthan oncosis, is the predominant mode of spontaneous deathof isolated adult rat cardiac myocytes in culture,” JapaneseCirculation Journal, vol. 65, no. 8, pp. 743–748, 2001.

[17] O. J. Muller, H. A. Katus, and R. Bekeredjian, “Targeting theheart with gene therapy-optimized gene delivery methods,”Cardiovascular Research, vol. 73, no. 3, pp. 453–462, 2007.

[18] C. A. Holladay, T. O’Brien, and A. Pandit, “Non-viral genetherapy for myocardial engineering,” Wiley InterdisciplinaryReviews: Nanomedicine and Nanobiotechnology, vol. 2, no. 3,pp. 232–248, 2010.

[19] K. Matsumoto and T. Nakamura, “Hepatocyte growth factor(HGF) as a tissue organizer for organogenesis and regenera-tion,” Biochemical and Biophysical Research Communications,vol. 239, no. 3, pp. 639–644, 1997.

[20] K. Matsumoto and T. Nakamura, “HGF: its organotrophic roleand therapeutic potential,” CIBA Foundation Symposia, no.212, pp. 198–211, 1997.

[21] N. Tomita, R. Morishita, J. Higaki, and T. Ogihara, “Novelmolecular therapeutic approach to cardiovascular diseasebased on hepatocyte growth factor,” Journal of Atherosclerosisand Thrombosis, vol. 7, no. 1, pp. 1–7, 2000.

[22] K. Komamura, J. Miyazaki, E. Imai, K. Matsumoto, T.Nakamura, and M. Hori, “Hepatocyte growth factor genetherapy for hypertension,” Methods in Molecular Biology, vol.423, pp. 393–404, 2008.

[23] K.-I. Kosai, K. Matsumoto, S. Nagata, Y. Tsujimoto, and T.Nakamura, “Abrogation of Fas-induced fulminant hepatic


failure in mice by hepatocyte growth factor,” Biochemical andBiophysical Research Communications, vol. 244, no. 3, pp. 683–690, 1998.

[24] T. Ueki, Y. Kaneda, H. Tsutsui et al., “Hepatocyte growth factorgene therapy of liver cirrhosis in rats,” Nature Medicine, vol. 5,no. 2, pp. 226–230, 1999.

[25] S. Mizuno, K. Matsumoto, and T. Nakamura, “HGF asa renotrophic and anti-fibrotic regulator in chronic renaldisease,” Frontiers in Bioscience, vol. 13, pp. 7072–7086, 2008.

[26] K. Matsumoto and T. Nakamura, “Hepatocyte growth factor:renotropic role and potential therapeutics for renal diseases,”Kidney International, vol. 59, no. 6, pp. 2023–2038, 2001.

[27] K. Tachibana, T. Uchida, K. Ogawa, N. Yamashita, and K.Tamura, “Induction of cell-membrane porosity by ultra-sound,” The Lancet, vol. 353, no. 9162, p. 1409, 1999.

[28] S. Mehier-Humbert, T. Bettinger, F. Yan, and R. H. Guy,“Plasma membrane poration induced by ultrasound exposure:implication for drug delivery,” Journal of Controlled Release,vol. 104, no. 1, pp. 213–222, 2005.

[29] F. Yang, N. Gu, D. Chen et al., “Experimental study on cell self-sealing during sonoporation,” Journal of Controlled Release,vol. 131, no. 3, pp. 205–210, 2008.

[30] D. M. Skyba, R. J. Price, A. Z. Linka, T. C. Skalak, and S. Kaul,“Direct in vivo visualization of intravascular destruction ofmicrobubbles by ultrasound and its local effects on tissue,”Circulation, vol. 98, no. 4, pp. 290–293, 1998.

[31] J. Song, J. C. Chappell, M. Qi, E. J. VanGieson, S. Kaul, andR. J. Price, “Influence of injection site, microvascular pressureand ultrasound variables on microbubble-mediated deliveryof microspheres to muscle,” Journal of the American College ofCardiology, vol. 39, no. 4, pp. 726–731, 2002.

[32] S. Hernot and A. L. Klibanov, “Microbubbles in ultrasound-triggered drug and gene delivery,” Advanced Drug DeliveryReviews, vol. 60, no. 10, pp. 1153–1166, 2008.

[33] M. R. Bohmer, C. H. T. Chlon, B. I. Raju, C. T. Chin,T. Shevchenko, and A. L. Klibanov, “Focused ultrasoundand microbubbles for enhanced extravasation,” Journal ofControlled Release, vol. 148, no. 1, pp. 18–24, 2010.

[34] I. Rosenthal, J. Z. Sostaric, and P. Riesz, “Sonodynamic thera-pya review of the synergistic effects of drugs and ultrasound,”Ultrasonics Sonochemistry, vol. 11, no. 6, pp. 349–363, 2004.

[35] J. M. Tsutsui, F. Xie, and R. T. Porter, “The use of microbubblesto target drug delivery,” Cardiovascular Ultrasound, vol. 2,article 23, 2004.

[36] C. M. H. Newman and T. Bettinger, “Gene therapy progressand prospects: ultrasound for gene transfer,” Gene Therapy,vol. 14, no. 6, pp. 465–475, 2007.

[37] C. R. Mayer and R. Bekeredjian, “Ultrasonic gene and drugdelivery to the cardiovascular system,” Advanced Drug DeliveryReviews, vol. 60, no. 10, pp. 1177–1192, 2008.

[38] S. Hernot and A. L. Klibanov, “Microbubbles in ultrasound-triggered drug and gene delivery,” Advanced Drug DeliveryReviews, vol. 60, no. 10, pp. 1153–1166, 2008.

[39] S. Tinkov, R. Bekeredjian, G. Winter, and C. Coester,“Microbubbles as ultrasound triggered drug carriers,” Journalof Pharmaceutical Sciences, vol. 98, no. 6, pp. 1935–1961, 2009.

[40] C.-Y. Lin, T.-M. Liu, C.-Y. Chen et al., “Quantitative andqualitative investigation into the impact of focused ultrasoundwith microbubbles on the triggered release of nanoparticlesfrom vasculature in mouse tumors,” Journal of ControlledRelease, vol. 146, no. 3, pp. 291–298, 2010.


Research Article

In Vitro Gene Delivery Mediated byAsialofetuin-Appended Cationic Liposomes Associatedwith γ-Cyclodextrin into Hepatocytes

Keiichi Motoyama,1 Yoshihiro Nakashima,1 Yukihiko Aramaki,2 Fumitoshi Hirayama,3

Kaneto Uekama,3 and Hidetoshi Arima1

1 Graduate School of Pharmaceutical Sciences, Kumamoto University, 5-1 Oe-honmachi, Kumamoto 862-0973, Japan2 School of Pharmacy, Tokyo University of Pharmacy and Life Sciences, 1432-1 Horinouchi, Hachioji, Tokyo 192-0392, Japan3 Faculty of Pharmaceutical Sciences, Sojo University, 4-22-1 Ikeda, Kumamoto 860-0082, Japan

Correspondence should be addressed to Hidetoshi Arima, [email protected]

Received 1 July 2010; Accepted 10 October 2010

Academic Editor: Ali Nokhodchi

Copyright © 2011 Keiichi Motoyama et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

The purpose of this study is to evaluate in vitro gene delivery mediated by asialofetuin-appended cationic liposomes (AF-liposomes) associating cyclodextrins (CyD/AF-liposomes) as a hepatocyte-selective nonviral vector. Of various CyDs, AF-liposomes associated with plasmid DNA (pDNA) and γ-cyclodextrin (γ-CyD) (pDNA/γ-CyD/AF-liposomes) showed the highestgene transfer activity in HepG2 cells without any significant cytotoxicity. In addition, γ-CyD enhanced the encapsulation ratioof pDNA with AF-liposomes, and also increased gene transfer activity as the entrapment ratio of pDNA into AF-liposomes wasincreased. γ-CyD stabilized the liposomal membrane of AF-liposomes and inhibited the release of calcein from AF-liposomes. Thestabilizing effect of γ-CyD may be, at least in part, involved in the enhancing gene transfer activity of pDNA/γ-CyD/AF-liposomes.Therefore, these results suggest the potential use of γ-CyD for an enhancer of transfection efficiency of AF-liposomes.

1. Introduction

The principle of somatic gene therapy is that genes can beintroduced into selected cells in the body in order to treatgenetic or acquired diseases. The liver may be potentially animportant target for gene therapy, because crucial diseasessuch as amyloidosis, primary biliary cirrhosis, familialhypercholesteremia, phenyl ketonuria, and virus hepatitisoccur in this organ [1]. In addition, the liver has the abilityto synthesize a wide variety of proteins, to perform variousposttranslational modifications, and to secrete them into theblood.

Of various nonviral methods, the lipofection method,by which cationic lipids (cationic liposomes) are used fortransfection and interact with plasmid DNA (pDNA) togive a lipoplex, has recently attracted attention [2]. Cationicliposomes have great advantages as gene delivery carrierssuch as (1) low cytotoxicity and immunogenicity [3], (2)

regulation of the pharmacokinetics through the modificationof particle size or lipids components of liposomes [4], (3)entrapment of pDNA into inner water phase of liposomesand suppression of DNA degradation by DNase [5], and (4)delivery of gene to target cells by the addition of target ligandsand/or antibody [6].

Asialofetuin (AF) is a glycoprotein that possesses threeasparagine-linked triantennary complex carbohydrate chainswith terminal N-acetylgalactosamine residues. The proteindisplays affinity to asialoglycoprotein receptor (ASGP-R) onhepatocytes and enters the cells through the receptor [7, 8].Thus, AF has been used as a ligand to deliver drugs tohepatocytes and a competitive inhibitor to ASGP-R [9, 10].In fact, the widespread use of AF-appended liposomes (AF-liposomes) as a hepatocyte-selective gene transfer carrier hasbeen reported [11, 12].

Cyclodextrins (CyDs) have recently been applied to genetransfer and oligonucleotide delivery [13–16]. CyDs are


cyclic (α-1,4)-linked oligosaccharides of α-D-glucopyranosecontaining a hydrophobic central cavity and hydrophilicouter surface, and they are known to be able to act asnovel host molecules by chemical modification [17]. Davisand his colleagues reported that the ternary complex ofa water-soluble β-CyD polymer with 6A,6D-dideoxy-6A,-6D-di-(2-aminoethanethio)-β-CyD and dimethylsuberimi-date (βCDP6), galactosylated, or transferrin polyethyleneglycol conjugates with adamantane, and pDNA possesseshigher transfection efficiency in hepatoma or leukemiacells, respectively, through receptor-mediated endocyto-sis [18, 19]. Recently, we reported the potential use ofPAMAM dendrimer functionalized with α-CyD (α-CDE)[20] and lactosylated α-CDE (Lac-α-CDE) as a hepatocytespecific gene delivery in vitro and in vivo [21]. Mean-while, Lawrencia et al. reported that lipoplex transfectionof pDNA with DOTAP (N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium methyl-sulfate) in the presenceof cholesterol, which is solubilized by methyl-β-cyclodextrin(methyl-β-CyD), has significantly improved transfectionefficiency in urothelial cells due to change in membranefluidity by methyl-β-CyD [22]. In addition, we previouslydemonstrated that intravenous injection of the pegylatedliposomes entrapping the doxorubicin (DOX) complex withγ-CyD in BALB/c mice bearing Colon-26 tumor cellsshowed DOX accumulation in tumor tissues and the potentantitumor effect, compared with those of DOX solutionand pegylated liposomes entrapping DOX alone [23]. Theselines of evidence suggest that transfection efficiency andpharmacokinetics of pDNA can be altered by the associationof CyDs with AF-liposomes.

Based on these backgrounds, the purpose of this study isto evaluate in vitro gene delivery of AF-liposomes associatedwith CyDs as a hepatocyte-selective nonviral vector inHepG2 cells. In addition, the mechanisms by which γ-CyDenhanced transfection efficiency of pDNA/AF-liposomeswere investigated in the view of a receptor recognition,physicochemical properties (particle size, ζ-potential, andencapsulation ratio), membrane fluidity, cellular uptake, andcytotoxicity of AF-liposomes.


2.1. Materials. Dilauroylphosphatidylcholine (DLPC), diol-eoylphosphatidylethanolamine (DOPE), dipalmitoylphos-phatidylethanolamine (DPPE), and diacylphosphatidyleth-anolamine-N-lissamine rhodamine B sulfonyl (RH-PE)were obtained from Avanti Polar-Lipid (Alabama). N-(α-Trimethylammonioacetyl)-didodecyl-D-glutamate chloride(TMAG) was purchased from Sogo Pharmaceutical (Tokyo,Japan). Asialofetuin (AF) and 2-mercaptoethanol wereobtained from Sigma Chemical (St. Louis, MO). N-hydroxysulfosuccinimide (Sulfo-NHS) was purchased fromFluka (Buchs, Switzerland). 1-Ethyl-3-(3-dimethylami-nopropyl) carbodiimide (EDC) was from Dojindo (Kum-amoto, Japan). 2-(N-morpholino) ethanesulfonic acid(MES) and 2-[4-(2-hydroxyethyl)-1-piperazinyl] ethanes-ulfonic acid (HEPES) were purchased from Nacalai Tesque

Table 1: Chemical structures of CyDs used in this study.

CyDs n R D.S.(a)

α-CyD 1 H

γ-CyD 3 H

HP-α-CyD(b) 1 H or CH2CH(CH3)OH 4.0

HP-β-CyD(c) 2 H or CH2CH(CH3)OH 4.8

HP-γ-CyD(d) 3 H or CH2CH(CH3)OH 4.3

DM-β-CyD(e) 2 H or CH3 14

(a) The average degree of substitution, (b) 2-Hydroxypropyl-α-CyD, (c)2-Hydroxypropyl-β-CyD, (d) 2-Hydroxypropyl-γ-CyD, and (e) 2, 6-Di-O-methyl-β-CyD.

(Kyoto, Japan). CyDs used in this study were suppliedby Nihon Shokuhin Kako (Tokyo, Japan) (Table 1). Theaverage degrees of substitution of 2-hydroxypropyl groupin 2-hydroxypropyl-α-CyD (HP-α-CyD), HP-β-CyD, andHP-γ-CyD are 4.0, 4.8 and 4.3, respectively. Fetal calf serum(FCS) was obtained from Nichirei (Tokyo, Japan). Dulbecco’smodified Eagle’s medium (DMEM) was purchased fromNissui Pharmaceuticals (Tokyo, Japan). Plasmid pRL-CMV-Luc vector encoding Renilla luciferase and having a CMVpromoter as well as pGL3-control vector encoding fireflyluciferase and having a SV40 promoter were obtainedfrom Promega (Tokyo, Japan). pEGFP N1 DNA encodingEGFP and having a CMV promoter was purchased from BDBioscience Clontech (San Jose, CA). These DNA vectors wereabbreviated to pDNA. The purification of pDNA amplifiedin bacteria was carried out using QIAGEN EndoFreeplasmid MAXI kit (<0.1 EU/μg endotoxin). PicogreendsDNA reagent and ULYSIS Alexa Fluor 488 (Alexa)Nucleic Acid Labeling Kit were purchased from MolecularProbes (Tokyo, Japan). Bovine serum albumin (BSA) wasobtained from Roche Diagnostics (Tokyo, Japan). Otherchemicals and solvents were of analytical reagent grade, anddeionized double-distilled water was used throughout thestudy.

2.2. Preparation of AF-Liposomes. Preparation of AF-liposomes was performed according to the method reportedby Hara et al. [11] with some modifications (Figure 1).Briefly, lipids mixtures, DLPC/TMAG/DOPE/DPPE(3/2/4/1, molar ratio, the amount of total lipids was30 μmol), were dissolved in chloroform, and the solvent wasremoved under reduced pressure by a rotary evaporator.After addition of 5 mL of 0.1 M MES buffer (activationbuffer, pH 5.0) containing 0.9% NaCl, the solution wassonicated by a bath type of sonicator (ultrasonic automaticwasher US-4, AS ONE, Osaka, Japan) for 5 min undergassing with nitrogen, and then small unilamellar vesicle,large unilamellar vesicle, and/or small oligolamellarliposomes were obtained. Next, 6 mg of AF was dissolvedin 1 mL of activation buffer, which contained 3.3 mg ofSulfo-NHS and 1.2 mg of EDC. After mixing for 15 minat room temperature, the reaction was stopped by theaddition of 4.2 μL of 2-mercaptothanol and obtainedSulfo-NHS-AF. Sulfo-NHS-AF was mixed with LUV andstirred for 2 h at room temperature. After addition of


AF-liposome

NH

Sulfo-NHS-AF

NH2

Cationic liposome

OH

SO3Na

OH

Sulfo-NHS

SO3Na

O

AF

AF

AF

O ON

O ON

C O

C O

C O

(a)

0

0.4

0.8

1.2

Con

cen

trat

ion

ofin

orga

nic

phos

phor

us

(mM

)

Con

cen

trat

ion

ofA

F(m

g/m

L)

0

0.02

0.04

0.06

010 20

Fraction number

30 40 50

PhospholipidAF

(b)

Figure 1: Preparation of asialofetuin-modified liposomes (AF-liposomes). (a) Preparation pathway of AF-liposomes. (b) Elutionprofiles of AF-liposomes determined by gel filtration chromatogra-phy.

hydroxylamine HCl, AF-liposomes were obtained. Toremove the free AF, we performed gel filtration usingSepharose CL-4B column (Amersham Pharmacia Biotech,Freiburg, Germany) and determined phospholipids andAF by the Bartlett method [24] and the Bradford method[25], respectively. The fractions number 26–30, whicheluted both phospholipids and AF, were collected as theAF-liposomes fraction (Figure 1(b)). The ζ-potential valueand particle size of AF-liposomes were 31.2 ± 0.1 mV and230.7± 4.2 nm, respectively. The amount of AF modificationin 1 μmol of AF-liposome lipids was 35.8 μg/μmol lipids,indicating that AF modification rate against DPPE inliposomes was 0.75%. Liposomes without AF modification

(N-liposomes) was prepared by the solution without Sulfo-NHS-AF, and the other procedure was the same as that ofAF-liposomes.

2.3. Preparation of pDNA/CyDs/AF-Liposomes. Preparationof pDNA/CyD/AF-liposomes was performed according tothe method reported by Hara et al. with some modifications[12]. Briefly, 2 μL of the solution containing pDNA (1 μg/μL)and CyDs (1 μM/μM lipids) dissolved in TE buffer wereadded to AF- or N-liposomes suspension. After mixing, thesolution was freeze-dried. Then, the sample was rehydratedwith THBS (10 mM, pH 7.5) for 30 min. After freezing-thawing for three times, the vesicles were extruded throughPVDF membranes (Nucleopore, Plesanton, CA) with poresof diameter 450 and 200 nm. The filtrates were usedfor further experiments as pDNA/CyDs/AF-liposomes orpDNA/CyDs/N-liposomes. The entrapment ratios of CyDswere evaluated by the anthrone-sulfuric acid method [26].Briefly, 3 mL of anthrone reagent was added to 0.5 mL ofthe suspension containing CyDs/liposomes. The tube wascovered with a glass ball and was heated for 10 min in boilingwater. After quenching with cold water, absorbance of thesuspension was measured by a U-2000A spectrophotometer(Hitachi, Tokyo, Japan) at 620 nm. The encapsulation ratiosof pDNA were determined by a fluorescent spectrometer F-4500 (Hitachi, Tokyo, Japan). Briefly, 350 μL of the suspen-sion containing pDNA/CyD/AF-liposomes in 10 mM THBS(pH 7.5) were mixed with 200 times diluted PicogreendsDNA reagent (350 μL). After incubation for 30 min at25◦C, fluorescent intensity (Fpo) was determined. Next, afteraddition of 20% of Triton-X (20 μL) to the sample, fluores-cent intensity (Fpt ) was determined and the encapsulationratio of pDNA was calculated as follows: encapsulation ratio(%) = [(Fpt · r − Fpo)/Fpt · r] · 100, where r is compensationcoefficient (r = 1.03). Particle size and ζ-potential valueof liposomes in 10 mM THBS (pH 7.5) were measured bya submicron particle analyzer N4 Plus (Beckman Coulter,Fullerton, CA) and ELS-8000 (Otsuka Electronics, Osaka,Japan), respectively.

2.4. Interaction of CyDs with AF-Liposomes. AF-liposomesencapsulating calcein were prepared by the freezing andthawing method after addition of 0.1 mM calcein in 10 mMTHBS (pH 7.5). The vesicles were extruded through twostacked polycarbonate membranes (Nucleopore, Plesanton,CA) with pores of diameter 1 μm. The sample was subjectedto 10 passes through the filter at 40◦C. The filtrateswere extruded through the polycarbonate membranes (poresize 0.2 μm) as described above. Two milliliters of CyDssolution adjusted at the appropriate concentration (5–20 mM) using 10 mM phosphate buffer were added to20 μL of the liposomal suspension, and then the resultingsuspension was incubated for 30 min at 25◦C. The flu-orescence intensity of calcein (Ft) was measured with afluorophotometer (Hitachi F-4500, Tokyo, Japan) at 25◦C;excitation and emission wavelengths were 490 and 520 nm,respectively. After addition of 20 μL of cobalt chloridesolution (10 mM) to the sample to quench the fluorescence


of nonencapsulated calcein, the intensity of fluorescence ofencapsulated calcein (Fin) was also determined. Then, theliposomes were completely disrupted by the addition of20 μL of Triton X-100 (20%) solution, and the intensitiesof fluorescence after quenching by cobalt chloride (Fq) weremeasured. Calcein encapsulation ratio was calculated by theequation as follows: encapsulation ratio (%) = [(Fin − Fq ·r)/(Ft − Fq · r)] · 100, where r is compensation coefficient(r = 1.04).

2.5. Cell Culture. HepG2 cells, a human hepatocellularcarcinoma cell line, A549 cells, a adenocarcinomic humanalveolar basal epithelial cells, and NIH3T3 cells, a mouseembryonic fibroblast cell line, were obtained from RikenBioresource Center (Tsukuba, Japan). HepG2, A549, andNIH3T3 cells were grown in DMEM, containing 1× 105 U/Lof penicillin, 0.1 g/L of streptomycin supplemented with10% FCS at 37◦C in a humidified 5% CO2 and 95% airatmosphere.

2.6. In Vitro Gene Transfer. In vitro transfection of thepDNA/CyDs/AF-liposomes was performed utilizing theluciferase expression of pDNA (pRL-CMV-Luc or pGL3-control vector) in HepG2, A549, and NIH3T3 cells. Thecells (2 × 105 cells per 24 well plate) were seeded 6 hbefore transfection and then washed twice with 500 μL ofserum-free medium. Two hundred μL of serum-free mediumcontaining pDNA/CyDs/AF-liposomes in the absence andpresence of AF as a competitor protein or BSA as a controlprotein were added to each dish and then incubated at37◦C for 3 h. After washing HepG2 cells with serum-freemedium twice, 500 μL of medium containing 10% FCS wereadded to each dish and then incubated at 37◦C for 21 h.After transfection, the gene expression was measured asfollows: Renilla and firefly luciferase contents in the celllysate were quantified by a luminometer (Lumat LB9506,EG&G Berthold Japan, Tokyo, Japan) using the PromegaRenilla and firefly luciferase assay reagent (Tokyo, Japan),respectively. It was confirmed that CyDs and AF-liposomeshad no influence on the luciferase assays under the presentexperimental conditions. Total protein content of the super-natant was determined by Bio-Rad protein assay kit (Bio-Rad Laboratories, Tokyo, Japan). EGFP-expressing cells weredetermined by a confocal laser scanning microscopy (CLSM,Olympus FV300-BXCarl Zeiss LSM-410, Tokyo, Japan) withan argon laser at 488 nm after fixation. Briefly, the cells(2 × 105 cells per 35 mm glass bottom dish) were seeded6 h before transfection and then washed twice with 500 μLof serum-free medium. Transfection with pEGFP N1 DNAwas performed using the same protocol as described above.The EGFP expression ratio was determined by the numberof EGFP-expressing cells per 100 cells. To observe the cel-lular uptake of Rhodamine-labeled AF-liposomes (RH-AF-liposomes) and Alexa-labeled pDNA (Alexa-pDNA), HepG2cells (2 × 105 cells/dish) were incubated with the Alexa-pDNA/CyDs/RH-AF-liposomes for 3 h. After incubation,the cells were rinsed with PBS (pH 7.4) twice and fixedin methanol at 4◦C for 5 min prior to observation bya CLSM.

2.7. Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR). Total RNA was isolated using an RNeasy Mini Kit(Qiagen, Tokyo, Japan) according to manufacturer’s proce-dure. The synthesis of the first-strand cDNA was carriedout with SuperScript III reverse transcriptase (Invitrogen,Carlsbad, CA). Approximately, 1 μM random primer wasannealed to 3 μg of total RNA and extended with 1 μLof reverse transcriptase in 10 μL of reaction containing4 μL of 5 × first-strand buffer, 1 μL of deoxyribonucleotidetriphosphate (dNTPs), and 1 μL of dithiothreitol. Reversetranscription was carried out at 42◦C for 50 min. Theexpression of mRNA transcripts of Renilla luciferase (for-ward: 5′-GGCTGACCGCCCAACGACCCCC-3′, reverse: 5′-GACGTCAATAGGGGGCGGACTTGG-3′) and human β-actin (forward: 5′- TCCTGTGGCATCCATCCACGAAACT-3′, reverse: 5′-GAAGCATTTGCGGTGGACGAT-3′) wasdetermined by RT-PCR. PCR amplification was carried outin a PCR Thermal Cycler (Takara Bio, Shiga, Japan). PCRwas conducted in a total volume of 100 μL with 2 μL of thecDNA solution, 2 μL of each 10 mM dNTP, 2.5 U of TaKaRaEx Taq DNA polymerase, and 500 nM of both forward andreverse primers. The thermal cycling conditions were set to95◦C for 11 min, followed by 20 cycles of amplification at94◦C for 1 min, 55◦C for 1 min, and 72◦C for 2 min fordenaturing, annealing, and extension. After the last cycle,the samples were incubated at 72◦C for 7 min. The amplifiedproducts were separated on 2% agarose gels by electrophore-sis and visualized with 0.1% ethidium bromide under UVlight.

2.8. Cytotoxicity. The effects of pDNA/CyDs/AF-liposomeson cell viability were measured as reported previously [27].The transfection was performed as described in the trans-fection section. After washing twice with Hanks’ balancedsalt solutions (HBSS, pH 7.4) to remove pDNA and/orAF-liposomes, 270 μL of fresh HBSS and 30 μL of WST-1reagent were added to the plates and incubated at 37◦C for30 min. The absorbance of the solution was measured at450 nm, with referring absorbance at 655 nm, with a Bio-RadModel 550 microplate reader (Bio-Rad Laboratories, Tokyo,Japan).

2.9. Membrane Fluidity of Liposomes. To evaluate the ther-modynamic characterization of liposomes in the pres-ence of CyDs, differential scanning calorimetry (DSC)measurements were performed. Briefly, N-liposome orDLPC-liposome suspension (5 μmol/mL of lipids) andCyDs (10–50 mM) solution were mixed, and the solutionwas analyzed by a Microcal MC2 scanning calorimeter(Northampton, MA) with a scanning rate of 1◦C/min inthe range of 4–70◦C. From the thermographs, membranephase-inversion temperature and its enthalpy (ΔHcal) werecalculated.

2.10. Statistical Analysis. Data are given as the mean ± SEM.Statistical significance of means for the studies was deter-mined by analysis of variance followed by Scheffe’s test. Pvalues for significance were set at .05.


20

40

60

80

100

Cal

cein

leak

aged

(%)

0 5Concentration of CyDs (mM)

10 15 20

α-CyDDM-β-CyDγ-CyD

HP-α-CyDHP-β-CyDHP-γ-CyD

Figure 2: Effects of CyDs on leakage of calcein from calcein-encapsulated AF-liposomes in isotonic Tris-HCl-buffered saline(10 mM, pH 7.5) at 25◦C. Calcein-encapsulated AF-liposomes wereincubated with various CyDs for 30 min, and the concentrationsof calcein released from AF-liposomes were determined using afluorospectrophotometer. Each point represents the mean ± SEMof 3 experiments.

2

4

6

8

Ren

illa

luci

fera

seac

tivi

ty(f

old)

∗

0

Wit

hou

tC

yD

HP-α

-CyD

HP-β

-CyD

γ-C

yD

HP-γ

-CyD

Figure 3: Gene transfer activity of pDNA/CyDs/AF-liposomesin HepG2 cells. pDNA was pRL-CMV. The charge ratio of AF-liposomes/pDNA was 1.6. CyDs were added to AF-liposomessuspension before freeze-drying. The concentration of CyD was1 μM/μM lipids. Cells were incubated with pDNA/γ-CyDs/AF-liposomes for 3 h in FCS-free medium. After washing twice, thecells were incubated for 21 h in culture medium supplemented with10% FCS. The luciferase activity in cell lysates was determinedusing a luminometer. Each value represents the mean ± SEM of 3–7 experiments. ∗P < .05 versus without CyD.

Renilla luciferasemRNA

β-actin mRNA

Wit

hou

tC

yD

Wit

hγ

-CyD

Wit

hH

P-γ

-CyD

(a)

100

200

300

Ren

illa

luci

fera

sem

RN

Ale

vel(

%of

con

trol

)

∗

0

Wit

hou

tC

yD

γ-C

yD

HP-γ

-CyD

(b)

Figure 4: Effect of γ-CyDs on Renilla luciferase mRNA levelsafter transfection of pDNA/AF-liposomes or pDNA/γ-CyDs/AF-liposomes in HepG2 cells. The charge ratio of AF-liposomes/pDNAwas 1.6. CyDs were added to AF-liposomes suspension beforefreeze-drying. The concentration of CyD was 1 μM/μM lipids. Cellswere incubated with pDNA/γ-CyDs/AF-liposomes for 3 h in FCS-free medium. After washing twice, the cells were incubated for21 h in culture medium supplemented with 10% FCS. The Renillaluciferase mRNA level in HepG2 cells was assayed by RT-PCR. Theluciferase mRNA level in each sample was normalized to abundanceof β-actin mRNA. Each value represents the mean ± SEM of 3–7experiments. ∗P < .05 versus without CyD.

3. Results

3.1. Interaction between AF-Liposomes and CyDs. CyDs havebeen reported to interact with cell membrane constituentssuch as cholesterol and phospholipids, resulting in the induc-tion of hemolysis of human and rabbit red blood cells athigh concentrations of CyDs [28–30]. In addition, CyDs arewell known to disrupt liposomal membranes, depending onCyD cavity sizes and membrane components [31, 32]. Then,we evaluated the interaction of CyDs with AF-liposomes


pEGFP-N1

DIC

Without CyD γ-CyD HP-γ-CyD

(a)

5

10

15

Fire

fly

luci

fera

seac

tivi

ty(R

LU/μ

gpr

otei

n)

∗

×104

0

Wit

hou

tC

yD

γ-C

yD

HP-γ

-CyD

(b)

Figure 5: Gene transfer efficiency of pDNA/AF-liposomes orpDNA/γ-CyDs/AF-liposomes in HepG2 cells. pDNA was (a)pEGFP-N1 or (b) pGL3-control vector. The charge ratio of AF-liposomes/pDNA was 1.6. γ-CyDs were added to AF-liposomessuspension before freeze-drying. The concentration of CyDs was1 μM/μM lipids. Cells were incubated with pDNA/γ-CyDs/AF-liposomes for 3 h in FCS-free medium. After washing twice, thecells were incubated for 21 h in culture medium supplemented with10% FCS. (a) Cells were determined by a confocal laser scanningmicroscopy. The percentage in parenthesis represents frequency rateof EGFP expression cells. (b) The luciferase activity in cell lysateswas determined using a luminometer. Each value represents themean ± SEM of 3–7 experiments. ∗P < .05 versus without CyD.

by measuring the leakage of calcein, a fluorescent marker,from calcein-encapsulated AF-liposomes in isotonic Tris-HCl-buffered saline (pH 7.5) (Figure 2). α-CyD and DM-β-CyD significantly increased calcein leakage from calcein-encapsulated AF-liposomes after incubation for 30 min ina concentration-dependent manner. On the other hand,calcein leakage in the presence of γ-CyD and three types ofHP-CyDs was low even at the concentration of 20 mM CyDs.

0.5

1

1.5

Luci

fera

seac

tivi

ty(f

old)

0Control

∗

+ AF + BSA

N-liposomeAF-liposome

Figure 6: Effects of competitors on gene transfer activity ofpDNA/γ-CyD/N-liposomes and AF-liposomes in HepG2 cells.pDNA was pRL-CMV. The charge ratio of AF-liposomes/pDNA was1.6. γ-CyD was added to AF-liposomes suspension before freeze-drying. The concentration of γ-CyD was 1 μM/μM lipids. Cellswere incubated with pDNA/γ-CyDs/AF-liposomes for 3 h in FCS-free medium. After washing twice, the cells were incubated for 21 hin culture medium supplemented with 10% FCS. The luciferaseactivity in cell lysates was determined using a luminometer. Theconcentrations of AF and BSA were 5 mg/mL. Each value representsthe mean ± SEM of 3–7 experiments. ∗P < .05 versus control.

These results suggest that the interaction of γ-CyD and HP-CyDs with AF-liposomes was weaker than that of α-CyD andDM-β-CyD.

3.2. In Vitro Gene Delivery of AF-Liposomes Associated withCyDs. Next, we evaluated in vitro gene transfer activity ofpDNA/CyDs/AF-liposomes in HepG2 cells, ASGP-R positivecells (Figure 3). Here, we used pRL-CMV (CMV promoter)as pDNA and the charge ratio (AF-liposomes/pDNA) of 1.6optimized by our previous study (data not shown). The genetransfer activity of pDNA/γ-CyD/AF-liposomes in HepG2cells was significantly higher than that of pDNA/HP-α-,HP-β-, and HP-γ-CyDs/AF-liposomes (Figure 3). Next, wemeasured the Renilla luciferase mRNA level after transfectionof pDNA/CyDs/AF-liposomes in HepG2 cells by the RT-PCR method (Figure 4). As predicted, γ-CyD significantlyincreased the luciferase expression, but HP-γ-CyD did not,suggesting that γ-CyD is involved in the enhancing effect onluciferase expression at or prior to a transcription process.

To evaluate the enhancing effects of γ-CyD on genetransfer activity of AF-liposomes associating pDNA encodingEGFP controlled by a CMV or SV40 promoter, we examinedEGFP and firefly luciferase gene expression after transfectionof pDNA/γ-CyDs/AF-liposomes in HepG2 cells (Figure 5).


2

4

6

8

10

Luci

fera

seac

tivi

ty(f

old)

∗

0

Wit

hou

tC

yD

γ-C

yD

HP-γ

-CyD

(a) A549 cells

2

4

6

8

10

Luci

fera

seac

tivi

ty(f

old)

∗

0

Wit

hou

tC

yD

γ-C

yD

HP-γ

-CyD

(b) NIH3T3 cells

Figure 7: Gene transfer activity of pDNA/CyDs/AF-liposomes in (a) A549 cells and (b) NIH3T3 cells. pDNA was pRL-CMV. The chargeratio of AF-liposomes/pDNA was 1.6. CyDs were added to AF-liposomes suspension before freeze-drying. The concentrations of γ-CyDswere 1 μM/μM lipids. Cells were incubated with pDNA/γ-CyDs/AF-liposomes for 3 h in FCS-free medium. After washing twice, the cellswere incubated for 21 h in culture medium supplemented with 10% FCS. The luciferase activity in cell lysates was determined using aluminometer. Each value represents the mean ± SEM of 3–7 experiments. ∗P < .05 versus without CyD.

Table 2: ζ-potential value and particle size of pDNA/AF-liposomesor pDNA/γ-CyDs/AF-liposomes.

CyDParticle size ζ-potential

(nm) (mV)

Without CyD 357.9 ± 17.7 21.5 ± 2.9

γ-CyD 277.9 ± 10.5∗ 17.0 ± 2.7

HP-γ-CyD 348.0 ± 16.5 22.3 ± 3.3

The charge ratio of AF-liposome/pDNA was 1.6. γ-CyDs were added to AF-liposome suspension before freeze-drying. The concentrations of γ-CyDswere 1 μM/μM lipids. The ζ-potential was measured by a light-scatteringmethod. The particle size was determined using a photon correlationspectroscopic analyzer. Each value represents the mean ± SEM of 3experiments. ∗P < .05 versus without CyD.

The extent of EGFP-expressing cells in the pDNA/AF-liposomes system without CyDs was found to be 8%,while that with γ-CyD and HP-γ-CyD were 19% and10%, respectively (Figure 5(a)). Additionally, the enhancingeffect of γ-CyD was observed in the pGL3-control vectorencoding firefly luciferase and having a SV40 promoter(Figure 5(b)). These results suggest that the enhancing effectof γ-CyD on gene transfer activity of AF-liposomes is a gene-and promoter-independent manner. To confirm whetherpDNA/γ-CyD/AF-liposomes have ASGP-R-mediated genetransfer activity, we performed transfection experiments in

HepG2 cells in the presence and absence of AF, as an ASGP-R competitive inhibitor. Here, we confirmed that ASGP-R areexpressed in HepG2 cells by the RT-PCR method (data notshown), which is consistent with previous findings [33–35].As shown in Figure 6, gene transfer activity of AF-liposomeswas markedly inhibited by the addition of AF, but not BSA, acontrol protein. These results suggest that AF-liposomes hadthe ASGP-R-mediated gene transfer activity. Interestingly,the similar enhancing effects of γ-CyD on Renilla luciferaseprotein expression after transfection of pDNA/AF-liposomeswere, however, observed in A549 cells and NIH3T3 cells,ASGP-R negative cells (Figure 7). These results suggest thatγ-CyD enhances the transfection efficiency of pDNA/AF-liposomes in ASGP-R-independent manner.

3.3. Cytotoxicity. To reveal cytotoxicity of pDNA/CyDs/AF-liposomes, we examined the WST-1 method (Figure 8).Although the cell viability after treatment with pDNA/AF-liposomes and pDNA/CyDs/AF-liposomes for 3 h slightlydecreased as a charge ratio of AF-liposomes/pDNA wasincreased in both HepG2 and NIH3T3 cells, that is,more than 80% of cell viability after application ofpDNA/CyDs/AF-liposomes was observed at the charge ratioof 1.6 used in the transfection study as described above.These results suggest that pDNA/CyDs/AF-liposomes havegreat advantages as a nonviral vector, that is, superior trans-fection efficiency and less cytotoxicity, and the enhancing


50

100

Cel

lvia

bilit

y(%

)

00

1

Charge ratio of AF-liposome/pDNA

2 3

(a) HepG2 cells

50

100

Cel

lvia

bilit

y(%

)

00

1

Charge ratio of AF-liposome/pDNA

2 3

Without CyDγ-CyDHP-γ-CyD

(b) NIH3T3 cells

Figure 8: Cytotoxicity of pDNA/AF-liposomes or pDNA/γ-CyDs/AF-liposomes with various charge ratios in (a) HepG2 and(b) NIH3T3 cells. The concentrations of γ-CyDs were 1 μM/μMlipids. Cells were incubated with pDNA/γ-CyDs/AF-liposomes for3 h in FCS-free medium. Cell viability was assayed by the WST-1method. Each point represents the mean ± SEM of 3 experiments.

effect of γ-CyD on gene transfer activity of pDNA/AF-liposomes is not associated with cytotoxicity.

3.4. Effects of γ-CyD on Physicochemical Properties ofpDNA/AF-Liposomes. To clarify physicochemical propertiesof the pDNA/AF-liposomes, we determined the parti-cle sizes and ζ-potential values of the pDNA/γ-CyD/AF-liposomes at the charge ratio of 1.6. The mean diameterof the pDNA/γ-CyD/AF-liposomes was smaller than thatof pDNA/AF-liposomes or pDNA/HP-γ-CyD/AF-liposomes(Table 2). Meanwhile, the ζ-potential values of pDNA/AF-liposomes, pDNA/γ-CyD/AF-liposomes, and pDNA/HP-γ-CyD/AF-liposomes were almost comparable (Table 2). Theseresults indicate that γ-CyD reduced the particle size ofAF-liposomes but did not change the ζ-potential value ofpDNA/CyD/AF-liposomes.

Table 3: Encapsulation ratios of pDNA and γ-CyDs into AF-liposomes.

CyD Encapsulation ratio ofpDNA (%)

Encapsulation ratio ofγ-CyDs (%)

Without CyD 42.4 ± 4.3 —

γ-CyD 58.2 ± 1.8∗ 10.2 ± 0.5

HP-γ-CyD 41.9 ± 3.3 11.0 ± 2.2

The charge ratio of AF-liposome/pDNA was 1.6. γ-CyDs were added to AF-liposome suspension before freeze-drying. The concentrations of γ-CyDswere 1 μM/μM lipids. The encapsulation ratios of pDNA were determinedusing Picogreen assay. The encapsulation ratios of γ-CyDs were determinedby an anthrone-sulfuric acid method. Each value represents the mean ±SEM of 3 experiments. ∗P < .05 versus without CyD.

Next, we examined the effects of γ-CyDs on encapsu-lation ratios of pDNA into AF-liposomes (Table 3). Theencapsulation ratio of pDNA in pDNA/γ-CyD/AF-liposomeswas significantly higher than those of pDNA/HP-γ-CyD/AF-liposomes and pDNA/AF-liposomes. Meanwhile, the encap-sulation ratios of γ-CyD and HP-γ-CyD into AF-liposomeswere approximately 10.2% and 11.0%, respectively. Theseresults suggest that γ-CyD improves the encapsulation ofpDNA into AF-liposomes, although the extent of γ-CyDencapsulation into AF-liposomes is not high. In addition,both the encapsulation ratio of pDNA into AF-liposomesand gene transfer activity of pDNA/AF-liposomes wereraised, as the number of the freeze-thaw cycle was increased,suggesting that the encapsulation of pDNA in AF-liposomesis correlated with the gene transfer activity of pDNA/AF-liposomes (Table 4).

3.5. Effects of γ-CyD on Membrane Fluidity of Liposomes. Itis known that membrane fluidity of liposomes affects therelease profiles as well as a retention time of drug encapsu-lated into liposomes. Therefore, we investigated the effectsof γ-CyD on membrane fluidity of AF-liposomes using aMicrocal MC2 scanning calorimeter. In the present study,we utilized N-liposomes to eliminate the effects of the heatdegeneration of AF in a DSC thermograph. Figure 9 showsthe effects of γ-CyD and HP-γ-CyD on DSC thermograms ofN-liposomes (5 mM of total lipids). The peak derived fromgel-to-fluid state transition in N-liposomes was observed at55◦C, while new peak was appeared at 42◦C in the presenceof 50 mM of γ-CyD (Figure 9(a)). On the other hand,no significant change in DSC thermographs was observedin the presence of HP-γ-CyD (Figure 9(b)). These resultssuggest that γ-CyD may affect the membrane fluidity of N-liposomes.

Generally, membrane fluidity and phase transition ofliposomes are determined by the intensity of lipid-lipidinteractions such as hydrophobic interaction, van der Waalsforces, and hydrogen bond. Therefore, to evaluate the effectsof γ-CyD on lipid-lipid interactions in AF-liposomes, we per-formed DSC analysis of DLPC-liposomes in the presence andabsence of γ-CyD. The reason why we used DLPC-liposomesis due to clear observation of lipid-lipid interaction usingDLPC composed of AF-liposomes. Figure 10 shows the


Exc

ess

hea

tca

paci

ty

Temperature (◦C)

10 20 30 40 50 60 70

0 mM

20 mM

50 mM

(a) γ-CyD

Exc

ess

hea

tca

paci

ty

Temperature (◦C)

10 20 30 40 50 60 70

0 mM

20 mM

50 mM

(b) HP-γ-CyD

Figure 9: DSC thermograms of the gel-to-fluid state transition of N-liposomes at various concentrations of γ-CyDs. γ-CyDs were added toN-liposomes suspension. The concentration of total lipids of N-liposomes was 5 mM. The experiments were performed in the range of 4◦Cto 70◦C using a Microcal MC2 apparatus. The temperature scanning rate was 1◦C/min.

Table 4: Encapsulation ratio of pDNA into AF-liposomes and genetransfer activity of pDNA/AF-liposomes at various cycles of freeze-thaw in HepG2 cells.

Freeze thawcycle

Encapsulation ratio ofpDNA (%)

Gene transfer activity(fold)

0 35.2 ± 0.9 0.56 ± 0.07

1 36.2 ± 0.5 0.59 ± 0.06

2 37.4 ± 2.1 0.76 ± 0.02

3 42.4 ± 4.3 1

5 44.1 ± 1.3 1.03 ± 0.02

The charge ratio of AF-liposomes/pDNA was 1.6. pDNA/AF-liposome wasprepared using a freeze-thaw method which was repeated from 0 to 5 cycles.Cells were incubated with pDNA/γ-CyDs/AF-liposome for 3 h in FCS-freemedium. After washing twice, the cells were incubated for 21 h in culturemedium supplemented with 10% FCS. The luciferase activity in cell lysateswas determined using a luminometer. Each value represents the mean ±SEM of 3 experiments.

effects of γ-CyD and HP-γ-CyD on DSC thermograms, phasetransition temperature, and enthalpy (ΔHcal) of DLPC-liposomes (5 mM of total lipids). The phase transitiontemperature (Tc) of DLPC-liposomes in the presence of γ-CyD was shifted to high temperature as the concentrationof γ-CyD was increased (Figures 10(a) and 10(c)). TheΔHcal value of DLPC-liposomes in the γ-CyD system wasdrastically elevated at 20 mM of γ-CyD (Figure 10(d)).Meanwhile, in the case of HP-γ-CyD, there was no significantchange in DSC thermograms, phase transition temperature,and the ΔHcal values (Figures 10(b), 10(c) and 10(d)). Takentogether, these results strongly suggest that γ-CyD enhancesthe lipid-lipid interaction of DLPC-liposomes, leading to themembrane stabilization of DLPC-liposomes.

3.6. Cellular Uptake of pDNA/AF-Liposomes. Next, we exam-ined the cellular uptake of pDNA/γ-CyD/AF-liposomes intoHepG2 cells using a CLSM. Figure 11 shows the CLSMimages for distribution of RH-AF-liposomes and Alexa-pDNA in HepG2 cells at 3 h after transfection. The strongfluorescence derived from RH-AF-liposomes and Alexa-pDNA in the presence of γ-CyD was mainly observed incytoplasm of HepG2 cells. Meanwhile, the fluorescence RH-AF-liposomes and Alexa-pDNA in the absence of CyD andwith HP-γ-CyD was mainly observed on cell surface. Hence,these results suggest that pDNA/γ-CyD/AF-liposomes can beinternalized into HepG2 cells to a larger extent, compared topDNA/AF-liposomes and pDNA/γ-CyD/AF-liposomes.

4. Discussion

In this study, we clarified that pDNA/γ-CyD/AF-liposomeshave potent hepatocyte-selective gene transfer activity andnegligible cytotoxicity, compared to pDNA/AF-liposomesand pDNA/HP-CyDs/AF-liposomes.

In cationic liposome-mediated gene transfection, lipidcomposition and lipid type are the most important physico-chemical factors, because they affect not only the interactionwith pDNA but also the affinity to target cells [36, 37].In the present study, we prepared AF-liposomes with alipid composition of TMAG/DOPE/DLPC/DPPE (2/4/3/1,molar ratio). DOPE is known to enhance the endosomalescape of pDNA due to its structural change into hexagonalII form at pH 5-6, an endosomal pH range, resulting indestabilizing endosomal membranes [38, 39]. TMAG, acationic lipid, makes it possible to interact with pDNA inAF-liposomes. Additionally, DPPE was used as a bindinglipid with AF. Actually, cationic liposomes composed of


Exc

ess

hea

tca

paci

ty

Temperature (◦C)

10 20 30 40 50 60 70

0 mM

20 mM

27.5 mM

35 mM

50 mM

(a) γ-CyD

Exc

ess

hea

tca

paci

ty

Temperature (◦C)

10 20 30 40 50 60 70

0 mM

20 mM

50 mM

(b) HP-γ-CyD

Gel

-to-

flu

idst

ate

tran

siti

onte

mp

erat

ure

(◦C

)

0

20

30

40

50

10

Concentration of γ-CyDs (mM)

10 20 30 40 50

γ-CyDHP-γ-CyD

(c)

ΔH

cal(

kcal

/mol

)

0

80

120

40

Concentration of γ-CyDs (mM)

10 20 30 40 50

γ-CyDHP-γ-CyD

(d)

Figure 10: DSC thermograms (a, b), gel-to-fluid state transition temperature (c), and ΔHcal value (d) of the gel-to-fluid state transition ofDLPC-liposomes at various concentrations of γ-CyDs. γ-CyDs were added to DLPC-liposomes suspension. The DLPC concentration was5 mM. The experiments were performed in the range of 4◦C to 70◦C using a Microcal MC2 apparatus. The temperature scanning rate was1◦C/min. The gel-fluid state transition temperature and ΔHcal were taken at a peak temperature and a peak area of the DSC thermograms,respectively. Each point represents the mean ± SEM of 3 experiments.

TMAG/DOPE/DLPC (1/2/2, molar ratio) are commerciallyavailable transfection reagents as GeneTransfer, which havealready been utilized in a clinical trial for a nonviral vectorto deliver the interferon-β gene for the treatment of braintumor in Japan [40]. Therefore, we used AF-liposomescomposed of these lipids in the present study.

The most important finding found in the present studyis that γ-CyD enhances transfection efficiency of pDNA/AF-liposomes in HepG2 cells (Figures 3–5) with negligiblecytotoxicity (Figure 8). The enhancing mechanisms of γ-CyD presumed are discussed as follows.

In the present study, we revealed that transfection effi-ciency of the pDNA/γ-CyD/AF-liposomes, not N-liposomes,

was inhibited by the addition of AF in HepG2 cells (Figure 6).Meanwhile, in NIH3T3 cells, transfection efficiency of thepDNA/AF-liposomes was not suppressed by the addition ofAF. These results strongly suggest that pDNA/γ-CyD/AF-liposomes can be entered HepG2 cells through ASGP-R-mediated endocytosis, consistent with Aramaki and hiscolleague’s report [41], and the enhancing effect of the γ-CyDmay by associated with the ASGP-R-mediated endocytosis.However, γ-CyD also enhanced gene transfer activity ofpDNA/AF-liposomes even in A549 cells and NIH3T3 cells,ASGP-R negative cells (Figure 7). These results suggest thatthe enhancing effect of γ-CyD on transfection efficiency ofpDNA/AF-liposomes is in an ASGP-R-independent manner.


RH-AF-liposome

Without CyD

(a)

γ-CyD

(b)

HP-γ-CyD

(c)

Alexa-pDNA

(d) (e) (f)

Merge

(g) (h) (i)

Figure 11: Confocal laser microscopic images for distribution of RH-AF-liposomes and Alexa-pDNA in HepG2 cells. Magnification: ×200.The charge ratio of AF-liposomes/pDNA was 1.6. γ-CyDs were added to AF-liposomes suspension before freeze-drying. The concentrationsof γ-CyDs were 1 μM/μM lipids. Cells were incubated with Alexa-pDNA/γ-CyDs/RH-AF-liposomes for 3 h in FCS-free medium. Afterwashing twice, the cells were observed using a confocal laser scanning microscopy.

The particle sizes and encapsulation ratio of pDNA/γ-CyD/AF-liposomes should be involved in the enhanc-ing effect of γ-CyD on gene transfer activity of AF-liposomes. The particle size of pDNA/γ-CyD/AF-liposomeswas decreased in the presence of γ-CyD, although the ζ-potential value of pDNA/γ-CyD/AF-liposomes was almostequivalent to that of pDNA/AF-liposomes and pDNA/HP-γ-CyD/AF-liposomes (Table 2), suggesting that γ-CyD inhibitsthe aggregation of pDNA/AF-liposomes, because the particlesize shows more than 200 nm, despite the fact that theliposomes were extruded through a filter membrane havinga pore size of 200 nm. Meanwhile, the encapsulation ratioof pDNA was significantly increased by adding γ-CyD topDNA/AF-liposomes (Table 3). Thus, these lines of evidencespeculate that addition of γ-CyD enhances cellular uptake ofpDNA/AF-liposomes. In fact, the CLSM study demonstrated

that cellular uptake of pDNA/γ-CyD/AF-liposomes washigher than that of pDNA/γ-CyD/AF-liposomes (Figure 11).Furthermore, we confirmed that transfection efficiency ofpDNA/γ-CyD/AF-liposomes was increased, as the encapsu-lation ratio of pDNA into pDNA/γ-CyD/AF-liposomes wasaugmented (Table 4). In view of the findings, the particle sizeof pDNA/γ-CyD/AF-liposomes and encapsulation ratio ofpDNA into pDNA/γ-CyD/AF-liposomes are crucial role forenhancing transfection efficiency of pDNA/AF-liposomes.

The important question regarding the enhancing effectof γ-CyD on transfection efficiency of pDNA/AF-liposomesstill remains, because three types of HP-CyDs did not havethe enhancing effect. To address this question, the DSCanalysis was performed. This study indicated that γ-CyD, butnot HP-γ-CyD, changed membrane fluidity and stabilizedthe N-liposomal membranes (Figure 9). In fact, γ-CyD


increased the Tc value of DLPC-liposomes, although HP-γ-CyD did not increase anymore (Figure 10). This increasein the Tc value induced by γ-CyD could be attributed to acompactness of lipid bilayer of DLPC liposomes. It is therebypossible that γ-CyD may increase Tc values of the otherliposomes such as DMPC, DPPC, and DSPC. Here, it is wellknown that γ-CyD is highly hydrophilic and surface inactive[17]. Therefore, we presumed that γ-CyD encapsulates theaqueous compartment of liposomes rather than in the bilayerof liposomes. Anyhow, it is clear that the magnification ofthe interaction of liposomal membranes containing DLPCwith HP-γ-CyD is weaker than that with γ-CyD, possiblydue to the steric hindrance of the HP group in a HP-γ-CyDmolecule. Taken together, it is likely that the stabilizing effectsof γ-CyD on AF-liposomal membrane may lead to inhibitionof pDNA leakage from AF-liposomes and increase in cellularuptake of pDNA, eventually leading to the enhancement of invitro transfection efficiency of pDNA/γ-CyD/AF-liposomesin cells.

Finally, we investigated the role of free γ-CyD on trans-fection efficiency of pDNA/γ-CyD/AF-liposomes in HepG2cells. The physical mixture of pDNA/AF-liposomes and γ-CyD in culture medium had no enhancing effect on trans-fection efficiency of pDNA/AF-liposomes (data not shown).Therefore, encapsulation of γ-CyD into AF-liposomes maybe pivotal for enhancing gene transfer activity. To revealthe detailed mechanism for the enhancing effect of γ-CyD associated in AF-liposomes on transfection efficiencyof pDNA/AF-liposomes, further elaborate study should benecessary.

5. Conclusions

In the present study, we demonstrated that γ-CyD enhancedgene transfer activity of pDNA/AF-liposomes in not onlyHepG2 cells but also A549 and NIH3T3 cells, probably due tovarious effects of γ-CyD on AF-liposomes such as inhibitionof aggregation of the liposomes, high encapsulation of pDNAinto the liposomes, and stabilization of lipid bilayer of theliposomes. Consequently, the potential use of γ-CyD couldbe expected as an enhancer of gene transfer activity of AF-liposomes. Also, these data may be useful for design of cell-specific cationic liposomes as a nonviral vector.

References

[1] S. E. Raper and J. M. Wilson, “Gene therapy for human liverdisease,” Progress in Liver Diseases, vol. 13, pp. 201–230, 1995.

[2] P. L. Felgner, T. R. Gadek, M. Holm et al., “Lipofection: ahighly efficient, lipid-mediated DNA-transfection procedure,”Proceedings of the National Academy of Sciences of the UnitedStates of America, vol. 84, no. 21, pp. 7413–7417, 1987.

[3] R. M. Abra, C. A. Hunt, and D. T. Lau, “Liposome dispositionin vivo VI: delivery to the lung,” Journal of PharmaceuticalSciences, vol. 73, no. 2, pp. 203–206, 1984.

[4] T. M. Allen and J. M. Everest, “Effect of liposome size anddrug release properties on pharmacokinetics of encapsulateddrug in rats,” Journal of Pharmacology and ExperimentalTherapeutics, vol. 226, no. 2, pp. 539–544, 1983.

[5] R. J. Mannino, E. S. Allebach, and W. A. Strohl, “Encapsu-lation of high molecular weight DNA in large unilamellarphospholipid vesicles. Dependence on the size of the DNA,”FEBS Letters, vol. 101, no. 2, pp. 229–232, 1979.

[6] N. Garcon, G. Gregoriadis, M. Taylor, and J. Summerfield,“Mannose-mediated targeted immunoadjuvant action of lipo-somes,” Immunology, vol. 64, no. 4, pp. 743–745, 1988.

[7] Y. Kamimoto, D. Tanabe, S. Tashiro, T. Hiraoka, and Y.Miyauchi, “Changes in receptor-mediated endocytosis in liversinusoidal cells after partial hepatectomy in the rat,” Liver, vol.14, no. 3, pp. 141–147, 1994.

[8] L. Roseng, H. Tolleshaug, and T. Berg, “Uptake, intracellulartransport, and degradation of polyethylene glycol- modifiedasialofetuin in hepatocytes,” The Journal of Biological Chem-istry, vol. 267, no. 32, pp. 22987–22993, 1992.

[9] S. Becker, M. Spiess, and H.-D. Klenk, “The asialoglycoproteinreceptor is a potential liver-specific receptor for Marburgvirus,” Journal of General Virology, vol. 76, no. 2, pp. 393–399,1995.

[10] A. G. Y. Chang and G. Y. Wu, “Gene therapy: applications tothe treatment of gastrointestinal and liver diseases,” Gastroen-terology, vol. 106, no. 4, pp. 1076–1084, 1994.

[11] T. Hara, H. Ishihara, Y. Aramaki, and S. Tsuchiya, “Specificuptake of asialofetuin-labeled liposomes by isolated hepato-cytes,” International Journal of Pharmaceutics, vol. 42, no. 1-3,pp. 69–75, 1988.

[12] T. Hara, H. Kuwasawa, Y. Aramaki et al., “Effects of fusogenicand DNA-binding amphiphilic compounds on the receptor-mediated gene transfer into hepatic cells by asialofetuin-labeled liposomes,” Biochimica et Biophysica Acta, vol. 1278,no. 1, pp. 51–58, 1996.

[13] S. Abdou, J. Collomb, F. Sallas, A. Marsura, and C. Finance,“Beta-cyclodextrin derivatives as carriers to enhance theantiviral activity of an antisense oligonucleotide directedtoward a coronavirus intergenic consensus sequence,” Archivesof Virology, vol. 142, no. 8, pp. 1585–1602, 1997.

[14] M. A. Croyle, B. J. Roessler, C.-P. Hsu, R. Sun, and G. L.Amidon, “Beta cyclodextrins enhance adenoviral-mediatedgene delivery to the intestine,” Pharmaceutical Research, vol.15, no. 9, pp. 1348–1355, 1998.

[15] M. E. Davis and M. E. Brewster, “Cyclodextrin-based phar-maceutics: past, present and future,” Nature Reviews DrugDiscovery, vol. 3, no. 12, pp. 1023–1035, 2004.

[16] B. J. Roessler, A. U. Bielinska, K. Janczak, I. Lee, andJ. R. Baker Jr., “Substituted β-cyclodextrins interact withPAMAM dendrimer-DNA complexes and modify transfectionefficiency,” Biochemical and Biophysical Research Communica-tions, vol. 283, no. 1, pp. 124–129, 2001.

[17] K. Uekama, F. Hirayama, and T. Irie, “Cyclodextrin drugcarrier systems,” Chemical Reviews, vol. 98, no. 5, pp. 2045–2076, 1998.

[18] N. C. Bellocq, S. H. Pun, G. S. Jensen, and M. E. Davis,“Transferrin-containing, cyclodextrin polymer-based parti-cles for tumor-targeted gene delivery,” Bioconjugate Chemistry,vol. 14, no. 6, pp. 1122–1132, 2003.

[19] S. H. Pun, N. C. Bellocq, A. Liu et al., “Cyclodextrin-modifiedpolyethylenimine polymers for gene delivery,” BioconjugateChemistry, vol. 15, no. 4, pp. 831–840, 2004.

[20] H. Arima, F. Kihara, F. Hirayama, and K. Uekama, “Enhance-ment of gene expression by polyamidoamine dendrimerconjugates with α-, β-, and γ-cyclodextrins,” BioconjugateChemistry, vol. 12, no. 4, pp. 476–484, 2001.


[21] H. Arima, S. Yamashita, Y. Mori et al., “In vitro and invivo gene delivery mediated by lactosylated dendrimer/α-cyclodextrin conjugates (G2) into hepatocytes,” Journal ofControlled Release, vol. 146, no. 1, pp. 106–117, 2010.

[22] C. Lawrencia, R. Mahendran, and K. Esuvaranathan, “Trans-fection of urothelial cells using methyl-β-cyclodextrin solubi-lized cholesterol and Dotap,” Gene Therapy, vol. 8, no. 10, pp.760–768, 2001.

[23] H. Arima, Y. Hagiwara, F. Hirayama, and K. Uekama,“Enhancement of antitumor effect of doxorubicin by itscomplexation with γ-cyclodextrin in pegylated liposomes,”Journal of Drug Targeting, vol. 14, no. 4, pp. 225–232, 2006.

[24] G. R. Bartlett, “Phosphorus assay in column chromatography,”The Journal of Biological Chemistry, vol. 234, no. 3, pp. 466–468, 1959.

[25] M. M. Bradford, “A rapid and sensitive method for thequantitation of microgram quantities of protein utilizing theprinciple of protein dye binding,” Analytical Biochemistry, vol.72, no. 1-2, pp. 248–254, 1976.

[26] L. H. Koehler, “Differentiation of carbohydrates by anthronereaction rate and color intensity,” Analytical Chemistry, vol. 24,no. 10, pp. 1576–1579, 1952.

[27] T. Tsutsumi, F. Hirayama, K. Uekama, and H. Arima, “Evalua-tion of polyamidoamine dendrimer/α-cyclodextrin conjugate(generation 3, G3) as a novel carrier for small interfering RNA(siRNA),” Journal of Controlled Release, vol. 119, no. 3, pp.349–359, 2007.

[28] K. Motoyama, H. Arima, H. Toyodome, T. Irie, F. Hirayama,and K. Uekama, “Effect of 2,6-di-O-methyl-α-cyclodextrin onhemolysis and morphological change in rabbit’s red bloodcells,” European Journal of Pharmaceutical Sciences, vol. 29, no.2, pp. 111–119, 2006.

[29] K. Motoyama, H. Toyodome, R. Onodera et al., “Involvementof lipid rafts of rabbit red blood cells in morphologicalchanges induced by methylated β-cyclodextrins,” Biologicaland Pharmaceutical Bulletin, vol. 32, no. 4, pp. 700–705, 2009.

[30] Y. Ohtani, T. Irie, K. Uekama, K. Fukunaga, and J. Pitha,“Differential effects of α-, β- and γ-cyclodextrins on humanerythrocytes,” European Journal of Biochemistry, vol. 186, no.1-2, pp. 17–22, 1989.

[31] K. Miyajima, K. Tomita, and M. Nakagaki, “Complex for-mation between Di- and Monophosphatidylcholines andcyclodextrin in water,” Chemical and Pharmaceutical Bulletin,vol. 33, pp. 2587–2590, 1985.

[32] J. Nishijo and H. Mizuno, “Interactions of cyclodextrins withDPPC liposomes. Differential scanning calorimetry studies,”Chemical and Pharmaceutical Bulletin, vol. 46, no. 1, pp. 120–124, 1998.

[33] J. Bischoff, S. Libresco, M. A. Shia, and H. F. Lodish, “The H1and H2 polypeptides associate to form the asialoglycoproteinreceptor in human hepatoma cells,” Journal of Cell Biology, vol.106, no. 4, pp. 1067–1074, 1988.

[34] D.-J. Peng, J. Sun, Y.-Z. Wang et al., “Inhibition of hepatocar-cinoma by systemic delivery of Apoptin gene via the hepaticasialoglycoprotein receptor,” Cancer Gene Therapy, vol. 14, no.1, pp. 66–73, 2007.

[35] K. Wada, H. Arima, T. Tsutsumi, F. Hirayama, and K.Uekama, “Enhancing effects of galactosylated dendrimer/α-cyclodextrin conjugates on gene transfer efficiency,” Biologicaland Pharmaceutical Bulletin, vol. 28, no. 3, pp. 500–505, 2005.

[36] P. L. Katsel and R. J. Greenstein, “Eukaryotic gene transferwith liposomes: effect of differences in lipid structure,”Biotechnology Annual Review, vol. 5, pp. 197–220, 2000.

[37] B. Ruozi, F. Forni, R. Battini, and M. A. Vandelli, “Cationicliposomes for gene transfection,” Journal of Drug Targeting,vol. 11, no. 7, pp. 407–414, 2003.

[38] H. Farhood, N. Serbina, and L. Huang, “The role of dioleoylphosphatidylethanolamine in cationic liposome mediatedgene transfer,” Biochimica et Biophysica Acta, vol. 1235, no. 2,pp. 289–295, 1995.

[39] S. W. Hui, M. Langner, Y.-L. Zhao, P. Ross, E. Hurley, andK. Chan, “The role of helper lipids in cationic liposome-mediated gene transfer,” Biophysical Journal, vol. 71, no. 2, pp.590–599, 1996.

[40] A. Natsume, M. Mizuno, Y. Ryuke, and J. Yoshida, “Antitumoreffect and cellular immunity activation by murine interferon-β gene transfer against intracerebral glioma in mouse,” GeneTherapy, vol. 6, no. 9, pp. 1626–1633, 1999.

[41] Y. Aramaki, A. Inaba, and S. Tsuchiya, “Selective incorporationof asialofetuin into hepatocyte of rat,” Biopharmaceutics andDrug Disposition, vol. 6, no. 4, pp. 389–400, 1985.


Research Article

Effects of Polyethylene Glycol Spacer Length and Ligand Densityon Folate Receptor Targeting of Liposomal Doxorubicin In Vitro

Kumi Kawano and Yoshie Maitani

Institute of Medicinal Chemistry, Hoshi University, Ebara 2-4-41, Shinagawa-ku, Tokyo 142-8501, Japan

Correspondence should be addressed to Kumi Kawano, [email protected]

Received 1 July 2010; Revised 22 November 2010; Accepted 22 November 2010

Academic Editor: Sophia Antimisiaris

Copyright © 2011 K. Kawano and Y. Maitani. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

The folate receptor is an attractive target for selective tumor delivery of liposomal doxorubicin (DXR) because it is abundantlyexpressed in a large percentage of tumors. This study examined the effect of polyethylene glycol (PEG) spacer length andfolate ligand density on the targeting ability of folate-modified liposomes. Liposomes were modified with folate-derivatizedPEG-distearoylphosphatidylethanolamine with PEG molecular weights of 2000, 3400, or 5000. The association of DXR-loadedliposomes with KB cells, which overexpress the folate receptor, was evaluated by flow cytometry at various ratios of folatemodification. A low ratio of folate modification with a sufficiently long PEG chain showed the highest folate receptor-mediatedassociation with the cells, but did not show the highest in vitro cytotoxicity. DXR release from folate-modified liposomes inendosomes might be different. These findings will be useful for designing folate receptor-targeting carriers.

1. Introduction

Antitumor drug delivery systems with nanoscopic dimen-sions have received much attention due to their unique accu-mulation behavior at the tumor site. Various nanoparticulatecarriers such as liposomes, polymer conjugates, polymericmicelles, and nanoparticles are utilized for selective deliveryof various anticancer drugs to tumors in a passive targetingmanner [1]. However, a more effective and active targetingsystem is needed to enhance the uptake of drugs usingnanocarriers into cancerous cells at the tumor site.

Receptor-mediated endocytosis pathways have beenexploited for tumor-specific targeting of nanocarriers andintracellular delivery of their contents. Modification ofcarriers with a ligand directed to an overexpressed receptor incancer cells can improve selectivity and facilitate the move-ment of carriers into the intracellular compartment. Onesuch candidate ligand is folic acid because the folate receptor-α is overexpressed in a number of human tumors, includingovarian, lung, brain, head and neck, and breast tumors [2–4]. Folic acid has been widely employed as a targeting moietyfor various anticancer drugs through covalent conjugationto anticancer drugs and nanocarriers [5–8]. Liposomes

modified with folic acid showed selective targeting towardhuman carcinomas along with enhancement of doxorubicin(DXR) cytotoxicity in vitro [9].

Ligand density per drug carrier and spacer lengthare important in designing suitable carriers for targeting.However, the optimal ligand density on liposomes is contro-versial. Different densities of folate in liposomes (ligand/totallipid molar ratio) ranging between 0.01% and 1.0% havebeen reported in the literature as sufficient to promoteliposome binding to the folate receptor on cells [10–12].These differences may be related to the accessibility of thefolate ligand [13] or to the differences in the polyethyleneglycol (PEG)-folate chemical linkage [14]. Because PEGy-lated liposomes, called sterically stabilized liposomes, reducethe association of liposome-modified ligands with theirreceptors by steric hindrance of the PEG polymer [13], weused non-PEGylated liposomes to examine the optimumnumber and spacer length of the targeting ligand.

In this study, folate-mediated association of DXR-loadedliposomes with human oral carcinoma KB cells, whichoverexpress the folate receptor, was evaluated in terms ofPEG spacer length and the ratio of modification with thefolate ligand. Enhanced association of DXR in KB cells was


shown with an extremely low ratio of folate modification anda sufficiently long PEG spacer length, but high cytotoxicity ofDXR was observed with a high ratio of folate modification.


2.1. Materials. Hydrogenated soybean phosphatidylcholine(HSPC), aminopoly(ethyleneglycol)-distearoylphosphatid-ylethanolamine (amino-PEG-DSPE, PEG mean molecularweight of 2000, 3400, and 5000), and methoxy-PEG5000-DSPE (mPEG5000-DSPE) were obtained from NOF Corpo-ration (Tokyo, Japan). Cholesterol (Ch), doxorubicin (DXR)hydrochloride, folic acid, and HPLC-grade acetonitrilewere purchased from Wako Pure Chemical Industries, Ltd.(Osaka, Japan). Folate-derivatized PEG-DSPE (F-PEG2000-,F-PEG3400-, and F-PEG5000-DSPE), which are conjugatesof folic acid and amino-PEG-DSPE, were synthesized asreported previously [13, 15]. Ionophore A23187 and 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlo-rate (DiI) were purchased from Sigma (St. Louis, MO,USA) and Lambda Probes and Diagnostics (Graz, Austria),respectively. Folate-deficient RPMI 1640 medium and fetalbovine serum (FBS) were obtained from Invitrogen Corp.,(Carlsbad, CA, USA). Other reagents used in this study werereagent grade.

2.2. Preparation of Folate-Modified Liposomes. Liposomeswere prepared from HSPC/Ch (55/45 mol/mol). All lipidswere dissolved in chloroform, which was removed by evapo-ration. Lipophilic fluorescent marker DiI-labeled liposomeswere prepared by the same procedure, but with the additionof DiI (0.4 mol% of total lipid) to the lipid mixture andwithout DXR loading. The film was hydrated with MgSO4

aqueous solution (300 mM, adjusted to pH 3.5 with HCl)and sonication. The resulting mean diameter of liposomeswas about 130 nm, as determined by the dynamic lightscattering method (ELS-800; Otsuka Electronics Co., Ltd.,Osaka, Japan) at 25◦C after diluting the liposome suspensionwith water.

DXR was encapsulated into liposomes using theionophore-mediated loading method [16, 17]. The MgSO4

gradient was formed by exchange of the external solutionwith sucrose buffer (300 mM sucrose, 20 mM HEPES, and15 mM EDTA; pH 7.4) by gel filtration chromatography.Subsequent addition of ionophore A23187 to the liposomedispersion results in the outward movement of 1 metal cationin exchange for 2 protons, thus establishing a transmem-brane pH gradient. A23187 was used at a concentration of0.1 μg/μmol lipid and liposomes were incubated with theionophore at 60◦C for 5 min prior to the addition of drug.DXR was then added to the liposomes at a final drug-to-lipid ratio of 0.2 : 1 (wt/wt) and incubated at 60◦C for20 min.

For comparison of loading procedures, DXR was encap-sulated in liposomes by the pH gradient method [18]. Briefly,the lipid film was hydrated with citrate buffer (300 mM; pH4.0) and sonicated. After the external pH was adjusted to 7.4,liposomes were incubated with DXR (drug : lipid = 0.2 : 1,wt/wt) at 60◦C for 20 min.

The folate ligand was inserted into preformed liposomesby the postinsertion technique [19]. Briefly, liposomes(DXR-loaded or DiI-labeled) were incubated with an aque-ous dispersion of F-PEG-DSPE (from 0.01 to 1 mol% oftotal lipid) at 60◦C for 1 h. In the case of unmodifiedliposomes (NF-L), water was added instead of F-PEG-DSPEsolution. Liposomes modified with F-PEG2000-, F-PEG3400-,F-PEG5000-, or mPEG5000-DSPE will henceforth be desig-nated as F2-L, F3-L, F5-L, and M5-L, respectively. Afterheating, the liposomes were cooled to room temperature.The suspension was then passed through a Sephadex G-50 column to remove any leaked DXR and unincorporatedfolate ligand. DXR loading efficiency was determined andsignificant DXR leakage was not observed with incubationof F-PEG-DSPE at these concentrations. DXR concentrationwas determined by measuring absorbance at 480 nm (UV-1700 Phamaspec, Shimadzu Corp., Kyoto, Japan).

2.3. In Vitro Assay for Drug Retention. The release of drugfrom the liposomes in phosphate-buffered saline (PBS, pH7.4 or 5.0) was monitored by a dialysis method. The dialysiswas done at 37◦C using seamless cellulose tube membranes(Spectrum, Houston, TX, USA) with a molecular weightcutoff of 300,000 Da and PBS as the sink solution. The initialconcentration of DXR-loaded liposomes was 0.2 mg/mL. Thesample volume in the dialysis bag was 1 mL, and the sinkvolume was 100 mL. The concentration of drug was analyzedat various times points during dialysis.

2.4. Cellular Association of Liposomes Determined by FlowCytometry. KB cells were obtained from the Cell ResourceCenter for Biomedical Research, Tohoku University (Miyagi,Japan). The cells were cultured in folate-deficient RPMI1640 medium with 10% heat-inactivated FBS and kanamycinsulfate (50 μg/mL) in a humidified atmosphere containing5% CO2 at 37◦C.

The cells were prepared by plating 3 × 105 cells/wellin a 12-well culture plate 1 day before the assay. Thecells were incubated with DXR-loaded liposomes or DiI-labeled liposomes containing 20 μg/mL DXR or 100 μg/mLlipid diluted in 1 mL of serum-free medium for 2 h orthe indicated time at 37◦C. For free ligand competitionstudies, 1 mM folic acid was added to the medium. Afterincubation, the cells were washed with cold PBS (pH 7.4),detached with 0.02% EDTA-PBS, and then suspended inPBS containing 0.1% bovine serum albumin and 1 mMEDTA. The suspended cells were directly introduced intoa FACSCalibur flow cytometer (Becton Dickinson, SanJose, CA) equipped with a 488 nm argon ion laser. Datafor 10,000 fluorescent events were obtained by recordingforward scatter, side scatter, and 585/42 nm fluorescence. Theautofluorescence of cells incubated with serum-free mediumwithout drug for 2 h was used as the control.

2.5. Cytotoxicity of Liposomes in KB Cells. KB cells wereincubated with DXR-loaded liposomes (20 μg/mL) for 2 h.After incubation, the cells were washed with cold PBSand cultured in fresh medium for 48 h. Cytotoxicity wasdetermined using the WST-8 assay (Dojindo Laboratories,


0

20

40

60

80

100D

XR

rele

ased

(%)

0 20 40 80 100

Time (h)

Figure 1: DXR release profile of liposomes loaded by theMgSO4/ionophore method (�) and pH gradient method usingcitrate buffer (�) in PBS (pH 7.4) at 37◦C. Each value representsthe mean ± SD (n = 3).

Kumamoto, Japan) based on enzymatic reduction of atetrazolium salt, WST-8, to water-soluble formazan. Thenumber of viable cells was then determined by absorbanceat 450 nm.


3.1. Characterization of DXR-Loaded and Folate-ModifiedLiposomes. For efficient drug delivery to the target site,drugs should be stably entrapped in liposomes. In thisstudy, an ionophore-mediated pH gradient method utilizingMgSO4 was applied to load DXR into liposomes becausethis method can effectively encapsulate drugs [17]. Morethan 95% of DXR was incorporated in liposomes using thissystem at a drug-to-total lipid ratio of 1 : 5 (wt : wt). Thedrug retention in the liposomes was examined by incubationin PBS (pH 7.4) at 37◦C. For comparison, DXR-liposomesloaded by the remote loading method using citrate bufferwere used. As shown in Figure 1, DXR-liposomes loadedusing MgSO4 showed significantly lower DXR leakage duringthe 72-h incubation compared with those loaded usingcitrate buffer, indicating that the DXR-liposomes producedby the ionophore/MgSO4 loading method were more stablethan those produced by the pH gradient method. Therefore,we applied the ionophore/MgSO4 method to load DXRinto liposomes for evaluation of folate receptor-targetedliposomes.

The average particle size of liposomes used in this studywas approximately 130 nm, and the folate modification didnot change the sizes of liposomes. More than 80% of folateligand was inserted in liposomes at each ratio, which wasconfirmed after the separation of folate-modified liposomesby ultracentrifugation (100,000 × g, 1 h, 4◦C).

3.2. Effects of Spacer Length and Modification Ratio of F-PEG-DSPE on Liposome Association with KB Cells. In thisstudy, the cellular association of folate-modified liposomeswas evaluated in KB cells with respect to PEG spacer lengthand modification ratio by flow cytometry based on DXRfluorescence (Figure 2(a)) and DiI-labeled liposomes (Fig-ure 2(b)). Folate modification with F-PEG2000-, F-PEG3400-,or F-PEG5000-DSPE (F2-, F3-, and F5-L) at 0.03 to 1.0 mol%enhanced the cellular association compared to that ofunmodified liposomes (NF-L), indicating that differences inthe density and PEG spacer length of folate ligands resultedin different liposome associations with KB cells. The highestassociation of liposomes was observed with 0.03 mol% folatemodification with the PEG5000 spacer, which was 1.7-foldand 160-fold higher than that of unmodified liposomes bymeasurement of DXR and DiI, respectively (Figure 2(a)inset and Figure 2(b)). The large discrepancy in the valuemight be due to the difference of distribution of DXRand DiI in liposomes, that is, DXR was entrapped in thewater phase of liposomes, but DiI was incorporated inthe liposomal membrane. Both the drug incorporated intoliposomes and the lipid membrane of liposomes revealed asimilar enhancement in association, suggesting that the DXRwas associated in the liposomal form, not as drug releasedfrom liposomes.

Next, we confirmed whether folate might mediate cellu-lar association with KB cells by mPEG-DSPE modificationand free ligand competition (Figure 3). F5-L with 0.03 mol%folate modification showed higher cellular association thanNF-L and M5-L (0.03 mol% mPEG-DSPE modification)did. Furthermore, the cellular association of F5-L could beblocked by 1 mM free folic acid and reduced to the level ofNF-L. These results indicated that enhancement was due tofolate-mediated cellular association.

The effect of incubation time on the cellular associa-tion of liposomes was then examined (Figure 4). As theincubation time increased, cellular associations increasedand higher association was observed with F5-L modifiedat 0.03 mol% than at 0.3 mol%. Cellular association ofliposomes modified at 0.3 mol% seemed to be saturatedafter a 2-h incubation. It has been reported that the folatereceptor recycling system is downregulated as a result ofsatisfaction of the cellular folate requirement [11]. Therefore,liposomes modified with more folate ligands would leadto a larger intracellular folate content than those withfewer targeting ligands. Our data showed that liposomescontaining fewer folate ligands per liposome had higherassociation efficiencies compared to liposomes containinglarge numbers of folate ligands per liposome. As a result,liposomes with minimal folate ligands may be efficient inenhancing drug accumulation in cells.

3.3. Effect of Folate Modification Ratio on Cytotoxicity. Theeffect of the folate modification ratio on cytotoxicity in KBcells was evaluated using the WST-8 assay (Figure 5). Cellviability was compared with untreated control. All folate-modified liposomes showed higher cytotoxicity than NF-L.The cytotoxicity of F5-L was sharply enhanced from 0 mol%to 0.03 mol% folate modification, which correlated with


0

50

100

150

200

250

300

Mea

nfl

uor

esce

nce

ofD

XR

-loa

ded

lipos

ome

0 0.2 0.4 0.6 0.8 1

F-PEG-DSPE modification (mol%)

NF-LF2-LF2-L

F3-LF5-L

0

20

40

60

80

100

120

Mea

nfl

uor

esce

nce

(DX

R)

0 0.1 0.2 0.3F-PEG5000-DSPE (mol%)

(a)

0

200

400

600

800

1000

1200

1400

Mea

nfl

uor

esce

nce

ofD

iI-l

abel

edlip

osom

e

0 0.1 0.2 0.3


NF-LF2-LF2-L

F3-LF5-L

(b)

Figure 2: Association of folate-modified liposomes with KB cellswith 2-h incubation was determined by fluorescence of DXR-loadedliposomes (a) and DiI-labeled liposomes (b) using flow cytometry.(a) Folate modification from 0.03 to 1.0 mol% of F2-, F3-, andF5-PEG-DSPE. Inset: F5-PEG-DSPE modification from 0.01 to0.3 mol%. (b) Folate modification from 0.01 to 0.3 mol% of F2-,F3-, and F5-PEG-DSPE.

M5-L(0.03 mol%)

F5-L(0.03 mol%)

NF-L

0 25 50 75 100

Mean fluorescence of DXR-loaded liposome

Folate-free medium1 mM folic acid

Figure 3: Association of DXR-loaded liposomes with KB cellswith 1-h incubation was determined by flow cytometry. Cells wereincubated with each liposome in folate-free medium or mediumcontaining 1 mM folic acid.

0

50

100

150

200

Mea

nfl

uor

esce

nce

ofD

XR

-loa

ded

lipos

ome

0 1 2 3 4

Time (h)

NF-LF5-L (0.03 mol%)F5-L (0.3 mol%)

Figure 4: Cellular association of DXR-loaded liposomes with timewas determined by flow cytometry. KB cells were incubated withF5-L modified at 0.03 or 0.3 mol% or without folate modification(NF-L).

the result of their cellular associations (Figure 2). However,modification with concentrations greater than 0.03 mol%gently enhanced the cytotoxicity, which did not correlatewith cellular associations. As in the case of F5-L, 0.3 mol%folate-modified F2-L and F3-L showed higher cytotoxicitythan with 0.03 mol% modification.


0

20

40

60

80

Cel

lvia

bilit

y(%

con

trol

)

0 0.1 0.2 0.3


NF-LF2-L

F3-LF5-L

Figure 5: Cytotoxicity of DXR-loaded liposomes against KB cells.Cells were incubated with each liposome at a DXR concentrationof 20 μg/mL for 2 h, then in fresh medium for 48 h. Each valuerepresents the mean ± SD (n = 3).

0

10

20

30

40

DX

Rre

leas

ed(%

)

0 10 20 30 40 50

Time (h)

NF-LF5-L (0.03 mol%)F5-L (0.3 mol%)

Figure 6: DXR release profile of folate-modified liposomes (F5-L) in PBS (pH 5.0) at 37◦C. Each point represents the mean of 2experiments.

The release of free drug from liposomes is involvedin cytotoxicity or antitumor activity [20, 21]. Thus, wemeasured DXR release from liposomes with different mod-ification ratios at pH 5.0, which resembled endosome con-tent (Figure 6). Liposomes with high modification showedslightly higher drug release than those with low or nomodification, although the release from all formulations wasvery low. Because DXR was stably loaded in the liposome

using ionophore/MgSO4, it may be difficult to evaluate drugrelease differences under these conditions. Taken together,the enhanced cytotoxicity might reflect changes in drugrelease from liposomes by folate modification. Furtherevaluation of folate-modified drug carriers will be needed tooptimize cellular association, cytotoxicity, and/or antitumoreffects.

4. Conclusions

In this study, the effects of PEG spacer length and liganddensity on folate receptor-targeted liposomes were evaluated.A low ratio of folate modification with a sufficiently longPEG chain (F-PEG5000-DSPE) increased folate receptor-mediated association, but a high ratio of folate modificationenhanced in vitro cytotoxicity. This information will beuseful for designing folate receptor-targeting carriers.

Acknowledgments

This work was supported in part by a Grant-in-Aid for YoungScientists (B) from the Ministry of Education, Culture,Sports, Science and Technology of Japan and by the OpenResearch Center Project. The authors would like to thank Mr.Ken Kajihara for his assistance in the experimental work.

References

[1] K. Cho, XU. Wang, S. Nie, Z. Chen, and D. M. Shin, “Ther-apeutic nanoparticles for drug delivery in cancer,” ClinicalCancer Research, vol. 14, no. 5, pp. 1310–1316, 2008.

[2] S. D. Weitman, R. H. Lark, L. R. Coney et al., “Distributionof the folate receptor GP38 in normal and malignant cell linesand tissues,” Cancer Research, vol. 52, no. 12, pp. 3396–3401,1992.

[3] M. Wu, W. Gunning, and M. Ratnam, “Expression of folatereceptor type α in relation to cell type, malignancy, and dif-ferentiation in ovary, uterus, and cervix,” Cancer EpidemiologyBiomarkers and Prevention, vol. 8, no. 9, pp. 775–782, 1999.

[4] N. Parker, M. J. Turk, E. Westrick, J. D. Lewis, P. S. Low,and C. P. Leamon, “Folate receptor expression in carcinomasand normal tissues determined by a quantitative radioligandbinding assay,” Analytical Biochemistry, vol. 338, no. 2, pp.284–293, 2005.

[5] S. Wang and P. S. Low, “Folate-mediated targeting of anti-neoplastic drugs, imaging agents, and nucleic acids to cancercells,” Journal of Controlled Release, vol. 53, no. 1–3, pp. 39–48,1998.

[6] X. Q. Pan, H. Wang, and R. J. Lee, “Antitumor activityof folate receptor-targeted liposomal doxorubicin in a KBoral carcinoma murine xenograft model,” PharmaceuticalResearch, vol. 20, no. 3, pp. 417–422, 2003.

[7] W. A. Henne, D. D. Doorneweerd, A. R. Hilgenbrink, S. A.Kularatne, and P. S. Low, “Synthesis and activity of a folatepeptide camptothecin prodrug,” Bioorganic and MedicinalChemistry Letters, vol. 16, no. 20, pp. 5350–5355, 2006.

[8] Y. Bae, N. Nishiyama, and K. Kataoka, “In vivo antitu-mor activity of the folate-conjugated pH-sensitive polymericmicelle selectively releasing adriamycin in the intracellularacidic compartments,” Bioconjugate Chemistry, vol. 18, no. 4,pp. 1131–1139, 2007.


[9] R. J. Lee and P. S. Low, “Folate-mediated tumor cell targetingof liposome-entrapped doxorubicin in vitro,” Biochimica etBiophysica Acta, vol. 1233, no. 2, pp. 134–144, 1995.

[10] J. A. Reddy, C. Abburi, H. Hofland et al., “Folate-targeted,cationic liposome-mediated gene transfer into disseminatedperitoneal tumors,” Gene Therapy, vol. 9, no. 22, pp. 1542–1560, 2002.

[11] J. M. Saul, A. Annapragada, J. V. Natarajan, and R. V.Bellamkonda, “Controlled targeting of liposomal doxorubicinvia the folate receptor in vitro,” Journal of Controlled Release,vol. 92, no. 1-2, pp. 49–67, 2003.

[12] H. Shmeeda, L. Mak, D. Tzemach, P. Astrahan, M. Tarshish,and A. Gabizon, “Intracellular uptake and intracavitary tar-geting of folate-conjugated liposomes in a mouse lymphomamodel with up-regulated folate receptors,” Molecular CancerTherapeutics, vol. 5, no. 4, pp. 818–824, 2006.

[13] A. Gabizon, A. T. Horowitz, D. Goren et al., “Targeting folatereceptor with folate linked to extremities of poly(ethyleneglycol)-grafted liposomes: in vitro studies,” BioconjugateChemistry, vol. 10, no. 2, pp. 289–298, 1999.

[14] C. P. Leamon and J. A. Reddy, “Folate-targeted chemotherapy,”Advanced Drug Delivery Reviews, vol. 56, no. 8, pp. 1127–1141,2004.

[15] T. Shiokawa, Y. Hattori, K. Kawano et al., “Effect of polyethy-lene glycol linker chain length of folate-linked microemulsionsloading aclacinomycln a on targeting ability and antitumoreffect in vitro and in vivo,” Clinical Cancer Research, vol. 11,no. 5, pp. 2018–2025, 2005.

[16] B. C. L. Cheung, T. H. T. Sun, J. M. Leenhouts, and P. R. Cullis,“Loading of doxorubicin into liposomes by forming Mn-drugcomplexes,” Biochimica et Biophysica Acta, vol. 1414, no. 1-2,pp. 205–216, 1998.

[17] S. A. Abraham, K. Edwards, G. Karlsson et al., “Formationof transition metal-doxorubicin complexes inside liposomes,”Biochimica et Biophysica Acta, vol. 1565, no. 1, pp. 41–54, 2002.

[18] X. Li, D. J. Hirsh, D. Cabral-Lilly et al., “Doxorubicin physicalstate in solution and inside liposomes loaded via a pHgradient,” Biochimica et Biophysica Acta, vol. 1415, no. 1, pp.23–40, 1998.

[19] P. S. Uster, T. M. Allen, B. E. Daniel, C. J. Mendez, M. S.Newman, and G. Z. Zhu, “Insertion of poly(ethylene glycol)derivatized phospholipid into pre-formed liposomes results inprolonged in vivo circulation time,” FEBS Letters, vol. 386, no.2-3, pp. 243–246, 1996.

[20] M. J. Kirchmeier, T. Ishida, J. Chevrette, and T. M. Allen,“Correlations between the rate of intracellular release of endo-cytosed liposomal doxorubicin and cytotoxicity as determinedby a new assay,” Journal of Liposome Research, vol. 11, no. 1, pp.15–29, 2001.

[21] M. J. W. Johnston, S. C. Semple, S. K. Klimuk et al., “Ther-apeutically optimized rates of drug release can be achievedby varying the drug-to-lipid ratio in liposomal vincristineformulations,” Biochimica et Biophysica Acta, vol. 1758, no. 1,pp. 55–64, 2006.


Review Article

Liposomes for Use in Gene Delivery

Daniel A. Balazs and WT. Godbey

Laboratory for Gene Therapy and Cellular Engineering, Department of Chemical and Biomolecular Engineering,Tulane University, 6823 St. Charles Avenue, 300 Lindy Boggs Center, New Orleans, LA 70118, USA

Correspondence should be addressed to WT. Godbey, [email protected]

Received 30 June 2010; Accepted 29 October 2010

Academic Editor: Seyed Moein Moghimi

Copyright © 2011 D. A. Balazs and WT. Godbey. This is an open access article distributed under the Creative CommonsAttribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work isproperly cited.

Liposomes have a wide array of uses that have been continuously expanded and improved upon since first being observed to self-assemble into vesicular structures. These arrangements can be found in many shapes and sizes depending on lipid composition.Liposomes are often used to deliver a molecular cargo such as DNA for therapeutic benefit. The lipids used to form such lipoplexescan be cationic, anionic, neutral, or a mixture thereof. Herein physical packing parameters and specific lipids used for gene deliverywill be discussed, with lipids classified according to overall charge.

1. Introduction

Liposomes are vesicular structures that can form via theaccumulation of lipids interacting with one another in anenergetically favorable manner. Depending upon the struc-ture and the composition of the bulk solution, liposomescan separate hydrophobic or hydrophilic molecules from thesolution. These vesicles are not rigid formations but ratherare fluid entities that are versatile supramolecular assemblies.Because they have dynamic properties and are relativelyeasy to manipulate, liposomes have been used widely in theanalytical sciences as well as for drug and gene delivery.Since their first published use in 1965 [1, 2], the value andpracticality of liposome functions have been recognized andcontinually improved upon.

The advances that brought about liposome-derived tech-nologies have been recognized as some of the cornerstones ofbionanotechnology [3]. The unique advantages imparted bylipid vesicles are their diverse range of morphologies, compo-sitions, abilities to envelope and protect many types of ther-apeutic biomolecules, lack of immunogenic response, lowcost, and their differential release characteristics [4–6]. Thesecharacteristics have led to applications in chemical and bio-chemical analytics, cosmetics, food technologies, and drugand gene delivery [7, 8]. There are numerous lipid formula-tions for each of these applications. However, this review willfocus primarily on the use of liposomes for gene delivery.

2. Characteristics

Liposomes are generally formed by the self-assembly of dis-solved lipid molecules, each of which contains a hydrophilichead group and hydrophobic tails. These lipids take onassociations which yield entropically favorable states of lowfree energy, in some cases forming bimolecular lipid leaflets(Figure 1). Such leaflets are characterized by hydrophobichydrocarbon tails facing each other and hydrophilic headgroups facing outward to associate with aqueous solution[9]. At this point, the bilayer formation is still energeticallyunfavorable because the hydrophobic parts of the moleculesare still in contact with water, a problem that is overcomethrough curvature of the forming bilayer membrane uponitself to form a vesicle with closed edges [10] (Figure 1).This free-energy-driven self-assembly is stable and hasbeen exploited as a powerful mechanism for engineeringliposomes specifically to the needs of a given system [11].

Lipid molecules used in liposomes are conserved entitieswith a head group and hydrophobic hydrocarbon tailsconnected via a backbone linker such as glycerol [12].Cationic lipids commonly attain a positive charge throughone or more amines present in the polar head group. Thepresence of positively charged amines facilitates binding withanions such as those found in DNA. The liposome thusformed is a result of energetic contributions by Van der Waalsforces and electrostatic binding to the DNA which partially


Figure 1: Certain amphipathic lipid molecules in aqueous solution spontaneously form leaflets, then bilayer membranes, and eventuallyliposomes.

dictates liposome shapes [13]. Because of the polyanionicnature of DNA, cationic (and neutral) lipids are typicallyused for gene delivery, while the use of anionic liposomeshas been fairly restricted to the delivery of other therapeuticmacromolecules [14].

Liposomes can exhibit a range of sizes and morphologiesupon the assembly of pure lipids or lipid mixtures suspendedin an aqueous medium [2]. A common morphology whichis analogous to the eukaryotic cellular membrane is the unil-amellar vesicle. This vesicle is characterized by a single bilayermembrane which encapsulates an internal aqueous solution,thus separating it from the external (bulk) solution [15].Both cationic amine head groups and anionic phospholipidhead groups can form these single-walled vesicles. Vesiclesizes fall into the nanometer to micrometer range: smallunilamellar vesicles are 20–200 nm, large unilamellar vesiclesare 200 nm–1 μm, and giant unilamellar vesicles are largerthan 1 μm [2].

Giant vesicles also include other morphologies suchas multilamellar, which consists of multiple concentricbilayers, oligolamellar, which consists of only two concentricbilayers, and multivesicular, which consists of multiplesmaller unilamellar vesicles inside of one giant one. With theexception of multilamellar vesicles, these other morphologiesare difficult to obtain without highly controlled processes forformation [2]. Giant vesicles also deserve special attentionbecause their sizes are large, ranging from 1 μm to morethan 100 μm [2]. These large vesicles are studied and wellcharacterized, partially due to the ease of observation viaoptical microscopy [10].

During the compaction of polynucleotides into liposo-mal assemblies, a number of structures have been knownto appear [5, 6, 16–19]. Each structure is formed in themost energetically favorable conformation based upon char-acteristics of the specific lipids used in the system [13]. Adependent term known as the structure-packing parametercan be used to suggest what shape the amphiphile willtake, depending on the ratio of size variables. The packingparameter is defined as

P = v

alc, (1)

where v: the volume of the hydrocarbon portion, a: theeffective area of the head group, and lc: the length of the lipidtail.

This correlation predicts a range of structures accordingto the following conditions [13, 20] (Figure 2):

P <13−→ spherical micelle,

13≤ P <

12−→ cylindrical micelle,

12≤ P < 1 −→ flexible bilayers, vesicles,

P = 1 −→ planar bilayers,

P > 1 −→ inverted micelles,(hexagonal (HII)phase

).

(2)

3. Cationic Lipids

A solution of cationic lipids, often formed with neutral hel-per lipids, can be mixed with DNA to form a positively char-ged complex termed a lipoplex [21]. Well-characterized andwidely used commercial reagents for cationic lipid transfec-tion include N-[1-(2,3-dioleyloxy)propyl]-N,N,N-trimethy-lammonium chloride(DOTMA) [22], [1,2-bis(oleoyloxy)-3-(trimethylammonio)propane] (DOTAP) [23], 3β[N-(N′,N′-dimethylaminoethane)-carbamoyl] cholesterol (DC-Chol) [24], and dioctadecylamidoglycylspermine (DOGS)[25]. Dioleoylphosphatidylethanolamine (DOPE), a neutrallipid, is often used in conjunction with cationic lipidsbecause of its membrane destabilizing effects at low pH,which aide in endolysosomal escape [26].

Many cationic lipid compounds have been formulatedsince the advent of DOTMA [27–31]. Each lipid has differentstructural aspects, such as head group size and hydrocarbontail length. These aspects confer distinct characteristics to thelipid/DNA complex, which in turn affect association withand uptake into the cell. However, the basic structure ofcationic lipids mimics the chemical and physical attributes ofbiological lipids [32]. The positive charge on the head groupfacilitates spontaneous electrostatic interaction with DNA, aswell as binding of the resulting lipoplexes to the negatively


Spherical micelles

P < 1/3

Cylindrical micelles

1/3 ≤ P < 1/2

1/2 ≤ P < 1

Flexible bilayers,Vesicles

Planar bilayers

P = 1

Inverted micelles(hexagonal (HII) phase)

P > 1

a

v lc

Figure 2: Structures predicted by the packing parameter P.

charged components of the cell membrane prior to cellularuptake [33, 34]. The use of a cation is a recurring themefor virtually all chemically mediated gene delivery vectors,including polymers, lipids, and nondegradable nanoparti-cles.

Between 8–18 carbons commonly comprise the hydro-carbon tails of lipids used for gene delivery. The tails aretypically saturated, but a single double bond is occasionallyseen. The combination of hydrocarbon chains in a lipidmixture can be symmetric or asymmetric. It has been shownthat certain asymmetric lipid mixtures with both shortersaturated carbon chains and long unsaturated carbon chains

produce relatively high transfection efficiencies as comparedto mixed formulations of symmetric cationic lipids [35].

Hydrophobic tails are not the only liposomal featuresthat play a role in effective gene delivery—ionizable headgroups are also involved. Some examples are the multivalentcationic lipids DOSPA and DOGS (covered in Section 3.2);both of which have a functionalized spermine head groupthat confers the ability to act as a buffer, such as in thecase where there is an influx of protons into a maturingendosome/endolysosome [36]. Such buffering could extendthe amount of time needed to activate acid hydrolasesand could explain why some multivalent cationic lipids


H3C

CH3

CH2

CH2

CH2H3C

H2C

HC O

O

N

Figure 3: The structure of DOTMA.

can exhibit higher transfection efficiencies versus theirmonovalent counterparts [25, 37].

3.1. Monovalent Cationic Lipids

3.1.1. DOTMA (see Figure 3). N-[1-(2,3-dioleyloxy) prop-yl]-N,N,N-trimethylammonium chloride, or DOTMA, wasone of the first synthesized and commercially availablecationic lipids used for gene delivery. Its structure consistsof 2 unsaturated oleoyl chains (C18 :Δ9), bound by anether bond to the three-carbon skeleton of a glycerol, witha quaternary amine as the cationic head group [22]. Ascompared to other methods of gene transfer used in thelate 1980s, DOTMA proved to facilitate up to 100-foldmore efficient gene delivery than the use of DEAE-dextrancoprecipitation or calcium phosphate [22]. The ability toentrap DNA or RNA in a liposome in a relatively simplefashion, with effective gene delivery to cells, significantlyinfluenced and improved the potential of nonviral agents forgene therapy [22, 38]. Based upon the use of comparativeprotein expression assays such as luciferase, β-galactosidase,or chloramphenicol acetyltransferase, initial success of invitro transfection of multiple cell lines with DOTMA sparkeda number of attempts to improve the lipid formulationand resulted in the creation of many effective formula-tions including such notable lipids as DOTAP [23] (seeSection 3.1.2) and DC-Chol [24] (Section 3.1.3).

Commercialization of DOTMA as Lipofectin involvedits coupling with DOPE (Section 4.1) in a 1/1 ratio dueto the ability of DOPE to increase transfection efficiencies.Once commercialized, improvements in Lipofectin weredesired, motivating others to add functional groups tothe DOTMA. Many alterations made in the four majormoieties of DOTMA (head group, linker, linkage bonds,and hydrocarbon chains) have reflected widespread efforts toreduce toxicity and increase transfection efficiencies [23, 39].These studies have suggested, however, that cytotoxicitiesassociated with the formulated monovalent lipids weredependent on plated cell density. Plate densities of 25%–35%, treated with cationic lipoplexes, yielded roughly halfthe amount of cell protein per plate versus controls. Near-confluent cell monolayers exhibited very little evidence ofcytotoxicity. These findings supported a need for manip-

ulations in the structural aspects of the lipids for loweredcytotoxicity in subconfluent populations [23]. Felgner et al.[40] also experimented with novel lipid formulations byaltering DOTMA to obtain a more robust understanding ofthe mechanism of biological action. The structural changesincluded different combinations of side chains and alkylattachments to the head groups, as well as the replacementof a methyl group on the quaternary amine of DOTMAwith a hydroxyl. Their report suggested that compoundswith such a hydroxyl modification display improved proteinexpression after transfection by two- to three-fold overthose observed following DOTMA-mediated transfections.Stabilization of the bilayer vesicles was purported to occuras a result of the hydroxyl group remaining in contact withthe aqueous layer surrounding the liposome. Compoundslacking this moiety were hypothesized to become entrenchedin the aliphatic region, thus destabilizing the membrane. Itwas also indicated that aliphatic chain length had a largeeffect on the efficacy of lipid vectors. As the lengths of thesaturated chains were increased in the DOTMA analogs,transfection efficiencies decreased. This was thought to bedue to increased bilayer stiffness, which may have preventedefficient fluid interactions with the endosomal membrane tothus hamper the release of the liposomes or plasmid DNAfrom the endosomal compartments.

3.1.2. DOTAP (see Figure 4). [1,2-bis(oleoyloxy)-3-(trime-thylammonio)propane], or DOTAP, was first synthesized byLeventis and Silvius in 1990 [23]. The molecule consists of aquaternary amine head group coupled to a glycerol backbonewith two oleoyl chains. The only differences between thismolecule and DOTMA are that ester bonds link the chainsto the backbone rather than ether bonds. It was originallyhypothesized that ester bonds, which are hydrolysable, couldrender the lipid biodegradable and reduce cytotoxicity. Thisstudy showed that the transfection activities and levels ofcytotoxicity associated with DOTAP/DOPE formulationsare not statistically different from those associated withDOTMA/DOPE composites. Notably, this type of mono-valent lipids also showed little to no cytotoxic effect onnear-confluent cell monolayers, in addition to exhibitingthe same lipoplex sensitivity at 25%–35% cell confluence asmentioned in Section 3.1.1 [23].


HC O

O

C

O

O

C

N

CH3

CH2

H2C

H3C

H3C

Figure 4: The structure of DOTAP.

O

O

NH

HN

H3C

H3C

Figure 5: The structure of DC-Chol.

The use of 100% DOTAP for gene delivery is inefficientdue to the density of positive charges on the liposomesurface, which possibly prevents counter ion exchange [41].DOTAP is completely protonated at pH 7.4 (which is notthe case for all other cationic lipids) [41], so it is possiblethat more energy is required to separate the DNA from thelipoplex for successful transfection [42]. Thus, for DOTAPto be more effective in gene delivery, it should be combinedwith a helper lipid, as seems to be the case for most cationiclipid formulations.

High temperature and long incubation times havebeen used to create lipoplexes that exhibit resistance toserum interaction [43]. Interestingly, this approach wasonly observed to affect monovalent cationic lipids such asDOTMA, DOTAP, or DC-Chol, as opposed to multivalentcationic lipids. The specific reasons for this phenomenonremain unclear. In fact, the specific mechanism behindserum inactivation of lipoplexes in general is as yet unex-plained. Several hypotheses have been offered as to themechanism, including the prevention of lipoplex binding tocell membranes by serum proteins [34, 43], the preventionof structural complex maturation by serum proteins bindingto cationic charges on the lipoplexes [43], and the disparityof endocytosis pathways—which have varying kinetics—that are used for lipoplex endocytosis, with the method ofendocytosis being regulated by the size of the lipoplexes oraggregates of lipoplexes plus serum proteins [34, 44].

3.1.3. DC-Chol (see Figure 5). 3β[N-(N’,N’-dimethylami-noethane)-carbamoyl]cholesterol, or DC-Chol, was firstsynthesized by Gao and Huang in 1991 [24]. DC-Cholcontains a cholesterol moiety attached by an ester bondto a hydrolysable dimethylethylenediamine. Cholesterol wasreportedly chosen for its biocompatibility and the stability itimparts to lipid membranes, an idea which was supportedby observed transfection activity of up to two- to four-fold

greater chloramphenicol acetyltransferase expression (CATassay). Additionally, DC-Chol was found to have a four-foldreduction in cytotoxicity versus Lipofectin in some cell lines[24].

In contrast to cationic liposomes containing fully chargedquaternary amines (e.g. DOTMA and DOTAP), DC-Chol, ina 1 : 1 lipid ratio with DOPE, contains a tertiary amine that ischarged on 50% of the liposome surface at pH 7.4 [45]. Thisfeature is thought to reduce the aggregation of lipoplexesleading to higher transgene expression [46]. The reductionin overall lipoplex charge can also aid in DNA dissociationduring gene delivery [41], which has been proven to benecessary for successful transfection [42].

3.2. Multivalent Cationic Lipids

3.2.1. DOSPA (see Figure 6). 2,3-dioleyloxy-N-[2(spermi-necarboxamido)ethyl]-N,N-dimethyl-l-propanaminium tri-fluoroacetate, or DOSPA, is another cationic lipid synthe-sized as a derivative of DOTMA. The structure is similarto DOTMA except for a spermine group which is boundvia a peptide bond to the hydrophobic chains. This cationiclipid, used with the neutral helper lipid DOPE at a 3 : 1ratio, is commercially available as the transfection reagentLipofectamine. In general, the addition of the sperminefunctional group allows for a more efficient packing of DNAin terms of liposome size. The efficient condensation ispossibly due to the many ammonium groups in spermine.It has been shown that spermine can interact via hydrogenbonds with the bases of DNA in such a way as to be attractedon one strand and wind around the major groove to interactwith complementary bases of the opposite strand [47].

3.2.2. DOGS (see Figure 7). Di-octadecyl-amido-glycyl-sper-mine, or DOGS, has a structure similar to DOSPA; both


NH2

NH2

CH2

CH3

CH2

CH2

CH2

NH3

NH3

H3C

H2C

HC O

O

NHNO

Figure 6: The structure of DOSPA.

HNO

N

O

NH3

NH3

NH2

NH2

Figure 7: The structure of DOGS.

molecules have a multivalent spermine head group and two18-carbon alkyl chains. However, the chains in DOGS aresaturated, are linked to the head group through a peptidebond, and lack a quaternary amine. DOGS is commerciallyavailable under the name Transfectam. This lipid has beenused to transfect many cell lines, with transgene expressionlevels more than 10-fold greater than those seen followingcalcium phosphate transfections [25]. In addition, Behr et al.showed that not only was DOGS very effective in deliveringthe CAT reporter plasmid, but it was also associated with nonoticeable cytotoxicity [25].

Much like the multivalent cationic lipid DOSPA, DOGSis very efficient at binding and packing DNA, a result ofthe spermine head group that so closely associates withDNA [25]. Characterization of the head group of DOGSwas determined to facilitate not only efficient condensationof DNA but also buffering of the endosomal compartment,which was thought to protect the delivered DNA fromdegradation by pH-sensitive nucleases [36]. DOGS is amultifaceted molecule in terms of buffering capacity. AtpH values lower than 4.6 all of the amino groups in thespermine are protonated, while at pH = 8 only two are

purportedly ionized, which promotes arrangement into alamellar structure [48]. The packing ability of DOGS is due,in part, to the dynamics of the large head group molecule andthe length of long unsaturated carbon chains.

3.3. Modifications for Improved Liposome-Mediated

Gene Delivery

3.3.1. Poly(ethylene) Glycol. Recent improvements in lipo-fection have facilitated protection from degradation invivo, due to surface modifications with polyethylene glycol(PEG). PEG presents many attractive qualities as a liposomalcoating, such as availability in a variety of molecular weights,lack of toxicity, ready excretion by the kidneys, and ease ofapplication [49]. Methods of modifying liposomal surfaceswith PEG include its physical adsorption onto the liposomalsurface and its covalent attachment onto premade liposomes[50].

It has been shown by Kim et al. [51] that PEGylatedlipoplexes yield increased transfection efficiencies in thepresence of serum as compared to liposomal transfectionmethods lacking such surface attachments. Additionally, the


PEGylated lipoplexes display improved stabilities and longercirculation times in the blood. It is thought that the PEGforms a steric barrier around the lipoplexes, which stiflesclearance due to reduced macrophage uptake [50], andmay allow the liposome to overcome aggregation problemsthrough mutually repulsive interactions between the PEGmolecules [52]. These characteristics increase bioavailability,facilitating higher transfection efficiencies due to improvedtissue distribution and larger available concentrations [53].

Because of the decreased immune responses andincreased circulation times associated with PEG-modifiedliposomes, these particles are sometimes referred to as“stealth liposomes.” However, such liposomes lack specificitywith regard to cellular targeting. Notably, Shi et al. foundthat PEGylation inhibited endocytosis of the lipoplexes ina fashion that was dependent upon the mole percentageof PEG on the liposome, as well as the identity of certainfunctional groups that were conjugated to the lipoplexes[54]. Additionally, upon incorporation into the cell, PEGworked to deter proper complex dissociation by stabilizing alamellar phase of DNA packing. As a result of these findings,a need has arisen for the creation of novel PEG-containingliposomes whereby the attached PEG is removed followingendocytosis via a hydrolysable connecting molecule.

3.3.2. Additions and Alternatives to Poly(ethylene) Glycol.Alternative liposomal formulations utilizing polymers otherthan PEG are being produced with the goal of creating ster-ically protected lipoplexes. Additional aims of such systemsinclude biocompatibility, flexible structure, and solubility inphysiological systems [50]. A report by Metselaar et al. onthe use of L-amino-acid-based polymers for lipoplex mod-ification found an extended circulation time and reducedclearance by macrophages at levels similar to those seenwith lipoplexes modified with PEG. Results suggested thatapproximately 10% of the injected dose of the L-amino-acid-modified complexes was still present in the blood oftreated rats after 48 hours [49]. These oligopeptides areattractive alternatives to PEG due to advantages such asincreased biodegradability and favorable pharmacokineticswhen lower concentrations are used per dose.

Liposomes can also be coupled to targeting moietiesthrough the use of PEG to impart attraction to affectedtissues for optimal routing and transfection. Targetingligands are selected based upon specific target cell receptors.The target cells can be normal or transformed (tumor) cells.Examples of such ligands include transferrin [55], a popularligand for delivery of anticancer drugs to solid tumors invivo, and haloperidol [56], a ligand that associates with sigmareceptors that are overexpressed in many types of cancer.

4. Neutral Lipids

4.1. DOPE and DOPC (see Figure 8). Most liposomal for-mulations used for gene delivery consist of a combinationof charged lipids and neutral helper lipids [12, 22–24, 26,28]. The neutral helper lipids used are often dioleoylphos-phatidylethanolamine (DOPE), which is the most widely

used neutral helper lipid, or dioleoylphosphatidylcholine(DOPC). Results have shown that the use of DOPE ver-sus DOPC as the helper lipid yields higher transfectionefficiencies in many cell types [28, 57], thought to bedue to a conformational shift to an inverted hexagonalpacking structure (Figure 2) that is imparted by DOPEat low pH. In contrast to the creation of repeated layersof DNA/lipids, as is the case in lamellar packing, theinverted hexagonal packing structure is similar to that ofa honeycomb of tubular structures which condense DNAinside the tubes through electrostatic interactions. The tubesaggregate due to Van der Waals interactions between thelipid tails that spread out to encircle each tube. Fusionand destabilization of the lipoplexes during transfection arethought to occur due to the exposure of the endosomalmembrane to invasive hydrocarbon chains [58]. Studieshave suggested that a hexagonal conformation allows forefficient escape of complexed DNA from endosomal vesiclesvia destabilization of the vesicle membrane [17, 59]. Withthe lysosomotropic agent chloroquine inhibiting the activityof DOPE-containing lipoplexes, it is reasonable to assumethat the membrane-destabilizing hexagonal conformationassociated with DOPE is brought about at acidic pH [26].

In DOTAP-mediated DNA-binding studies, it wasdiscovered that liposomes—formulated without DOPE—would not effectively complex with DNA to neutralize it untila 2 : 1 N : P ratio was reached due to an inability to displacecounter ions bound to the cationic lipid head groups [41].In contrast, complexes with a 1 : 1 ratio of DOTAP/DOPEcontinuously neutralized and complexed with the negativelycharged DNA at all charge ratios. This is possibly dueto salt bridges more easily forming between the positivelycharged head groups of the cationic lipids and the phosphategroups of DOPE moieties. This association would force theprimary amine of DOPE to stabilize itself in the plane ofthe liposome surface and allow for more close interactionswith the negatively charged phosphate of the DNA. DOPEcould also facilitate counter ion release from the positivelycharged lipid head group, thus lowering the energy requiredfor binding DNA [41]. Circular dichroism has been used toindicate that the use of DOPE as a helper lipid allows formuch closer contact and packing of DNA helices [41].

DC-Chol and other cholesterol derivatives have beenincorporated into lipoplex assembly for increased transfec-tion efficiency in vivo [60, 61]. Galactosylated cholesterolderivatives have been shown to lower cytotoxicity levels andimprove transfection efficiencies in human hepatoma cells(Hep G2), likely due to the affinity of cellular receptorsfor galactosylated ligands [62]. This result indicates thatlipoplexes can be formulated for cell-specific uptake throughthe addition of specific ligands.

5. Anionic Lipids

In general, gene delivery by anionic lipids is not very effi-cient. The negatively charged head group prevents efficientDNA compaction due to repulsive electrostatic forces thatoccur between the phosphate backbone of DNA and theanionic head groups of the lipids. However, due to the fact


HC O

O

C

O

O

C

O OP

O

CH2 CH2 CH2H3N

H2C

O−

(a)

N

CH3

CH3

H3C

HC O

O

C

O

O

C

O OP

O

CH2 CH2 CH2

H2C

O−

(b)

Figure 8: The structures of two neutral lipids. (a) DOPE (b) DOPC.

that cationic liposomes can be inactivated in the presenceof serum, are unstable upon storage, and exhibit somecytotoxicity both in vitro and in vivo, anionic liposomeshave been studied as potential gene delivery vehicles [63–65]. Formation of DNA-containing liposomes using anioniclipids can be brought about through the use of divalentcations to negate the mutual electrostatic repulsion andfacilitate lipoplex assembly [8]. Anionic lipoplexes are com-posed of physiologically safe components including anioniclipids, cations, and plasmid DNA [66]. Commonly usedlipids in this category are phospholipids that can be foundnaturally in cellular membranes such as phosphatidic acid,phosphatidylglycerol, and phosphatidylserine (Figure 9). Aswith the lipids presented earlier, anionic lipids can containany of a wide range of fatty acid chains in the hydrophobicregion. The specific fatty acids incorporated are responsiblefor the fluidic characteristics of the liposome in terms ofphase behavior and elasticity [2]. Perhaps due to the naturalpresence of these specific phospholipids in the host cellmembrane, gene delivery via lipoplexes with net negativesurface potentials has been associated with lower clearanceand phagocytosis by macrophages, which is consistent withfavorable biocompatibility [67].

Various anionic liposomes have been characterized forgene delivery in a small number of cell types includingCHO cells and primary hippocampal neurons [8, 66, 68,69]. While such investigations are novel, overall knowledge

regarding anionic lipofection is as yet limited due to a lackof extensive testing; DNA entrapment in anionic liposomesis still inefficient, and cytotoxicity data remain inadequate.

Divalent cations can be incorporated into the system toenable the condensation of nucleic acids prior to envelop-ment by anionic lipids. Several divalent cations have beentested for use in anionic lipoplexes such as Ca2+, Mg2+, Mn2+,and Ba2+, but it has been observed that the use of Ca2+

yielded the highest transfection efficiency due to its higherDNA binding affinity [70, 71]. An investigation conductedby Srinivasan and Burgess confirmed that Ca2+ was themost effective cation for DNA compaction as compared toNa+ and Mg2+ [66]. This affinity is potentially a result ofthe smaller hydrodynamic radius of calcium which gives alarger charge per unit surface area. The use of Ca2+ not onlyovercame the strong electrostatic repulsion between the DNAand the lipids, but also promoted uptake of the lipoplexesby the cell [8]. However, the use of high concentrations ofcalcium (in excess of 25 mM) was shown to be detrimentalto transfection efficiency because of the creation of aggregatelipoplexes, having particle sizes of 500 nm and higher [66].Optimum transfection efficiency is achieved with particlessizes of about 200 nm due to factors thought to be relatedto clathrin-mediated uptake [72].

Mixtures of the anionic lipid dioleoylphosphatidylglyc-erol (DOPG) and the neutral lipid DOPE have beeninvestigated to determine an optimal ratio for transfection


HC

H2C

R2

R1

CH2

O

O

C

O

O

C

O P

O

O

O

H

(a)

CH2 CH2CHHO

OH HC

H2C

R2

R1

CH2

O

O

C

O

O

C

O P

O

O

O

(b)

NH3

OOC CH2CH

HC

H2C

R2

R1

CH2

O

O

C

O

O

C

O P

O

O

O

(c)

Figure 9: Anionic Lipids. (a) Phosphatidic acid (pH = 7). (b)Phosphatidylglycerol. (c) Phosphatidylserine.

[66]. It was suggested that a 1 : 4 ratio of DOPG toDOPE was a proper balance to allow the negatively chargedphospholipids to form lipoplexes while still having enough ofthe neutrally charged phospholipids to allow for endosomalescape. DOPG has a packing parameter less than 1 and tendsto form flexible bilayers and vesicles (Figure 2) [73]. Thischaracteristic can be contrasted to that of DOPE, which has apacking parameter greater than one and is known to adoptan inverted hexagonal structure that promotes membranedestabilization [13, 70]. Transmission electron microscopyrevealed that this particular formulation yields liposomesof a spherical multilamellar structure [66]. However, uponrelocation to the late endosome or endolysosome, thelipoplex may alter its morphology due to the effects of pHupon the DOPE. The 1 : 4 ratio was seen to exhibit highertransfection efficiency and cell viability versus the cationicformulation Lipofectamine 2000 [66].

Despite some favorable investigations into the use ofanionic liposomes for gene delivery, there are some potentialdownfalls associated with systemic delivery that must befurther explored. Some studies have indicated that, uponexposure to certain plasma lipoproteins, destabilization andleakage of liposomal contents can occur. For example, inliposomes lacking cholesterol, high density lipoprotein cancause some disintegration of the liposome [74]. However,in liposomes which do contain cholesterol, low densitylipoproteins can also cause leakage of contents [75]. Char-acterization studies like these are very useful in terms of

determining what mole percentages and types of lipids mustbe taken away or added to liposomal formulations to obtainmaximum delivery of a desired cargo.

6. Concluding Remarks

An abundance of uses for liposomes has been investigatedsince their introduction into the scientific literature in the1960s. These studies have highlighted both the self-assemblyof various lipid formulations and dynamic properties of cel-lular membranes as they interact with the local environment.Not only have mechanisms of membrane transport andpharmaceutical cargo delivery via liposomes been elucidated,but analytical uses such as immunoassays and biosensorshave also been developed.

At the rudimentary level, most lipids that self assembleinto useful shapes are amphipathic, containing both ahydrophilic head group and a hydrophobic lipid tail group.The shapes that are formed are determined by the typesof lipids used, which, in turn, provide various optionsregarding delivery. The cationic head groups appear to bebetter suited for DNA delivery due to the natural chargeattraction between negatively charged phosphate groups andthe positively charged head groups. Anionic head groups areperhaps better suited for drug delivery. However, this doesnot preclude their use as gene delivery vehicles as work withdivalent cations has shown.

One must keep in mind all of the variables that comeinto play when using different gene delivery vectors. Thereis no concrete comparison that can easily be made to suggestthat one liposomal vector is better than another for all celltypes, environments, and applications. While some of thelipids presented above were originally found to yield little-to-no cytotoxicity for a given cell type, the observationdoes not necessarily hold true when they are applied todifferent cell types [23–25]. Improvements and adjustmentsto these formulations are constantly being explored throughthe addition of different lipids, targeting molecules, orshielding moieties designed to prevent clearance in vivo. Theidentification of the optimal gene delivery vector continuesto be an elusive process, and liposomes are but a fraction ofall the vehicles that are being examined.

References


[2] A. Jesorka and O. Orwar, “Liposomes: technologies and ana-lytical applications,” Annual Review of Analytical Chemistry,vol. 1, no. 1, pp. 801–832, 2008.

[3] P. R. Leduc, M. S. Wong, P. M. Ferreira et al., “Towards an invivo biologically inspired nanofactory,” Nature Nanotechnol-ogy, vol. 2, no. 1, pp. 3–7, 2007.

[4] T. Montier, T. Benvegnu, P.-A. Jaffres, J.-J. Yaouanc, and P.Lehn, “Progress in cationic lipid-mediated gene transfection:a series of bioinspired lipids as an example,” Current GeneTherapy, vol. 8, no. 5, pp. 296–312, 2008.


[5] C. Tros de Ilarduya, Y. Sun, and N. Duzgunes, “Genedelivery by lipoplexes and polyplexes,” European Journal ofPharmaceutical Sciences, vol. 40, no. 3, pp. 159–170, 2010.

[6] I. Koltover, T. Salditt, and C. R. Safinya, “Phase diagram,stability, and overcharging of lamellar cationic lipid- DNAself-assembled complexes,” Biophysical Journal, vol. 77, no. 2,pp. 915–924, 1999.

[7] G. Gregoriadis, Liposome Technology, Taylor &Francis/Informa, New York, NY, USA, 2006.

[8] S. D. Patil, D. G. Rhodes, and D. J. Burgess, “Anionic liposomaldelivery system for DNA transfection,” The AAPS Journal, vol.6, no. 4, article no. e29, 2004.

[9] D.-G. Margineanu, “Equilibrium and non-equilibriumapproaches in biomembrane thermodynamics,” ArchivesInternationales de Physiologie et de Biochimie, vol. 95, no. 5,pp. 381–422, 1987.

[10] S. Svetina and B. Zeks, “Shape behavior of lipid vesicles as thebasis of some cellular processes,” Anatomical Record, vol. 268,no. 3, pp. 215–225, 2002.

[11] M. Shimomura and T. Sawadaishi, “Bottom-up strategy ofmaterials fabrication: a new trend in nanotechnology of softmaterials,” Current Opinion in Colloid and Interface Science,vol. 6, no. 1, pp. 11–16, 2001.

[12] P. P. Karmali and A. Chaudhuri, “Cationic liposomes asnon-viral carriers of gene medicines: resolved issues, openquestions, and future promises,” Medicinal Research Reviews,vol. 27, no. 5, pp. 696–722, 2007.

[13] J. N. Israelachvili, Intermolecular and Surface Forces, AcademicPress, London, UK, 2nd edition, 1991.

[14] E. Mayhew and D. Papajadjopoulos, “Therapeutic applica-tions of liposomes,” in Liposomes, M. J. Ostro, Ed., pp. 289–341, Marcel Dekker, New York, NY, USA, 1983.

[15] D. Lasic, Liposomes: From Physics to Applications, Elsevier,Amsterdam, The Netherlands, 1993.

[16] B. Sternberg, “New structures in complex formation betweenDNA and cationic liposomes visualized by freeze-fractureelectron microscopy,” FEBS Letters, vol. 356, no. 2-3, pp. 361–366, 1994.

[17] I. Koltover, T. Salditt, J. O. Radler, and C. R. Safinya,“An inverted hexagonal phase of cationic liposome-DNAcomplexes related to DNA release and delivery,” Science, vol.281, no. 5373, pp. 78–81, 1998.

[18] J. O. Radler, I. Koltover, T. Salditt, and C. R. Safinya, “Structureof DNA-cationic liposome complexes: DNA intercalationin multilamellar membranes in distinct interhelical packingregimes,” Science, vol. 275, no. 5301, pp. 810–814, 1997.

[19] J. Gustafsson, G. Arvidson, G. Karlsson, and M. Almgren,“Complexes between cationic liposomes and DNA visualizedby cryo-TEM,” Biochimica et Biophysica Acta, vol. 1235, no. 2,pp. 305–312, 1995.

[20] W.-L. Hsu, H.-L. Chen, W. Liou, H.-K. Lin, and W.-L. Liu,“Mesomorphic complexes of DNA with the mixtures of acationic surfactant and a neutral lipid,” Langmuir, vol. 21, no.21, pp. 9426–9431, 2005.

[21] L. Wasungu and D. Hoekstra, “Cationic lipids, lipoplexes andintracellular delivery of genes,” Journal of Controlled Release,vol. 116, no. 2, pp. 255–264, 2006.

[22] P. L. Felgner, T. R. Gadek, M. Holm et al., “Lipofection: ahighly efficient, lipid-mediated DNA-transfection procedure,”Proceedings of the National Academy of Sciences of the UnitedStates of America, vol. 84, no. 21, pp. 7413–7417, 1987.

[23] R. Leventis and J. R. Silvius, “Interactions of mammalian cellswith lipid dispersions containing novel metabolizable cationic

amphiphiles,” Biochimica et Biophysica Acta, vol. 1023, no. 1,pp. 124–132, 1990.

[24] X. Gao and L. Huang, “A novel cationic liposome reagent forefficient transfection of mammalian cells,” Biochemical andBiophysical Research Communications, vol. 179, no. 1, pp. 280–285, 1991.

[25] J.-P. Behr, B. Demeneix, J.-P. Loeffler, and J. Perez-Mutul,“Efficient gene transfer into mammalian primary endocrinecells with lipopolyamine-coated DNA,” Proceedings of theNational Academy of Sciences of the United States of America,vol. 86, no. 18, pp. 6982–6986, 1989.

[26] H. Farhood, N. Serbina, and L. Huang, “The role of dioleoylphosphatidylethanolamine in cationic liposome mediatedgene transfer,” Biochimica et Biophysica Acta, vol. 1235, no. 2,pp. 289–295, 1995.

[27] J.-P. Behr, “Gene transfer with synthetic cationic amphiphiles:prospects for gene therapy,” Bioconjugate Chemistry, vol. 5, no.5, pp. 382–389, 1994.

[28] H. Farhood, X. Gao, K. Son et al., “Cationic liposomes fordirect gene transfer in therapy of cancer and other diseases,”Annals of the New York Academy of Sciences, vol. 716, pp. 23–35, 1994.

[29] A. Kikuchi, Y. Aoki, S. Sugaya et al., “Development of novelcationic liposomes for efficient gene transfer into peritonealdisseminated tumor,” Human Gene Therapy, vol. 10, no. 6, pp.947–955, 1999.

[30] C. J. Wheeler, P. L. Felgner, Y. J. Tsai et al., “A novel cationiclipid greatly enhances plasmid DNA delivery and expression inmouse lung,” Proceedings of the National Academy of Sciences ofthe United States of America, vol. 93, no. 21, pp. 11454–11459,1996.

[31] V. Budker, V. Gurevich, J. E. Hagstrom, F. Bortzov, and J.A. Wolff, “pH-sensitive, cationic liposomes: a new syntheticvirus-like vector,” Nature Biotechnology, vol. 14, no. 6, pp. 760–764, 1996.

[32] N. Maurer, A. Mori, L. Palmer et al., “Lipid-based systemsfor the intracellular delivery of genetic drugs,” MolecularMembrane Biology, vol. 16, no. 1, pp. 129–140, 1999.

[33] A. Elouahabi and J.-M. Ruysschaert, “Formation and intra-cellular trafficking of lipoplexes and polyplexes,” MolecularTherapy, vol. 11, no. 3, pp. 336–347, 2005.

[34] P. Pires, S. Simoes, S. Nir, R. Gaspar, N. Duzgunes, and M. C.Pedroso De Lima, “Interaction of cationic liposomes and theirDNA complexes with monocytic leukemia cells,” Biochimica etBiophysica Acta, vol. 1418, no. 1, pp. 71–84, 1999.

[35] M. E. Ferrari, D. Rusalov, J. Enas, and C. J. Wheeler,“Synergy between cationic lipid and co-lipid determines themacroscopic structure and transfection activity of lipoplexes,”Nucleic Acids Research, vol. 30, no. 8, pp. 1808–1816, 2002.

[36] J.-S. Remy, C. Sirlin, P. Vierling, and J.-P. Behr, “Genetransfer with a series of lipophilic DNA-binding molecules,”Bioconjugate Chemistry, vol. 5, no. 6, pp. 647–654, 1994.

[37] E. Uchida, H. Mizuguchi, A. Ishii-Watabe, and T. Hayakawa,“Comparison of the efficiency and safety of non-viral vector-mediated gene transfer into a wide range of human cells,”Biological and Pharmaceutical Bulletin, vol. 25, no. 7, pp. 891–897, 2002.

[38] R. W. Malone, P. L. Felgner, and I. M. Verma, “Cationicliposome-mediated RNA transfection,” Proceedings of theNational Academy of Sciences of the United States of America,vol. 86, no. 16, pp. 6077–6081, 1989.

[39] T. Ren, Y. K. Song, G. Zhang, and D. Liu, “Structural basisof DOTMA for its high intravenous transfection activity inmouse,” Gene Therapy, vol. 7, no. 9, pp. 764–768, 2000.


[40] J. H. Felgner, R. Kumar, C. N. Sridhar et al., “Enhanced genedelivery and mechanism studies with a novel series of cationiclipid formulations,” Journal of Biological Chemistry, vol. 269,no. 4, pp. 2550–2561, 1994.

[41] N. J. Zuidam and Y. Barenholz, “Electrostatic and structuralproperties of complexes involving plasmid DNA and cationiclipids commonly used for gene delivery,” Biochimica etBiophysica Acta, vol. 1368, no. 1, pp. 115–128, 1998.

[42] J. Zabner, A. J. Fasbender, T. Moninger, K. A. Poellinger, andM. J. Welsh, “Cellular and molecular barriers to gene transferby a cationic lipid,” Journal of Biological Chemistry, vol. 270,no. 32, pp. 18997–19007, 1995.

[43] J.-P. Yang and L. Huang, “Time-dependent maturation ofcationic liposome-DNA complex for serum resistance,” GeneTherapy, vol. 5, no. 3, pp. 380–387, 1998.

[44] C. Marchini, M. Montani, A. Amici et al., “Structural stabilityand increase in size rationalize the efficiency of lipoplexes inserum,” Langmuir, vol. 25, no. 5, pp. 3013–3021, 2009.

[45] N. J. Zuidam and Y. Barenholz, “Electrostatic parametersof cationic liposomes commonly used for gene delivery asdetermined by 4-heptadecyl-7-hydroxycoumarin,” Biochimicaet Biophysica Acta, vol. 1329, no. 2, pp. 211–222, 1997.

[46] P. S. Ajmani and J. A. Hughes, “3β [N-(NM′, N′-dimethylaminoethane)-carbamoyl] cholesterol (DC-chol)-mediated gene delivery to primary rat neurons: characteriza-tion and mechanism,” Neurochemical Research, vol. 24, no. 5,pp. 699–703, 1999.

[47] S. Jain, G. Zon, and M. Sundaralingam, “Base only bindingof spermine in the deep groove of the A-DNA octamerd(GTGTACAC),” Biochemistry, vol. 28, no. 6, pp. 2360–2364,1989.

[48] T. Boukhnikachvili, O. Aguerre-Chariol, M. Airiau, S. Lesieur,M. Ollivon, and J. Vacus, “Structure of in-serum transfectingDNA-cationic lipid complexes,” FEBS Letters, vol. 409, no. 2,pp. 188–194, 1997.

[49] J. M. Metselaar, P. Bruin, L. W. T. De Boer et al., “Anovel family of L-amino acid-based biodegradable polymer-lipid conjugates for the development of long-circulatingliposomes with effective drug-targeting capacity,” BioconjugateChemistry, vol. 14, no. 6, pp. 1156–1164, 2003.

[50] M. L. Immordino, F. Dosio, and L. Cattel, “Stealth liposomes:review of the basic science, rationale, and clinical applications,existing and potential,” International Journal of Nanomedicine,vol. 1, no. 3, pp. 297–315, 2006.

[51] J.-K. Kim, S.-H. Choi, C.-O. Kim, J.-S. Park, W.-S. Ahn,and C.-K. Kim, “Enhancement of polyethylene glycol (PEG)-modified cationic liposome-mediated gene deliveries: effectson serum stability and transfection efficiency,” Journal ofPharmacy and Pharmacology, vol. 55, no. 4, pp. 453–460, 2003.

[52] D. Needham, T. J. McIntosh, and D. D. Lasic, “Repulsiveinteractions and mechanical stability of polymer-grafted lipidmembranes,” Biochimica et Biophysica Acta, vol. 1108, no. 1,pp. 40–48, 1992.

[53] M. DeCastro, Y. Saijoh, and G. C. Schoenwolf, “Optimizedcationic lipid-based gene delivery reagents for use in devel-oping vertebrate embryos,” Developmental Dynamics, vol. 235,no. 8, pp. 2210–2219, 2006.

[54] F. Shi, L. Wasungu, A. Nomden et al., “Interference ofpoly(ethylene glycol)-lipid analogues with cationic-lipid-mediated delivery of oligonucleotides; role of lipid exchange-ability and non-lamellar transitions,” Biochemical Journal, vol.366, no. 1, pp. 333–341, 2002.

[55] O. Ishida, K. Maruyama, H. Tanahashi et al., “Liposomes bear-ing polyethyleneglycol-coupled transferrin with intracellular

targeting property to the solid tumors in vivo,” PharmaceuticalResearch, vol. 18, no. 7, pp. 1042–1048, 2001.

[56] A. Mukherjee, T. K. Prasad, N. M. Rao, and R. Banerjee,“Haloperidol-associated stealth liposomes: a potent carrierfor delivering genes to human breast cancer cells,” Journal ofBiological Chemistry, vol. 280, no. 16, pp. 15619–15627, 2005.

[57] S. Simoes, V. Slepushkin, R. Caspar, M. C. Pedroso de Lima,and N. Duzgunes, “Gene delivery by negatively chargedternary complexes of DNA, cationic liposomes and transferrinor fusigenic peptides,” Gene Therapy, vol. 5, no. 7, pp. 955–964,1998.

[58] S. Chesnoy and L. Huang, “Structure and function of lipid-DNA complexes for gene delivery,” Annual Review of Bio-physics and Biomolecular Structure, vol. 29, pp. 27–47, 2000.

[59] I. S. Zuhorn, U. Bakowsky, E. Polushkin et al., “Nonbilayerphase of lipoplex-membrane mixture determines endosomalescape of genetic cargo and transfection efficiency,” MolecularTherapy, vol. 11, no. 5, pp. 801–810, 2005.

[60] M. J. Bennett, M. H. Nantz, R. P. Balasubramaniam, D. C.Gruenert, and R. W. Malone, “Cholesterol enhances cationicliposome-mediated DNA transfection of human respiratoryepithelial cells,” Bioscience Reports, vol. 15, no. 1, pp. 47–53,1995.

[61] K. Hong, W. Zheng, A. Baker, and D. Papahadjopoulos,“Stabilization of cationic liposome-plasmid DNA complexesby polyamines and poly(ethylene glycol)-phospholipid conju-gates for efficient in vivo gene delivery,” FEBS Letters, vol. 400,no. 2, pp. 233–237, 1997.

[62] S. Kawakami, F. Yamashita, M. Nishikawa, Y. Takakura, and M.Hashida, “Asialoglycoprotein receptor-mediated gene transferusing novel galactosylated cationic liposomes,” Biochemicaland Biophysical Research Communications, vol. 252, no. 1, pp.78–83, 1998.

[63] J. Marshall, N. S. Yew, S. J. Eastman, C. Jiang, R. K. Scheule,and S. H. Cheng, “Cationic lipid-mediated gene delivery to theairways,” in Nonviral Vectors for Gene Therapy, L. Huang, M.-C. Hung, and E. Wagner, Eds., pp. 39–68, Academic Press, SanDiego, Calif, USA, 1999.

[64] K. Lappalainen, I. Jaaskelainen, K. Syrjanen, A. Urtti, and S.Syrjanen, “Comparison of cell proliferation and toxicity assaysusing two cationic liposomes,” Pharmaceutical Research, vol.11, no. 8, pp. 1127–1131, 1994.

[65] S. Dokka, D. Toledo, X. Shi, V. Castranova, and Y. Rojanasakul,“Oxygen radical-mediated pulmonary toxicity induced bysome cationic liposomes,” Pharmaceutical Research, vol. 17,no. 5, pp. 521–525, 2000.

[66] C. Srinivasan and D. J. Burgess, “Optimization and charac-terization of anionic lipoplexes for gene delivery,” Journal ofControlled Release, vol. 136, no. 1, pp. 62–70, 2009.

[67] F. Roerdink, N. M. Wassef, E. C. Richardson, and C. R. Alving,“Effects of negatively charged lipids on phagocytosis of lipo-somes opsonized by complement,” Biochimica et BiophysicaActa, vol. 734, no. 1, pp. 33–39, 1983.

[68] A. Lakkaraju, J. M. Dubinsky, W. C. Low, and Y.-E. Rahman,“Neurons are protected from excitotoxic death by p53 anti-sense oligonucleotides delivered in anionic liposomes,” Journalof Biological Chemistry, vol. 276, no. 34, pp. 32000–32007,2001.

[69] S. D. Patil, D. G. Rhodes, and D. J. Burgess, “Biophysical char-acterization of anionic lipoplexes,” Biochimica et BiophysicaActa, vol. 1711, no. 1, pp. 1–11, 2005.

[70] D. D. Lasic and N. S. Templeton, “Liposomes in gene therapy,”Advanced Drug Delivery Reviews, vol. 20, no. 2-3, pp. 221–266,1996.


[71] V. I. Kulkarni, V. S. Shenoy, S. S. Dodiya, T. H. Rajyaguru,and R. R. Murthy, “Role of calcium in gene delivery,” ExpertOpinion on Drug Delivery, vol. 3, no. 2, pp. 235–245, 2006.

[72] J. Rejman, V. Oberle, I. S. Zuhorn, and D. Hoekstra,“Size-dependent internalization of particles via the pathwaysof clathrin-and caveolae-mediated endocytosis,” BiochemicalJournal, vol. 377, no. 1, pp. 159–169, 2004.

[73] R. Koynova and R. C. MacDonald, “Natural lipid extracts andbiomembrane-mimicking lipid compositions are disposedto form nonlamellar phases, and they release DNA fromlipoplexes most efficiently,” Biochimica et Biophysica Acta, vol.1768, no. 10, pp. 2373–2382, 2007.

[74] G. Scherphof, F. Roerdink, M. Waite, and J. Parks, “Disintegra-tion of phosphatidylcholine liposomes in plasma as a resultof interaction with high-density lipoproteins,” Biochimica etBiophysica Acta, vol. 542, no. 2, pp. 296–307, 1978.

[75] S. J. Comiskey and T. D. Heath, “Serum-induced leakage ofnegatively charged liposomes at nanomolar lipid concentra-tions,” Biochemistry, vol. 29, no. 15, pp. 3626–3631, 1990.


Research Article

Liposome Model Systems to Study the Endosomal Escapeof Cell-Penetrating Peptides: Transport across PhospholipidMembranes Induced by a Proton Gradient

Fatemeh Madani,1 Alex Peralvarez-Marın,2, 3 and Astrid Graslund1

1 Department of Biochemistry and Biophysics, Arrhenius Laboratories for Natural Sciences, Stockholm University,10691 Stockholm, Sweden

2 Department of Anesthesia, Brigham and Women’s Hospital, Boston, MA 02115, USA3 Centre d’Estudis Biofısics, Universitat Autonoma de Barcelona, 08193 Bellaterra, Spain

Correspondence should be addressed to Astrid Graslund, [email protected]

Received 30 June 2010; Accepted 17 November 2010

Academic Editor: Ali Nokhodchi

Copyright © 2011 Fatemeh Madani et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

Detergent-mediated reconstitution of bacteriorhodopsin (BR) into large unilamellar vesicles (LUVs) was investigated, and theeffects were carefully characterized for every step of the procedure. LUVs were prepared by the extrusion method, and theirsize and stability were examined by dynamic light scattering. BR was incorporated into the LUVs using the detergent-mediatedreconstitution method and octyl glucoside (OG) as detergent. The result of measuring pH outside the LUVs suggested that in thepresence of light, BR pumps protons from the outside to the inside of the LUVs, creating acidic pH inside the vesicles. LUVs with20% negatively charged headgroups were used to model endosomes with BR incorporated into the membrane. The fluorescein-labeled cell-penetrating peptide penetratin was entrapped inside these BR-containing LUVs. The light-induced proton pumpingactivity of BR has allowed us to observe the translocation of fluorescein-labeled penetratin across the vesicle membrane.

1. Introduction

Live cells are protected from the surrounding environmentby the cell membrane, which only allows compounds with asmall molecular size to pass this barrier into the cell. Somedrug molecules, on the other hand, are large hydrophilicmolecules, and this creates major limitations for theirpenetration through the cell membrane. With discoveryof cell-penetrating peptides (CPPs), the transport of suchmolecules can be accomplished. Generally, CPPs are definedas short, water soluble and partly hydrophobic and/orpolybasic peptides (at most 30–35 amino acid residues)with a net positive charge at physiological pH [1]. Thisnew class of peptides was introduced in the late 1980s bythe discovery of the human immunodeficiency virus type 1(HIV-1) encoded Tat peptide [2, 3] and the amphiphilicDrosophila Antennapedia homeodomain-derived 16 aminoacid penetratin peptide (pAntp), which was discovered

somewhat later [4–7]. These two peptides are the mostextensively studied of all CPPs.

The main feature of CPPs is that they are able to penetratethe cell membrane at low micromolar concentrations in vivoand in vitro without using any chiral receptors and withoutcausing irreversible membrane damage. These peptides arecapable of internalizing electrostatically or covalently boundbiologically active cargoes such as drugs, with high efficiencyand low toxicity [1, 8].

Despite many studies made on CPPs, the mecha-nism(s) by which CPPs enter the cells has not been com-pletely resolved. There is some evidence for both energy-independent processes and endocytosis in internalizationof CPPs. Presently, endocytosis, composed of two steps,endocytotic entry followed by endosomal escape, is believedto be the most common uptake mechanism at low CPPconcentrations [8, 9].


Model biomembranes or lipid bilayers are efficient modelsystems to investigate the CPPs translocation mechanism(s).Large unilamellar vesicles (LUVs) are among the most com-monly used model membranes in lipid-peptide interactionstudies [10].

Here, we have performed experiments to study thebackground mechanism(s) of endosomal escape. Cell mem-branes are normally weakly negatively charged and consistof different phospholipid molecules and associated proteinsand proteoglycans. The lipids used in our study (a mix-ture of zwitterionic POPC and negatively charged POPGphospholipids) have been chosen to mimic cell membranes.Bacteriorhodopsin (BR) reconstituted into LUVs with 20%negatively charged phospholipid are used to model theendosomes. The LUVs were prepared by the extrusionmethod, and their size and stability were carefully examinedby dynamic light scattering (DLS).

BR is an integral membrane protein of about 26 KDafound in Halobacterium salinarium. There are various meth-ods to reconstitute membrane proteins into the vesiclesincluding organic solvent-mediated reconstitution, directincorporation into preformed liposomes, mechanical means,and the detergent-mediated reconstitution method. Amongthese methods, detergent-mediated reconstitution is themost common and successful technique to incorporatemembrane proteins into vesicles [11]. The final orientationof the protein incorporated into the vesicle bilayers dependson several factors; one of the most critical is the detergentcomposition in the proteoliposomes [12]. When BR absorbslight, it pumps protons in a direction that depends onthe direction of protein insertion into the membrane andgenerates an H+ gradient and membrane potential [13]. Thedetergent-mediated reconstitution method can provide 95%inside-out orientation of BR in the bilayer indicating thatBR pumps protons from the outside to the inside of vesicles[11]. In the following, some practical aspects crucial for thereproducibility of the method are described. Furthermore,we have studied the translocation ability of fluorescein-labeled penetratin in the presence of a pH gradient acrossan LUV membrane.


2.1. Materials. 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-choline (POPC) and 1-palmitoyl-2-oleoyl-sn-glycero-3[phospho-rac-(1-glycerol)] (POPG) used in this study wereobtained from Avanti Polar Lipids (Alabaster, Alabama,USA) and were used without any extra purification. Thedetergent n-octyl-β-D-glucopyranoside (OG) was fromGlycon Biochemicals (Luckenwalde, Germany). PD-10desalting columns were purchased from GE Healthcare(Buckinghamshire, UK). Bio-Beads were from BIO-RAD(California, USA). Fluorescein-labeled penetratin wasproduced by Neosystem Laboratories (Strasbourg, France).Halobacterium salinarum strain S9 was a generous giftfrom Professor Esteve Padros (Universitat Autonoma deBarcelona, Spain). Bacteriorhodopsin (BR) was producedand purified essentially according to a published protocol

[14]. A UV-Vis absorption spectrum of the purified BRwas recorded within the 800–250 nm range to check thepurity of the sample and to calculate the concentration(ε = 62700 M−1 cm−1 at 568 nm, MW = 26000 Da).Aliquots at the desired concentration were stored at −20◦C.

2.2. Vesicle Preparation. The extrusion method is a commonmethod for vesicle preparation, which produces LUVs witha narrow size distribution [15]. We used a hand-drivenextrusion apparatus with one milliliter capacity. In thismethod, 20% negatively charged LUVs are prepared bydissolving the lipids (neutral POPC and negatively chargedPOPG) at the total concentration of 20 mM in chloroformto obtain a homogeneous mixture of the lipids. Then, thesolvent is removed by evaporation under high vacuumfor 3 hr. The resulting dried lipid film is resuspended byadding a buffer solution (20 mM phosphate buffer, 100 mMKCl, pH 7.2). This liposomal suspension is then vortexedfor 10 minutes followed by 5 freeze-thaw cycles to reducethe lamellarity and obtain more aqueous trapped volumes.After the freezing and thawing cycles, the lipid suspensioncontaining multilamellar vesicles is pushed through twopolycarbonate filters (100 nm pore size) 20 times by using anAvanti manual extruder. This results in LUVs with a well-defined and homogeneous size.

2.3. Reconstitution of BR into LUVs: Detergent-Mediated Re-constitution Method. The preparation of BR-reconstitutedLUVs consists of three steps: vesicle solubilization, BR addi-tion, and detergent removal [11, 12, 16].

2.3.1. Vesicle Solubilization. LUVs prepared by extrusionmethod were diluted in the buffer used for their preparationto the desired concentration. Here, we have used 2.3 mLvesicle suspension of 5 mM phospholipid concentration.After the addition of the detergent, LUV solubilization takesplace in three stages (Figure 1); first the detergent monomersdiffuse among bilayers, and at the same time there are somefree detergent monomers in the solution (stage I). The per-meability, size, and stability of the LUVs will change. Furtheraddition of detergent saturates the vesicle bilayer. At stage II,when free detergent monomer concentration reaches its cmcvalue, transition from monomers to mixed lipid/detergentmicelles will occur. At this step, both saturated vesicles andmixed micelles coexist. Stage III is the point where all LUVshave disappeared and only mixed micelles are present in thesolution.

The choice of detergent and its concentration affect thisthree-stage mechanism. In the present paper, octyl glucoside(OG) has been used. OG is a nonionic detergent with a cmcvalue of about 25 mM that facilitates its removal [17]. Here,after adding OG, the final concentrations of lipid and OGwere 4.8 mM and 25.6 mM, respectively.

2.3.2. BR Addition. After 5–10 min of the vesicle solubiliza-tion, BR monomers resulting from detergent solubilizationof purple membrane (BR 1 mg/mL, OG 100 mM) were addedto the solubilized LUVs suspension and incubated for 5 to


Lipid

Detergent

Onset ofsolubilization

Totalsolubilization

Solubility(detergent concentration)

Addition of BR

Detergent removal

LUV reconstituted BR

Figure 1: Scheme for the detergent-mediated reconstitution of BR into LUVs (after [11]). Stage I–III: Gradual addition of detergent to LUVs.For optimal reconstitution efficiency, BR should be added during stage II. Detergent is removed by Bio-Beads, and the result is an LUV withincorporated BR.

10 minutes. The resulting suspension should be a mixture ofBR/lipid/detergent vesicles and lipid/detergent micelles withthe final concentrations of 4 μM, 4.3 mM, and 29 mM forBR, lipid, and detergent, respectively. At this stage, BR maybe incorporated into the vesicles which have been saturatedand destabilized by the detergent. As suggested also in [11],by varying the detergent/lipid ratio in the BR incorporationprocess, we found that the partly detergent-saturated LUVsare optimal in reconstitution of BR. The detergent-BR-phospholipid mixtures were kept at room temperature for5 min to 15 min, and the detergent was then removed.

2.3.3. Detergent Removal. The method of detergent removalhighly affects the results of the reconstitution process.High proton pumping activity of BR-reconstituted vesiclesrequires sealed vesicles which result from removing allresidual detergents from the suspension. Any remainingdetergent may alter the size, permeability, and stability of thevesicles produced by detergent removal from mixed micelles.In addition, the rate of detergent removal is another factoraffecting the reconstitution process. It was observed that bothinitial rates and total amounts of H+ pumping decrease whenthe rate of detergent removal increases [12, 16].

There are several strategies to remove detergent frommixed lipid/protein/detergent vesicles. The nature of the

detergent affects the method that has to be employed. Bio-Beads can absorb almost any kind of detergents with awide range of cmc values. For example, Triton-X with alow cmc value cannot be easily removed by the dialysismethod. However, absorption by hydrophobic Bio-Beadsmay efficiently remove even low cmc value detergents [18].

Detergent removal should best be performed in twosteps: first wet Bio-Beads (80 mg/mL) were directly added tothe BR-lipid-detergent suspension. The mixture was lightlystirred at room temperature. Transition from micellar tolamellar may take place at this stage. After 3 hr of incubationat room temperature, a second portion of slightly wet beadswas added and mixed overnight with a small shaker and therate of around 400 rpm to remove residual detergents. At theend, two PD-10 columns were used to remove Bio-Beads andresidual detergents from the sample.

2.4. pH Measurement. In order to monitor the pH changesoutside the vesicle, we prepared an experiment using aXenon lamp to illuminate the sample and the pH meter(PHM 93 Reference pH meter and Thermo Scientific model320 electrode) to record the values of the pH. The BR-reconstituted vesicle suspension was equilibrated in 120 mMKCl pH 7.4 buffer using a PD-10 column. The BR-sample waskept in the dark at least 30 min to ensure the dark adaptationof the sample, and the pH was recorded in the dark as


the baseline. Light-induced pH changes of BR-reconstitutedLUVs were measured in a cuvette under agitation.

2.5. Preparation of CPPs-Entrapping LUVs. A 20 μM fluo-rescein-labeled penetratin solution was prepared in 20 mMpotassium phosphate, 100 mM KCl (pH 7.2), and 100 mMpotassium iodide (KI) used as a quencher. LUVs containingthe peptide were prepared as described earlier by using thissolution as buffer. At this stage, BR may be introducedinto the LUVs according to the procedure described above.Finally, the LUV suspension was washed twice using two PD-10 columns to remove non-encapsulated fluorescein-labeledpenetratin and quencher from the outside of the LUVs. It isimportant to remove components outside the vesicles (e.g.,peptides or quencher) after the detergent removal stage sincedetergent changes the membrane permeability, and it is notworth removing them before this stage. KI was used toquench and minimize the background fluorescence intensity.Thus, any increase in background fluorescence is due to theleakage of the labeled peptide from the LUVs.

At the end of this preparation, the sample had a totallipid concentration estimated to about 2.3 mM. Based onvesicle geometry (diameter 100 nm) each vesicle containedabout 105 lipids. This would yield an approximate vesicleconcentration of 2.3∗ 10−8 M. In BR-containing vesicles, theBR concentration in the sample was estimated by measuringthe light absorption after treatment with Triton-X, using theε of 62700 M−1 cm−1 at 568 nm. The concentration of BRwas found to be around 1 μM corresponding to about 40 BRmolecules inserted per vesicle.

2.6. Fluorescence Spectroscopy and CPP Leakage Study. Tostudy the effect of a pH gradient on the CPP escape fromLUVs, we used fluorescence spectroscopy. Fluorescence wasmeasured in a Horiba Jobin Yvon Fluorolog-3 spectrometerusing the DataMax operating software and with a 4 ∗10 mm quartz cuvette. The sample was excited at 494 nm,and its emission was scanned from 505 to 550 nm with 1 nmemission and excitation bandwidths. All experiments wererun at 20◦C.

2.7. Circular Dichroism (CD) Spectroscopy. CD was used todetermine the secondary structure of the BR reconstitutedin the vesicles. CD spectra were recorded on a ChirascanCD spectrometer at 20◦C. Wavelengths between 190 nmand 260 nm were recorded, using a bandwidth of 2 nm.A quartz cuvette with an optical path length of 2 mmwas used, requiring approximately 500 μL of sample. Thetemperature was adjusted using a TC 125 temperaturecontrol. The background spectra of the vesicle solution weresubtracted from the peptide spectra. Spectra were collectedand averaged over ten measurements.

2.8. Dynamic Light Scattering (DLS). DLS was used todetermine the hydrodynamic radius of the vesicle and BR-reconstituted vesicles. Measurements were carried out usinga light scattering instrument ALV/CGS-3 equipped witha Light Scattering Electronics and Multiple Tau Digital

1E + 041E + 031E + 021E + 01

Unweighted radius (nm)−2E − 01

0E + 00

2E − 01

4E − 01

6E − 01

8E − 01

1E + 00

1.2E + 00

Nor

mal

ized

prob

abili

ty

(a)

1E + 081E + 061E + 041E + 021E + 00

Unweighted radius (nm)−2E − 01

0E + 00

2E − 01

4E − 01

6E − 01

8E − 01

1E + 00

1.2E + 00

Nor

mal

ized

prob

abili

ty

(b)

Figure 2: Unweighted size distribution for (a) LUVs and (b) BR-reconstituted LUVs.

correlator ALV/LSE-5004. Correlation data were acquiredtypically for 3 runs each for 30 sec. Correlation functionsat 150◦ were recorded at the temperature of 20◦C using aJulabo temperature control. The hydrodynamic radius wascalculated using the ALV software, unweighted fitting.


We prepared 20% negatively charged LUVs composed of80% POPC and 20% POPG by the extrusion technique.BR was reconstituted into the LUVs using the detergent-mediated reconstitution method. The resulting LUVs with-out and with BR were characterized using dynamic lightscattering. As shown in Figures 2(a) and 2(b), both sampleshave a relatively homogenous population with slightlydifferent vesicle sizes. According to the detergent-mediatedreconstitution method [11], BR is oriented in the membraneof LUVs such that it pumps protons from the outside to theinside of the vesicles upon illumination.

In addition, we used CD spectroscopy to determinewhether the reconstitution process affected the secondarystructure of BR (Figure 3). The secondary structure of BRwas dominated by α-helix both in free solution and whenreconstituted into the LUVs.

Upon illumination of the sample, the measured pHincreased and reached a maximum on the outside of theLUVs (Figure 4) indicating that most of the BR moleculesare properly oriented towards the interior of the vesicles.


255245235225215205195

Wavelength (nm)

−100

−50

0

50

100

150

[θ]/

10−3

(deg

cm2dm

ol−1

)

Figure 3: Circular dichroism spectra of 2 μM BR in 100 mM OGdetergent (black) and reconstituted LUVs (red) at 20◦C.

Light Dark Light Dark

400350300250200150100500

Time (min)

0

0.1

0.2

0.3

0.4

0.5

ΔpH

Figure 4: pH changes outside the BR-reconstituted vesicle as a func-tion of illumination time. Conditions: 20 mM potassium phosphatebuffer and 100 mM KCl, pH 7.2 inside the 20% negatively chargedLUVs, 120 mM KCl outside the LUVs, pH = 7.4, 25◦C.

Translocation of protons from the outside to the insidevesicles resulted in a more acidic pH inside the LUVs. Whenillumination was discontinued, the measured pH outside theLUVs decreased, indicating that protons leaked out againacross the membrane and reached an equilibrium (Figure 4).As shown in this figure, the proton pumping process can berepeated with the same sample.

We have repeated the experiment in the absence of BR toinvestigate whether this effect observed is due to the protonpumping of BR or some other effects. No changes in pH wereobserved upon illumination of LUVs in the absence of BRwhich indicates that light-induced pH changes are indeeddue to the proton pumping of BR (data not shown).

change in pH (ΔpH) outside the vesicles can be used tocalculate the corresponding ΔpH inside the vesicles based onproton concentration and the estimated inner volume of allvesicles in the solution. A ΔpH outside the vesicles of +0.2after 25 min corresponds to almost −2 pH units inside thevesicles under the conditions used here.

We also evaluated the effect of the pH gradient onthe translocation abilities of the fluorescein-labeled CPPpenetratin. BR with the inside-out orientation was recon-stituted into LUVs. Upon illumination, BR pumps protonsinto the LUVs creating a pH gradient over the membrane.Fluorescein-labeled penetratin together with KI as a quench-er was enclosed in the BR-reconstituted LUVs. Figure 5

Background5 min in dark60 min in dark

545535525515505

Wavelength (nm)

0

10000

20000

30000

40000

50000

60000

70000

Inte

nsi

ty

(a)

Background20 min45 min

65 min85 min105 min

545535525515505

Wavelength (nm)

0

20000

40000

60000

80000

100000

120000

Inte

nsi

ty

(b)

Figure 5: Fluorescence changes of the sample containing BR-LUVswith fluorescein-labeled penetratin and fluorescence quencher KIinside the vesicles. Changes in fluorescence intensity between 505and 550 nm (excitation wavelength 494 nm) were recorded in the(a) absence of illumination and (b) in the presence of light forindicated periods of time. Conditions: 20 mM potassium phos-phate, 100 mM KCl buffer, 20 μM fluorescein-labeled penetratin,and 100 mM KI inside the LUVs, total lipid concentration 2.3 mM,pH 7.4, and 20◦C.

shows the fluorescence intensity changes of the samplecontaining BR-reconstituted LUVs and fluorescein-labeledpenetratin together with fluorescence quencher KI inside theLUVs.

In the dark, we observed no changes in the fluorescenceintensity, indicating insignificant leakage of the peptide outof the LUVs. The peptides are not able to translocate acrossthe membrane without any promoting proton gradient(Figure 5(a)).


Efficient peptide escape was observed in the presence ofthe light. A significant increase in the fluorescence intensitywas observed when the sample was illuminated. This resultindicates that a pH gradient across the membrane enhancesthe vesicular escape for the examined fluorescein-labeledCPP (Figure 5(b)).

Longer period of illumination leads to more leakage ofthe CPP. However, after around 100 min, it reaches an almoststable condition, corresponding to the transport of around30% of the fluorescein-labeled penetratin out of the LUVs(data not shown). In this experiment, a 100% release wasachieved when the fluorescence intensity was measured afteraddition of 10% (w/v) Triton X-100 detergent.

4. Conclusions

Our studies show that there are several factors affectingthe result of the proton pumping experiment, starting fromvesicle preparation to the detergent removal. The mostimportant of these factors concerns the permeability andstability of the LUVs which strongly affects the protonpumping activity and hence pH gradient of the resulting BR-vesicles. Leaking vesicles display lower pH gradient due to theproton leakage from the membrane. Further, it is importantto use lipids with high purity and to ensure complete removalof the detergent. Finally, one should examine the vesicles byDLS to verify their homogeneity and size.

The degree of orientation of the BR incorporated intothe LUVs also affects the proton pumping efficiency. It hasbeen shown that 95% inside-out orientation will be achievedusing the detergent-mediated reconstitution method. How-ever, this percentage strongly depends on the experimentalconditions, for example, detergent to lipid ratio and the timepoint, where BR will be added to the LUVs [11].

Overall, our observations are in agreement with the ear-lier preliminary results with labeled penetratin by Bjorklundet al. [19].

Use of an ionophore nigericin is another alternative tocreate acidic pH inside the vesicles [20]. It works by exchang-ing K+ for H+ across the vesicle membrane and creating atransmembrane pH gradient. However, the effect of nigericinis dependent on the presence of high concentrations of aK+ salt inside the vesicles. To create a transmembrane saltgradient, metal ions have to be removed from outside thevesicles by passing through the columns equilibrated by highconcentrations of, for example, sucrose. High-concentratedsugar and metal ions may destabilize the vesicles resulting inleakage of the protons and hence decreasing the pH gradient.

The light-induced BR proton pumping experiment hasthe advantages that (1) it does not require any special bufferwhich alters the vesicle stability, (2) one is able to controlpumping activity by the illumination time period, and (3)several experiments can be carried out with the same samplerepeating dark-illumination cycles. The present studies alsosuggest a general mechanism by which positively chargedmolecules, other than peptides, may enter into cells byendocytotic uptake followed by escape from the acidifiedendosome.

Acknowledgments

This study was supported by the Swedish Research Council(to A. Graslund) and the Swedish Foundation for StrategicResearch (Project no. MDB09-0015). The authors want tothank Professor Esteve Padros from Universitat Autonomade Barcelona for the generous gift of the strain S9 ofH. salinarum. The authors also want to acknowledge thefunding from the European Union (Marie Curie ActionPIOF-GA-2009-237120 to A. Peralvarez-Marın).

References

[1] P. Jarver and U. Langel, “Cell-penetrating peptides—a briefintroduction,” Biochimica et Biophysica Acta, vol. 1758, no. 3,pp. 260–263, 2006.

[2] M. Green and P. M. Loewenstein, “Autonomous functionaldomains of chemically synthesized human immunodeficiencyvirus tat trans-activator protein,” Cell, vol. 55, no. 6, pp. 1179–1188, 1988.

[3] E. Vives, P. Brodin, and B. Lebleu, “A truncated HIV-1 Tatprotein basic domain rapidly translocates through the plasmamembrane and accumulates in the cell nucleus,” Journal ofBiological Chemistry, vol. 272, no. 25, pp. 16010–16017, 1997.

[4] A. Joliot, C. Pernelle, H. Deagostini-Bazin, and A. Prochiantz,“Antennapedia homeobox peptide regulates neural morpho-genesis,” Proceedings of the National Academy of Sciences of theUnited States of America, vol. 88, no. 5, pp. 1864–1868, 1991.

[5] A. H. Joliot, A. Triller, M. Volovitch, C. Pernelle, and A.Prochiantz, “α-2,8-polysialic acid is the neuronal surfacereceptor of antennapedia homeobox peptide,” New Biologist,vol. 3, no. 11, pp. 1121–1134, 1991.

[6] D. Derossi, A. H. Joliot, G. Chassaing, and A. Prochiantz, “Thethird helix of the Antennapedia homeodomain translocatesthrough biological membranes,” Journal of Biological Chem-istry, vol. 269, no. 14, pp. 10444–10450, 1994.

[7] D. Derossi, S. Calvet, A. Trembleau, A. Brunissen, G.Chassaing, and A. Prochiantz, “Cell internalization of thethird helix of the antennapedia homeodomain is receptor-independent,” Journal of Biological Chemistry, vol. 271, no. 30,pp. 18188–18193, 1996.

[8] S. El-Andaloussi, T. Holm, and U. Langel, “Cell-penetratingpeptides: mechanisms and applications,” Current Pharmaceu-tical Design, vol. 11, no. 28, pp. 3597–3611, 2005.

[9] I. Nakase, H. Hirose, G. Tanaka et al., “Cell-surface accumu-lation of flock house virus-derived peptide leads to efficientinternalization via macropinocytosis,” Molecular Therapy, vol.17, no. 11, pp. 1868–1876, 2009.

[10] M. Magzoub and A. Graslund, “Cell-penetrating peptides:small from inception to application,” Quarterly Reviews ofBiophysics, vol. 37, no. 2, pp. 147–195, 2004.

[11] J. L. Rigaud, B. Pitard, and D. Levy, “Reconstitution ofmembrane proteins into liposomes: application to energy-transducing membrane proteins,” Biochimica et BiophysicaActa, vol. 1231, no. 3, pp. 223–246, 1995.

[12] J. L. Rigaud, M. T. Paternostre, and A. Bluzat, “Mechanismsof membrane protein insertion into liposomes during recon-stitution procedures involving the use of detergents. 2. Incor-poration of the light-driven proton pump bacteriorhodopsin,”Biochemistry, vol. 27, no. 8, pp. 2677–2688, 1988.

[13] J. Heberle, “Proton transfer reactions across bacteriorhodop-sin and along the membrane,” Biochimica et Biophysica Acta,vol. 1458, no. 1, pp. 135–147, 2000.


[14] D. Oesterhelt and W. Stoeckenius, “Isolation of the cellmembrane of Halobacterium halobium and its fractionationinto red and purple membrane,” Methods in Enzymology, vol.31, pp. 667–678, 1974.

[15] L. D. Mayer, M. J. Hope, and P. R. Cullis, “Vesicles of variablesizes produced by a rapid extrusion procedure,” Biochimica etBiophysica Acta, vol. 858, no. 1, pp. 161–168, 1986.

[16] M. T. Paternostre, M. Roux, and J. L. Rigaud, “Mecha-nisms of membrane protein insertion into liposomes duringreconstitution procedures involving the use of detergents. 1.Solubilization of large unilamellar liposomes (prepared byreverse-phase evaporation) by Triton X-100, octyl glucoside,and sodium cholate,” Biochemistry, vol. 27, no. 8, pp. 2668–2677, 1988.

[17] M. L. Jackson and B. J. Litman, “Rhodopsin-phospholipidreconstitution by dialysis removal of octyl glucoside,” Bio-chemistry, vol. 21, no. 22, pp. 5601–5608, 1982.

[18] J. L. Rigaud, D. Levy, G. Mosser, and O. Lambert, “Deter-gent removal by non-polar polystyrene beads: applicationsto membrane protein reconstitution and two-dimensionalcrystallization,” European Biophysics Journal, vol. 27, no. 4, pp.305–319, 1998.

[19] J. Bjorklund, H. Biverstahl, A. Graslund, L. Maler, and P.Brzezinski, “Real-time transmembrane translocation of pene-tratin driven by light-generated proton pumping,” BiophysicalJournal, vol. 91, no. 4, pp. L29–L31, 2006.

[20] M. Magzoub, A. Pramanik, and A. Graslund, “Modeling theendosomal escape of cell-penetrating peptides: transmem-brane pH gradient driven translocation across phospholipidbilayers,” Biochemistry, vol. 44, no. 45, pp. 14890–14897, 2005.


Research Article

Liposomal Tumor Targeting in Drug Delivery UtilizingMMP-2- and MMP-9-Binding Ligands

Oula Penate Medina,1 Merja Haikola,2 Marja Tahtinen,3 Ilkka Simpura,3 Sami Kaukinen,3

Heli Valtanen,3 Ying Zhu,3 Sari Kuosmanen,3 Wei Cao,1 Justus Reunanen,4 Tuula Nurminen,4

Per E. J. Saris,4 Peter Smith-Jones,1 Michelle Bradbury,1 Steven Larson,1 and Kalevi Kairemo5

1 Department of Radiology, Sloan Kettering Institute for Cancer Research, 1275 York Ave., New York, NY 10065, USA2 Division of Pharmaceutical Technology, University of Helsinki, P.O. Box 56, FI-00014, Finland3 CTT Cancer Targeting Technologies Ltd., Viikinkaari 4 C, FI-00790 Helsinki, Finland4 Department of Food and Environmental Sciences, University of Helsinki, P.O. Box 56, FI-00014, Finland5 International Comprehensive Cancer Center Docrates, Saukonpaadenranta 2, FI-00180 Helsinki, Finland

Correspondence should be addressed to Oula Penate Medina, [email protected]

Received 2 August 2010; Revised 20 October 2010; Accepted 19 November 2010

Academic Editor: Sophia Antimisiaris

Copyright © 2011 Oula Penate Medina et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

Nanotechnology offers an alternative to conventional treatment options by enabling different drug delivery and controlled-releasedelivery strategies. Liposomes being especially biodegradable and in most cases essentially nontoxic offer a versatile platform forseveral different delivery approaches that can potentially enhance the delivery and targeting of therapies to tumors. Liposomespenetrate tumors spontaneously as a result of fenestrated blood vessels within tumors, leading to known enhanced permeabilityand subsequent drug retention effects. In addition, liposomes can be used to carry radioactive moieties, such as radiotracers, whichcan be bound at multiple locations within liposomes, making them attractive carriers for molecular imaging applications. Phagedisplay is a technique that can deliver various high-affinity and selectivity peptides to different targets. In this study, gelatinase-binding peptides, found by phage display, were attached to liposomes by covalent peptide-PEG-PE anchor creating a targeted drugdelivery vehicle. Gelatinases as extracellular targets for tumor targeting offer a viable alternative for tumor targeting. Our findingsshow that targeted drug delivery is more efficient than non-targeted drug delivery.

1. Introduction

Liposomal nanotechnology provides a versatile platform forexploring several approaches that can potentially enhancethe delivery and targeting of therapies to tumors. As abiodegradable and essentially nontoxic platform, liposomescan be used to encapsulate both hydrophilic and hydropho-bic materials and be utilized as drug carriers in drug deliverysystems (DDSs). In addition, liposomes can be used tocarry radioactive moieties, such as radiotracers, which canbe bound at multiple locations within liposomes, makingthem attractive carriers for molecular imaging applications.In this study, gelatinase-binding peptides were attachedto liposomes for synthesizing a targeted drug deliveryvehicle.

For active targeting or drug delivery applications or both,intraliposomal encapsylation of multiple targeting agents ortherapies can be (i) to the lipid bilayer, which can bindhydrophobic conjugates; (ii) to hydrated compartments forwater-soluble components; (iii) by covalent binding directlyor by utilizing spacer to the outer lipid leaflet [1]. Deliveryof these nanoformulations to the reticuloendothelial system(RES) is readily achieved since, given their larger size, theRES traps most conventional liposomes that are not shieldedby polyethylene glycol chains (PEGs) or other similar stericwater carrying substance. RES uptake can be increased byaltering particle surface chemistry and charge, for instance,by adding positively charged lipids or biologically activatingproteins or sugars on the surface of the liposomes. Forpurposes of agent delivery to target organs other than the


RES, long-circulating liposomes have been developed bymodifying the liposomal surface [2]. Determination of thein vivo biodistribution and targeting kinetics of liposome-encapsulated drugs is required for the assessment of drugbioavailability.

The most commonly used nanoformulated drug isCaelyx/Doxil, a liposomal doxorubicin product. It has nearlysupplanted doxorubicin in the therapy of ovarian cancer,breast cancer, and Kaposi’s sarcoma. It differs from theformer generation liposomal delivery systems, as the outersurface of Caelyx/Doxil is coated with PEG chains thatprotect the liposomes from being opsonized by componentsof the immune system in the circulation. These stealth-typeliposomes have longer circulation half-times than those foruncoated liposomes. In addition, they are safer than thenative drugs themselves (e.g., Caelyx/Doxil is not cardiotoxic,a major concern for native doxorubicin delivery).

For cancer-based applications, peptides that can selec-tively detect and target metastatic disease and tumor inva-sive potential may offer critical prognostic information.Metastatic invasion is promoted by the attachment of tumorcells to the extracellular matrix, the degradation of matrixcomponents by tumor-associated proteases, and the cellularmovement into the area modified by protease activity. Matrixmetalloproteases (MMPs) represent a family of enzymescapable of degrading the basement membrane and extracel-lular matrix (ECM), thus contributing to tissue remodelingand cell migration [3–5]. This family of enzymes can cleaveECM proteins, as well as alter the integrity of basementmembranes that serve as barriers to cell movement. This isnormally a tightly regulated process, with the presence ofactivated metalloproteinases occurring only under specificconditions.

MMPs may be divided into subgroups, one comprised oftype IV collagenases (gelatinases) such as MMP-2 and MMP-9, which play major roles in tumor growth, angiogenesis,and metastatic disease. These gelatinases degrade type IVcollagen (and its breakdown product, gelatin) and comprisethe primary structural component of the ECM, enablingtumor cells to gain access to the rest of the body. Overex-pression and/or prognostic significance of gelatinases havebeen examined in a range of cancer types, including ovariancancer [6–9], endometrial and cervical cancer [10, 11], andbreast cancer [12, 13]. High expression levels of gelatinasesin breast and ovarian cancers, for instance, are known to beassociated with unfavorable prognoses.

In this study, binding peptides (BPs) extracted fromMMP-9 were attached to liposomes for synthesizing atargeted drug delivery vehicle. Downregulation of MMP-9 is known to exert inhibitory effects on endothelial cellmigration and tube formation [14]. Intriguingly MMP-2 hasbeen shown to dock on tumor cell-surface integrins, whichmakes gelatinases even more interesting as a target [15].Adenoviral-mediated MMP-9 downregulation was shownto retard endothelial cell migration in cell wounding andspheroid migration assays, resulting in reduced capillary-like tube formation [16]. MMP-2- or MMP-9-deficient micewere found to exhibit abnormal angiogenic properties [17].Further, in human gliomas, immunohistochemical findings

suggested that neoplastic and endothelial cells expressingMMP-9 protein may be associated with tumoral angiogenesis[18].

One of the first known specific gelatinase inhibitors, acyclic MMP-9-binding peptide identified by random phagedisplay libraries (i.e., CTTHWGFTLC peptide later CTT1),has previously been shown to have high affinity not onlyto MMP-9, but also to MMP-2 [19]. The peptide activelyinhibits endothelial and tumor cell migration in vitro andtumor progression in in vivo murine models [19]. Specif-ically, CTT-displaying phages block the formation of newblood vessels, resulting in tumor size reductions and pro-longed overall survival. These findings highlight the potentialof CTT peptide for targeting chemotherapeutics or otherimaging probes to the tumor neovasculature. CTT-peptidehas been used for liposomal drug delivery in vitro and in vivo[20, 21] and has been shown to be effective in improvingselective localization of chemotherapies such as doxorubicinin human tumor cells. In this work we modified a CTTpeptide with a peptide linker that bears a tyrosine moiety foriodination procedures and attached several additional aminoacids (GRENYHG) to enhance freedom of peptide bindingin order to create GRENYHGCTTHWGFTLC-peptide (i.e.,CTT2-peptide) (Figure 1). The synthesis of CTT2-peptideenabled us to retain bioactivity that would otherwise not bepresent if CTT peptide itself was directly linked to lipids orPEG spacers. This is the first study, to our knowledge, thathas utilized a peptide derived from a synthetic phage displaylibrary for constructing a more selective liposomal deliverysystem for targeting extracellular target molecules.

We initially present the synthesis of PEG-PE-CTT2peptide-bound micelles and liposomes. The feasibility of uti-lizing micellar and liposomal nanoformulations as therapeu-tic delivery vehicles to achieve efficacy in ovarian carcinomamodels was explored by attaching the radioiodinated CTTpeptide tracer, 125I-CTT2 peptide, to these platforms andloading them with doxorubicin, an inherently fluorescentchemotherapeutic agent. Biodistribution studies of bothtargeted nanoformulations were performed in normal andimmunosuppressed subcutaneous human xenograft modelsusing the CTT2-peptide.


Reagents. All reagents, unless stated otherwise, wereobtained from Sigma-Aldrich (St Louis, Missouri, USA)and culture media from Gibco Life Technologies (Paisley,Scotland). PEG-PE-NHS was from Avanti Polar Lipids Inc.(Alabama, USA) as all the other lipids used in this study.

2.1. Synthesis of Peptides. Peptides were synthesized on anApplied Biosystems 433A (Foster City, CA, USA) automaticsynthesizer using Fmoc-chemistry. For disulfide generation,peptides were dissolved at 1 mg/ml in 0.05 M ammoniumacetate (pH 8) and mixed with H2O2 for 40 min at roomtemperature so that 0.5 ml of 3% H2O2 was added per100 mg peptide. The peptides were purified by reversed phaseHPLC, and the molecular weight was identified by massspectrometry analysis.


HN

HN

HN

HN

HN

S

S

HO

HO

H2NOC

OH

OHN

N

CONH2

CO2H

NH

NH

NH

NH

NH

NH

HN

HN

HN

HN

NH

NH

NH

NH

NH

O

O O

O

O

O

O

O

O

O

O

O

O

O

O

O

H2N

H2N

HN

I

Figure 1: Chemical structure of 125I-CTT2-peptide. CTT2-peptide is a 17-amino acid peptide with a disulphide bridge between the twocysteines. The amino terminal end of the peptide is amidated to increase its stability. Upon iodination, peptide labeling occurs on thearomatic ring of the tyrosine amino acid.

NH

NH

NH

NH

NH

HN

HN

HN

HN

HN

HN

HN

HN

HN

NH

NH

NH

NH

NH

NH

NH

HN

HNO

O

O

O

OO

OO

O

O O

O O

O

O

O

OO

O

O

OO

O

O

OO

O

S

S

HO

HO

H2N H2NOC

OH

OHN

N

CONH2

CO2H

P−O

n = 77

DSPE PEG3400 CTT2||

Figure 2: Chemical structure of CTT2-PEG3400-DSPE. CTT2-PEG3400-DSPE was synthesized by coupling CTT2-peptide to PEG3400-DSPE, followed by purification of the reaction product from the initial mixture.

2.2. Synthesis of DSPE-PEG3400-CTT2. Coupling bioactivepeptides to PEGylated lipids can alter the pharmacokineticsand dynamics of these peptides. For pharmaceutical for-mulation purposes, CTT2-peptide (Figure 1) was covalentlyattached to the PEG phospholipid (Figure 2).

In this procedure, the peptide called CTT2 (cyclo-GRENYHGCTTHWGFTLC- NH2) was covalently attachedto PEG phospholipids through the chemical reactionbetween the terminal amine of the peptide and the functionalNHS (hydroxysuccinimidyl) group at the end of the PEGphospholipid polymer chain. The reaction between theterminal amine and the active succinimidyl ester of the PEGcarboxylic acid produced a stable amide linkage. Differentmolar ratios of the peptide and the PEG phospholipid, as wellas the reaction times, were varied to optimize the couplingreaction. Up to several hundred CTT2-PEG-lipid moleculescan be attached to the surface of each liposome.

CTT2 peptide (8.8 mg) and DSPE-PEG3400-NHS(100 mg) were dissolved in 2 milliliters (ml) dimethyl-form-amide. CTT2 peptide solution (500 μl) was mixed with600 μl of DSPE-PEG3400-NHS solution and incubated for21 hours (hrs). Samples were then precipitated by additionof diethylether and centrifuged (13200 rpm for 10 min). The

supernatant was decanted and the solid residue was stored at−70◦C.

For all studies, samples were reconstituted by adding100 μl methanol and 25 μl of 1 M sodium hydroxide, followedby 250 μl of 1% TFA in water after two hours. Analysis ofthese samples was performed after centrifugation (4200 rpm20 min) using a C-18 RP-HPLC by initially precipitating thepurified product with excess diethylether. The solid residueswere dissolved in 1500 μl methanol and analyzed by thin layerchromatography (TLC). Reaction yields for CTT2 peptide-DSPE-PEG3400-NHS coupling averaged 6.0 mg.

2.3. Preparation of Liposomes

2.3.1. CTT2-Micelles. Monomers or CTT2-PEG3400-DSPE(i.e., CTT2-PEG-lipid) spontaneously formed micelles ∼14 nm in diameter (i.e., CTT2-micelles) in aqueous solution,with DSPE lipid chains forming the hydrophobic core andPEGylated CTT2-peptide forming the hydrophilic surfaceof the micelle. CTT2-micelles were covalently labeled withradioiodine, I-125 (125I, half-life = 13 hrs), to determinetime-varying tissue distributions and tumor uptakes. Radio-chemical purity of ∼90% was achieved.


Table 1: Tumor uptake of various liposomal constructs.

Caelyx CTT2-SL liposome

Targeted formulation No Yes

Concentration-targetedformulation

— 0.2%

Analyte Doxorubicin Doxorubicin

Time point (hours) 6 6

Tumor uptake (%ID/gram)

8.1% 19.0%

Table 2

Lipid percent (%)

CTT2-SL liposome

DSPE-mPEG2000 3.2 mg/ml 1.2 mM 5.5%

HSPC 9.6 mg/ml 12.2 mM 56.2%

Cholesterol 3.2 mg/ml 8.2 mM 38.1%

CTT2-PEG-lipid 0.2 mg/ml 0.04 mM 0.2%

Total lipids 16.2 mg/ml 21.6 mM 100.00%

Caelyx

DSPE-mPEG2000 3.2 mg/ml 1.2 mM 5.5%

HSPC 9.6 mg/ml 12.2 mM 56.3%

Cholesterol 3.2 mg/ml 8.2 mM 38.2%

Total lipids 16.0 mg/ml 21.6 mM 100.0%

2.3.2. CTT2-SL Liposomes. CTT-2-peptide-targeted lipo-somes were synthesized either by incorporating CTT2-PEG-lipid onto the surface of commercially available liposomes orby combining CTT2-micelles with liposomal formulations.Prior studies have shown that incubation of certain lipidswith liposomes can result in intraliposomal inclusion of theselipids as a consequence of micellar-liposomal fusion [23,24]. This spontaneous process, occurring when lipid con-centrations exceed critical micellar concentrations (CMC),has been used as a postinsertion technique with preformedliposomes to produce immunoliposomes [25] and liposomescoated with peptides or oligosaccharides [26, 27]. CTT2-micelles were combined with the commercially availablenanoformulated drug, Caelyx (PEGylated liposomal doxoru-bicin HCl), to form CTT2-peptide-targeted Caelyx (CTT2-SL liposome). This method provides a CTT2-PEG-lipidcontent of ∼0.2% of all lipids on the resulting liposome sur-face; CTT2-peptide-lipid concentrations are essentially themaximum achievable concentrations using CTT2-micellemethodologies as Caelyx liposomes are PEGylated. The lipidcomponents of CTT2-SL liposome and Caelyx that were usedfor these studies are listed in Table 2.

CTT2-SL liposomes were made by pipetting the above-mentioned lipid mixture except the CTT2-PEG-lipid, to around bottomed flask, dried under nitrogen and lyophilizedfor 2 h to remove trace amounts of chloroform. Doxorubicinliposomes were prepared by using standard pH gradienttechnique [1].

To synthesize CTT2-PEG-3400-DSPE Caelyx/doxil-lipo-somes, CTT2-PEG-DSPE (1 mg) was suspended in 400 μlof buffer (100 mM histidine, 55 mM sucrose, pH 6.5), and

100 μl of this CTT2-PEG-DSPE micelle suspension wasadded to 1 ml Doxil/Caelyx solution or internally preparedsimilar to doxil-liposomes (Ortho Biotech). In vivo murinestudies were performed after incubating the mixture for30 min at 60◦C.

The incorporation efficiency, the percentage of totalactivity contained in the liposome fractions, was measuredby using radioisotope-labeled peptide and gel filtration toseparate the unreacted micelle from the liposome; optimalreaction conditions were found to be 60◦C at 30 min (nearly100% efficient).

The doxorubicin leakage from the liposomes after theincorporation experiments was determined by comparingthe amount of free doxorubicin versus liposome-bound dox-orubicin before and after the experiment. The leakage wasfound to be minimal (the leakage before the incorporationwas in average 4.5% and after the reaction in average 4.2%).

2.4. Radiolabeling of Peptides. Radiolabeling of peptidesand all liposomal formulations with iodine-125 (125I) wasperformed using the IODOGEN (Pierce, Rockford, IL).The CTT2-PEG3400-DSPE peptide was labeled with 125Iusing iodogen as a catalyst. 5 MBq of Na125I (Amersham,Buckinghamshire, England) in 0.5 ml PBS was added toa tube containing 10 μg dried iodogen and 100 μg CTT2-PEG3400-DSPE peptide construct. The mixture was incu-bated for 20 min at room temperature. The 125I-boundparticle fractions were purified by elution from PD-10columns. The activity of the peptide was determined in agamma counter (Cobra II, Packard Instruments).

2.5. Animal Models and Tumor Inoculation. The mice werecared for according to the instructions of the animal facility,and the experiments were approved by an ethical commit-tee of Helsinki University, Finland. Male athymic nu/numice (6–8 weeks old, Harland) were provided with waterand maintained on regular diets ad libitum. Subcutaneoushuman serous ovarian carcinoma (OV-90) xenograft modelswere generated by coinjecting equal volumes of cells (∼ 5 ×106/100 μl phosphate buffered saline, PBS) and matrigel sub-cutaneously into the hindlegs of nude mice. Average tumorvolumes of 65 mm3–200 mm3 were used for all studies.

2.6. In Vivo Biodistribution and Pharmacokinetics. Followingsingle i.v. tracer doses of purified 125I-CTT1-peptide(∼40 μg/mouse), 125I-CTT2-peptide (∼40 μg/mouse), orCTT2-micelles (200 μg/mouse; 200 kBq), the percentageof the injected dose per gram of tissue (%ID/g) values,corrected for radioactive decay to the time of injection,were measured in tumor and major tissues/organs (heart,liver, kidney, lung, muscle, brain, spleen, and tumor) bysacrificing groups of normal mice or mice bearing serousovarian hindleg xenografts (OV-90) at specified time points.

Additional distribution data was measured in immuno-suppressed mice (n = 6/group) bearing subcutaneoushuman mucinous ovarian tumors (A2780) using singlebolus injections of CTT2-SL liposome or Caelyx (9 mg/kg,calculated doxorubicin equivalents). Lyophilized tissue andplasma were extracted in acid alcohol, and their doxorubicin


Figure 3: Schematic illustration of CTT2-PEG-3400-DSPE lipo-some [22].

0

50

100

150

200

250

300

350

Muscle CTT-2 CTT-1

Con

trol

(%)

Liver

Figure 4: Hepatic accumulation of 125I-CTT2-peptides in normalmice. Liver accumulation of peptides per gram of tissue in normalmice (n = 5) relative to muscle (control). All values are expressed asthe percentage of the control (% control) ± SD.

concentrations were determined using a Varian spectroflu-orometer. Doxorubicin fluorescence intensities (a.u.) weremeasured at 590 nm using excitation wavelengths of 470 nm,and comparing these intensities against standard samplescontaining known amounts of doxorubicin. Doxorubicinconcentrations in tumor (μg doxorubicin per gram drytissue) were expressed at each time point when delivered asCTT2-SL liposome or Caelyx.

2.7. Efficacy Studies

2.7.1. Doxorubicin Administered as CTT2-SL Liposomes andCaelyx. Therapeutic efficacy studies were conducted insubcutaneous A2780 xenografts using doxorubicin, adminis-tered as either CTT2-SL liposomes or Caelyx. Commerciallyavailable nonliposomal (“free”) drug (i.e., doxorubicin) andsaline dilution buffer were used as treatment controls. A2780ovarian cancer cells (5 × 106 in 100 μl PBS) were injectedsubcutaneously into the posterior flanks of NMRI nudemice (n = 40). Mice received i.v. bolus injections ofCTT2-SL liposome, Caelyx, doxorubicin, and buffer. CTT2-SL liposomes were injected when tumor volumes reached65 mm3, while administration of Caelyx and doxorubicin todifferent xenograft mice was offset in time from CTT2-SL

81012141618202224

(%ID

/g)

tiss

ue

0246

30m

in18

0m

in

30m

in18

0m

in

30m

in18

0m

in

30m

in18

0m

in

30m

in18

0m

in

30m

in18

0m

in

30m

in18

0m

in

30m

in18

0m

in

30m

in18

0m

in

Blood Heart Liver Kidney Lungs Muscle Brain Spleen OV-90

Tissues

125I-CTT2-peptide

Figure 5: Tissue distribution of a single dose of 125I-CTT2-peptidein immunosuppressed OV-90 xenograft mice. Blood and majororgans/tissues were collected at 0.5 hr and 3 hrs p.i. 125I-CTT2-peptide (40 μg/mouse, n = 5) and their radioactivities weremeasured. Results are expressed as percentage of injected dose pergram tissue (%ID/g). All values are given as mean ± SD.

liposomes by 3 and 6 days, respectively. Doxorubicin, CTT2-SL liposomes, and Caelyx were injected at doses of 9 mg/kgeach. Mouse body weights were monitored throughout thestudy period.

Aforementioned treatments were used to collect twoindependent biodistribution data sets in immunosuppressedOV-90 xenograft mice (n = 5/group). In one set ofstudies, CTT2-SL liposomes were injected using lower dosesof doxorubicin (5 mg/kg) compared to Caelyx (9 mg/kg).Doxorubicin was also administered to a second group ofmice (n-3 per group) in the form of CTT2-SL-DSPE-PEG3400 liposomes or CTT2-Caelyx-like liposomes. Theselatter formulations were bolus injected using 9 mg/kg (cal-culated doxorubicin equivalents). Harvesting, weighing, andcounting of blood, tumor, and major organs in a scintillationγ-counter were performed for all studies at specified timepoints. Doxorubicin was extracted from these formulations,and concentrations were analyzed using HPLC.


3.1. Biodistribution and Clearance Studies. The initial reasonto create the CTT-2 peptide was to make a peptide that wasmore easily iodinated and that offered a spacer that wascomfortably used for linking purposes without destroyingthe bioactivity of the peptide.

In nontumor-bearing mice, greater liver accumulation ofthe CTT1-peptide was observed than with the CTT2-peptide(Figure 4). This was thought to be secondary to the increasedrelative hydrophobicity of the former peptide construct. Allother tissues analyzed demonstrated no significant differencein the magnitude of uptake of these peptides (data notshown) [21]. The CTT2-peptide was thus selected for furtherstudies given its more rapid hepatic clearance.

In OV-90 xenograft models, substantially higher uptakeof 125I-CTT2-peptide was measured in all organs/tissues(Figure 5), particularly in the xenograft, with tumor-to-blood ratios ∼23 detected at 3 hrs postinjection (p.i.). This


0

2

4

6

8

10

12

14

16

18

20

1.5

h6

h24

h

1.5

h6

h24

h

1.5

h6

h24

h

1.5

h6

h24

h

1.5

h6

h24

h

1.5

h6

h24

h

1.5

h6

h24

h

1.5

h6

h24

h

1.5

h6

h24

h

Blood Heart Liver Kidney Lungs Muscle Brain Spleen

(%ID

/g)

OV-90

125I-CTT2-micelle

Figure 6: Tissue distribution of 125I-CTT2-micelle in OV-90 tumor mice. %ID/g values after i.v. injection of CTT2-micelles (200 μg/mouse,n = 5). All values are expressed as mean ± SD.

coupled with the poor prognosis of this disease in humans,showed the potential to improve treatment response usingCTT2-peptide targeted delivery, and the need to ensurecontrolled and sustained drug release led us to extend thismodel to investigate tumor uptake with micellar and lipo-somal formulations (CTT2-micelles and CTT2-liposomes).The amount of CTT2-bound peptide available for liposomaltargeting activity was found to be 500 based on the measuredaverage size and surface area of the resulting peptide-boundliposomal product by dynamic light scattering and theaforementioned reaction conditions.

For doxorubicin-containing liposomes, doxorubicinleakage after peptide attachment was assessed by comparingfree and liposomal doxorubicin on the basis of fluorometricanalysis. Leakage was found to be minimal, with leakagebefore and after incorporation averaging 4.5% and 4.2%,respectively.

Both OV-90 and mucinous ovarian carcinomas (A2780)were thus selected as xenograft models for subsequentnanoformulation studies. In OV-90 tumor mice, cleartargeting of CTT2-micelles was observed, reaching maxi-mum values of 17.6% of the injected dose per gram (%ID/g)of tumor at 6 hrs p.i. (Figure 6).

Doxorubicin concentrations (μg doxorubicin per gramdry tissue), in the form of CTT2-SL (targeted) and SL(nontargeted Caelyx/Doxil) liposomes, were measured asa function of time p.i. in A2780 xenografts, as shown inFigure 7. Doxorubicin was delivered more efficiently and at afaster rate to tumors using CTT2-SL liposomal formulationscompared to Caelyx, with significantly elevated tumorallevels of doxorubicin observed 3 days after drug injection.

CTT2-SL liposomal antitumor efficacy data following i.v.bolus injections of CTT2-SL liposome, Caelyx, doxorubicin,and buffer in A2780 xenografts is shown in Figure 8.Live mice exhibiting tumor sizes exceeding 1000 mm3 weresacrificed, including those at days 15 and 24 followingi.v. administration of buffer or doxorubicin, respectively.

0

20

40

60

80

2 6 24 48 72 96

Time (hours)

SLCTT2-CL

(μg/

g)dr

yti

ssu

e

Figure 7: Comparison of doxorubicin concentrations in tumorsafter a single i.v. injection of CTT2-SL liposome or Caelyx. A2780xenografts (n = 6) were collected at 2, 6, 24, 48, 72, and 96 hoursafter CTT2-liposome or Caelyx injection, and their doxorubicincontent was measured using spectrophotometry. Results are shownas μg drug per gram of dry tissue. All values are expressed as themean ± SD. The differences in each time point are near significant.The overall difference in the AUC is significant.

Importantly, 80% of mice treated with CTT-SL liposomesand 50% treated with Caelyx were alive at 24 days followinginitiation of treatment. Treatment with CTT2-SL liposomeswas therefore found to increase mean survival times of miceby 38% from 27.9 to 38.6 days.

Mouse body weights were monitored throughout thestudy period (Figure 9). Each doxorubicin regimen (CTT2-SL liposome, Caelyx, and Doxorubicin) induced a slightweight decrease with a maximum loss of about 10% at day9. However, one week later, body weights returned to initiallevels.

Given the overall improved survival found followingtreatment of A2780 xenografts with CTT2-SL liposomes,studies were extended to assess treatment response in


0123456789

10

0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60

Days post-treatment

Nu

mbe

rof

surv

ivin

gm

ice

CTT2-SLSLDoxorubicinBuffer

Figure 8: Kaplan-Meier plot of the survival of tumor bearingmice. Mice were treated with doxorubicin (9 mg/kg), administeredeither as CTT2-SL liposome or Caelyx. Controls were injected withdoxorubicin (9 mg/kg) or saline dilution buffer. Injections for eachtreatment group were made at day 0, day 3 and day 6, respectively.

85

90

95

100

105

110

115

0 3 6 9 12 15 18 21 24 27 32

Days post-treatment

Bod

yw

eigh

ts(%

init

ial)

CTT-SLSLDoxorubicinBuffer

Figure 9: Mouse body weight changes in each treatment groupduring the first 32 days of the trial. Mice were treated with 9 mg/kgdoxorubicin (calculated doxorubicin equivalents) or saline dilutionbuffer at day 0, 3 and 6. All values are expressed as mean of 9 mice.

OV-90 xenograft models. As seen in Figure 10, despitethe lower doses of CTT2-SL liposomes administered,efficient targeting, along with rising concentrations ofdoxorubicin was measured using this serous ovarian model(Figure 10(b)) over a 6-hr time interval relative to theuntargeted formulation.

Additional serum and tumor uptake measurements con-ducted with CTT2-SL DSPE-PEG3400 liposomes are shownin Figure 11. Initial serum doxorubicin concentrations werefound to be lower for CTT2-SL-DSPE-PEG3400 liposomesthan for untargeted liposomes (i.e., PEG-liposomes), but theoverall kinetic profile of the two liposomal formulations wasessentially equivalent over time. Figure 11(b)demonstratestime-dependent changes in the total amount of doxorubicinfound in tumor tissue. Unlike prior studies performed withCTT2-SL liposomes and Caelyx, where maximal differences

20

40

60

80

100

120

140

160

0

Serum

SL

CTT-SL

(μg/

mL)

0.5 hrs 6 hrs

(a)

0

2

4

6

8

10

12

14

Tumor

SLCTT-SL

Dox

oru

bici

n(%

ID/g

tiss

ue)

0.5 hrs 6 hrs

(b)

Figure 10: Concentrations of doxorubicin in (a) serum and (b) OV-90 xenografts in mice treated with CTT2-SL liposome and Caelyx at0.5 and 6 hours. Data are represented as a mean of 5 mice ± SD.

in tumor tissue accumulation were detectable at 6 hours,doxorubicin accumulations in the present study weresimilar for both liposomal products at earlier time points.However, at 16 hours p.i., a clear difference is observedin the accumulated doxorubicin tumor concentrations,confirming earlier findings that efficacy improves withCTT2-peptide-bound liposomal delivery systems. Theextended times of accumulation may be a consequence ofthe different liposomal formulations used. Doxorubicinconcentrations, in the form of CTT2-SL-DSPE-PEG3400liposomes, continued to rise at later time points, as againstthe notable decreases in tumor concentrations observed with


0

20

40

60

80

100

0 5 10 15 20

Time (hours)

Serum

CTT2 liposomePEG liposome

Dox

oru

bici

n(μ

g/m

L)

(a)

0

1

2

3

4

0 5 10 15 20

Time (hours)

Tumor

CTT2 liposomePEG liposome

Dox

oru

bici

n(μ

g/g)

(b)

Figure 11: Serum doxorubicin levels. Concentration of doxoru-bicin in (a) serum and (b) OV-90 xenograft mice (n = 3) treatedwith CTT2-SL-DSPE-PEG3400. Data are represented as a mean ±SEM.

the untargeted CTT2-Caelyx-like liposomes. Future kineticstudies should monitor time-varying changes in tumordoxorubicin concentrations (in the form of CTT2-peptidetargeted liposomes) at delayed time intervals (i.e., >16 hrsp.i.) in order to determine whether antitumor efficacystudies could benefit from employing a dosing regimenreflecting longer, sustained tumor concentrations.

4. Conclusions

Gelatinases, as extracellular targets, offer a viable alternativefor tumor targeting. In gelatinase-expressing tumors, suchas OV-90, targeted liposomal constructs, 125I-CTT2-SL anddoxorubicin-containing CTT2, were found to be promisingnanotherapeutic delivery vehicles for achieving therapeuticefficacy. Table 1 summarizes the tumor uptakes of various

targeted and nontargeted liposomal formulations. Differ-ences in tumor uptake were observed range ovarian cancermodels, with the largest uptake values (i.e., ∼17% ID/g at6 hrs) achieved in OV-90 hindlimb xenografts using CTT2-peptide-bound liposomes (∼500 peptides per liposome).Further, CTT2-bound micelles and liposomes, as well asthe CTT2 peptide, demonstrated equivalent overall tumoruptake values, suggesting similar bioactivity. However, toachieve controlled and sustained drug release, we chose ananoformulation instead of a prodrug approach (i.e., drug-peptide coupling). Our findings show that the utilization ofthese targeted nanoformulations results in a more efficientmethod for delivering therapeutics than passive (i.e., non-targeted) liposomal products (i.e., Caelyx). The developmentof CTT2-peptide-bound liposomes as a clinically promisingtargeting therapeutic that has the potential to improve drugdelivery to human ovarian cancers will rest on the additionalassessment of shelf and in vivo stability studies and formaltoxicity testing.

References

[1] V. Torchilin and V. Weissig, Liposomes: A Practical Approach,Oxford University Press, Oxford, UK, 2nd edition, 2003.

[2] G. Bendas, “Immunoliposomes: a promising approach totargeting cancer therapy,” BioDrugs, vol. 15, no. 4, pp. 215–224, 2001.

[3] H. Nagaset and J. F. Woessner Jr., “Matrix metalloproteinases.Minireview,” The Journal of Biological Chemistry, vol. 274, no.31, pp. 21491–21494, 1999.

[4] S. Curran and G. I. Murray, “Matrix metalloproteinasesmolec-ular aspects of their roles in tumour invasion and metastasis,”European Journal of Cancer, vol. 36, no. 13, pp. 1621–1630,2000.

[5] L. J. McCawley and L. M. Matrisian, “Matrix metallo-proteinases: they’re not just for matrix anymore!,” CurrentOpinion in Cell Biology, vol. 13, no. 5, pp. 534–540, 2001.

[6] B. Schmalfeldt, D. Prechtel, K. Harting et al., “Increasedexpression of matrix metalloproteinases (MMP)-2, MMP-9,and the urokinase-type plasminogen activator is associatedwith progression from benign to advanced ovarian cancer,”Clinical Cancer Research, vol. 7, no. 8, pp. 2396–2404, 2001.

[7] X. Wu, H. Li, L. Kang, L. Li, W. Wang, and B. Shan, “Activatedmatrix metalloproteinase-2 - A potential marker of prognosisfor epithelial ovarian cancer,” Gynecologic Oncology, vol. 84,no. 1, pp. 126–134, 2002.

[8] S. Ozalp, H. M. Tanir, O. T. Yalcin, S. Kabukcuoglu, U. Oner,and M. Uray, “Prognostic value of matrix metalloproteinase-9 (gelatinase-B) expression in epithelial ovarian tumors,”European Journal of Gynaecological Oncology, vol. 24, no. 5, pp.417–420, 2003.

[9] P. L. Torng, T. L. Mao, W. Y. Chan, S. C. Huang, and C.T. Lin, “Prognostic significance of stromal metalloproteinase-2 in ovarian adenocarcinoma and its relation to carcinomaprogression,” Gynecologic Oncology, vol. 92, no. 2, pp. 559–567, 2004.

[10] A. Lopata, F. Agresta, M. A. Quinn, C. Smith, A. G. Ostor,and L. A. Salamonsen, “Detection of endometrial cancer bydetermination of matrix metalloproteinases in the uterinecavity,” Gynecologic Oncology, vol. 90, no. 2, pp. 318–324, 2003.


[11] B. Davidson, I. Goldberg, J. Kopolovic et al., “MMP-2 andTIMP-2 expression correlates with poor prognosis in cervicalcarcinoma—a clinicopathologic study using immunohisto-chemistry and mRNA in situ hybridization,” GynecologicOncology, vol. 73, no. 3, pp. 372–382, 1999.

[12] A. Talvensaari-Mattila, P. Paakko, and T. Turpeenniemi-Hujanen, “Matrix metalloproteinase-2 (MMP-2) is associatedwith survival in breast carcinoma,” British Journal of Cancer,vol. 89, no. 7, pp. 1270–1275, 2003.

[13] S. M. Ranuncolo, E. Armanasco, C. Cresta, E. B. De Kier Joffe,and L. Puricelli, “Plasma MMP-9 (92 kDa-MMP) activity isuseful in the follow-up and in the assessment of prognosisin breast cancer patients,” International Journal of Cancer, vol.106, no. 5, pp. 745–751, 2003.

[14] M. Toth, D. C. Gervasi, and R. Fridman, “Phorbol ester-induced cell surface association of matrix metalloproteinase-9 in human MCF10A breast epithelial cells,” Cancer Research,vol. 57, no. 15, pp. 3159–3167, 1997.

[15] P. C. Brooks, S. Silletti, T. L. von Schalscha, M. Friedlander,and D. A. Cheresh, “Disruption of angiogenesis by PEX, anoncatalytic metalloproteinase fragment with integrin bind-ing activity,” Cell, vol. 92, no. 3, pp. 391–400, 1998.

[16] U. Jadhav, S. Chigurupati, S. S. Lakka, and S. Mohanam, “Inhi-bition of matrix metalloproteinase-9 reduces in vitro invasionand angiogenesis in human microvascular endothelial cells,”International Journal of Oncology, vol. 25, no. 5, pp. 1407–1414, 2004.

[17] M. Nguyen, J. Arkell, and C. J. Jackson, “Human endothelialgelatinases and angiogenesis,” International Journal of Bio-chemistry and Cell Biology, vol. 33, no. 10, pp. 960–970, 2001.

[18] K. Komatsu, Y. Nakanishi, N. Nemoto, T. Hori, T. Sawada, andM. Kobayashi, “Expression and quantitative analysis of matrixmetalloproteinase-2 and -9 in human gliomas,” Brain TumorPathology, vol. 21, no. 3, pp. 105–112, 2004.

[19] E. Koivunen, W. Arap, H. Valtanen et al., “Cancer therapy witha novel tumor-targeting gelatinase inhibitor selected by phagepeptide display,” Nature Biotechnology, vol. 17, pp. 768–774,1999.

[20] O. Penate Medina, T. Soderlund, L. J. Laakkonen, E. K. J.Tuominen, E. Koivunen, and P. K. J. Kinnunen, “Binding ofnovel peptide inhibitors of type IV collagenases to phospho-lipid membranes and use in liposome targeting to tumor cellsin vitro,” Cancer Research, vol. 61, no. 10, pp. 3978–3985, 2001.

[21] O. Penate Medina, K. Kairemo, H. Valtanen et al., “Radionu-clide imaging of tumor xenografts in mice using a gelatinase-targeting peptide,” Anticancer Research, vol. 25, no. 1, pp. 33–42, 2005.

[22] O. P. Medina, Y. Zhu, and K. Kairamo, “Targeted liposomaldrug delivery in cancer,” Current Pharmaceutical Design, vol.10, no. 24, pp. 2981–2989, 2004.

[23] S. Kanda, K. Inoue, S. Nojima et al., “Incorporation ofganglioside and spin-labelled ganglioside analogue into celland liposome membranes,” The Journal of Biochemistry, vol.91, pp. 2095–2098, 1982.

[24] P. S. Uster, T. M. Allen, B. E. Daniel, C. J. Mendez, M. S.Newman, and G. Z. Zhu, “Insertion of poly(ethylene glycol)derivatized phospholipid into pre-formed liposomes results inprolonged in vivo circulation time,” FEBS Letters, vol. 386, no.2-3, pp. 243–246, 1996.

[25] T. Ishida, D. L. Iden, and T. M. Allen, “A combinatorialapproach to producing sterically stabilized (Stealth) immuno-liposomal drugs,” FEBS Letters, vol. 460, no. 1, pp. 129–133,1999.

[26] S. Zalipsky, N. Mullah, J. A. Harding, J. Gittelman, L. Guo,and S. A. DeFrees, “Poly(ethylene glycol)-grafted liposomeswith oligopeptide or oligosaccharide ligands appended to thetermini of the polymer chains,” Bioconjugate Chemistry, vol. 8,no. 2, pp. 111–118, 1997.

[27] S. Dagar, M. Sekosan, B. S. Lee, I. Rubinstein, and H.Onyuksel, “VIP receptors as molecular targets of breast cancer:implications for targeted imaging and drug delivery,” Journalof Controlled Release, vol. 74, no. 1–3, pp. 129–134, 2001.


Research Article

Antibody-Hapten Recognition at the Surface ofFunctionalized Liposomes Studied by SPR: Steric Hindrance ofPegylated Phospholipids in Stealth Liposomes Prepared forTargeted Radionuclide Delivery

Eliot. P. Botosoa,1 Mike Maillasson,1, 2 Marie Mougin-Degraef,1 Patricia Remaud-Le Saec,1

Jean-Francois Gestin,1 Yannick Jacques,1, 2 Jacques Barbet,1 and Alain Faivre-Chauvet1

1 Centre de Recherche en Cancerologie Nantes-Angers (CRCNA), Universite de Nantes, Inserm, UMR 892,Institut de Recherche Therapeutique de l’Universite de Nantes, 8 quai Moncousu, BP 70721, 44007 Nantes Cedex 1, France

2 Plateforme Interactome & Puces a Proteines Biogenouest, Institut de Recherche Therapeutique de l’Universite de Nantes,8 quai Moncousu, BP 70721, 44007 Nantes Cedex 1, France

Correspondence should be addressed to Eliot. P. Botosoa, [email protected]

Received 30 June 2010; Revised 6 October 2010; Accepted 9 December 2010

Academic Editor: Alekha K. Dash

Copyright © 2011 Eliot. P. Botosoa et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

Targeted PEGylated liposomes could increase the amount of drugs or radionuclides delivered to tumor cells. They show favorablestability and pharmacokinetics, but steric hindrance of the PEG chains can block the binding of the targeting moiety. Here,specific interactions between an antihapten antibody (clone 734, specific for the DTPA-indium complex) and DTPA-indium-tagged liposomes were characterized by surface plasmon resonance (SPR). Non-PEGylated liposomes fused on CM5 chips whereasPEGylated liposomes did not. By contrast, both PEGylated and non-PEGylated liposomes attached to L1 chips without fusion. SPRbinding kinetics showed that, in the absence of PEG, the antibody binds the hapten at the surface of lipid bilayers with the affinityof the soluble hapten. The incorporation of PEGylated lipids hinders antibody binding to extents depending on PEGylated lipidfraction and PEG molecular weight. SPR on immobilized liposomes thus appears as a useful technique to optimize formulationsof liposomes for targeted therapy.

1. Introduction

The development of liposomes capable of targeting cells hasbeen an objective since the 80s [1, 2]. The most prevalentmethod is to conjugate antibodies or antibody-based con-structs (e.g., fragments or single chain Fv) directly ontheir surface (i.e., immunoliposomes). However, the abilityof immunotargeted liposomes to deliver high doses ofdrugs or radioactivity to tumor cells in vivo remains to bedemonstrated, partly because it is difficult to include allnecessary features, that is, long circulation times, stabledrug encapsulation or radiolabeling with high activities, andefficient antibody targeting in the liposomes preparation [3].

Other antibody constructs, such as bispecific antibodies,provide an alternative way to specifically target liposomesto cancer cells [4]. The bispecific antibody is used here asa pretargeting agent. It recognizes both a tumor-specificantigen and a small molecule (the hapten) used as a tag to theliposome membrane. The pretargeting system presents theadvantage of using a stable bispecific antibody and liposomesthat can be loaded extemporaneously with drugs or radionu-clides, whereas stability and loading of immunoliposomesmay be a problem. We have developed a liposomes radiola-beling method which is based on an active-loading approachfor obtaining high specific activity-labeled liposomes [5].Thus, the use of liposome as delivery systems represents


an attractive alternative to vehicle important quantities ofradionuclides.

Recent formulations of liposomes prevent their opson-ization by serum proteins and thus enhance residencetime in the bloodstream. This is obtained by the additionof PEG functionalized lipids in their composition [6–8].Different PEGylated liposomes formulations bearing theDTPA-indium hapten at their surface have been tested. SuchPEGylated liposomes, also referred to as stealth liposomes,containing doxorubicin and a few other drugs have beenapproved for marketing. Liposomes containing 1.5%, 5%,or 8% PE-PEG were analysed for blood clearance over24 h after injection in mice. Rapid elimination of con-ventional liposomes and 1.5% PEGylated liposomes wasobserved. Incorporation of 5% PEG in liposome consid-erably increased the retention time in bloodstream. Theexperiment showed identical half life and clearance (13,06 hand 0.16 mL/h or 13,89 h and 0.20 mL/h, resp.) for 5 and 8%DSPE-PEG, indicating that 5% DSPE-PEG is sufficient toobtain a maximum blood residence time [9]. Nevertheless,preliminary in vivo results have shown an improvementby only a factor of 1.7 between passive tumor targeting(absence of bispecific antibody) and active targeting of theliposomes by prior injection of a bispecific antibody bindingcarcinomembryonic antigen (CEA) on one arm and theDTPA-indium hapten on the other, in a model of CEA-positive tumor xenografts in the mouse. Passive targetingof the liposomes through the well-known enhancementpermeability and retention effect [5] is very significant, and,therefore, to be interesting, active targeting of the liposomesto the tumor sites must be more efficient than what weobserved with these hapten-tagged PEGylated liposomes. It islong known that PEGylation can hinder specific recognitionbetween immunoliposomes and target cells [10]. Sterichindrance may also be the reason for the poor enhancementof tumor uptake caused by the bispecific antibody. Sincethis phenomenon has never been studied in a quantitativemanner, we decided to use surface plasmon resonance(SPR) to characterize the specific interactions between theantihapten antibody and hapten-tagged liposome as a modelof specific immunologic interaction at the liposome surfacein the presence of varying amounts of PEGylated lipidsand various PEG chain lengths. SPR is a technique thatis frequently applied for measuring binding rate constantsbetween two interacting entities, generally proteins. Its mostobvious advantages over other techniques are: direct andrapid determination of association and dissociation ratesof binding process and no need of labeling liposomes.Several studies have demonstrated that the technique issensitive enough to monitor interactions between solutesand lipid bilayers like liposomes. Artificial bilayer lipidmembranes (BLMs) have been extensively used to mimicbiological cell membranes for studying membrane processessuch as signal transduction, ligand-receptor interactions,and ion transport through cell membranes [11–13]. Recentadvances in the preparation of stable membrane-like surfacesand the commercialization of sensor chips has enabledwidespread use of SPR in analyzing these protein-membraneinteractions in an environment that closely resembles our

in vivo situation [14–16]. In this study, tethered bilayermembrane on CM5 chips and nonfused liposomes immo-bilized on L1 chips have been used to monitor by SPR thebinding of antibodies to conventional and PEGylated DTPA-indium-tagged liposomes. We compared several liposomesformulations composed of distearoylphosphatidylcholine(DSPC), cholesterol (Chol), DSPE-DTPA that varied in theirPEG content and molecular weight (2000, 1000, or 750).Binding kinetics of a specific anti-indium-DTPA antibody(clone 734) were monitored using the BIAcore system andthe kinetic parameters were calculated by curve fitting.


2.1. Materials and Equipment. The purified MAb 734 IgG,with binding specific for the DTPA-indium complex, waskindly provided by IBC Pharmaceuticals (Morris Plains, NJ).

All chemicals were dissolved in sterile water (versol orversylene, FRESENIUS, France). Phosphate buffered saline(PBS 9.55 g·L−1, PBS DULBECCO) was supplied by BIO-CHROM AG, (Berlin, Germany). 0.4 M N-ethyl-N-(3-dim-ethylaminopropyl)-carbodiimide hydrochloride and 0.1 MN-hydroxysuccinimide (NHS) were obtained from GEHealthcare. Dimyristoyl-L-α-phosphatidylethanolamine (D-MPE), Triton-X100 (t-octylphenoxypolyethoxyethanol) andstable indium-115 chloride (115In) were purchased, res-pectively, from Sigma-Aldrich and Sigma Ultra. Radioactiveindium-111 chloride (111In) was purchased from Mallinc-krodt (Petten, The Netherlands).

Other phospholipids used to prepare liposomes were:1,2-Distearoyl-sn-glycerol-3-phophoethanolamine-N-[Me-thoxy(Polyethylene glycol)-2000] M.W : 2805.54 (DSPE-PEG2000), 1,2-Distearoyl-sn-glycerol–3-phophoethanolam-ine-N-[Methoxy(Polyethylene glycol)-1000] M.W : 1631.37(DSPE-PEG1000), 1,2-Distearoyl-sn-glycerol-3-phophoeth-anolamine-N-[Methoxy(Polyethylene glycol)-750] M.W :1528 (DSPE-PEG750), and 1,2-Distearoyl-sn-glycerol-3-ph-ophoethanolamine-N-[Methoxy(Polyethylene glycol)-550]M.W : 1351.78 (DSPE-PEG550) were purchased from AvantiPolar Lipids (Alabaster, AL, USA). (DSPE-DTPA) was syn-thesized by Ecole Nationale Superieure de Chimie de Rennes(France). Vesicle extruder and filter supports were purchasedfrom Avanti Polar Lipids, Inc. Polycarbonate membranes forvesicle extrusion (100 nm or 200 nm pore size, Nucleopore)were from Whatman. All phospholipids were dissolved in9 : 1 chloroform/methanol mixture (HPLC grade, Carlo Erbaand Fisher Scientific, resp.).

2.2. MAb 734 Equilibrium Binding Assays in Coated Tubes.Avidin-coated tubes saturated with bovine serum albu-min (BSA) were used for equilibrium affinity constantdeterminations of the anti-DTPA-indium antibody (MAb734) in competition experiments between DTPA-111In asa tracer and stable DTPA-metal complexes. Briefly, 1 mLof a 50 ng/mL solution of biotinylated Mab 734 Fab frag-ment was incubated overnight at 4◦C in the avidin coatedtubes. Just before use, the tubes were washed with NaCl0.9%-Tween 20 0.05%. DTPA (0.1 nmol) was labeled with


commercial indium-111 chloride (5 × 107 cpm) and usedas a tracer (15000 cpm in a total incubation volume of0.3 mL). Incubation with varying concentrations of stableDTPA-metal or EDTA-metal competitors was performedovernight at 4◦C in PBS supplemented with BSA. Tubeswere then counted after two rapid washes with 2 mL ofNaCl-Tween.

2.3. DMPE Solubilization for CM5 Coating. Dimyristoyl-L-α-phosphatidylethanolamine (DMPE) was thoroughlymixed with PBS containing 1% Triton-X100 (t-octylphenox-ypoly-ethoxyethanol) to a final concentration of 1 mg·mL−1,followed by at least three freeze-thaw cycles, ultrasonication,and incubation at 55◦C.

2.4. Liposomes Preparation and Characterization. For vesiclespreparation, the desired phospholipids (DSPC) in organicsolvent CHCl3/MeOH (9 : 1) were transferred to a 10 mLround bottom flask and the solvent was evaporated todryness. PBS was then added to the lipid film for a final lipidconcentration of 20 μmol·mL−1.

Large unilamellar vesicles (LUVs) composed of DSPC,DSPC/DSPE-DTPA (98 : 2 molar ratio) or DSPC/Chol/DSPE-DTPA (68 : 30.5 : 1.5 molar ratio) were preparedaccording to the lipid film hydration method [17] fol-lowed by extrusion. Typically, for the nonfused liposomes,13.5 μmol of phospholipids, 6.6 μmol of cholesterol, and0.3 μmol of phospholipids coupled to the chelating agent(DTPA) were dissolved in chloroform/methanol (9 : 1 v/v)in a 10 mL round bottom flask. DSPE-PEG2000 (0.5 mol%,1.5 mol%, 2.5 mol%, 3.5 mol%, or 5 mol%) were included inthe preparation according to the necessity of experimenta-tion.

A thin dry film of lipids was obtained by evaporationof the solvents in a rotary evaporator. Hydration of the drylipids was accomplished by addition of 1 mL of aqueousphase and maintained above the gel crystal transitiontemperature of the lipids during all the hydration procedure.To this effect, the flask containing the liposomes suspensionwas mixed during 2 h on a rotary evaporation system withoutvacuum, at room temperature for conventional liposomes(DSPC), and 74◦C for DSPC/Chol/DSPE-DTPA PEGylatedliposomes. Typically, the final concentration of the liposomessuspension was 20 μmol of lipids per mL of aqueous phase.

To obtain small and homogeneous vesicles, the liposomessuspension was sonicated times to time in a bath-type son-icator then 20 times extruded through Nucleopore 100 nmpolycarbonate filters using a manual thermostat heatedextrusion device at room temperature for conventionalliposomes and at 74◦C for PEGylated liposomes [18]. Thesize and polydispersity of the vesicles were measured bydynamic laser light-scattering system using a Malvern HighPerformance Particle Sizer (HPPS-ET, Instrument SA, UK).Measurements were performed in triplicate after dilutionof the suspension in water. The mean size were 101 ±4 nm (polydispersity index <0.1) for conventional liposomesand 107 ± 3 nm (polydispersity index <0.1) for PEGylatedliposomes with all concentrations of PEG2000.

2.5. 115 In Loading Procedure. DTPA functionalized lipo-somes were prepared in citrate (0.10 M)/acetate (0.15 M)buffer, pH = 5.3. Nonradioactive indium (115In) chloridein HCl 0.02 N was added with a ratio of 10 indium molarequivalents per mole of lipids, and the mixture was incubatedfor 2 hours at 37◦C. Then, 115In-loaded liposomes wereseparated from free indium by gel filtration chromatographyusing a PD-10 column eluted in PBS.

2.6. Formation of Lipid Planar Bilayers on CM5 Chips. Freelyaccessible terminal carboxyl groups of the dextran layer wereactivated with N-ethyl-N’-(3-dimethylaminopropyl)-carbo-diimide hydrochloride) (EDC) and N-hydroxysuccinimide(NHS). The primary amine of dimyristoyl-L-α-phosph-atidylethanolamine (DMPE) was then reacted with theactivated succinimide esters overnight at 55◦C. This reactionyields the proximal monolayer of the lipid membrane that iscovalently attached to the dextran layer on the gold surface.Then, the DMPE-coated surface was thoroughly rinsed withdistilled water and the chip was docked again in the BIAcoreinstrument. All flow cells were washed three times with30 μL 100 mM NaOH (flow rate 30 μL/min). DMPE couplingprovides the highly hydrophobic surface necessary for thesubsequent functionalization of the tethered membrane byspontaneous vesicle spreading. Lipid vesicles or liposomeswere then spread on the DMPE layer to constitute the bilayer.Briefly, liposomes (1 mg/mL in PBS) were injected over thehydrophobic surface for four to ten minutes at a flow rate of10 μL/min.

2.7. Binding of Intact Liposomes to L1 Chips. The BIAcore3000 instrument equipped with the L1 chip was usedfor Surface Plasmon Resonance measurement of antibodybinding to nonfused liposomes. The surface of the chip wasconditioned with three consecutive injections for 1 min at30 μL/min of isopropanol/50 mM NaOH (2/3, v/v).

Liposomes (1 mg/mL in PBS) were deposited on threeflow cells for 10 min at flow rate of 5 μL/min. The liposomessurface was washed with NaOH 100 mM for 1 min at30 μL/min. Bound liposomes could be removed from theL1 surface at the end of the experiments by two 1-minuteinjections of 50 mM NaOH : isopropanol (2/3, v/v) followedby two injections of Chaps 2% (w/v). Surface binding of L1biosensor chip can be regenerated as often as needed.

2.8. Atomic Force Microscopy (AFM). A multitask AFMCP was used for AFM imaging in the tapping mode andtopographic measurements. Typically for the analysis, weobserved the presence of a tethered lipid bilayer modifiedarea and a nonmodified surface.

2.9. Kinetic Measurements. For all measurements, the flowrate was fixed at 60 μL/min. Serial two-fold dilutions ofMAb 734 were prepared (750 nM to 0.78 nM) and injectedover on either the tethered planar bilayer on CM5 sensorchip or nonfused liposomes immobilized on the L1 sensorchip. Dilutions of MAb 734 were injected from low to highconcentration in a single-cycle kinetic (SCK) mode with


Mab 734 competition binding

0

10

20

30

40

50

60

70

80

DTPA-metal concentration (M)

Bou

nd/

tota

l(%

)

DTPA-indiumEDTA-indium

×10−510.10.010.0010.00010.000010.000001

Figure 1: MAb 734 binding with DTPA-indium hapten. Inhibitionof DTPA-111Indium binding to biotinylated MAb734 coated toavidin tubes as a function of DTPA-indium or EDTA-indiumconcentration. The equilibrium affinity constants were calculatedfrom the data using standard Scatchard analysis.

association phases monitored for 3 min at 60 μL/min andallowing 4 min dissociation phases.

2.10. Data Analysis. The resulting sensorgrams were fittedusing a mathematical program based on single cycle kineticmodel implemented in BIAeval 4.1 software (BIAcore).

3. Results

3.1. MAb 734—DTPA-Indium Binding Characterization. TheMAb 734 was originally screened for its binding to solu-ble DTPA-indium complex [19]. Competition experiments(Figure 1) using tubes coated with MAb 734 allowed theequilibrium dissociation constant to be determined as0.3 nM at 4◦C.

3.2. CM5 Bilayer

3.2.1. Formation of Lipid Planar Bilayers on CM5 SensorChip. DMPE was used for the setup of tethered artificialmembranes by chemical coupling of the primary aminogroups with succinimide esters of the dextran carboxylategroups. Then, the three formulations of liposomes, PEGy-lated, and conventional were spread on the DMPE monolayerafter rinsing with PBS. Figure 2 shows that liposomes give astable signal of 1100 RU. This value is in agreement with theexpected RU of the second monolayer coating of the CM5sensor chip functionalized with DMPE.

Figure 3 emphasizes that (a) DSPC-containing lipidvesicles spread and formed a 900 RU tethered planar bilayer-and (b) DSPC/DSPE-DTPA-Indium-containing lipid vesi-cles also spread but formed a 1100 RU tethered planar bilayerwhereas (c) DSPC/DSPE-DTPA-Indium/DSPE-PEG2000-containing lipid vesicles were not able to spread on theDMPE-monolayer.

200 400 600 800 1000 1200

Time (s)

−200

0

200

400

600

800

1000

1200

1400

Res

pon

se(R

U)

Figure 2: Tethered planar bilayer formation on DMPE function-alized CM5 chip. A BIAcore CM5 chip was coated with DMPE asdescribed in materials and methods. The functionalized chip wastreated with DSPC/DSPE-DTPA containing vesicles at a 5 μL/minflow rate. The sensorgram shows the spreading of liposomes to alevel of 1100 RU and the formation of a stable planar bilayer.

Our formulation (DSPC/DSPE-DTPA-Indium-contain-ing lipid vesicles) induced a gain of 200 RU compared tothe standard formulation (DSPC-containing lipid vesicles)(Figures 3(a) and 3(b)).

Although conventional liposomes (non-PEGylated) gavesatisfactory results, it was not possible to create a lipid planarbilayer with the PEGylated liposomes (Figure 3(c)). This canbe explained by the fact that PEG-chains constitute a barrieragainst spreading on the DMPE layer.

A CM5 sensor chip coated with the model bilayer ob-tained by fusing conventional liposomes to the DMPElayer was examined with an Atomic Force Microscope(AFM). Images clearly showed the topographical structureof lipid planar bilayers overlaying the dextran matrix and noliposomes stuck to the dextran as single particles confirmingprevious findings on either the characterization of planarsupported bilayers [20] or the behavior adopted by liposomesadsorbed on CM5 and L1 sensor chips with modified dextranmatrix [15, 21]. We could assume that liposomes fuse toform a lipid planar bilayer on the top of the dextran matrixwhich is the upper component of CM5 chips. Moreover, ithas already been demonstrated that liposomes adsorbed onL1 chips surfaces may remain as intact single vesicles.

3.2.2. MAb 734 Binding to DTPA-Indium Coupled to LipidPlanar Bilayer. The affinity of MAb 734 to the DTPA-indiumfunctionalized lipid planar bilayer was tested on a BIAcore3000 instrument in a single cycle kinetic model. The antibodybound the DTPA-indium hapten coupled to the DSPE layerwithout binding to the control DSPC layer. The binding wasfollowed in real-time by a sensorgram resulting from thesingle-cycle kinetics assays on the functionalized bilayer aftersubtraction of the control DSPC flow cell. The interaction ishighly specific in this range of concentration. The associationand dissociation rate constants were calculated using thesingle cycle kinetics model also called “titration kineticsmodel.” The kinetic constants for MAb 734 binding to


200 400 600 800 1000 1200

Time (s)

−200

0

200

400

600

800

1000

1200

1400

Res

pon

se(R

U)

(a)

200 400 600 800 1000 1200

Time (s)

−200

0

200

400

600

800

1000

1200

1400

Res

pon

se(R

U)

(b)

−50

0

50

100

150

200

250

300

350

200 400 600 800 1000 1200

Time (s)

Res

pon

se(R

U)

(c)

Figure 3: Formation of tethered planar bilayers on DMPE-functionalized CM5 chip with liposomes of different lipid compositions. ADMPE-coated chip was treated with liposomes prepared with 4 different lipid compositions: (a) DSPC-containing lipid vesicles thatspread and formed a 900 RU tethered planar bilayer, (b) DSPC/DSPE-DTPA-indium-containing lipid vesicles that spread and formed a1100 RU tethered planar bilayer, and (c) DSPC/DSPE-DTPA-indium/DSPE-PEG-containing lipid vesicles that were unable to spread on thehydrophobic DMPE monolayer.

the DTPA-indium-functionalized bilayer calculated usingthis procedure of global fitting are listed in Table 1. Theratio of the kinetic constants (koff/kon) provided a KD value,1.6 nM, similar to those determined in the equilibriumbinding experiments.

3.3. L1 Chip. A control surface was prepared by loadingvesicles without DTPA-Indium on the first flow cell of aL1 sensor chip, resulting in the deposition of 15000 RU. Asindicated in Figure 4, conventional and PEGylated liposomeswere properly adsorbed on the sensor chip with approxi-mately the same level of absorbance (15–16000 RU). It wasnot possible to rule out vesicle fusion on the surface ofthe L1 chip by a direct observation with the AFM. Weassume that the liposomes are adsorbed intact. In addition,previous published data “strongly” suggested that vesiclesremain intact once bound to the lipophilic anchors onthe surface of L1 chips [22, 23]. Therefore, these vesiclesstill maintain their biophysical properties. This finding hasbeen confirmed by the characterization of calcein-loadedimmobilized liposomes [24].

Then, MAb 734 binding kinetics to the surface ofL1 chip which has been coated with different types ofliposomes (DSPC/Chol/DSPE-DTPA (68 : 30.5 : 1.5), DSPC/Chol/DSPE-DTPA/DSPE-PEG2000 (63 : 30.5 : 1.5 : 5)) wasmonitored as above.

The calculated affinity constant (KD = 1.6 × 10−9 M)for conventional liposomes was exactly the same as thevalue obtained with planar bilayer formed by conventionalliposomes on CM5 (Table 1). The surface of the first flowcell coated with vesicles without DTPA-indium was used asa nonspecific binding control. In addition, antibody bindingwas observed only on liposomes surfaces functionalized withDTPA loaded with indium (data not shown).

3.3.1. Influence of DSPE-PEG Molar Fraction. The bindingresponses shown in Figure 5 illustrate that DTPA-indium-functionalized liposomes prepared with DSPE-PEG2000 atvarious concentrations (0.5%–1.5%–2.5%–.5%) bound theantibody with an affinity that strongly decreased withthe DSPE-PEG2000 fraction (MAb 734 antibody-binding


Table 1: Analysis of MAb 734 binding on tethered DSPC/DSPE-DTPA-indium bilayer created on CM5 chip and of MAb 734 bindingto DTPA-indium-tagged liposomes adsorbed onto a L1 chip via kon (1/Ms), koff (1/s), Rmax (RU), and KD (M) parameters. MAb 734was prepared in two-fold serial concentration series (720 nM to 2.185 nM) and was injected at a flow rate 60 μL/min across the liposomesadsorbed L1 chip surface for 3 min. At the end of each injection, the dissociation phase was set to 4 min. Liposomes binding assay was onlyperformed once with CM5 and duplicated with L1 chip. kon and koff values were obtained by nonlinear regression of experimental data fittedwith the SCK mathematical model. The score of χ2 (Chi2) is <5; it means that the model used adequately describes our data. kon, koff, andKD values are the average ± standard deviations.

Chip kon (1/Ms) koff (1/s) Rmax KD (M)

CM5 5.57 105 8.99 10−4 80.7 1.60 10−9

L1 5.08± 0.04 105 8.03± 0.03 10−4 1970 1.58± 0.01 10−9

Table 2: Analysis of MAb 734 binding with DSPC/Chol/DSPE-DTPA-indium/DSPE-PEG2000 liposomes via kon (1/Ms), koff (1/s),and KD (M). The amount of DSPE-PEG contained in PEGylatedliposomes varied from 0% to 5%. Liposomes were immobilized onL1 chip. The length of PEG chain was fixed unchanged at 2000. TheDTPA amount was fixed. The score of χ2 (Chi2) is <5; it meansthat the model used adequately describes our data. All experimentswere duplicated. kon, koff, and KD values are the average ± standarddeviations.

Formulations(%DSPE-PEG2000)

kon (1/Ms) koff (1/s) KD (M)

0% 5.08± 0.04 105 8.03± 0.03 10−4 1.59± 0.01 10−9

0.5% 3.31± 0.38 105 8.03± 0.07 10−4 2.45± 0.26 10−9

1.5% 1.52± 0.23 105 1.21± 0.13 10−3 8.04± 0.39 10−9

2.5% 6.01± 0.07 104 2.06± 0.05 10−3 3.28± 0.12 10−8

3.5% 1.02± 0.25 104 2.68± 0.02 10−3 2.78± 0.63 10−7

5% / / /

responses were normalized for the level of liposomes cap-tured on each surface, making it possible to compare thebinding results directly). This effect was observed when theDTPA-indium hapten was directly coupled to DSPE (DSPE-DTPA). Using the same fitting procedure (SCK model),the kinetic constants for MAb734 binding were calculatedfor the different percentages of PEGylated lipid in theliposomes preparations (Table 2). Thus, the higher affinitywas observed for the non-PEGylated liposomes. When theconcentration of DSPE-PEG2000 was increased from 0 to3.5%, kon reduced about 40 times while koff increased about3 times. The Rmax value seems to follow the same decreaseas the association rate. From these findings, we can assumethat the DTPA-indium haptens are masked by the PEGylatedchains of DSPE-PEG2000. More precisely, the diminutionof association rate constant and Rmax can be almost totallyexplained by the steric effects of the PEGylated chains thatreduce the diffusion factor. The faster dissociation rates maybe attributed to a decrease of rebinding during the dissoci-ation phase that can also be explained by steric hindrance.Both effects combined in increasing the dissociation constantfrom 1.6 nM in the absence of DSPE-PEG to over 1 μM with3.5% of DSPE-PEG2000.

18202224262830323436×103

0 200 400 600 800 1000 1200 1400 1600 1800

Time (s)

Res

pon

se(R

U)

Injectliposome

Stop injection 20 mM NaOH50 mM NaOH

100 mM NaOH

Figure 4: Deposition and absorption of liposomes on an L1 chip.Liposomes were deposited on the L1 chip for 5 min at 5 μL/minin PBS. The concentration of lipids was 1 mg/mL. The flow ratewas changed to 30 μL/min after the deposition, and liposomes werewashed with three consecutive injections of NaOH in increasingconcentration (20, 50, and 100 mM). All vesicles were in PBS duringthe injection. A control surface was prepared by loading PEGylatedliposomes without DTPA-indium on the first flow cell of a L1 sensorchip (DSPC/Chol/DSPE-PEG2000).

3.3.2. Influence of DSPE-PEG Chain Size. The bindingresponses shown in Table 3 emphasize that DTPA-indium-functionalized liposomes prepared with DSPE-PEG at vari-ous sizes (DSPE-PEG750–DSPE-PEG1000) bound the anti-body with a much higher affinity for DSPE-PEG750 com-pared to DSPE-PEG2000 at the same concentration (2.5%of DSPE-PEG). The antibody was also able to bind theliposomes with a higher concentration of DSPE-PEG750 (6%and 8% of DSPE-PEG). Liposomes PEGylated with PEG1000gave an affinity which was intermediate but higher than theone obtained with DSPE-PEG2000.

4. Discussion

The fundamental properties of unconjugated liposomes(e.g., size, surface charge, PEGylation, and membranefluidity) that largely determine their fate in vivo havebeen identified [10, 25], and their effect on liposomesbiodistributions and pharmacokinetics has been studied andunderstood to a great extent. However, the presence ofsurface conjugated-ligands (antibodies, protein fragments,and haptens) in targeted liposomes introduces additionalcomplexities in their interactions with the biological milieu.


Table 3: Data of KD (M) and Rmax (RU) resulted from MAb 734 binding with DSPC/Chol/DSPE-DTPA-indium/DSPE-PEG750 andDSPC/Chol/DSPE-DTPA-indium/DSPE-PEG1000 liposomes in order to emphasize the influence of PEG chain size. Three differentconcentrations of DSPE-PEG (2.5%, 6%, and 8%) were studied for each size of PEG chain. The score of χ2 (Chi2) is <5; it means thatthe model used adequately describes our data. Experiments were duplicated. KD value is the average ± standard deviations.

%DSPE-PEG 2.5% 6% 8%

PEG Size KD (M) Rmax (RU) KD (M) Rmax (RU) KD (M) Rmax (RU)

750 2.29± 0.11 10−9 1081 8.57± 0.08 10−9 902 2.51± 0.15 10−8 845

1000 4.25± 0.21 10−9 688 5.48 ± 0.22 10−8 529 1.15± 0.44 10−7 424

Time (s)

−200

0

200

400

600

800

1000

1200

1400

Res

pon

se(R

U)

0 500 1000 1500 2000 2500 3000 3500

DSPE/chol/DSPE-DTPA/DSPE-PEG (0.5%)DSPE/chol/DSPE-DTPA/DSPE-PEG (1.5%)DSPE/chol/DSPE-DTPA/DSPE-PEG (2.5%)DSPE/chol/DSPE-DTPA/DSPE-PEG (3.5%)

Figure 5: MAb 734 binding to DTPA-indium-tagged liposomesadsorbed onto a L1 chip. Kinetic titration series of MAb 734on DSPC/DSPE-DTPA-indium-containing liposomes with variousDSPE-PEG 2000 concentrations deposited on an L1 chip. Theliposomes were deposited on the L1 Chip at the same level ofResonance Units, approximately 15 000 RU. For each formulation,MAb 734 was prepared in two-fold serial concentration series(720 nM to 2.185 nM) and was injected at a 60 μL/min flow rateacross the liposomes adsorbed on the L1 chip surface for 3 min. Atthe end of each injection, the dissociation phase was set to 4 min.

A better understanding of these interactions will result inbetter targeted liposomes for maximum targeting specificity.At this point, it is clear that addition of PEG to the liposomessurface is needed to prevent opsonisation, fast uptake ofthe liposomes in the liver, and rapid clearance. Introducingimmunospecific ligands in the liposome membrane cantarget the liposomes and their content, but PEG chainsinterfere with antibody recognition. This steric hindranceis observed when an antibody is attached to the liposomessurface [10] but also, as shown in this paper, when theliposome is tagged with a small molecular weight ligand to berecognized by an antibody or, as described by Cao and Suresh[4], by a bispecific antibody. This study also demonstratesthat biosensors and SPR may be used to quantify thisphenomenon of steric hindrance as a function of the fractionof PEGylated lipids added to the liposomes and as a functionof the length of the PEG chains.

Using CM5 chips, tethered bilayers are obtained uponaddition of non-PEGylated liposomes. CM5 chip providesus with the development of a tethered bilayer obtained by

the spreading of non-PEGylated liposomes that yielded usto determine information on kinetics and thermodynamics.This model offered the advantages of controlling moreprecisely the bilayer formation and above all the quantityand the saturation degree of hapten used for the calculationof binding rates. The antihapten antibody then binds to thehapten coupled to phospholipids and incorporated into theliposomes preparation. The binding kinetics between MAb734 and the indium-DTPA hapten bound to these non-PEGylated phospholipid bilayers showed specific bindingwith an affinity value of 1.6× 10−9 M, close to that measuredin a completely different system of immobilized antibodyand soluble radiolabeled indium-DTPA hapten, which maybe considered as a mirror situation. This first experimentprovides us with reference-binding rates and dissociationconstants obtained both by kinetics and equilibrium mea-surements. Unfortunately, DPSE-PEG containing liposomesappeared not to bind and fuse on the surface of theseCM5 chips. This may be easily explained by the hydrophilicbarrier created by PEG chains at the surface of the liposomespreventing the interaction with the hydrophobic surface ofthe chip. Thus, L1 chips were used and shown capableof binding both conventional and PEGylated liposomes,independently of the DSPE-PEG2000 molar ratio, PEG chainlength, and the addition of hapten-bearing phospholipids.Although AFM could not demonstrate directly that theliposomes attached to the L1 chips remain intact, indirectevidence has been published in the literature in favor of thishypothesis [22–24]. L1 chips were coated with liposomesprepared with different lipid compositions to very similarRU signals. It is difficult to ascertain that this means thatthe number of liposomes or the number of indium-DTPAmolecules attached to the chip is identical with PEGylatedand non-PEGylated liposomes. However, this means that theorders of magnitude of these numbers are similar. Besides,the kinetic analysis used to derive binding constants is notdependent on this number. These differences in bindingaffinities must reflect steric hindrance from the PEG chainsand not a problem of liposomes capture. Thus bindingof preformed liposomes of various compositions to L1chips provides a robust and versatile system for in vitrobinding studies of antibodies directed against liposome-bound antigens, using SPR.

The incorporation of PEGylated lipids in the liposomemembrane hinders antibody binding in a PEGylated lipidfraction-dependent manner. The concentration and thechain length of PEGylated lipids are limiting factors forantibody-liposome interaction. Results from these binding


studies show that hapten-tagged vesicles prepared withDSPE-PEG2000 at a ratio equal to or greater than 5%are poorly recognized by the antihapten antibody. Therelative affinity of the antibody for hapten-bearing liposomesdecreased from 1.6 × 10−9 M to 1.1 × 10−7 M when thePEGylated lipid percentage increased from 0% to 5%. Wehave looked into the variations of speed and diffusioncoefficients in real-time. Rate constants measured by SPRrevealed decreased diffusion coefficients of the antibodywith vesicles containing various concentrations of DSPE-PEG2000 that translated in kon values decreasing with thePEG concentration. Similarly, Rmax value showed a decreaseof the availability of haptens.

All these phenomena could be physically explained bythe relative pliability and flexibility of PEG chain and itsunspecific interactions in equilibrium with the positivecharge molecules present at the surface of liposomes. PEGchain with a relatively long size induces a steric hindrance, adiffusing difficulty of the antibodies to the hapten site. Themasking of these haptens results in kon decreasing value anda decrease of Rmax. Nevertheless, once the antibodies arebound to their haptens the intensity of the interactions is notmodified and only the rebinding capacity is diminished, thatis traduced by a less increasing of koff value.

Liposomes prepared with shorter PEG chains—DSPE-PEG750 or DSPE-PEG1000—have more tightly bound theanti-hapten antibody, thus showing less steric hindrance.The shorter the chain, the easier the antibody MAb 734diffuses, the higher the kon, and the higher the affinity for theantibody-hapten interaction. These three effects may be theresult of two phenomena: PEG chain sweeping is influencedby the chain length and the formation of a tight mesh of PEGchains is influenced by the PEG concentration. Clearly, sizeand concentration of PEG chains limit the mass transfer ofantibodies to their binding sites. Shorter PEG chains—evenat high concentrations—should improve in vivo targetingbecause the affinity decreased only to 2.5 × 10−8 M and5.5× 10−8 M with, respectively, 8% of DSPE-PEG750 or 6%of DSPE-PEG1000.

As the shorter PEG chains are more rigid, they are lesscapable of sweeping and unspecifically interacting with theliposomes. The less PEG chain interfere with the diffusionof antibodies and the masking of haptens, the less kineticsconstant are modified and the more PEG concentrationcould be increased without loss in affinity.

The first in vivo pretargeting experiments, with PEGy-lated radiolabeled liposomes prepared with 5% of DSPE-PEG2000 and bispecific antibodies, showed encouraging butinsufficient results with a tumor uptake increased by a factorof 1.7, compared to passive targeting of conventional lipo-somes. The results of this study by SPR reveal and quantifya large loss of hapten-antibody affinity with such a formula-tion. The pharmacokinetic parameters of the other formula-tions have been evaluated (data not shown) showing reason-ably long circulation times, particularly with liposomes con-taining 6% or 8% of shorter PEG chains (750 or 1000). Sincethese liposomes formulations show reduced steric hindrancefor the hapten—antibody interactions, they will be tested forpretargeted delivery of radionuclides to tumors in vivo.

5. Conclusion

SPR biosensors, such as BIAcore, are most often used tomeasure the binding kinetics and affinity constants of molec-ular interactions. We describe here an application that couldhave a significant impact on the study of antibody/liposomeinteractions. Earlier studies demonstrated the feasibilityof biosensor simulation for acquiring binding data andpredicting targeting performance. The study of PEGylatedliposome formulations with variable PEG 2000 fractionsand different PEG chain sizes (PEG 1000 and PEG 750)by SPR prompted us to perform further pharmacokineticsexperiments to obtain the necessary information to improveimmunotargeting in vivo. This method will also be applied toother kinds of particulate nanovectors.

Abbreviations Used

AFM: Atomic force microscopy

BLM: Bilayer lipid membrane

CM5: Carboxymethylated dextran matrix2pt] DMPE: Dimyristoyl-L-α-phosphatidylethanolamine

DTPA: Diethylenetriaminepentaacetate

DTPA-In: Diethylenetriaminepentaacetate indiumcomplex

DSPC: Distearyl-L-α-phosphatidylcholine

DSPE: Distearyl-L-α-phosphatidylethanolamine

Chol: Cholesterol

EDC: N-ethyl-N-(-3-dimethylaminopropyl)-carbodiimide hydrochloride

FRET: Fluorescence resonance energy transfer

koff: Dissociation rate

kon: Association rate

KD: Dissociation constant

LUV: Large unilamelar vesicle

MAb: Monoclonal antibody

NHS: N-hydroxysuccinimide

PBS: Phosphate buffered saline

PEG: Polyethylene glycol

PK: Pharmacokinetic

RU: Resonance unit

SPR: Surface plasmon resonance

SCK: Single cycle kinetic

Resp. Diff.: Response differential.

Acknowledgments

This work was supported in part by the EC through the FP7collaborative project TARCC, the French Agence Nationalede la Recherche through the PCV grant VecRIT, and byINCa through the Action Concertee Incitative 2007 “Inno-vative delivery systems for cancer radionuclide therapy.” Theauthors also thank Dr. Cedric Gaillard from “Plate-formeRIO BIBS INRA de Nantes” for all AFM experiments.


References

[1] J. Barbet, P. Machy, and L. D. Leserman, “Monoclonalantibody covalently coupled to liposomes: specific targeting tocells,” Journal of Supramolecular and Cellular Biochemistry, vol.16, no. 3, pp. 243–258, 1981.

[2] J. Connor, S. Sullivan, and L. Huang, “Monoclonal antibodyand liposomes,” Pharmacology and Therapeutics, vol. 28, no. 3,pp. 341–365, 1985.

[3] W. W. Cheng and T. M. Allen, “The use of single chainFv as targeting agents for immunoliposomes: an updateon immunoliposomal drugs for cancer treatment,” ExpertOpinion on Drug Delivery, vol. 7, no. 4, pp. 461–478, 2010.

[4] Y. Cao and M. R. Suresh, “Bispecific MAb aided liposomaldrug delivery,” Journal of Drug Targeting, vol. 8, no. 4, pp. 257–266, 2000.

[5] H. Maeda, J. Wu, T. Sawa, Y. Matsumura, and K. Hori, “Tumorvascular permeability and the EPR effect in macromoleculartherapeutics: a review,” Journal of Controlled Release, vol. 65,no. 1-2, pp. 271–284, 2000.

[6] M. C. Woodle, “Sterically stabilized liposome therapeutics,”Advanced Drug Delivery Reviews, vol. 16, no. 2-3, pp. 249–265,1995.

[7] P. Srinath and P. V. Diwan, “Stealth liposomes—an overview,”Indian Journal of Pharmacology, vol. 26, no. 3, pp. 179–184,1994.

[8] M. Mougin-Degraef, Etude de faisabilite du radiomarquageet du preciblage de liposomes pour la radioimmunotherapie,Nantes, France, 2005.

[9] M. Mougin-Degraef, C. Bourdeau, E. Jestin et al., “Doublyradiolabeled liposomes for pretargeted radioimmunotherapy,”International Journal of Pharmaceutics, vol. 344, no. 1-2, pp.110–117, 2007.

[10] A. L. Klibanov, K. Maruyama, A. M. Beckerleg, V. P. Torchilin,and L. Huang, “Activity of amphipathic poly(ethylene glycol)5000 to prolong the circulation time of liposomes dependson the liposome size and is unfavorable for immunoliposomebinding to target,” Biochimica et Biophysica Acta, vol. 1062, no.2, pp. 142–148, 1991.

[11] A. Ottova-Leitmannova and H. Ti Tien, “Bilayer lipid mem-branes: an experimental system for biomolecular electronicdevices development,” Progress in Surface Science, vol. 41, no.4, pp. 337–445, 1992.

[12] H. T. Tien, “Self-assembled lipid bilayers as a smart materialfor nanotechnology,” Materials Science and Engineering C, vol.3, no. 1, pp. 7–12, 1995.

[13] A. Janshoff and C. Steinem, “Transport across artificialmembranes-an analytical perspective,” Analytical and Bioan-alytical Chemistry, vol. 385, no. 3, pp. 433–451, 2006.

[14] R. Karlsson, “SPR for molecular interaction analysis: a reviewof emerging application areas,” Journal of Molecular Recogni-tion, vol. 17, no. 3, pp. 151–161, 2004.

[15] M. Besenicar, P. Macek, J. H. Lakey, and G. Anderluh, “Sur-face plasmon resonance in protein-membrane interactions,”Chemistry and Physics of Lipids, vol. 141, no. 1-2, pp. 169–178,2006.

[16] V. Silin and A. Plant, “Biotechnological applications of surfaceplasmon resonance,” Trends in Biotechnology, vol. 15, no. 9, pp.353–359, 1997.


[18] M. J. Hope, M. B. Bally, G. Webb, and P. R. Cullis, “Productionof large unilamellar vesicles by a rapid extrusion procedure.Characterization of size distribution, trapped volume andability to maintain a membrane potential,” Biochimica etBiophysica Acta, vol. 812, no. 1, pp. 55–65, 1985.

[19] J. M. Le Doussal, A. Gruaz-Guyon, M. Martin, E. Gautherot,M. Delaage, and J. Barbet, “Targeting of indium 111-labeledbivalent hapten to human melanoma mediated by bispecificmonoclonal antibody conjugates: imaging of tumors hostedin nude mice,” Cancer Research, vol. 50, no. 11, pp. 3445–3452,1990.

[20] S. N. Ratnayaka, R. J. Wysocki, and S. S. Saavedra, “Prepa-ration and characterization of asymmetric planar supportedbilayers composed of poly(bis-sorbylphosphatidylcholine) onn-octadecyltrichlorosilane SAMs,” Journal of Colloid and Inter-face Science, vol. 327, no. 1, pp. 63–74, 2008.

[21] E. M. Erb, X. Chen, S. Allen et al., “Characterization ofthe surfaces generated by liposome binding to the modifieddextran matrix of a surface plasmon resonance sensor chip,”Analytical Biochemistry, vol. 280, no. 1, pp. 29–35, 2000.

[22] G. Anderluh, M. Besenicar, A. Kladnik, J. H. Lakey, and P.Macek, “Properties of nonfused liposomes immobilized on anL1 Biacore chip and their permeabilization by a eukaryoticpore-forming toxin,” Analytical Biochemistry, vol. 344, no. 1,pp. 43–52, 2005.

[23] M. A. Cooper, A. Hansson, S. Lofas, and D. H. Williams, “Avesicle capture sensor chip for kinetic analysis of interactionswith membrane-bound receptors,” Analytical Biochemistry,vol. 277, no. 2, pp. 196–205, 2000.

[24] M. Dalla Serra and G. Menestrina, “Liposomes in the studyof pore-forming toxins,” Methods in Enzymology, vol. 372, pp.99–124, 2003.

[25] A. Gabizon and D. Papahadjopoulos, “The role of surfacecharge and hydrophilic groups on liposome clearance in vivo,”Biochimica et Biophysica Acta, vol. 1103, no. 1, pp. 94–100,1992.


Review Article

A Review on Composite Liposomal Technologies forSpecialized Drug Delivery

Maluta S. Mufamadi,1 Viness Pillay,1 Yahya E. Choonara,1 Lisa C. Du Toit,1 Girish Modi,2

Dinesh Naidoo,3 and Valence M. K. Ndesendo1

1 Department of Pharmacy and Pharmacology, University of the Witwatersrand, 7 York Road, Parktown,Johannesburg 2193, South Africa

2 Department of Neurology, University of the Witwatersrand, 7 York Road, Parktown, Johannesburg 2193, South Africa3 Department of Neurosurgery, University of the Witwatersrand, 7 York Road, Parktown, Johannesburg 2193, South Africa

Correspondence should be addressed to Viness Pillay, [email protected]

Received 28 July 2010; Revised 23 November 2010; Accepted 7 December 2010

Academic Editor: Guru V. Betageri

Copyright © 2011 Maluta S. Mufamadi et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

The combination of liposomes with polymeric scaffolds could revolutionize the current state of drug delivery technology. Althoughliposomes have been extensively studied as a promising drug delivery model for bioactive compounds, there still remain majordrawbacks for widespread pharmaceutical application. Two approaches for overcoming the factors related to the suboptimalefficacy of liposomes in drug delivery have been suggested. The first entails modifying the liposome surface with functionalmoieties, while the second involves integration of pre-encapsulated drug-loaded liposomes within depot polymeric scaffolds.This attempts to provide ingenious solutions to the limitations of conventional liposomes such as short plasma half-lives,toxicity, stability, and poor control of drug release over prolonged periods. This review delineates the key advances in compositetechnologies that merge the concepts of depot polymeric scaffolds with liposome technology to overcome the limitations ofconventional liposomes for pharmaceutical applications.

1. Introduction

Over the past few decades, liposomes have received wide-spread attention as a carrier system for therapeuticallyactive compounds, due to their unique characteristics suchas capability to incorporate hydrophilic and hydrophobicdrugs, good biocompatibility, low toxicity, lack of immunesystem activation, and targeted delivery of bioactive com-pounds to the site of action [1–4]. Additionally, someachievements since the discovery of liposomes are controlledsize from microscale to nanoscale and surface-engineeredpolymer conjugates functionalized with peptide, protein,and antibody [5, 6]. Although liposomes have been exten-sively studied as promising carriers for therapeutically activecompounds, some of the major drawback for liposomes usedin pharmaceutics are the rapid degradation due to the reticu-loendothelial system (RES) and inability to achieve sustaineddrug delivery over a prolonged period of time [7]. New

approaches are needed to overcome these challenges. Twopolymeric approaches have been suggested thus far. The firstapproach involves modification of the surface of liposomeswith hydrophilic polymers such polyethylene glycol (PEG)while the second one is to integrate the pre-encapsulateddrug-loaded liposomes within depot polymer-based systems[3]. A study conducted by Stenekes and coworkers [8]reported the success of using temporary depot of polymericmaterials to control the release of the loaded liposomes forpharmaceutical applications. This achievement leads to newapplications, which requires collaborative research amongpharmaceuticals, biomaterials, chemistry, molecular, and cellbiology. Numerous studies in this context have been reportedin the literature dealing with temporary depot delivery sys-tem to control the release of pre-encapsulated drug-loadedliposomes [9–12]. This system was developed to integrate theadvantages while avoid the disadvantages of both liposome-based and polymeric-based systems. The liposome-based


systems are known to possess limitations such as instability,short half-life, and rapid clearance. However, they aremore biocompatible than the polymer-based systems [13].On other hand, the polymer-based systems are known tobe more stable and provide improved sustained deliverycompared to liposome-based systems. However, one of themajor setbacks is poor biocompatibility which is associatedwith loss of the bioactive (i.e., the drug) during fabricatingconditions such as heat of sonication or exposure to organicsolvents [3, 11]. The benefits of a composite system, however,include improvement of liposome stability, the ability of theliposome to control drug release over a prolonged period oftime, and preservation of the bioactiveness of the drugs inpolymeric-based technology. In addition, increased efficacymay be achieved from this integrated delivery system whencompared to that of purely polymeric-based or liposome-based systems. The aim of this article therefore, is to reviewthe current liposome-based and polymeric-based technolo-gies, as well as the integration of liposome-based technologywithin temporary depot polymeric-based technology forsustained drug release. The discussion will focus on differenttypes of liposome-based technology and depot polymericscaffold technologies, various methods for embedding drug-loaded liposomes within a depot, and various approachesreported to control the rate of sustained drug release withindepot systems over a prolonged period of time.

2. Liposome-Based Technology

A liposome is a tiny vesicle consisting of an aqueous coreentrapped within one or more natural phospholipids form-ing closed bilayered structures (Figure 1) [5]. Liposomeshave been extensively used as potential delivery systemsfor a variety of compounds primarily due to their highdegree of biocompatibility and the enormous diversity ofstructures and compositions [14, 15]. The lipid componentsof liposomes are predominantly phosphatidylcholine derivedfrom egg or soybean lecithins [15]. Liposomes are biphasica feature that renders them the ability to act as carriersfor both lipophilic and hydrophilic drugs. It has beenobserved that drug molecules are located differently in theliposomal environment and depending upon their solubil-ity and partitioning characteristics, they exhibit differententrapment and release properties [15, 16]. Lipophilic drugsare generally entrapped almost completely in the lipidbilayers of liposomes and since they are poorly water soluble,problems like loss of an entrapped drug on storage are rarelyencountered. Hydrophilic drugs may either be entrappedinside the aqueous cores of liposomes or be located in theexternal water phase. Noteworthy is that the encapsulationpercentage of hydrophilic drugs by liposomes depends onthe bilayer composition and preparation procedure of theliposomes [17, 18].

Since liposome discovery by Bangham and coworkers [5],several different embodiments of liposome-based technologyhave been developed to meet diverse pharmaceutical criteria[7]. Liposome-based technology has progressed from thefirst generation “conventional vesicles,” to stealth liposomes,

targeted liposomes, and more recently stimuli-sensitive lipo-somes [3, 19]. Essentially, liposomes are classified accordingto their size range, being 50–5000 nm in diameter. Thisresulted into two categories of liposomes namely multilamel-lar vesicles and unilamellar vesicles [19]. Unilamellar vesiclesconsist of single bilayer with a size range of 50–250 nm whilemultilamellar vesicles consist of two or more lipid bilayerswith a size range of 500–5000 nm [3, 20].

2.1. Conventional Liposomes. Conventional liposome-basedtechnology is the first generation of liposome to be usedin pharmaceutical applications [3, 21, 22]. Conventionalliposome formulations are mainly comprised of naturalphospholipids or lipids such as 1,2-distearoryl-sn-glycero-3-phosphatidyl choline (DSPC), sphingomyelin, egg phos-phatidylcholine, and monosialoganglioside. Since this for-mulation is made up of phospholipids only, liposomalformulations have encountered many challenges; one of themajor ones being the instability in plasma, which resultsin short blood circulation half-life [7, 23–25] Liposomesthat are negatively or positively charged have been reportedto have shorter half-lives, are toxic, and rapidly removedfrom the circulation [23, 26, 27]. Several other attemptsto overcome these challenges have been made, specificallyin the manipulation of the lipid membrane. One of theattempts focused on the manipulation of cholesterol. Addi-tion of cholesterol to conventional formulations reducesrapid release of the encapsulated bioactive compound intothe plasma [28]. Furthermore, studies by Tran and cowork-ers [29] demonstrated liposome stability after additionof “helper” lipids such as cholesterol and 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE). Harashima andcoworkers [20] demonstrated that phagocytosis of liposomeswas due to the size of the liposome formulation. Larger sizeor multilamellar liposomes with a size range of 500–5000 nmwere the first to be eliminated from the systemic circulation.Nanosized liposomes or small unilamellar vesicles with asize range of 20–50 nm were only developed later [7, 20,30]. The following drugs: Ambisone, Myocet, Daunoxome,and Daunorubicin have received clinical approval usingconventional liposome technologies [31–33]. Although smallunilamilar liposomes were reported to have potential for adecreased microphage uptake, insufficient drug entrapmentis still a major disadvantage. On the basis of this study,the success of cholesterol and others phospholipids did notcompletely overcome the major challenges.

2.2. Stealth Liposomes. Stealth liposome technology is oneof the most often used liposome-based systems for deliveryof active molecules [3, 22]. This strategy was developed toovercome most of the challenges encountered by conven-tional liposome technology such as the inability to evadeinterception by the immune system, toxicity due to chargedliposomes, low blood circulation half-life, and steric stability[7, 22, 26]. Stealth liposome strategy was achieved simply bymodifying the surface of the liposome membrane, a processthat was achieved by engineering hydrophilic polymerconjugates [34]. The employed hydrophilic polymers were


(a) (b) (c)

(d) (e) (f)

Figure 1: Schematic representation of liposome-based systems. (a) Conventional liposomes. (b) Stealth liposome coated with a polymericconjugate such as PEG. (c) Stealth liposome coupled with a functionalized ligand. (d) Liposome with a single ligand and antibody.(e) Duplicated ligand with repeated peptide sequence. (f) Liposome loaded with perfluorocarbon gas (adapted from Zucker et al. [16]).

either natural or synthetic polymers such polyethylene glycol(PEG), chitosan, silk-fibroin, and polyvinyl alcohol (PVA)[35–38]. Several properties that would add advantages topolymeric conjugate were considered such as high biocom-patibility, nontoxicity, low immunogenicity, and antigenicity[3, 35]. Although the majority of hydrophilic polymersmeet the above criteria, PEG remains the most widely usedpolymer conjugate. It is specifically employed to increase thehydrophilicity of the liposome surface via a cross-linked lipid[39, 40]. PEGylated liposomal doxorubicin (DOXIL/Caelyx)is the exceptional example of stealth liposome technology tobe approved by both the USA Food and Drug Administration(FDA) and Europe Federation [41]. Although prominentresults were achieved from this model such as reductionof macrophage uptake, long circulation, and low toxicity,passive targeting is still a major disadvantage since liposomescan deliver active molecules not only to abnormal cells butalso to sensitive normal cells [7, 42]. Figure 2 depicts aschematic for a PEGylated liposome.

2.3. Targeted Liposomes. Targeted liposome based systemwas suggested after conventional stealth liposome failedto evade uptake of active molecules by sensitive normalcells or nonspecific targets in vivo [43, 44]. Unlike stealth

liposome, site-specific targeting liposome has been engi-neered or functionalized with different types of targetingmoieties such antibodies, peptide, glycoprotein, oligopep-tide, polysaccharide, growth factors, folic acid, carbohydrate,and receptors [45–50]. In addition, targeted ligand canfurther increase the rate of liposomal drug accumulation inthe ideal tissues/cells via overexpressed receptors, antigen,and unregulated selectin [51–55]. Peptides, protein, andantibodies have been most extensively studied as a ligandfor directing drug-loaded liposomes into sites of action, dueto their molecule structures, which are essentially composedof known amino acid sequences. Furthermore, it has beenpostulated that ligands can be conjugated onto pegylatedliposomes via different types of coupling methods, suchas covalent and noncovalent binding. Covalent couplingoccurs when novel ligands are indirectly engineered onthe surface of liposome through a hydrophobic anchorvia thioether, hydrazone bonds, avidin–biotin interaction,cross-linking between carboxylic acids and/or amines [56].Noncovalent coupling is observed when novel ligands aredirectly added to the mixture of phospholipids during theliposomal formulation [15]. Li et al. [48] attempted togenerate dual ligand liposome conjugates aimed at targetingmultiple receptor types on the cell surface receptors. Ex vivostudies demonstrated the success of the dual ligand approach


Figure 2: Schematic depicting of a stealth PEGylated liposome(Adapted from Rai et al. [58]).

in improving the selectivity when compared to a singleligand approach. In another study, Ying and coworkers [50]formulated dual targeted liposomes with various targetedmoieties such as p-aminophenyl-α-D-manno-pyranoside(MAN) and transferrin (TF). The study was conductedboth ex vivo (in C6 glioma cells) and in vivo (in C6brain glioma-bearing rats). The following were compared:free daunorubicin, daunorubicin liposomes, daunorubicinliposomes modified with MAN, and daunorubicin liposomesmodified with TF as the controls, and daunorubicin lipo-somes modified with MAN and TF. Daunorubicin liposomesmodified with dual ligands such as MAN and TF showeda more significant increase in therapeutic efficacy, whencompared with the drug alone, drug-loaded liposome, orsingle ligand modified surface of the liposome. However,the efficacy of these approaches faces limitations becauseprotein circulation and gene expression cannot be sustainedfor long periods of time [7]. Doxorubicin-loaded liposomeswere surface engineered with monoclonal antibody and arenow commercially available [57]. The overall advantage ofthis model of liposome is an increase in active molecules ordrug reach targeted cells via endocytosis [7].

In another study, Nallamothu and coworkers [59]demonstrated the usefulness of Combretastatin A4 as novelantivascular agent. This compound portrays its anticanceractivity by inducing irreversible vascular shutdown in solidtumors [60]. Despite its anticancer potential, the drughas shown to have several undesirable side effects to theunderlying normal tissues [61]. These problems may bealleviated by targeting the drug specifically to the solid tumorvasculature. Studies have shown that certain cell adhesionmolecules such as αvβ3 integrin receptors are overexpressedon actively proliferating endothelium of the tumor vascu-lature [62, 63]. These surface markers discriminate tumorendothelial cells from the normal endothelial cells and canbe used as a target for antivascular drug delivery [59].Nallamothu and coworkers [59] could demonstrate thatpeptides with Arginine-Glycine-Aspartine (A-G-A) amino

A-G- ApeptidePEG

Liposome bilayer

Figure 3: A schematic representation of the targeted liposomedelivery system depicting the cyclic RGD peptides that targets theαvβ3 integrin receptors on the vascular tumor cells (adapted fromNallamothu et al. [59]).

acid sequence constrained in a cyclic polyethylene-glycol(PEG)-based liposome framework can bind to the αvβ3

integrin receptors. Basing on this analogy, they could designa targeted liposome delivery system for combretastatin A4with cyclic (RDG) peptides as targeting ligands (Figure 3).Targeting of combretastatin A4 to irradiated tumors usingthis delivery system resulted into significant tumor growthdelay [59].

2.4. Other Types of Liposomes

2.4.1. Virosomes and Stimuli-Responsive Liposomes. Liposo-mal technologies, such as conventional, stealth, and targetedliposomes have already received clinical approval [64, 65].New generation types of liposome have been developed toincrease bioactive molecule delivery to the cytoplasm byescape endosome [1, 66, 67]. New approaches that employliposomes as pharmaceutical carriers are virosomes andstimuli-type liposomes. The stimulating agents in this caseinclude pH, light, magnetism, temperature, and ultrasonicwaves. A virosome (Figure 4) is another type of liposomeformulation. It comprises noncovalent coupling of a lipo-some and a fusogenic viral envelop [68]. A stimuli-sensitiveliposome is a type of liposome that generally depends ondifferent environmental factors in order to trigger drug, pro-tein, and gene delivery. A study conducted by Schroeder et al.[69], Liu and coworkers [67], and Lentacker and coworkers[70] demonstrated that the exposure of the liposome loadedwith perfluorocarbons gas to ultrasound waves triggereddrug and gene delivery into the cytoplasm of the targetedcells through cell membrane pores. Their data demonstratedthat the liposome-loaded magnetic agents triggered drugdelivery to the specific site in vivo, using an externally appliedmagnetic field. The enhancement of endosomal release ofdrug-loaded liposome into the cytoplasm was also reported


Hemagglutinin

Neuraminidase

Phosphatidylcholine

Phosphatidylethanolamine

Figure 4: A schematic representation of a virosome (source: PevionBiotech Ltd. [73]).

to be influenced by the utilization of pH-sensitive liposomesor by attachment of pH-sensitive fusogenic peptide ligands[1, 71, 72]. Most recently, a review article published byChen and coworkers [4] described the generation of stableliposomes utilizing lyophilization techniques, which may bea promising future model for liposome production.

2.4.2. Gene-Based Liposomes. The characterization of humangenome coupled with recombinant DNA technology has cre-ated opportunities for gene therapy that never existed before[74]. Candidate diseases for such technology include cancer[75], arteriosclerosis [76], cystic fibrosis [77], haemophilia,sickle cell anaemia, and other genetic diseases. Ideally, theadministration of the gene of interest should result inthe expression of the therapeutic protein. However, thedelivery of the large anionic bioactive DNA across cell hasbeen one of the most difficult endeavors. DNA is easilydegraded by circulating and intracellular deoxyribonucle-ases. Notwithstanding, it must also be delivered intact acrossthe cell and nucleolar membranes to the nucleus [74].Liposomes have thus proved to achieve efficient intracellulardelivery of DNA [78, 79]. Such liposomes are preparedfrom phospholipids with an amine hydrophilic head group.The amines may be either quaternary ammonium, tertiary,secondary, or primary, and the liposomes prepared in thisway are commonly referred to as cationic liposomes, sincethey possess a positive surface charge at physiological pH.The use of cationic liposomes as gene delivery systems wasfirstly enforced in the late 1980s when in vitro studies byFelgner and coworkers [80] could demonstrate that thecomplexation of genes with liposomes may promote geneuptake by cells in vitro. Since then, cationic liposomes ofvarying description have been used to promote the cellularuptake of DNA with resultant therapeutic protein expressionby various organs in vivo. Figure 5 depicts a schematicrepresentation of a DNA-liposome complex.

Although the experimental data have demonstrated thatcationic liposomes can facilitate the transfer of DNA intolive mammalian cells, there are still major problems thatneed to be overcome in order to effectively achieve thegoal. These include a reduction in the rapid clearanceof cationic liposomes and the production of efficientlytargeted liposomes. At the cellular level, the problems may be

Figure 5: A schematic representation of a DNA-liposome complex(adapted from Uchegbu [74]).

overcome by improving receptor mediated uptake employingappropriate ligands. The endowment of liposomes withendosomal escape mechanisms, coupled with more efficienttranslocation of DNA to the nucleus and the efficientdissociation of the liposome complex just before the entryof free DNA into the nucleus might provide an optimalcornerstone solution to the problem. This proposition isdepicted in Figure 6.

3. Temporary Depot Polymeric-Based Systemsfor Liposomal Coupling

Polymer-based systems, such as hydrogel or prefabricatedscaffolds have been used as depots for drugs, regenerativecells, protein, growth factor, and pre-encapsulated drug-loaded liposome for sustained release [8, 12, 81–85]. Variouspolymers have been researched for this application basedon their fundamental properties such as biodegradability,biocompatibility, nontoxicity, and the noninflammatorytendency. Natural and synthetic biodegradable polymericsystems such chitosan, collagen, gelatin, fibrin, alginate,dextran, carbopol, and polyvinyl alcohol have been employedas temporary depot-forming agents since they meet most ofthe above requirements [11, 84, 86, 87].

3.1. Injectable Polymeric Scaffolds. The strategy for gener-ating an ideal depot for an active compound or bioactivemolecule-loaded liposome with the benefit of in local drugretention and sustained release over prolonged time hasrecently received much attention in both pharmaceuticaland bioengineering research [85, 88]. The in-situ forminginjectable polymer was among the most successful models,since it was able to encapsulate protein and/or bioactivemolecules or function as a pre-encapsulated drug-loadedliposomal formulation that was in liquid form [89, 90].This solution or suspension mixture could then be injectedinto the target organ with a needle to form a semisolid


scaffold and finally an implant. The success in shifting fromliquid formulation to semisolid and finally to an implantwas a result of various desirable polymeric properties andstimulating agents such as water, light, temperature, and pH,that facilitated such processes within the polymer such asprecipitation, cross-linking, and polymerization [88, 91–93].Since the majority of hydrogels were composed of naturalor synthetic biodegradable polymers, bioactive moleculeswere released via passive diffusion, matrix pore formation,or polymeric degradation [94–97]. Furthermore, semisolidimplant formation was reported as being dependant onthe polymeric state such as phase inversion, low-glasstransition temperature, or on hydrogels that formed by theaid of cross-linking reagents and chemo- or thermosensi-tazation [98, 99]. In addition, the system could deliver drugdirectly or indirectly to the targeted sites, through subcu-taneous injection and/or intratumoral injection (Figure 7)[93]. Overall, the semisolid temporary depots offer sev-eral advantages such as enhanced local drug retention,sustained release, and potential for long-term storage.However, repeated injections and passive drug release arestill a factor that limits their use as ideal pharmaceuticalcarriers.

3.2. Prefabricated Polymeric Scaffolds. Prefabricated poly-meric scaffolds have gained a lot of attention as depots fordelivery of bioactive molecules, regenerative cells, growthfactors, and pre-encapsulated bioactive loaded liposome[100, 101]. Unlike injectable in situ scaffolds in which asemisolid scaffold is achieved after injection, prefabricatedpolymer scaffold solid depot materials are formed outsidethe body, then surgically implanted [102]. In additional,pre-fabrication polymeric scaffold can be designed to meetthe required characteristics of an ideal scaffold. Desir-able attributes of an ideal scaffold are: three-dimensionalstructure, appropriate surface chemistry, fabrication frommaterials which are biodegradable or bioresorbable, shouldnot induce any adverse response, scaled pore capacity, andhighly reproducible shapes and size [99, 101]. Differentfabrication techniques have been used to achieve the abovecriteria, such as fiber bonding, emulsion freeze drying,solvent casting, high-pressure processing, gas foaming, andelectrospinning [102–105]. Various polymers that have beenresearched for this application are either biodegradable ornondegradable, synthetic or natural, or a combination ofthe two [9, 106]. The major challenge of prefabricatedpolymeric scaffolds is that a nondegradable polymeric devicerequires surgical removal at the end of treatment, whichis often known to be associated with pain [107]. However,the benefit on sustained release for the pre-encapsulateddrug loaded scaffold over a long period of time has beenreported and declared successful [96]. Stenekes and cowork-ers [8] demonstrated that liposome embedded inside abiodegaradble depot polymeric scaffold was able to sustaineddrug release over a prolonged period of time (Figure 8).In addition, the released liposome was found intact aftermany days storage within the inside depot polymericscaffold.

4. Natural Product-Based Liposomal DrugDelivery Systems

4.1. Collagen-Based Liposomal Drug Delivery Systems. Col-lagen is a major natural protein component in mammalsthat is fabricated from glycine-proline-(hydroxy) prolinerepeats to form a triple helix molecular structure [84]. Sofar, nineteen types of collagen molecules have been isolated,characterized, and reported in both medical and pharma-ceutical applications [108–110]. Collagen has been widelyused in pharmaceutical applications due to the fulfillment ofmany requirements of a drug delivery system such as goodbiocompatibility, low antigenicity, and degradability uponimplantation [111]. Furthermore, collagen gels are one of thefirst natural polymers to be used as a promising matrix fordrug delivery and tissue engineering [112]. Biodegradablecollagen-based systems have served as 3D scaffold for cellculture, survival of transfected fibroblasts, and gene therapy[81, 113]. In this case, collagen scaffolds were fabricatedthrough introducing various chemical cross-linking agents(i.e., glutaraldehyde, formaldehyde, carbodiimide) or byphysical treatments (i.e., UV irradiation, freeze-drying, andheating) [109, 114–117]. The combination of liposomes andcollagen-based technologies has been long achieved sincethe early 80s [112]. In this case, drugs and other bioactiveagents were firstly encapsulated in the liposomes and thenembedded inside a depot composed of collagen-based sys-tems, including scaffolds and gels. The combination of thesetwo technologies (i.e., liposomes and collagen-based system)has improved storage stability, prolonged the drug releaserate, and increased the therapeutic efficacy [84, 118, 119].In addition, a study that was conducted by Marston et al.[120], demonstrated that temperature sensitive liposomesand collagen may thermally trigger the release of calciumand phosphate salts. Multiple collagen-based system forpharmaceutical carriers or medicinal applications are cur-rently available for clinical purposes [121]. Figure 9 depictsa schematic representation of collagen-based liposome.

4.2. Gelatin-Based Liposomal Drug Delivery Systems. Gelatinis a common natural polymer or protein which is normallyproduced by denaturing collagen [122]. It has been used inpharmaceutical and medical applications due to its outstand-ing properties such as biodegradability, biocompatibility, andlow antigenicity [100]. In addition, gelatin can be easy tomanipulate due to its isoelectric point that allows it to changefrom negative to positive charge in an appropriate physiolog-ical environment or during the fabrication, a property thathas found it being very attractive to many pharmaceuticalresearchers [123]. Gelatin is one of the natural polymersused as support material for gene delivery, cell culture,and more recently tissue engineering. Gelatin-based systemshave the ability to control release of bioactive agents suchas drugs, protein, and dual growth factors [95, 100, 124].It has been reported that it is possible to incorporateliposome-loaded bioactive compounds into PEG-gelatin gelwhich function as porous scaffold gelatin-based temporarydepots with controlled drug release over prolonged periods


Targeting ligand& stealth facility

DNA-liposome complex

Translocation to nucleus

Membrane receptors

(4)

Endosomal escapeand complex dissociation

(3)

Endosomal uptake(2)

Binding ligand/receptor(1)

Figure 6: A schematic depicting the optimization of liposomal gene delivery (source: Uchegbu [74]).

of time [125, 126]. However, some setbacks have beenidentified, and they are said be associated with the useof gelatin-based systems in pharmaceutical applications.These setbacks include poor mechanical strength and inef-fectiveness in the management of infected sites [108]. Acombination of a collagen-based system with liposomes hasbeen proposed to achieve the stability of the system andcontrolled release profiles of the incorporated compounds.The success of these formulations, (i.e., gelatin, hydrogel,and scaffolds) was enhanced by cross-linking agents such asglutaraldehyde, sugar, and enzyme transglutaminase. It wasalso discovered that the cross-linking density of gelatin wasable to affect the rate of degradation and rate of bioactiveagents release from gelatin vehicles or from liposomesembedded inside gelatin-based systems [127–130]. Anotherstudy by Peptu and coworkers [83] reported a controlledrelease of liposome-encapsulated calcein fluoroscence dyeor calcein labeled with rhodamine from temporary depotof gelatin-based system which is made up of Gelatin-carboxymethylcellulose films. In the same study, the releaserate of loaded liposome was found to depend mostly onthe quantity of liposomes entrapped inside the films, degreeof swelling of the film, film network density, and the filmgeometry which was supported by glutaraldehyde cross-linking agents. In a similar study, DiTzio and coworkers [125]demonstrated the success of prevention of bacterial adhesionto catheters by ciprofloxacin-loaded liposomes which wereentrapped inside a poly(ethylene glycol-)gelatin hydrogel.

Another study by Burke and coworkers [126] demonstratedthat there was a successive release of oxidizing reagent(sodium periodate) from thermal liposome entrapped insidea stimuli-responsive gelatinous derivative hydrogel. In gen-eral, the combination of collagen with liposome has beenreported to improve liposome stability and the controlledrelease of incorporated bioactive agents within liposomeformulations.

4.3. Chitosan-Based Liposomal Drug Delivery Systems. Chi-tosan is a natural linear bio-polyaminosaccharide polymerobtained by N-deacetylation of chitin, which is fabricatedfrom the exoskeleton of marine crustaceans such as shrimps,crabs, prawns, and fungi [87, 131]. It has been broadlyinvestigated in pharmaceutical applications as a bioactivemolecule delivery method or as depot of pharmaceuticalcarriers due to its desirable properties such as mucoadhe-siveness, biodegradability, biocompatibility, and nontoxicity[132–135]. The combination of chitosan with liposometechnologies is considered as being a promising approachin the drug delivery arena. More recently, chitosan tech-nology has been reported as being a depot for liposomaldrug delivery systems in the form of porous hydrogel orscaffold. Chitosan-based hydrogels were generate with orwithout a cross-linking agent such as glutaraldehyde or byinteracting with different types of divalent and polyvalentanions [12, 136, 137]. Novel in situ gelling formulationsof hydrogels such as thermosensitive and mucobioadhesive


Liposome

Bioactive agent

Stimuli-sensitive hydrogel Stimuli-sensitive hydrogel and liposomes

Subcuntaneous injection Intratumoral injection

Figure 7: Schematic depicting drug delivery from pre-encapsulated drug-loaded liposomes incorporated within an injectable hydrogel-basedsystem (adapted from Ta et al. [93]).

hydrogels have been recently been proposed as a depotfor liposomes for sustained drug release over a prolongperiod of time [12, 138]. Chitosan scaffold matrix can befabricated with unique structure by simple approaches suchlyophilization technique, by use of crosslinked agents ofchitosan solution/hydrogels followed by incubation in theliquid nitrogen, or by employing liquid carbon dioxide,solid-liquid separation, and, most recently, supercriticalimmersion precipitation techniques [11, 139–141]. Drugssuch as cytarabine that have been pre-encapsulated inliposomes and then incorporated within chitosan hydrogelshave been proven to be suitable model for drug delivery withsustained drug release in vivo at body temperature [12].

4.4. Fibrin-Based Liposomal Drug Delivery Systems. Fibrinis a biodegradable polymer obtained by polymerization offibrinogen in the presence of thrombin enzyme [142]. Theconcept of developing fibrin-based technology as a tempo-rary depot in both pharmaceutical and bioengineering fieldshas received considerable attention over the past decades[82, 143]. The unique properties of the fibrin-based systemssuch biodegradability and nontoxicity, have been reportedto influence the delivery efficiency of growth factors, genes,proteins, various cells and drugs [144–150]. The fabricationof semirigid fibrin scaffold upon injection has been achievedunder physiological conditions at the site of interest withrapid polymerization [147]. Furthermore, fibrin scaffoldshave also been used as temporary depots for drug deliveryvehicles by incorporation of drug-loaded liposomes alone,or by incorporation of liposomes into a chitosan matrix(containing bioactive agent molecules such as protein, drugsand genes) within the depot composed of the fibrin-basedsystems. The combination of two widespread devices, fibrinand liposome technologies, resulted in sustained bioactiveagent release over prolonged periods of time [11, 146, 150–152].

4.5. Alginate-Based Liposomal Drug Delivery Systems. Algi-nate also serves as an example of a naturally occurring

linear polysaccharide. It is extracted from seaweed, algae,and bacteria [153–155].The fundamental chemical structureof alginate is composed of (1–4)-b-D-mannuronic acid (M)and (1–4)-a-L-guluronic acid (G) units in the form ofhomopolymeric (MM- or GG-blocks) and heteropolymericsequences (MG or GM-blocks) [156]. Alginate and theirderivates are widely used by many pharmaceutical scientistsfor drug delivery and tissue engineering applications dueto its many unique properties such as biocompatibility,biodegradability, low toxicity, non-immunogenicity, watersolubility, relatively low cost, gelling ability, stabilizing prop-erties, and high viscosity in aqueous solutions [157, 158].Since alginate is anionic, fabrication of alginate hydrogelshas successively been achieved through a reaction withcross-linking agents such as divalent or trivalent cationsmainly calcium ions, water-soluble carbodiimide, and/orglutaraldehyde [159]. The cross-linking methodology wasconducted at room temperature and physiological pH [160].The success in fabricating highly porous 3D alginate scaffoldshas been through lyophilization [161]. Thus far, alginate-based systems have been successfully used as a matrix forthe encapsulation of stem cells and for controlled release ofproteins, genes, and drugs [162–166]. In addition, alginate-based systems have been used as depots for bioactive agent-loaded liposomes, for slow drug release [9, 167]. Highlyincreased efficacy has been reported from these integrateddelivery systems when compared to polymeric-based systemsor liposome-based systems alone [168, 169]. Machluf andcoworkers [170] have reported radio labeled protein releasefrom liposomes encapsulated within microspheres of thecalcium-crosslinked alginate. Another study by Hara andMiyake [171] demonstrated the release of Calcein (whichis a fluorescent dye) and Insulin from calcium alginate gel-entrapped large multilamellar liposomal vesicles in vivo.

4.6. Dextran-Based Liposomal Drug Delivery Systems. Dex-tran is a natural linear polymer of glucose linked by a 1–6 linked-glucoyranoside, and some branching of 1,3 linkedside-chains [172]. Dextran is synthesized from sucrose by


Bioactive loadedliposome polymer scaffold

Bioactive-loadedliposomes fixated in scaffold

Bioactive-loadedliposomes released

(i) Diffusion(ii) Scaffold erosion

Target Cell

Endosome

LyosomeDrug

Nucleus

Mitochondria

Recyclingendosome

12

3

45

6

7

Prefabricated

Endolysosome

Figure 8: Schematic depicting drug delivery from pre-encapsulated drug-loaded liposomes incorporated within a prefabricated polymeric-based depot system with eventual entry through a cell membrane (adapted from Stenekes et al. [8]).

certain lactic-acid bacteria, the best-known being Leuconos-toc mesenteroides and Streptococcus mutans. There are twocommercial preparations available, namely dextran 40 kilo-daltons (kDa) (Rheomacrodex) and dextran 70 Kilodaltons(kDa) (Macrodex) [173, 174]. In pharmaceutics, dextranhas been used as model of drug delivery due to its uniquecharacteristics that differentiate it from other types ofpolysaccharide. This include water solubility, biocompatibil-ity, and biodegradability [175]. In recent studies, dextranhas been regarded as a potential polysaccharide polymerthat can sustain the delivery of both proteins, vaccines, anddrugs [176–179]. Interleukin-2, which is a highly effectiveanticancer drug, is among the success obtained in deliveringa combination of drug-loaded liposome and injectabledextran hydrogel [180]. Injectable and degradable dextran-based systems for drug delivery were generated by a cross-linking reaction with photo-polymerization or free radicalpolymerization [181]. In another study by Yeo and Kohane[182], it was demonstrated that it is possible to fabricatedextran-based hydrogel using dextran derivatives such ascarboxymethyldextran derived by aldehyde-modification orcarboxymethylcellulose. In the same study, dextran-basedsystems were reported to inhibit peritoneal adhesions dueto cytotoxicity. Cytotoxicity study was demonstrated inmesothelial cells and macrophages, and it’s reported tobe associated with a crosslinked agent [182]. A study byStenekes and coworkers [8] demonstrated the successiveencapsulation of a drug-loaded liposome depot into a dex-tran polymer-based material. The polymeric-based materials

were fabricated using a two phase system, the first phase waswater and poly(ethylene glycol) and the second one watermethacyrlated dextran. The slower degradation of dextranpolymeric material resulted in sustained liposome releaseover a period of 100 days [8]. Liposomes released from depotwere reported to be intact, and there was no significantchange in liposomal size. In a gene therapy study by Liptayand coworkers [183], it was reported that recombinantDNA (which contains chloramphenicol acetyltransferase)was successively encapsulated in cationic liposomes and thenintegrated within dextran. This system was reported to bea suitable delivery system since it could stop transfectionefficiency within the colon epithelium wall in vivo [183].

5. Liposomal Drug Delivery Systems Based onSynthetic Polymers

5.1. Carbopol-Based Liposomal Drug Delivery Systems. Car-bopol hydrogel formulation is a synthetic type of hydro-gel, which is a polyacrylic acid derivative. Carbopol 980,Carbopol 974NF resin, and Carbopol 940 have been widelyused as pharmaceutical carriers due to their outstandingproperties such as bioadhesivity, biocompatibility, and lowtoxicity [184–186]. Carbopol can swell quickly in waterand adhere to the intestinal mucus because the functionalcarboxylic acid groups (–COOH) can form hydrogen bridgesto interpenetrate the mucus layer [187, 188]. Furthermore,carbopol can inhibit the activity of the dominant enzymes in


the gastrointestinal tract due to the possession of carboxylicgroups in its structure [187]. In a study that was conductedby Tang and coworkers [186], the formulation of Carbopolcontaining superporous hydrogel composites showed thatswelling was due to ionic strength in salt, sensitive at differentpH values. In recent studies, Hosny [189, 190] reportedthe possibility of incorporating drug-loaded liposome withinCarbopol hydrogel-based system which acted as a temporarydepot. They conducted the study in vitro with the aimof improving low viscosity and poor sustainability releaseover a prolonged period of time, which is associated withliposome setbacks. The results suggested that the degreeof encapsulation and prolongation of drug release rate ofeither drugs or loaded liposomes in temporary depots ofCarbopol depends to a great extent on the properties of thevesicles, such as charge and rigidity. Various drugs such asciprofloxacin and galifloxacin have been reported to havebeen employed in this system, by firstly being encapsulatedwithin liposomes and then integrated within the temporarydepot of the Carbopol-based system. These studies revealedthat loaded liposome integrated within Carbopol-basedsystem was a suitable model of drug delivery for both ocularand vaginal disorders [189–192].

5.2. Polyvinyl Alcohol-Based Liposomal Drug Delivery Systems.Polyvinyl alcohol (PVA) is a water soluble highly hydrophilicsynthetic polymer, with a molecular mass of 80 killodaltons(KDa). PVA can be used in a widely range of applicationssuch industrial, commercial, medical, and food products[193, 194]. In addition, PVA has gained a lot of attention inpharmaceutical applications due to some attractive proper-ties such as low toxicity, excellent film-forming, biodegrad-ability, emulsifying capacity, biocompatibility, and adhesiveproperties [195, 196]. PVA-based hydrogel or scaffolds havebeen fabricated using chemical cross-linking agents such ascitric acid derivative, glutaraldehyde, and formaldehyde, orby physical cross-linking processes such as ultraviolet photo-cross-linking, freezing-thawing, and radiation [126, 197,198]. Various studies have been performed on the effects ofPVA-based polymers on the release rate of pre-encapsulateddrug-loaded liposomes. In these combination systems, PVAwas postulated to enhance liposome viscosity, making themmore stable and less permeable, thus providing a sustainedrelease liposomal delivery system [185]. A recent studyconducted by Litvinchuk and coworkers [199] demonstratedthat the success of calcein-loaded liposome embedded insidea temporary depot was influenced by photocross-linking.In the same study, the fluorescence intensity was reportedto result in a sustained release effect as observed fromday 0 to 120, in both phosphate buffer saline and bloodplasma in vitro. Overall, the study demonstrated that PVAas a temporary depot offers several advantages to liposomedelivery systems. These include liposome stability, viscosity,and sustained drug release over prolonged periods of time.Ciprofloxacin, a synthetic chemotherapeutic antibiotic wasamong the drugs that were reported to have been successfullyintegrated into liposome and PVA-based delivery systems[185].

6. Techniques for Embedding Drug-LoadedLiposomes within Depot Polymeric-BasedSystems

Different techniques of loading the drug within temporarydepot polymeric-based systems either by using natural orsynthetic polymers have been reported by many researchers[8, 11, 12, 118, 185]. However, several disadvantages werefound to be associated with this approach such as lossof the efficacy of the drugs during the fabrication processdue to the acidic, basic, and/or toxic effect of the solventsemployed, heat of sonication, or biochemical interactionswith polymeric-based materials such human fibrin gel [11,200]. To avoid these setbacks, new techniques were suggestedby firstly pre-encapsulating the drugs within liposomeand then embedding the drug-loaded liposome into thetemporary depot polymeric-based system. This approachattracted many researchers as it improved drug deliveryand at the same time preserved drug bioactivity [11, 36,185, 189, 201]. The success of this technique was alsoreported after pre-encapsulating drug-loaded liposomes intofibrinogen solution, then injecting the mixture into porouschitosan films [11, 201]. Another approach using syntheticPVA was made in which thin films of liposomes werehydrated above their glass transition temperature togetherwith PVA as the hydration solution in order to enhanceliposome entrapment into the temporary depot of PVA-based system [185]. Thermosensitive hydrogel was alsoinvestigated using a chitosan derivative, which is temperaturesensitive. In this case, drug-loaded liposome was mixedwith prechilled solutions of chitosan solution until an iso-osmotic pressure was achieved within the chitosan solution[12]. In another study that was conducted by Gobin andcoworkers [36], it was demonstrated that drug-loaded lipo-somes were incorporated within a polymeric-based systemwith agitation and subsequently lyophilized after beingfrozen overnight at −80◦C. Tabandeh and Aboufazelia [118]suggested a nitrogen refrigeration approach. In this case, pre-encapsulated drug-loaded liposomes were mixed togetherwith collagen solution and then frozen in liquid nitrogenfor 24 hours. Since soluble collagen was used in this study,adequate concentrations of collagen were suggested in orderto facilitate the drug release and avoid the chain mobilityassociated with collagen.

A more recent study has demonstrated an enhanced pro-cess of drug-encapsulated liposome into Carbopol hydrogelby using deionized water as a vehicle (i.e., employing ahydration approach) [189]. This involved the developmentof an effective prolonged-release liposomal hydrogel for-mulation containing ciprofloxacin for ocular therapy. Drugdelivery in ocular therapy has for long been a difficult taskto accomplish because of the poor drug bioavailability thatis mainly due to the precorneal loss factors. These factorsinclude tear dynamics, insufficient residence time in theconjunctival sac, and nonproductive absorption [185, 202].Thus far, fluoroquinolones have shown excellent activityagainst most of the frequently occurring Gram-positive andGram-negative ocular pathogens [189]. Earlier generationsof fluoroquinolones (e.g., ofloxacin) were often encountered


Sgc8-TMR

PEG

Dextrin-FITC

Lipid

sgc8 sgc8

CholChol

HSPC: hydrogenated soy phosphatidyl choline

Chol: cholesterol

MPEG-DSPE: methoxypoly (ethylene glycol)-

distearoylphosphatidylethanolamine

MalPEG: maleimide-terminated PEG-DSPE

Figure 9: A schematic representation of a collagen-based liposome (source: Kang et al. [121]).

with a problem of developing resistance at a fast rate [203,204]. Ciprofloxacin is active against a broad spectrum ofaerobic Gram-positive and Gram-negative bacteria. Resis-tance to this drug develops slowly and has shown to causea minimal toxicity [189]. It is currently the drug of choiceas an anti-infective ocular agent [205, 206]. Efficacy ofthe marketed ophthalmic fluoroquinolone products, mostlyaqueous solutions, is limited by poor ocular bioavailability,compelling the frequent dosing regimen, and uncompro-mised patient compliance [207, 208]. Thus, prolonged-release ciprofloxacin liposomal hydrogel has proven to be asuitable delivery system for ocular infections.

7. Modulating Drug Release from Liposomeswithin Polymeric Depot Systems

Sustained release of therapeutically active compounds loadedwith liposome in a depot incorporated into polymeric-basedsystem offers the possibility of reducing the dosing frequency,which may lead to the reduction of side effects and thereforesustained drug action [12]. A study that was conductedby Machluf and coworkers [170] demonstrated that radio-labeled protein-loaded liposomes could be embedded withintwo membrane layers of a polymeric-based system such ascalcium cross-linked alginate and alginate integrated withpoly(l-lysine) for sustained release of radio-labeled bovineserum albumin both in vitro and in vivo. In another setof studies, it was postulated that the success of liposomerelease from polymeric-based systems could be due to meshsize of the matrix, size of liposome, diffusion, chemical,pH, and/or enzyme factor [8, 82, 112, 209]. In yet anotherstudy by Dhoot and Wheatley [168], liposome releasefrom barium-alginate depots was reported to be influencedby the cross-linking ions. Leakiness of liposomes duringthe encapsulation process was due to high lipid content(i.e., cholesterol) during liposome fabrication for which

a high liposomal escape was also observed. In comparingthe liposome and degradable system to the liposome andnondegradable polymer-based systems, the results indicatedthat the liposomal release for the first system was dueto degradation of the polymeric matrix, while for thesecond system an insufficient release was observed duringthe same period of study [210]. Nixon and Yeung [164]conducted a study together with Stenekes and coworkers[8] in which they could demonstrate that liposomes withlow and high membrane fluidity were successfully releasedfrom a polymeric-based system in their intact form andwith preserved size for approximately 60 days. Although pre-encapsulated drug-loaded liposome could show controlleddrug release from the depot, majority of these studies haveshown that the obtained drug release profiles depended to agreater extent on the liposomal burst effect rather than thediffusion process [11, 170, 201].

8. The Successes and Challenges Emergingfrom Composite Liposome andPolymeric-Based Technologies

The combination of liposome-based system and polymeric-based system for sustained release of therapeutically activecompounds has been demonstrated to be successful inpharmaceutical applications. Sustained release profiles of dif-ferent bioactive molecules such as gene, drugs, protein, andgrowth factor from liposome encapsulated in both naturalor synthetic biodegradable polymeric material have beenobtained [12, 169, 171]. The success of this drug deliverycombination depends mostly on encapsulation efficacy andthe type of drug release profile that is obtained. Efficiencyin encapsulating drug-loaded liposome was reported to bedependent on several techniques, such as cross-linking agents(glutaraldehyde, formaldehyde, carbodiimide) or physical


treatments (i.e., UV irradiation, freeze-drying), during fab-rication process [152, 160]. Sustained release kinetics of thepre-encapsulated drug-loaded liposome depends most onthe degradation rate of the polymeric materials. This systemhas added a remarkable advantage to both technologies(i.e., liposome-based and polymeric-based), though more soto the liposome technology since polymeric materials aremore stable than liposomes. The following properties wereachieved by embedding the liposomes into a polymeric-based system: (i) sustained release over prolonged periodsof time, (ii) improved viscosity, (iii) stability of liposome,and (iv) improved half-life for both the drugs and liposome.In polymeric-based system incorporated with liposomes,drug delivery efficacy and preservation of drug bioactivityhas been achieved. This is due to the fact that liposomeshave a higher degree of biocompatibility when comparedto polymeric materials [8, 36]. Although this compositesystem demonstrated improved success, there are still somemajor challenges that need to be overcome. Incorporation oftoxic organic solvent or high heat during fabrication processcan inhibit the activity of some bioactive molecules such asprotein [11, 200]. Furthermore, since drug-loaded liposomerelease profiles seem to depend most on degradation ofpolymeric materials, majority of drug-loaded liposome mayremain enmeshed within the depot or insufficient initialrelease at commencement of treatment may be a problem. Atthe same time, high overdose may occur during high degra-dation period. In either case, degradable polymeric materialhas demonstrated more efficacy than nondegradable poly-meric material since with the latter depot, insufficient drugrelease was reported [210].

9. Future Perspective

Significant development has been reported on combinationof the liposome-based technology with temporary depotpolymeric-based technology in sustaining drug release overprolonged periods of time. However, combination of bothdrug delivery technologies into a single model of drugdelivery has been reported to be associated with inad-equate drug release. Since both materials can be easilymanipulated, design of a new ideal temporary depot ofthe polymeric-based technologies to enhance therapeuticefficacy or improve the drug release profile is of a greatinterest. Integration of the more advanced types of liposome-based technologies such as targeted- or stimuli-sensitiveliposomes in this system can enhance therapeutic efficacy.In addition, targeted liposome formulations, with targetedmoieties such as antibodies, peptide, glycoprotein, polysac-charide, growth factors, carbohydrate, and receptors mayincrease liposomal drug accumulation in the tissues/cells viaoverexpressed receptors, antigen, and unregulated selectins.Sensitivity of liposomes to pH, light, magnetism, tem-perature, and ultrasonic waves can enhance therapeuticefficacy. Some polymeric systems have demonstrated somedisadvantages in this application such as nondegradabilitythat results in insufficient drug release. The use of a combi-nation liposomal-based system with natural and/or synthetic

polymeric biodegradable and/or nondegradable polymersmay add strength to the depot while improving liposomalrelease profile. Although organic solvent are normally addedduring fabrication, nontoxicity should be rigorously assessedin ex vivo studies. In summary, the combination system, asa model of sustained release of drug-loaded liposome fromtemporary polymeric depots, has been declared successfulbut system improvements are demanded. Since this system isimplantable, it may be useful in future for the managementof chronic diseases such as Aid Dementia Complex, Tuber-culosis, Cancer, or Neurodegenerative disorders, such asParkinson’s and Alzheimer’s disease, which normally requireregular doses over prolonged periods of time.

Acknowledgments

This work was supported by the National Research Foun-dation (NRF) and the Faculty of Health Science IndividualResearch Grant of the University of the Witwatersrand,Johannesburg, South Africa.

References

[1] E. Mastrobattista, G. A. Koning, L. van Bloois, A. C. S. Filipe,W. Jiskoot, and G. Storm, “Functional characterization of anendosome-disruptive peptide and its application in cytosolicdelivery of immunoliposome-entrapped proteins,” Journal ofBiological Chemistry, vol. 277, no. 30, pp. 27135–27143, 2002.

[2] A. Schnyder and J. Huwyler, “Drug transport to brain withtargeted liposomes,” NeuroRx, vol. 2, no. 1, pp. 99–107, 2005.

[3] M. L. Immordino, F. Dosio, and L. Cattel, “Stealth lipo-somes: review of the basic science, rationale, and clinicalapplications, existing and potential,” International Journal ofNanomedicine, vol. 1, no. 3, pp. 297–315, 2006.

[4] C. Chen, D. Han, C. Cai, and X. Tang, “An overview ofliposome lyophilization and its future potential,” Journal ofControlled Release, vol. 142, no. 3, pp. 299–311, 2010.

[5] A. D. Bangham, M. W. Hill, and G. A. Miller, “Preparationand use of liposomes as models of biological membranes,”in Methods in Membrane Biology, vol. 1, pp. 61–68, PlenumPress, New York, NY, USA, 1974.

[6] A. Yousefi, F. Esmaeili, S. Rahimian, F. Atyabi, and R. Dinar-vand, “Preparation and in vitro evaluation of a pegylatednano-liposomal formulation containing docetaxel,” ScientiaPharmaceutica, vol. 77, no. 2, pp. 453–464, 2009.

[7] V. P. Torchilin, “Recent advances with liposomes as pharma-ceutical carriers,” Nature Reviews Drug Discovery, vol. 4, no.2, pp. 145–160, 2005.

[8] R. J. H. Stenekes, A. E. Loebis, C. M. Fernandes, D. J.A. Crommelin, and W. E. Hennink, “Controlled release ofliposomes from biodegradable dextran microspheres: a noveldelivery concept,” Pharmaceutical Research, vol. 17, no. 6, pp.690–695, 2000.

[9] M. Hara and J. Miyake, “Calcium alginate gel-entrappedliposomes,” Materials Science and Engineering C, vol. 17, no.1-2, pp. 101–105, 2001.

[10] D. G. Wallace and J. Rosenblatt, “Collagen gel systems forsustained delivery and tissue engineering,” Advanced DrugDelivery Reviews, vol. 55, no. 12, pp. 1631–1649, 2003.

[11] T. W. Chung, M. C. Yang, and W. J. Tsai, “A fibrin encap-sulated liposomes-in-chitosan matrix (FLCM) for delivering


water-soluble drugs: influences of the surface properties ofliposomes and the crosslinked fibrin network,” InternationalJournal of Pharmaceutics, vol. 311, no. 1-2, pp. 122–129, 2006.

[12] R. Mulik, V. Kulkarni, and R. S. R. Murthy, “Chitosan-basedthermosensitive hydrogel containing liposomes for sustaineddelivery of cytarabine,” Drug Development and IndustrialPharmacy, vol. 35, no. 1, pp. 49–56, 2009.

[13] R. I. Mahato, “Water insoluble and soluble lipids for genedelivery,” Advanced Drug Delivery Reviews, vol. 57, no. 5, pp.699–712, 2005.

[14] J. K. Vasir, M. K. Reddy et al., “Multifunctional water-solublepolymers for drug delivry,” Current Nanoscience, vol. 1, pp.47–64, 2005.

[15] J. Y. Fang, T. L. Hwang, and Y. L. Huang, “Liposomes asvehicles for enhancing drug delivery via skin routes,” CurrentNanoscience, vol. 2, no. 1, pp. 55–70, 2006.

[16] D. Zucker, D. Marcus, Y. Barenholz, and A. Goldblum, “Lipo-some drugs’ loading efficiency: a working model based onloading conditions and drug’s physicochemical properties,”Journal of Controlled Release, vol. 139, no. 1, pp. 73–80, 2009.

[17] M. Manconi, C. Sinico, D. Valenti, G. Loy, and A. M.Fadda, “Niosomes as carriers for tretinoin. I. Preparation andproperties,” International Journal of Pharmaceutics, vol. 234,no. 1-2, pp. 237–248, 2002.

[18] M. Johnsson and K. Edwards, “Liposomes, disks, andspherical micelles: aggregate structure in mixtures ofgel phase phosphatidylcholines and poly(ethylene glycol)-phospholipids,” Biophysical Journal, vol. 85, no. 6, pp. 3839–3847, 2003.

[19] D. J. Bharali, M. Khalil, M. Gurbuz, T. M. Simone, andS. A. Mousa, “Nanoparticles and cancer therapy: a concisereview with emphasis on dendrimers,” International Journalof Nanomedicine, vol. 4, no. 1, pp. 1–7, 2009.

[20] H. Harashima, K. Sakata, K. Funato, and H. Kiwada,“Enhanced hepatic uptake of liposomes through comple-ment activation depending on the size of liposomes,” Phar-maceutical Research, vol. 11, no. 3, pp. 402–406, 1994.

[21] R. M. Abra, R. B. Bankert, F. Chen et al., “The next generationof liposome delivery systems: recent experience with tumor-targeted, sterically-stabilized immunoliposomes and active-loading gradients,” Journal of Liposome Research, vol. 12, no.1-2, pp. 1–3, 2002.

[22] L. Cattel, M. Ceruti, and F. Dosio, “From conventional tostealth liposomes: a new frontier in cancer chemotherapy,”Journal of Chemotherapy, vol. 16, no. 4, pp. 94–97, 2004.

[23] J. Senior and G. Gregoriadis, “Is half-life of circulatingliposomes determined by changes in their permeability?”FEBS Letters, vol. 145, no. 1, pp. 109–114, 1982.

[24] M. M. Frank, “The reticuloendothelial system and blood-steam clearance,” Journal of Laboratory and Clinical Medicine,vol. 122, no. 5, pp. 487–488, 1993.

[25] E. L. Riche, B. W. Erickson, and M. J. Cho, “Novellong-circulating liposomes containing peptide library-lipidconjugates: synthesis and in vivo behavior,” Journal of DrugTargeting, vol. 12, no. 6, pp. 355–361, 2004.

[26] S. J. H. Soenen, A. R. Brisson, and M. De Cuyper, “Addressingthe problem of cationic lipid-mediated toxicity: the magne-toliposome model,” Biomaterials, vol. 30, no. 22, pp. 3691–3701, 2009.

[27] K. Nishikawa, H. Arai, and K. Inoue, “Scavenger receptor-mediated uptake and metabolism of lipid vesicles containingacidic phospholipids by mouse peritoneal macrophages,”Journal of Biological Chemistry, vol. 265, no. 9, pp. 5226–5231, 1990.

[28] J. Damen, J. Regts, and G. Scherphof, “Transfer and exchangeof phospholipid between small unilamellar liposomes and ratplasma high density lipoproteins. Dependence on cholesterolcontent and phospholipid composition,” Biochimica et Bio-physica Acta, vol. 665, no. 3, pp. 538–545, 1981.

[29] M. A. Tran, R. J. Watts, and G. P. Robertson, “Use of lipo-somes as drug delivery vehicles for treatment of melanoma,”Pigment Cell and Melanoma Research, vol. 22, no. 4, pp. 388–399, 2009.

[30] A. Gabizon and D. Papahadjopoulos, “Liposome formula-tions with prolonged circulation time in blood and enhanceduptake by tumors,” Proceedings of the National Academy ofSciences of the United States of America, vol. 85, no. 18, pp.6949–6953, 1988.

[31] S. Mondal, P. Bhattacharya, M. Rahaman, N. Ali, and R.P. Goswami, “A curative immune profile one week aftertreatment of Indian Kala-Azar patients predicts success witha short-course liposomal amphotericin B therapy,” PLoSNeglected Tropical Diseases, vol. 4, no. 7, Article ID e764,2010.

[32] T. M. Allen and F. J. Martin, “Advantages of liposomaldelivery systems for anthracyclines,” Seminars in Oncology,vol. 31, no. 13, pp. 5–15, 2004.

[33] P. R. Veerareddy and V. Vobalaboina, “Lipid-based formula-tions of amphotericin B,” Drugs of Today, vol. 40, no. 2, pp.133–145, 2004.

[34] S. D. Li and L. Huang, “Stealth nanoparticles: high densitybut sheddable PEG is a key for tumor targeting,” Journal ofControlled Release, vol. 145, no. 3, pp. 178–181, 2010.

[35] L. Ruizhen, G. Lu, Y. Xiangliang, and X. Huibi, “Chitosan asa condensing agent induces high gene transfection efficiencyand low cytotoxicity of liposome,” Journal of Bioscience andBioengineering, vol. 111, no. 1, pp. 98–103, 2011.

[36] A. S. Gobin, R. Rhea, R. A. Newman, and A. B. Mathur, “Silk-fibroin-coated liposomes for long-term and targeted drugdelivery,” International Journal of Nanomedicine, vol. 1, no.1, pp. 81–87, 2006.

[37] K. Nakano, Y. Tozuka, and H. Takeuchi, “Effect of surfaceproperties of liposomes coated with a modified polyvinylalcohol (PVA-R) on the interaction with macrophage cells,”International Journal of Pharmaceutics, vol. 354, no. 1-2, pp.174–179, 2008.

[38] Y. Wang, S. Tu, R. Li, X. Yang, L. Liu, and Q. Zhang, “Choles-terol succinyl chitosan anchored liposomes: preparation,characterization, physical stability, and drug release behav-ior,” Nanomedicine: Nanotechnology, Biology, and Medicine,vol. 6, no. 3, pp. 471–477, 2010.

[39] C. Allen, N. Dos Santos, R. Gallagher et al., “Controlling thephysical behavior and biological performance of liposomeformulations through use of surface grafted poly(ethyleneglycol),” Bioscience Reports, vol. 22, no. 2, pp. 225–250, 2002.

[40] F. Atyabi, A. Farkhondehfai, F. Esmaeili, and R. Dinar-vand, “Preparation of pegylated nano-liposomal formulationcontaining SN-38: in vitro characterization and in vivobiodistribution in mice,” Acta Pharmaceutica, vol. 59, no. 2,pp. 133–144, 2009.

[41] S. E. Krown, D. W. Northfelt, D. Osoba, and J. S. Stewart,“Use of liposomal anthracyclines in Kaposi’s sarcoma,”Seminars in Oncology, vol. 31, no. 13, pp. 36–52, 2004.

[42] G. L. Scherphof, J. Dijkstra, and H. H. Spanjer, “Uptakeand intracellular processing of targeted and nontargetedliposomes by rat Kupffer cells in vivo and in vitro,” Annalsof the New York Academy of Sciences, vol. 446, pp. 368–384,1985.


[43] P. Sapra and T. M. Allen, “Ligand-targeted liposomal anti-cancer drugs,” Progress in Lipid Research, vol. 42, no. 5, pp.439–462, 2003.

[44] O. P. Medina, Y. Zhu, and K. Kairamo, “Targeted liposomaldrug delivery in cancer,” Current Pharmaceutical Design, vol.10, no. 24, pp. 2981–2989, 2004.

[45] C. K. Song, S. H. Jung, D. D. Kim, K. S. Jeong, B. C.Shin, and H. Seong, “Disaccharide-modified liposomes andtheir in vitro intracellular uptake,” International Journal ofPharmaceutics, vol. 380, no. 1-2, pp. 161–169, 2009.

[46] K. Takara, H. Hatakeyama, N. Ohga, K. Hida, and H.Harashima, “Design of a dual-ligand system using a specificligand and cell penetrating peptide, resulting in a synergisticeffect on selectivity and cellular uptake,” International Journalof Pharmaceutics, vol. 396, no. 1-2, pp. 143–148, 2010.

[47] H. Shmeeda, Y. Amitay, J. Gorin et al., “Delivery of zoledronicacid encapsulated in folate-targeted liposome results inpotent in vitro cytotoxic activity on tumor cells,” Journal ofControlled Release, 2010.

[48] X. Li, L. Ding, Y. Xu, Y. Wang, and Q. Ping, “Targeteddelivery of doxorubicin using stealth liposomes modifiedwith transferrin,” International Journal of Pharmaceutics, vol.373, no. 1-2, pp. 116–123, 2009.

[49] V. P. Torchilin, “Tat peptide-mediated intracellular deliveryof pharmaceutical nanocarriers,” Advanced Drug DeliveryReviews, vol. 60, no. 4-5, pp. 548–558, 2008.

[50] X. Ying, HE. Wen, W. L. Lu et al., “Dual-targeting daunoru-bicin liposomes improve the therapeutic efficacy of brainglioma in animals,” Journal of Controlled Release, vol. 141, no.2, pp. 183–192, 2010.

[51] D. Simonis, M. Schlesinger, C. Seelandt, L. Borsig, and G.Bendas, “Analysis of SM4 sulfatide as a P-selectin ligand usingmodel membranes,” Biophysical Chemistry, vol. 150, no. 1–3,pp. 98–104, 2010.

[52] M. N. Hossen, K. Kajimoto, H. Akita, M. Hyodo, T. Ishitsuka,and H. Harashima, “Ligand-based targeted delivery of apeptide modified nanocarrier to endothelial cells in adiposetissue,” Journal of Controlled Release, vol. 147, no. 2, pp. 261–268, 2010.

[53] S. Hua, H. I. Chang, N. M. Davies, and P. J. Cabot, “Tar-geting of ICAM-1-directed immunoliposomes specifically toactivated endothelial cells with low cellular uptake: use of anoptimized procedure for the coupling of low concentrationsof antibody to liposomes,” Journal of Liposome Research. Inpress.

[54] B. Yu, H. C. Tai, W. Xue, L. J. Lee, and R. J. Lee, “Receptor-targeted nanocarriers for therapeutic delivery to cancer,”Molecular Membrane Biology, vol. 27, no. 7, pp. 286–298,2010.

[55] K. M. Stewart, K. L. Horton, and S. O. Kelley, “Cell-penetrating peptides as delivery vehicles for biology andmedicine,” Organic and Biomolecular Chemistry, vol. 6, no.13, pp. 2242–2255, 2008.

[56] L. Nobs, F. Buchegger, R. Gurny, and E. Allemann, “Currentmethods for attaching targeting ligands to liposomes andnanoparticles,” Journal of Pharmaceutical Sciences, vol. 93, no.8, pp. 1980–1992, 2004.

[57] A. N. Lukyanov, T. A. Elbayoumi, A. R. Chakilam, andV. P. Torchilin, “Tumor-targeted liposomes: doxorubicin-loaded long-circulating liposomes modified with anti-cancerantibody,” Journal of Controlled Release, vol. 100, no. 1, pp.135–144, 2004.

[58] P. Rai, D. Vance, V. Poon, J. Mogridge, and R. S. Kane, “Stableand potent polyvalent anthrax toxin inhibitors: raft-inspired

domain formation in liposomes that contain PEGylatedlipids,” Chemistry: A European Journal, vol. 14, no. 26, pp.7748–7751, 2008.

[59] R. Nallamothu, G. C. Wood, C. B. Pattillo et al., “Atumor vasculature targeted liposome delivery system forcombretastatin A4: design, characterization, and in vitroevaluation,” AAPS PharmSciTech, vol. 7, no. 2, pp. E1–E10,2006.

[60] M. Zoldakova, Z. Kornyei, A. Brown, B. Biersack, E.Madarasz, and R. Schobert, “Effects of a combretastatin A4analogous chalcone and its Pt-complex on cancer cells: acomparative study of uptake, cell cycle and damage to cellularcompartments,” Biochemical Pharmacology, vol. 80, no. 10,pp. 1487–1496, 2010.

[61] S. L. Young and D. J. Chaplin, “Combretastatin A4 phos-phate: background and current clinical status,” Expert Opin-ion on Investigational Drugs, vol. 13, no. 9, pp. 1171–1182,2004.

[62] C. C. Kumar, L. Armstrong, Z. Yin et al., “Targeting inte-grins αβ and αβ for blocking tumor-induced angiogenesis,”Advances in Experimental Medicine and Biology, vol. 476, pp.169–180, 2000.

[63] R. O. Hynes, “A reevaluation of integrins as regulators ofangiogenesis,” Nature Medicine, vol. 8, no. 9, pp. 918–921,2002.

[64] E. S. Kim, C. Lu, F. R. Khuri et al., “A phase II study ofSTEALTH cisplatin (SPI-77) in patients with advanced non-small cell lung cancer,” Lung Cancer, vol. 34, no. 3, pp. 427–432, 2001.

[65] P. Goyal, K. Goyal, S. G. V. Kumar, A. Singh, OM. P. Katare,and D. N. Mishra, “Liposomal drug delivery systems—clinical applications,” Acta Pharmaceutica, vol. 55, no. 1, pp.1–25, 2005.

[66] P. Pradhan, J. Giri, F. Rieken et al., “Targeted temperaturesensitive magnetic liposomes for thermo-chemotherapy,”Journal of Controlled Release, vol. 142, no. 1, pp. 108–121,2010.

[67] Y. Liu, H. Miyoshi, and M. Nakamura, “Encapsulated ultra-sound microbubbles: therapeutic application in drug/genedelivery,” Journal of Controlled Release, vol. 114, no. 1, pp. 89–99, 2006.

[68] Y. Kaneda, “Virosomes: evolution of the liposome as atargeted drug delivery system,” Advanced Drug DeliveryReviews, vol. 43, no. 2-3, pp. 197–205, 2000.

[69] A. Schroeder, J. Kost, and Y. Barenholz, “Ultrasound, lipo-somes, and drug delivery: principles for using ultrasound tocontrol the release of drugs from liposomes,” Chemistry andPhysics of Lipids, vol. 162, no. 1-2, pp. 1–16, 2009.

[70] I. Lentacker, N. Wang, R. E. Vandenbroucke, J. Demeester,S. C. De Smedt, and N. N. Sanders, “Ultrasound exposureof lipoplex loaded microbubbles facilitates direct cytoplasmicentry of the lipoplexes,” Molecular Pharmaceutics, vol. 6, no.2, pp. 457–467, 2009.

[71] M. A. Bellavance, M. B. Poirier, and D. Fortin, “Uptakeand intracellular release kinetics of liposome formulations inglioma cells,” International Journal of Pharmaceutics, vol. 395,no. 1-2, pp. 251–259, 2010.

[72] S. Anabousi, U. Bakowsky, M. Schneider, H. Huwer, C.-M.Lehr, and C. Ehrhardt, “In vitro assessment of transferrin-conjugated liposomes as drug delivery systems for inhalationtherapy of lung cancer,” European Journal of PharmaceuticalSciences, vol. 29, no. 5, pp. 367–374, 2006.

[73] Pevion Biotech Ltd., “Virosomes,” July 2010, http://www.pevion.com/index.php?page=651.


[74] I. F. Uchegbu, “Parenteral drug delivery: 1,” PharmaceuticalJournal, vol. 263, no. 7060, pp. 309–318, 1999.

[75] C. R. Dass and P. F. M. Choong, “Selective gene delivery forcancer therapy using cationic liposomes: in vivo proof ofapplicability,” Journal of Controlled Release, vol. 113, no. 2,pp. 155–163, 2006.

[76] L. J. Feldman and G. Steg, “Optimal techniques for arterialgene transfer,” Cardiovascular Research, vol. 35, no. 3, pp.391–404, 1997.

[77] U. Griesenbach, A. Chonn, R. Cassady et al., “Compari-son between intratracheal and intravenous administrationof liposome-DNA complexes for cystic fibrosis lung genetherapy,” Gene Therapy, vol. 5, no. 2, pp. 181–188, 1998.

[78] J. Smith, Y. Zhang, and R. Niven, “Toward development of anon-viral gene therapeutic,” Advanced Drug Delivery Reviews,vol. 26, no. 2-3, pp. 135–150, 1997.

[79] B. K. Kim, K. O. Doh, J. H. Nam et al., “Synthesis ofnovel cholesterol-based cationic lipids for gene delivery,”Bioorganic and Medicinal Chemistry Letters, vol. 19, no. 11,pp. 2986–2989, 2009.

[80] P. L. Felgner, T. R. Gadek, M. Holm et al., “Lipofection:a highly efficient, lipid-mediated DNA-transfection proce-dure,” Proceedings of the National Academy of Sciences of theUnited States of America, vol. 84, no. 21, pp. 7413–7417, 1987.

[81] K. Wolf, S. Alexander, V. Schacht et al., “Collagen-based cellmigration models in vitro and in vivo,” Seminars in Cell &Developmental Biology, vol. 20, no. 8, pp. 931–941, 2009.

[82] S. Meyenburg, H. Lilie, S. Panzner, and R. Rudolph, “Fibrinencapsulated liposomes as protein delivery systemStudies onthe in vitro release behavior,” Journal of Controlled Release,vol. 69, no. 1, pp. 159–168, 2000.

[83] C. Peptu, M. Popa, and S. G. Antimisiaris, “Release ofliposome-encapsulated calcein from liposome entrappinggelatin-carboxymethylcellulose films: a presentation of dif-ferent possibilities,” Journal of Nanoscience and Nanotechnol-ogy, vol. 8, no. 5, pp. 2249–2258, 2008.

[84] C. Kojima, S. Tsumura, A. Harada, and K. Kono, “A collagen-mimic dendrimer capable of controlled release,” Journal ofthe American Chemical Society, vol. 131, no. 17, pp. 6052–6053, 2009.

[85] A. E. Hafeman, K. J. Zienkiewicz, E. Carney et al.,“Local delivery of tobramycin from injectable biodegradablepolyurethane scaffolds,” Journal of Biomaterials Science, Poly-mer Edition, vol. 21, no. 1, pp. 95–112, 2010.

[86] K. Kawakami, Y. Nishihara, and K. Hirano, “Effect ofhydrophilic polymers on physical stability of liposomedispersions,” Journal of Physical Chemistry B, vol. 105, no. 12,pp. 2374–2385, 2001.

[87] J. Berger, M. Reist, J. M. Mayer, O. Felt, N. A. Peppas, and R.Gurny, “Structure and interactions in covalently and ionicallycrosslinked chitosan hydrogels for biomedical applications,”European Journal of Pharmaceutics and Biopharmaceutics, vol.57, no. 1, pp. 19–34, 2004.

[88] A. A. Exner and G. M. Saidel, “Drug-eluting polymerimplants in cancer therapy,” Expert Opinion on Drug Delivery,vol. 5, no. 7, pp. 775–788, 2008.

[89] R. B. Patel, L. Solorio, H. Wu, T. Krupka, and A. A. Exner,“Effect of injection site on in situ implant formation and drugrelease in vivo,” Journal of Controlled Release, vol. 147, no. 3,pp. 350–358, 2010.

[90] C. M. Paleos, D. Tsiourvas, and Z. Sideratou, “Hydro-gen bonding interactions of liposomes simulating cell-cellrecognition. A review,” Origins of Life and Evolution of theBiosphere, vol. 34, no. 1-2, pp. 195–213, 2004.

[91] K. Deligkaris, T. S. Tadele, W. Olthuis, and A. van denBerg, “Hydrogel-based devices for biomedical applications,”Sensors and Actuators B, vol. 147, no. 2, pp. 765–774, 2010.

[92] W. Y. Lee, Y. H. Chang, Y. C. Yeh et al., “The use of injectablespherically symmetric cell aggregates self-assembled in athermo-responsive hydrogel for enhanced cell transplanta-tion,” Biomaterials, vol. 30, no. 29, pp. 5505–5513, 2009.

[93] H. T. Ta, C. R. Dass, and D. E. Dunstan, “Injectable chitosanhydrogels for localised cancer therapy,” Journal of ControlledRelease, vol. 126, no. 3, pp. 205–216, 2008.

[94] N. Bhattarai, J. Gunn, and M. Zhang, “Chitosan-basedhydrogels for controlled, localized drug delivery,” AdvancedDrug Delivery Reviews, vol. 62, no. 1, pp. 83–99, 2010.

[95] T. A. Holland, Y. Tabata, and A. G. Mikos, “Dual growthfactor delivery from degradable oligo(poly(ethylene glycol)fumarate) hydrogel scaffolds for cartilage tissue engineering,”Journal of Controlled Release, vol. 101, no. 1–3, pp. 111–125,2005.

[96] M. D. Krebs, E. Salter, E. Chen, K. A. Sutter, and E. Alsberg,“Calcium phosphate-DNA nanoparticle gene delivery fromalginate hydrogels induces in vivo osteogenesis,” Journal ofBiomedical Materials Research Part A, vol. 92, no. 3, pp. 1131–1138, 2010.

[97] L. Yu, G. T. Chang, H. Zhang, and J. D. Ding, “Injectableblock copolymer hydrogels for sustained release of a PEGy-lated drug,” International Journal of Pharmaceutics, vol. 348,no. 1-2, pp. 95–106, 2008.

[98] H. F. Zhang, H. Zhong, L. L. Zhang et al., “Modulate thephase transition temperature of hydrogels with both ther-mosensitivity and biodegradability,” Carbohydrate Polymers,vol. 79, no. 1, pp. 131–136, 2010.

[99] A. J. McHugh, “The role of polymer membrane formation insustained release drug delivery systems,” Journal of ControlledRelease, vol. 109, no. 1–3, pp. 211–221, 2005.

[100] A. Narita, M. Takahara, T. Ogino, S. Fukushima, Y. Kimura,and Y. Tabata, “Effect of gelatin hydrogel incorporatingfibroblast growth factor 2 on human meniscal cells in anorgan culture model,” The Knee, vol. 16, no. 4, pp. 285–289,2009.

[101] H. Tabesh, G. H. Amoabediny, N. S. Nik et al., “The roleof biodegradable engineered scaffolds seeded with Schwanncells for spinal cord regeneration,” Neurochemistry Interna-tional, vol. 54, no. 2, pp. 73–83, 2009.

[102] H. J. Chung and T. G. Park, “Surface engineered and drugreleasing pre-fabricated scaffolds for tissue engineering,”Advanced Drug Delivery Reviews, vol. 59, no. 4-5, pp. 249–262, 2007.

[103] D. Sin, X. Miao, G. Liu et al., “Polyurethane (PU) scaffoldsprepared by solvent casting/particulate leaching (SCPL)combined with centrifugation,” Materials Science and Engi-neering C, vol. 30, no. 1, pp. 78–85, 2010.

[104] N. Bhardwaj and S. C. Kundu, “Electrospinning: a fascinatingfiber fabrication technique,” Biotechnology Advances, vol. 28,no. 3, pp. 325–347, 2010.

[105] A. Salerno, S. Zeppetelli, E. D. Maio, S. Iannace, and P. A.Netti, “Novel 3D porous multi-phase composite scaffoldsbased on PCL, thermoplastic zein and ha prepared via super-critical CO2 foaming for bone regeneration,” CompositesScience and Technology, vol. 70, no. 13, pp. 1838–1846, 2010.

[106] A. Ghaffari, M. Oskoui, K. Helali, K. Bayati, and M. Rafiee-Tehrani, “Pectin/chitosan/Eudragit RS mixed-film coatingfor bimodal drug delivery from theophylline pellets: prepara-tion and evaluation,” Acta Pharmaceutica, vol. 56, no. 3, pp.299–310, 2006.


[107] D. Eglin and M. Alini, “Degradable polymeric materials forosteosynthesis: tutorial,” European Cells & Materials, vol. 16,pp. 80–91, 2008.

[108] R. Parenteau-Bareil, R. Gauvin, and F. Berthod, “Collagen-based biomaterials for tissue engineering applications,”Materials, vol. 3, no. 3, pp. 1863–1887, 2010.

[109] H. Chen and Z. H. Shana, “Stabilization of collagen by cross-linking with oxazolidine E-resorcinol,” International Journalof Biological Macromolecules, vol. 46, no. 5, pp. 535–539,2010.

[110] C. Holladay, M. Keeney, U. Greiser, M. Murphy, T. O’Brien,and A. Pandit, “A matrix reservoir for improved control ofnon-viral gene delivery,” Journal of Controlled Release, vol.136, no. 3, pp. 220–225, 2009.

[111] C. Yang, P. J. Hillas, J. A. Baez et al., “The applicationof recombinant human collagen in tissue engineering,”BioDrugs, vol. 18, no. 2, pp. 103–119, 2004.

[112] A. L. Weiner, S. S. Carpenter-Green, and E. C. Soehngen,“Liposome-collagen gel matrix: a novel sustained drugdelivery system,” Journal of Pharmaceutical Sciences, vol. 74,no. 9, pp. 922–925, 1985.

[113] C. Holladay, M. Keeney, U. Greiser, M. Murphy, T. O’Brien,and A. Pandit, “A matrix reservoir for improved control ofnon-viral gene delivery,” Journal of Controlled Release, vol.136, no. 3, pp. 220–225, 2009.

[114] Q. Lu, K. Hu, Q. Feng, and F. Cui, “Growth of fibroblast andvascular smooth muscle cells in fibroin/collagen scaffold,”Materials Science and Engineering C, vol. 29, no. 7, pp. 2239–2245, 2009.

[115] N. Davidenko, J. J. Campbell, E. S. Thian, C. J. Watson, and R.E. Cameron, “Collagen-hyaluronic acid scaffolds for adiposetissue engineering,” Acta Biomaterialia, vol. 6, pp. 3957–3968,2010.

[116] M. Kikuchi, H. N. Matsumoto, T. Yamada, Y. Koyama,K. Takakuda, and J. Tanaka, “Glutaraldehyde cross-linkedhydroxyapatite/collagen self-organized nanocomposites,”Biomaterials, vol. 25, no. 1, pp. 63–69, 2004.

[117] C. M. Tierney, M. J. Jaasma, and F. J. O’Brien, “Osteoblastactivity on collagen-GAG scaffolds is affected by collagenand GAG concentrations,” Journal of Biomedical MaterialsResearch Part A, vol. 91, no. 1, pp. 92–101, 2009.

[118] H. Tabandeh, R. Aboufazelia et al., “Liposomes dispersedin two soluble types of collagens and the effect of collagenson the release rate of entrapped sodium shromate,” IranianJournal of Pharmaceutical Research, vol. 2, pp. 161–165, 2003.

[119] A. W. Pederson, J. W. Ruberti, and P. B. Messersmith, “Ther-mal assembly of a biomimetic mineral/collagen composite,”Biomaterials, vol. 24, no. 26, pp. 4881–4890, 2003.

[120] W. A. Marston, A. Isala, R. S. Hill, R. Mendes, and M.-A.Minsley, “Initial report of the use of an injectable porcinecollagen-derived matrix to stimulate healing of diabetic footwounds in humans,” Wound Repair and Regeneration, vol. 13,no. 3, pp. 243–247, 2005.

[121] H. Kang, M. B. O’Donoghue, H. Liu, and W. Tan, “Aliposome-based nanostructure for aptamer directed deliv-ery,” Chemical Communications, vol. 46, no. 2, pp. 249–251,2010.

[122] S. Hao, L. Li, X. Yang et al., “The characteristics of gelatinextracted from sturgeon (Acipenser baeri) skin using variouspretreatments,” Food Chemistry, vol. 115, no. 1, pp. 124–128,2009.

[123] S. Young, M. Wong, Y. Tabata, and A. G. Mikos, “Gelatinas a delivery vehicle for the controlled release of bioactive

molecules,” Journal of Controlled Release, vol. 109, no. 1–3,pp. 256–274, 2005.

[124] K. Ofokansi, G. Winter, G. Fricker, and C. Coester, “Matrix-loaded biodegradable gelatin nanoparticles as new approachto improve drug loading and delivery,” European Journal ofPharmaceutics and Biopharmaceutics, vol. 76, pp. 1–9, 2010.

[125] V. DiTizio, G. W. Ferguson, M. W. Mittelman, A. E. Khoury,A. W. Bruce, and F. DiCosmo, “A liposomal hydrogel for theprevention of bacterial adhesion to catheters,” Biomaterials,vol. 19, no. 20, pp. 1877–1884, 1998.

[126] S. A. Burke, M. Ritter-Jones, B. P. Lee, and P. B. Messersmith,“Thermal gelation and tissue adhesion of biomimetic hydro-gels,” Biomedical Materials, vol. 2, no. 4, pp. 203–210, 2007.

[127] A. Samad, Y. Sultana, R. K. Khar, K. Chuttani, and A. K.Mishra, “Gelatin microspheres of rifampicin cross-linkedwith sucrose using thermal gelation method for the treat-ment of tuberculosis,” Journal of Microencapsulation, vol. 26,no. 1, pp. 83–89, 2009.

[128] X. Zhang, M. D. Do, P. Casey et al., “Chemical cross-linkinggelatin with natural phenolic compounds as studied by high-resolution NMR spectroscopy,” Biomacromolecules, vol. 11,no. 4, pp. 1125–1132, 2010.

[129] K. Kuwahara, Z. Yang, G. C. Slack, M. E. Nimni, andB. Han, “Cell delivery using an injectable and adhesivetransglutaminase-gelatin gel,” Tissue Engimeering, Part C, vol.16, no. 4, pp. 609–618, 2010.

[130] F. Cheng, Y. B. Choy, H. Choi, and K. Kim, “Modelingof small-molecule release from crosslinked hydrogel micro-spheres: effect of crosslinking and enzymatic degradation ofhydrogel matrix,” International Journal of Pharmaceutics, vol.403, no. 1-2, pp. 90–95, 2011.

[131] C. K. S. Pillai, W. Paul, and C. P. Sharma, “Chitin and chi-tosan polymers: chemistry, solubility and fiber formation,”Progress in Polymer Science, vol. 34, no. 7, pp. 641–678, 2009.

[132] M. Prabaharan, “Review paper: chitosan derivatives aspromising materials for controlled drug delivery,” Journal ofBiomaterials Applications, vol. 23, no. 1, pp. 5–36, 2008.

[133] J. H. Park, G. Saravanakumar, K. Kim, and I. C. Kwon,“Targeted delivery of low molecular drugs using chitosan andits derivatives,” Advanced Drug Delivery Reviews, vol. 62, no.1, pp. 28–41, 2010.

[134] T. Kean and M. Thanou, “Biodegradation, biodistributionand toxicity of chitosan,” Advanced Drug Delivery Reviews,vol. 62, no. 1, pp. 3–11, 2010.

[135] S. Mao, W. Sun, and T. Kissel, “Chitosan-based formulationsfor delivery of DNA and siRNA,” Advanced Drug DeliveryReviews, vol. 62, no. 1, pp. 12–27, 2010.

[136] H. S. Kas, “Chitosan: properties, preparations and applica-tion to microparticulate systems,” Journal of Microencapsula-tion, vol. 14, no. 6, pp. 689–711, 1997.

[137] R. Hejazi and M. Amiji, “Chitosan-based gastrointestinaldelivery systems,” Journal of Controlled Release, vol. 89, no.2, pp. 151–165, 2003.

[138] L. Illum, “Nasal drug delivery—possibilities, problems andsolutions,” Journal of Controlled Release, vol. 87, no. 1–3, pp.187–198, 2003.

[139] C. Ji, A. Barrett, L. A. Poole-Warren, N. R. Foster, and F.Dehghani, “The development of a dense gas solvent exchangeprocess for the impregnation of pharmaceuticals into porouschitosan,” International Journal of Pharmaceutics, vol. 391,no. 1-2, pp. 187–196, 2010.

[140] H. J. Chun, G. W. Kim, and C. H. Kim, “Fabrication ofporous chitosan scaffold in order to improve biocompatibil-ity,” Journal of Physics and Chemistry of Solids, vol. 69, no. 5-6,pp. 1573–1576, 2008.


[141] A. R. C. Duarte, J. F. Mano, and R. L. Reis, “Preparation ofchitosan scaffolds for tissue engineering using supercriticalfluid technology,” Materials Science Forum, vol. 636-637, pp.22–25, 2010.

[142] D. Le Nihouannen, L. L. Guehennec, T. Rouillon et al.,“Micro-architecture of calcium phosphate granules andfibrin glue composites for bone tissue engineering,” Bioma-terials, vol. 27, no. 13, pp. 2716–2722, 2006.

[143] T. Sasagawa, T. Shimizu, S. Sekiya et al., “Design of prevascu-larized three-dimensional cell-dense tissues using a cell sheetstacking manipulation technology,” Biomaterials, vol. 31, no.7, pp. 1646–1654, 2010.

[144] P. P. Spicer and A. G. Mikos, “Fibrin glue as a drug deliverysystem,” Journal of Controlled Release, vol. 148, no. 1, pp. 49–55, 2010.

[145] N. F. Huang, J. Chu, R. J. Lee, and S. Li, “Biophysical andchemical effects of fibrin on mesenchymal stromal cell geneexpression,” Acta Biomaterialia, vol. 6, pp. 3947–3956, 2010.

[146] A. des Rieux, A. Shikanov, and L. D. Shea, “Fibrin hydrogelsfor non-viral vector delivery in vitro,” Journal of ControlledRelease, vol. 136, no. 2, pp. 148–154, 2009.

[147] K. L. Christman, A. J. Vardanian, Q. Fang, R. E. Sievers, H.H. Fok, and R. J. Lee, “Injectable fibrin scaffold improves celltransplant survival, reduces infarct expansion, and inducesneovasculature formation in ischemic myocardium,” Journalof the American College of Cardiology, vol. 44, no. 3, pp. 654–660, 2004.

[148] A. S. C. Soon, S. E. Stabenfeldt, W. E. Brown, and T.H. Barker, “Engineering fibrin matrices: the engagementof polymerization pockets through fibrin knob technologyfor the delivery and retention of therapeutic proteins,”Biomaterials, vol. 31, no. 7, pp. 1944–1954, 2010.

[149] E. Briganti, D. Spiller, C. Mirtelli et al., “A compositefibrin-based scaffold for controlled delivery of bioactive pro-angiogenetic growth factors,” Journal of Controlled Release,vol. 142, no. 1, pp. 14–21, 2010.

[150] P. Lei, R. M. Padmashali, and S. T. Andreadis, “Cell-controlled and spatially arrayed gene delivery from fibrinhydrogels,” Biomaterials, vol. 30, no. 22, pp. 3790–3799, 2009.

[151] S. S. Wang, M. C. Yang, and T. W. Chung, “Lipo-somes/chitosan scaffold/human fibrin gel composite sys-tems for delivering hydrophilic drugs—release behaviors ofTirofiban in vitro,” Drug Delivery, vol. 15, no. 3, pp. 149–157,2008.

[152] M. Kulkarni, A. Breen, U. Greiser, T. O’Brien, and A. Pandit,“Fibrin-lipoplex system for controlled topical delivery ofmultiple genes,” Biomacromolecules, vol. 10, no. 6, pp. 1650–1654, 2009.

[153] W. R. Gombotz and S. F. Wee, “Protein release from alginatematrices,” Advanced Drug Delivery Reviews, vol. 31, no. 3, pp.267–285, 1998.

[154] N. Saude, H. Cheze-Lange, D. Beunard et al., “Alginateproduction by Azotobacter vinelandii in a membrane biore-actor,” Process Biochemistry, vol. 38, no. 2, pp. 273–278, 2002.

[155] S. M. Willerth and S. E. Sakiyama-Elbert, “Approaches toneural tissue engineering using scaffolds for drug delivery,”Advanced Drug Delivery Reviews, vol. 59, no. 4-5, pp. 325–338, 2007.

[156] C. Gao, M. Liu, S. Chen, S. Jin, and J. Chen, “Preparationof oxidized sodium alginate-graft-poly((2-dimethylamino)ethyl methacrylate) gel beads and in vitro controlled releasebehavior of BSA,” International Journal of Pharmaceutics, vol.371, no. 1-2, pp. 16–24, 2009.

[157] M. George and T. E. Abraham, “Polyionic hydrocolloidsfor the intestinal delivery of protein drugs: alginate andchitosan—a review,” Journal of Controlled Release, vol. 114,no. 1, pp. 1–14, 2006.

[158] T. Li, X. W. Shi, Y. M. Du, and Y. F. Tang, “Quaternizedchitosan/alginate nanoparticles for protein delivery,” Journalof Biomedical Materials Research Part A, vol. 83, no. 2, pp.383–390, 2007.

[159] J. B. Xu, J. P. Bartley, and R. A. Johnson, “Preparation andcharacterization of alginate hydrogel membranes crosslinkedusing a water-soluble carbodiimide,” Journal of AppliedPolymer Science, vol. 90, no. 3, pp. 747–753, 2003.

[160] O. Jeon, K. H. Bouhadir, J. M. Mansour, and E. Alsberg,“Photocrosslinked alginate hydrogels with tunable biodegra-dation rates and mechanical properties,” Biomaterials, vol.30, no. 14, pp. 2724–2734, 2009.

[161] N. Mohan and P. D. Nair, “Novel porous, polysaccharidescaffolds for tissue engineering applications,” Trends inBiomaterials and Artificial Organs, vol. 18, no. 2, pp. 219–224,2005.

[162] O. Khanna, M. L. Moya, E. C. Opara, and E. M. Brey,“Synthesis of multilayered alginate microcapsules for thesustained release of fibroblast growth factor-1,” Journal ofBiomedical Materials Research Part A, vol. 95 A, no. 2, pp.632–640, 2010.

[163] N. Wang, G. Adams, L. Buttery, F. H. Falcone, and S.Stolnik, “Alginate encapsulation technology supports embry-onic stem cells differentiation into insulin-producing cells,”Journal of Biotechnology, vol. 144, no. 4, pp. 304–312, 2009.

[164] J. R. Nixon and V. W. Yeung, “Preparation of microencap-sulated liposomes, II. Systems containing nylon-gelatin andnylon-gelatin-acacia walling material,” Journal of Microen-capsulation, vol. 6, no. 1, pp. 43–52, 1989.

[165] H. K. Tilakaratne, S. K. Hunter, M. E. Andracki, J. A. Benda,and V. G. J. Rodgers, “Characterizing short-term releaseand neovascularization potential of multi-protein growthsupplement delivered via alginate hollow fiber devices,”Biomaterials, vol. 28, no. 1, pp. 89–98, 2007.

[166] H. J. Kong, E. S. Kim, Y. C. Huang, and D. J. Mooney, “Designof biodegradable hydrogel for the local and sustained deliveryof angiogenic plasmid DNA,” Pharmaceutical Research, vol.25, no. 5, pp. 1230–1238, 2008.

[167] M. Monshipouri and A. S. Rudolph, “Liposome-encapsulated alginate: controlled hydrogel particle formationand release,” Journal of Microencapsulation, vol. 12, no. 2, pp.117–127, 1995.

[168] N. O. Dhoot and M. A. Wheatley, “Microencapsulatedliposomes in controlled drug delivery: strategies to modulatedrug release and eliminate the burst effect,” Journal ofPharmaceutical Sciences, vol. 92, no. 3, pp. 679–689, 2003.

[169] C. Dai, B. Wang, H. Zhao, B. Li, and J. Wang, “Preparationand characterization of liposomes-in-alginate (LIA) forprotein delivery system,” Colloids and Surfaces B, vol. 47, no.2, pp. 205–210, 2006.

[170] M. Machluf, O. Regev, Y. Peled, J. Kost, and S. Cohen, “Char-acterization of microencapsulated liposome systems for thecontrolled delivery of liposome-associated macromolecules,”Journal of Controlled Release, vol. 43, no. 1, pp. 35–45, 1997.

[171] M. Hara and J. Miyake, “Calcium alginate gel-entrappedliposomes,” Materials Science and Engineering C, vol. 17, no.1-2, pp. 101–105, 2001.

[172] R. Mehvar, “Dextrans for targeted and sustained deliveryof therapeutic and imaging agents,” Journal of ControlledRelease, vol. 69, no. 1, pp. 1–25, 2000.


[173] G. Sun, Y. I. Shen, C. C. Ho, S. Kusuma, and S. Gerecht,“Functional groups affect physical and biological propertiesof dextran-based hydrogels,” Journal of Biomedical MaterialsResearch Part A, vol. 93, no. 3, pp. 1080–1090, 2010.

[174] S. Hornig, H. Bunjes, and T. Heinze, “Preparation andcharacterization of nanoparticles based on dextran-drugconjugates,” Journal of Colloid and Interface Science, vol. 338,no. 1, pp. 56–62, 2009.

[175] P. K. Shrivastava and S. K. Shrivastava, “Dextran carriermacromolecule for colon specific delivery of celecoxib,”Current Drug Delivery, vol. 7, no. 2, pp. 144–151, 2010.

[176] J. Qi, P. Yao, F. He, C. Yu, and C. Huang, “Nanoparticles withdextran/chitosan shell and BSA/chitosan core-Doxorubicinloading and delivery,” International Journal of Pharmaceutics,vol. 393, no. 1-2, pp. 177–185, 2010.

[177] S. R. van Tomme and W. E. Hennink, “Biodegradable dextranhydrogels for protein delivery applications,” Expert Review ofMedical Devices, vol. 4, no. 2, pp. 147–164, 2007.

[178] T. Jin, J. Zhu, F. Wu, W. Yuan, L. L. Geng, and H. Zhu,“Preparing polymer-based sustained-release systems withoutexposing proteins to water-oil or water-air interfaces andcross-linking reagents,” Journal of Controlled Release, vol. 128,no. 1, pp. 50–59, 2008.

[179] E. M. Bachelder, T. T. Beaudette, K. E. Broaders et al.,“In vitro analysis of acetalated dextran microparticles as apotent delivery platform for vaccine adjuvants,” MolecularPharmaceutics, vol. 7, no. 3, pp. 826–835, 2010.

[180] C. J. De Groot, J. A. Cadee, J. W. Koten, W. E. Hennink, andW. Den Otter, “Therapeutic efficacy of IL-2-loaded hydrogelsin a mouse tumor model,” International Journal of Cancer,vol. 98, no. 1, pp. 134–140, 2002.

[181] J. Maia, L. Ferreira, R. Carvalho, M. A. Ramos, and M. H.Gil, “Synthesis and characterization of new injectable anddegradable dextran-based hydrogels,” Polymer, vol. 46, no.23, pp. 9604–9614, 2005.

[182] Y. Yeo and D. S. Kohane, “Polymers in the prevention ofperitoneal adhesions,” European Journal of Pharmaceutics andBiopharmaceutics, vol. 68, no. 1, pp. 57–66, 2008.

[183] S. Liptay, H. Weidenbach, G. Adler, and R. M. Schmid,“Colon epithelium can be transiently transfected with lipo-somes, calcium phosphate precipitation and DEAE dextranin vivo,” Digestion, vol. 59, no. 2, pp. 142–147, 1998.

[184] A. M. Durrani, N. M. Davies, M. Thomas, and I. W. Kellaway,“Pilocarpine bioavailability from a mucoadhesive liposomalopthalmic drug delivery system,” International Journal ofPharmaceutics, vol. 88, no. 1–3, pp. 409–415, 1992.

[185] L. Budai, M. Hajdu, M. Budai et al., “Gels and liposomesin optimized ocular drug delivery: studies on ciprofloxacinformulations,” International Journal of Pharmaceutics, vol.343, no. 1-2, pp. 34–40, 2007.

[186] C. Tang, L. Yin, J. Yu, C. Yin, and Y. Pei, “Swelling behaviorand biocompatibility of carbopol-containing superporoushydrogel composites,” Journal of Applied Polymer Science, vol.104, no. 5, pp. 2785–2791, 2007.

[187] H. E. Junginger and J. C. Verhoef, “Macromolecules as safepenetration enhancers for hydrophilic drugs—a fiction?”Pharmaceutical Science and Technology Today, vol. 1, no. 9,pp. 370–376, 1998.

[188] G. Rajput, F. Majmudar et al., “Stomach-specific mucoad-hesive microsphere as a controlled drug delivery system,”Systematic Reviews in Pharmacy, vol. 1, no. 1, pp. 70–78,2010.

[189] K. M. Hosny, “Ciprofloxacin as ocular liposomal hydrogel,”AAPS PharmSciTech, vol. 11, no. 1, pp. 241–246, 2010.

[190] K. M. Hosny, “Optimization of gatifloxacin liposomalhydrogel for enhanced transcorneal permeation,” Journal ofLiposome Research, vol. 20, no. 1, pp. 31–37, 2010.

[191] Z. Pavelic, N. Skalko-Basnet, J. Filipovic-Grcic, A. Martinac,and I. Jalsenjak, “Development and in vitro evaluation of aliposomal vaginal delivery system for acyclovir,” Journal ofControlled Release, vol. 106, no. 1-2, pp. 34–43, 2005.

[192] S. Mourtas, S. Fotopoulou, S. Duraj, V. Sfika, C. Tsakiroglou,and S. G. Antimisiaris, “Liposomal drugs dispersed inhydrogels. Effect of liposome, drug and gel properties ondrug release kinetics,” Colloids and Surfaces B, vol. 55, no. 2,pp. 212–221, 2007.

[193] C. C. DeMerlis and D. R. Schoneker, “Review of the oraltoxicity of polyvinyl alcohol (PVA),” Food and ChemicalToxicology, vol. 41, no. 3, pp. 319–326, 2003.

[194] D. J. Macquarrie and J. J. E. Hardy, “Applications of func-tionalized chitosan in catalysis,” Industrial and EngineeringChemistry Research, vol. 44, no. 23, pp. 8499–8520, 2005.

[195] S. Horiike, S. Matsuzawa, and K. Yamaura, “Prepara-tion of chemically crosslinked gels with maleate-denaturedpoly(vinyl alcohol) and its application to drug release,”Journal of Applied Polymer Science, vol. 84, no. 6, pp. 1178–1184, 2002.

[196] S. L. Bourke, M. Al-Khalili, T. Briggs, B. B. Michniak, J. Kohn,and L. A. Poole-Warren, “A photo-crosslinked poly(vinylalcohol) hydrogel growth factor release vehicle for woundhealing applications,” AAPS PharmSciTech, vol. 5, no. 4, pp.1–11, 2003.

[197] C. Yang, P. J. Hillas, J. A. Baez et al., “The applicationof recombinant human collagen in tissue engineering,”BioDrugs, vol. 18, no. 2, pp. 103–119, 2004.

[198] W. Abdelwahed, G. Degobert, and H. Fessi, “Freeze-dryingof nanocapsules: impact of annealing on the drying process,”International Journal of Pharmaceutics, vol. 324, no. 1, pp. 74–82, 2006.

[199] S. Litvinchuk, Z. Lu, P. Rigler, T. D. Hirt, and W. Meier,“Calcein release from polymeric vesicles in blood plasma andPVA hydrogel,” Pharmaceutical Research, vol. 26, no. 7, pp.1711–1717, 2009.

[200] M. Sandor, A. Riechel, I. Kaplan, and E. Mathiowitz, “Effectof lecithin and MgCO3 as additives on the enzymatic activ-ity of carbonic anhydrase encapsulated in poly(lactide-co-glycolide) (PLGA) microspheres,” Biochimica et BiophysicaActa, vol. 1570, no. 1, pp. 63–74, 2002.

[201] S.-S. Wang, M.-C. Yang, and T.-W. Chung, “Liposomes/chitosan scaffold/human fibrin gel composite systems fordelivering hydrophilic drugs—release behaviors of Tirofibanin vitro,” Drug Delivery, vol. 15, no. 3, pp. 149–157, 2008.

[202] Y. Sultana, R. Jain, M. Aqil, and A. Ali, “Review of ocular drugdelivery,” Current Drug Delivery, vol. 3, no. 2, pp. 207–217,2006.

[203] N. A. Chaudhry, H. W. Flynn, T. G. Murray, H. Tabandeh, M.O. Mello, and D. Miller, “Emerging ciprofloxacin-resistantPseudomonas aeruginosa,” American Journal of Ophthalmol-ogy, vol. 128, no. 4, pp. 509–510, 1999.

[204] A. Jauch, M. Fsadni, and G. Gamba, “Meta-analysis of sixclinical phase III studies comparing lomefloxacin 0.3% eyedrops twice daily to five standard antibiotics in patients withacute bacterial conjunctivitis,” Graefe’s Archive for Clinicaland Experimental Ophthalmology, vol. 237, no. 9, pp. 705–713, 1999.

[205] D. M. Campoli-Richards, J. P. Monk, A. Price, P. Benfield,P. A. Todd, and A. Ward, “Ciprofloxacin. A review of


its antibacterial activity, pharmacokinetic properties andtherapeutic use,” Drugs, vol. 35, no. 4, pp. 373–447, 1988.

[206] P. C. Appelbaum and P. A. Hunter, “The fluoroquinoloneantibacterials: past, present and future perspectives,” Interna-tional Journal of Antimicrobial Agents, vol. 16, no. 1, pp. 5–15,2000.

[207] H.-H. Lin, S.-M. Ko, L.-R. Hsu, and Y.-H. Tsai, “Thepreparation of norfloxacin-loaded liposomes and their in-vitro evaluation in pig’s eye,” Journal of Pharmacy andPharmacology, vol. 48, no. 8, pp. 801–805, 1996.

[208] B. Wiechens, D. Neumann, J. B. Grammer, U. Pleyer,J. Hedderich, and G. I.W. Duncker, “Retinal toxicity ofliposome-incorporated and free ofloxacin after intravitrealinjection in rabbit eyes,” International Ophthalmology, vol.22, no. 3, pp. 133–143, 1999.

[209] P. G. Kibat, Y. Igari, M. A. Wheatley, H. N. Eisen, andR. Langer, “Enzymatically activated microencapsulated lipo-somes can provide pulsatile drug release,” FASEB Journal, vol.4, no. 8, pp. 2533–2539, 1990.

[210] W. N. E. van Dijk-Wolthuis, J. A. M. Hoogeboom, M. J. vanSteenbergen, S. K. Y. Tsang, and W. E. Hennink, “Degra-dation and release behavior of dextran-based hydrogels,”Macromolecules, vol. 30, no. 16, pp. 4639–4645, 1997.


Review Article

Recent Applications of Liposomes in Ophthalmic Drug Delivery

Gyan P. Mishra, Mahuya Bagui, Viral Tamboli, and Ashim K. Mitra

Division of Pharmaceutical Sciences, School of Pharmacy, University of Missouri-Kansas City, 2464 Charlotte Street, Kansas City,MO 64108-2718, USA

Correspondence should be addressed to Ashim K. Mitra, [email protected]

Received 3 August 2010; Revised 30 November 2010; Accepted 22 December 2010

Academic Editor: Rita Cortesi

Copyright © 2011 Gyan P. Mishra et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Liposomal formulations were significantly explored over the last decade for the ophthalmic drug delivery applications. Theseformulations are mainly composed of phosphatidylcholine (PC) and other constituents such as cholesterol and lipid-conjugatedhydrophilic polymers. Liposomes are biodegradable and biocompatible in nature. Current approaches for topical delivery ofliposomes are focused on improving the corneal adhesion and permeation by incorporating various bioadhesive and penetrationenhancing polymers. In the case of posterior segment disorders improvement in intravitreal half life and targeted drug deliveryto the retina is achieved by liposomes. In this paper we have attempted to summarize the applications of liposomes in the field ofophthalmic drug delivery by citing numerous investigators over the last decade.

1. Introduction

Ocular drug delivery is challenging in terms of achievingoptimum drug concentration due to unique protectivemechanisms of the eye. Development of a drug deliverysystem for attaining therapeutic concentration at the targetsite requires a comprehensive understanding of static anddynamic barriers of the eye. The eye has two broadlydefined segments, (a) anterior segment, and (b) posteriorsegment. Anterior segment is the front one-third of theeye that includes the optical structure in front of vitreoushumor: cornea, pupil, aqueous humor, iris, lens and ciliarybody. Posterior segment is the back two-thirds of theeye that mainly includes sclera, choroid, retina, vitreoushumor, macula, and optical nerve. The common routesof drug administration for the treatment of eye disordersare topical, systemic, periocular, and intravitreal. Topicaladministration is the most preferred route because of highestpatient compliance and least invasive nature. Upon topicalinstillation, absorption of drugs takes place either throughcorneal route (cornea, aqueous humor, intraocular tissues)or noncorneal route (conjunctiva, sclera, choroid/RPE). Thecornea can be mainly divided into the epithelium, stromaand endothelium, where each layer offers a different polarityand a potential rate-limiting structure for drug permeation.

The non-corneal route involves absorption across the scleraand conjunctiva into the intraocular tissues. However,a small fraction of the topically applied drugs, generally lessthan 5%, reaches the intraocular tissues. Factors responsiblefor poor ocular bioavailability following topical instillationare precorneal drainage and lipoidal nature of the cornealepithelium. In addition, a major fraction of drug reachesthe systemic circulation through conjunctival vessels andnasolacrimal duct, which leads to severe adverse effects.Consequently, topical route has met with limited success inattaining therapeutic drug concentrations in the posteriorsegment. Systemic administration can provide therapeuticlevels in the posterior segment, but administration of highdoses is necessary, which often leads to severe side effects.Blood-aqueous barrier and blood-retinal barrier are thetwo major barriers for anterior segment and posteriorsegment ocular drug delivery, respectively, after systemicadministration. The tight junctional complexes located in thetwo discrete cell layers, the endothelium of the iris/ciliaryblood vessels, and the nonpigmented ciliary epithelium offerblood-aqueous barrier which prevents the entry of solutesinto the aqueous humor. Blood retinal barrier is composedof two types of cells, that is, retinal capillary endothelial cellsand retinal pigment epithelium (RPE) cells which preventsthe entry of solute into the retina. Intravitreal administration


Lipid layer

Lipid layer

Aqueouscore

Phospolipidbilayer

Unilamellar liposomes Multilamellar liposomes

SUV LUV GUV MLV

Figure 1: Schematic representation of basic structures and differenttypes of liposomes.

requires frequent administration which may lead to highsusceptibility for vitreous hemorrhage, retinal detachmentand endophthalmitis. These side effects can be minimizedby developing delivery systems which provide controlledand targeted drug delivery for prolonged periods [1–3].Conventional ophthalmic formulations such as solutionsand suspensions exhibit poor bioavailability. Over the lastdecade, numerous drug delivery systems have been exploredto overcome the limitation of conventional dosage forms.Novel formulations such as nanoparticles, liposomes, den-drimers, and niosomes were developed to enhance drugbioavailability and to minimize adverse effects [4, 5]. Amongthem liposomal formulations were widely explored in the lastdecade for drug delivery applications.

In 1965 liposomes were first introduced as the drugdelivery carriers [6]. Liposomes are usually within the sizerange of 10 nm to 1 μm or greater. These vesicular systems arecomposed of aqueous core enclosed by phospholipid bilayersof natural or synthetic origin. Liposomes are structurallyclassified on the basis of lipid bilayers such as small unilamel-lar vesicles (SUVs) or multilamellar vesicles (MLVs). Fur-thermore, on the basis of size, liposomes are classified intosmall unilamellar vesicles (SUVs), giant unilamellar vesicles(GUVs), and large unilamellar vesicles (LUVs) (Figure 1).Unilamellar vesicles are composed of single layer of lipid suchas lecithin or phosphatidylglycerol encapsulating aqueousinterior core. Multilamellar vesicle is composed of variouslayers of lipid bilayers [7–9]. MLVs are metastable energyconfiguration having different facets depending upon thepolydispersity of the liposomal formulation. Various types ofliposomes with size are summarized in Table 1.

Drug loading capacity of liposomes depends on manyfactors such as size of liposomes, types of lipid utilizedfor preparation, and physicochemical properties of ther-apeutic agent itself. For example, being the smallest insize entrapping efficiency for SUVs is poor in comparisonto MLVs. However, LUVs provide a balance between sizeand drug loading capacity. Liposomes are advantageous in

Table 1: Size of different types of liposomes [12].

Vesicle type Size

SUVs ∼20 nm to ∼200 nm

LUVs ∼200 nm to ∼1 μm

MLVs >0.5μm

GUVs >1 μm

encapsulating both lipophilic and hydrophilic molecules.Hydrophilic drugs are entrapped in the aqueous layer, whilehydrophobic drugs are stuck in the lipid bilayers. Loadingcapacity of ionic molecules can be further improved by usingcationic or anionic lipids for the preparation of liposomes[11].

Majority of liposomal formulations utilize phosphatidyl-choline (PC) and other constituents such as cholesterol andlipid-conjugated hydrophilic polymers as the main ingre-dients. Incorporation of cholesterol enhances the stabilityby improving the rigidity of the membrane. Stability ofliposomes depends upon the various properties such assurface charge, size, surface hydration, and fluidity of lipidbilayers. Surface charge determines interaction of liposomeswith ocular membrane. Positively charged liposomes displaybetter corneal permeation than the neutral and nega-tively charged liposomes. Neutral liposomes upon systemicadministration evade the elimination by reticuloendothelialsystem (RES). However, these vesicles possess higher self-aggregation tendency. In contrast, negatively and positivelycharged liposomes exhibit lower aggregation tendency butundergo rapid clearance by RES cells due to higher interac-tion with serum proteins. In addition, size of the liposomescan also regulate the clearance by RES. Liposomes of size lessthan 100 nm generally exhibit significantly higher circulationtime due to decrease in opsonization of liposomes withserum protein [13].

Amphiphilic nature of phospholipids allows thesemolecules to form lipid bilayers. This unique feature isutilized for the preparation of liposomes. In general, hydra-tion of phospholipids results in the formation of MLVs,which can be processed into SUVs with proper sonication.However, addition of aqueous solution of surfactant abovethe critical micelle concentration results in the formationof phospholipids micelles. After the dialysis of surfactantaggregation of micelles form LUVs, critical micelle concen-trations of amphiphiles which can form micelles are fourto five orders of magnitude higher than the phospholipidswhich form liposomes [12]. Numerous methods have beenreported to prepare liposomes. Most commonly solventevaporation method, reverse phase evaporation method anddetergent dialysis method are employed [14]. The encap-sulated drug from liposome can be released either throughpassive diffusion, vesicle erosion, or vesicle retention. Inpassive diffusion, drug molecules tend to penetrate throughthe lipid layers of liposome to reach extra vesicular layereither by diffusion or convection mechanism. The rate ofdiffusion depends on the size, lipid composition, and theproperties of the drug itself [15–17]. Unilamellar liposomesexhibit faster release rate than multilamellar ones because


in multilayered liposome, drug diffusion occurs througha series of barriers; hence, the drug release is delayed.Phospholipase and high-density lipoprotein present in bloodplasma can damage phospholipid layers of liposome andthus results in vesicle erosion and releases the encapsulateddrug into the cell. The drug release rate depends on theextent of liposomal membrane damage [18]. Liposome-cell interactions depend on several factors like size, surfacecharge, composition of liposomes, targeting ligand on thesurface of liposome, and biological environment. Lipo-somes can interact with cells by four different mechanisms:adsorption, fusion, lipid exchange and endocytosis (receptormediated). Liposomes can be specifically or nonspecificallyadsorbed onto the cell surface or can be fused with cellmembranes, and release encapsulated drug inside the cell.During adsorption, liposomes can release encapsulated drugin front of cell membrane, and released drug can enter cellvia micropinocytosis. They can also be engulfed inside thecell by specific or nonspecific endocytosis process. Negativelycharged liposomes have been found to be more efficientthan neutral liposomes for internalization into the cellsby endocytosis process. Liposomes bind to the receptorpresent in the invaginations of cellular membrane and areinternalized into the cell by endocytotic pathway. Afterendocytosis, they can fuse with the endosomal membraneto form endosome which can be delivered to lysosomes. Inlysosomes, the presence of peptidase and hydrolase degradesthe liposomes and their content. To avoid this degradationand thus to increase cytoplasmic bioavailability, stimuli-responsive liposomes (such as pH or temperature) havebeen developed. pH-sensitive liposomes can undergo fusionwith endosomal membrane and release their content directlyinto cytosol. In some cases liposomes become destabilizedinside the endosome and release their content, or theydestabilize endosomal membrane resulting in leakage ofencapsulated content into cytosol [19, 20]. In this paper wehave attempted to summarize the application of liposomes inthe field of ophthalmic drug delivery attempted by numerousinvestigators over the last decade.

2. Application of Liposomes inOphthalmic Drug Delivery

Liposomes have been investigated for ophthalmic drugdelivery since it offers advantages as a carrier system.It is a biodegradable and biocompatible nanocarrier. Itcan enhance the permeation of poorly absorbed drugmolecules by binding to the corneal surface and improvingresidence time. It can encapsulate both the hydrophilicand hydrophobic drug molecules. In addition, liposomescan improve pharmacokinetic profile, enhance therapeuticeffect, and reduce toxicity associated with higher dose.Owing to their versatile nature, liposomes have been widelyinvestigated for the treatment of both anterior and pos-terior segment eye disorders. Current approaches for theanterior segment drug delivery are focused on improvingcorneal adhesion and permeation by incorporating variousbioadhesive and penetration enhancing polymers. However,in the case of posterior segment disorders, improvement of

intravitreal half-life and targeted drug delivery to the retinais necessary. Currently verteporfin is being used clinicallyin photodynamic therapy for the treatment of subfovealchoroidal neovascularization (CNV), ocular histoplasmosis,or pathological myopia effectively. Verteporfin is a light-activated drug which is administered by intravenous infu-sion. In photodynamic therapy, after the drug is injected,a low-energy laser is applied to the retina through thecontact lens in order to activate verteporfin that resultsin closure of the abnormal blood vessels. Unfortunately,photodynamic therapy usually does not permanently closethe abnormal vessels and choroidal neovessels reappear afterseveral months. Another liposomal photosensitizing agent,rostaporfin, was evaluated for the treatment of age-relatedmacular degeneration. It is now under phase 3 clinical trial.Rostaporfin requires less frequent administration comparedto verteporfin. Liposome technology has been exploredfor ophthalmic drug delivery. However, there are someissues to be addressed such as formulation, and storage ofliposomes is very difficult, and they are known to cause long-term side effects. Intravitreal administration of liposomeshas resulted in vitreal condensation, vitreal bodies in thelower part of eye, and retinal abnormalities. Therefore, allthese factors should be taken into account while developingliposomal formulation for ophthalmic application [21–25].Recent applications of liposomal formulations encapsulatingvarious therapeutic molecules are summarized in Table 2.

3. Topical Applications

In 1981, Samolin et al. investigated the role of liposomesin ophthalmic drug delivery. Since then several investigatorsproposed strategies to enhance absorption of drugs havingpoor physicochemical properties. Studies performed bySchaeffer and Krohn suggested the role of charge and sizein transcorneal permeation. Investigators observed four-fold higher in vitro corneal flux from penicillin G-loadedSUVs. They reported corneal permeation in the order ofSUV+ > MLV− > SUV− > SUV > MLV free drug. Thesestudies explored the role of vesicle type on transcorneal per-meation across the excised rabbit cornea [36]. Role of physic-ochemical property of entrapped drug was elucidated byother investigators. Liposomal formulation of TA producedtwofold increase in drug concentration in both the corneaand aqueous humor in the rabbit model. On the contrary,liposomal formulation of hydrophilic drug, that is, dihy-drostreptomycin sulfate, did not improve the corneal per-meation [37, 38]. Considering these findings, it was evidentthat both vesicle type and physicochemical property of drugsignificantly affects the transcorneal flux of the formulation.

Earlier investigation by Fitzgerald et al. was significantin exploring the clearance of liposomes by gamma scintig-raphy following topical administration in the rabbit model.These investigators reported, SUVs with positive charge hadimproved the corneal retention by interacting with negativelycharged corneal surface. Since then, approaches based onpositively charged liposomes were explored considerably.Researchers also explored immunoliposomes, lectin func-tionalized liposomes, and positively charged lipid analogs.


Table 2: Application of liposomes for the delivery of various drug molecules.

Drug Formulation Result Year Ref

GCV Liposomes

In vitro transcornealpermeation and in vivoocular pharmacokineticswas improved

2007 [26]

CiprofloxacinLiposomalhydrogel

Fivefold highertranscorneal permeationthan the liposomes alone

2010 [27]

Levofloxacin

Liposomesattached tothe contactlens

Drug was releasedfollowing first-orderkinetics for more than 6days and formulation hadshowed activity against S.aureus.

2007 [28]

Herpessimplex virusantigens

Periocularvaccine

Treated rabbits showedanti-gB immune responseand protected againstreactivation of HSVinfection

2006 [29]

Acetazolamide

Neutral- andsurface-chargedliposomes

Positively chargedliposomes reduced IOP andexhibited prolonged effectthan negatively chargedliposomes

2007 [30]

Tacrolimus Liposomes

More than 50 ng/mLvitreous concentration wasmaintained for 2 weeks andreduced drug relatedtoxicity

2010 [31]

Vasoactiveintestinalpeptide

Rhodamine-conjugatedliposomes

Liposomes wereinternalized by retinalMuller glial cells, residentmacrophages; majority ofthe liposomes reached thecervical lymph nodes andresulted in slower releaseand long-term expressioninside the eye

2007 [32]

Clodronate LiposomesEffectively inhibitinfiltration of ED2-positivemacrophages

2005 [33]

Plasmid DNACationicliposomes

Significantly increasedtransfection efficiency ofpDNA

2004 [34]

TherapeuticDNA

Cationiclipoplexes

Achieved good vitreousmobility with moderatelypegylated cationiclipoplexes with size lessthan 500 nm

2005 [35]

Among these approaches only immunoliposomes did notimproved liposome-corneal interaction. However, lectin andlipid analog-based approaches are not explored considerablyin the field of ophthalmic drug delivery [39].

Approach of utilizing chitosan in the formulation wasreported to be advantageous in improving the precornealresidence time due to its mucoadhesive nature. Degrada-tion of chitosan into oligosaccharides is mediated through

lysozymes, and degradation products are nontoxic in nature[40, 41]. Biodegradable nature is advantageous for selectingchitosan in the formulation of ocular drug delivery sys-tems. Topical administration of chitosan-coated liposomes(chitosomes) improves precorneal retention and also slowsdown drug metabolism at the precorneal epithelial surface.Chitosan-based mucoadhesive liposomal formulation ofCPX was prepared and evaluated by Mehanna et al. Reverse


phase evaporation technique was utilized for the preparationof liposomes, which were further coated with chitosanof different molecular weights. The authors reported thatliposomes coated with high molecular weight chitosan weresmaller in size due to complete coverage of liposomal surface,which acted as a physical barrier to inhibit aggregation. Inaddition, authors determined lower encapsulation efficiency(EE) of 49.93% for coated liposomes in comparison touncoated negative and neutral liposomes with 71.4% and53.2% EE, respectively, due to electrostatic repulsion betweenchitosan and cationic drug. The effect of liposomal surfacecharge on the particle size was also determined. Negativelycharged liposomes were larger in diameter due to predomi-nantly electrostatic attraction between the positively chargedchitosan and negatively charged phospholipids. Rheologicalstudies revealed ideal pseudoelastic behavior of chitosomesand higher apparent viscosity than the liposome dispersion.The author suggested that pseudoelastic property of chito-some provides prolonged retention and stability of tear film.Moreover, in vitro release studies with chitosomes exhibitedslower drug release rate in comparison to free liposomes dueto additional diffusion barrier for drug molecule. Ex vivocorneal permeation studies across isolated rabbit cornea sug-gested that due to absorption enhancing nature of chitosanrelative permeability of chitosomes was 1.74-fold higherthan free drug. Furthermore, in vitro antibacterial studiesrevealed that chitosomes exhibited enhanced antibacterialactivity than the marketed aqueous solution against referenceand clinically isolated strains of Pseudomonas aeruginosa andStaphylococcus aureus. Authors suggested the electrostaticinteraction of positively charged chitosan and negativelycharged bacterial cell wall enhanced the antibacterial actionof liposomal formulation. Comparative single dose in vivostudy performed on bacterial conjunctivitis rabbit modelrevealed that chitosomes inhibited the growth of Pseu-domonas aeruginosa for 24 h. It was reported that marketedproduct (Clioxan) is comparatively less effective and requiresfrequent administration. These investigators demonstratedthe role of medium molecular weight chitosan. However,other studies suggest the advantages of water-soluble lowmolecular weight chitosan as potential biopolymer forcoating liposomes [42]. Application of LCH was advanta-geous in eliminating the aggregation behavior of chitosanat physiological pH that had dramatically influenced invivo performance of the liposomal formulation. Investigatorreported higher ex vivo corneal penetration across excisedrabbit cornea in the case of LCH-coated liposomes as shownin Figure 2. However, higher concentration of LCH (0.25%and 0.5% w/w) did not show significant change in particlesize. Researchers suggested that a loose coating layer isresponsible for aggregation of vesicles which resulted inhigher particle size in the case of 0.1% w/v LCH. Moreover,the drug release at 6 h was 38.9% in noncoated liposomeswhereas only 25.4% drug release was observed in liposomescoated with 0.25% w/v chitosan solution. Both nontreatedand treated group did not demonstrate any abnormality ofthe corneal and conjunctival epithelial cells. In addition, noocular irritation and inflammatory response was observed.In vitro precorneal retention studies in rabbits showed that

40036032028024020016012080400

Time (min)

0

10

20

30

40

Cu

mu

lati

vepe

net

rati

on(μ

g/cm

2)

SolutionNon-L

LCHL2LCHL3

Figure 2: Corneal penetration profiles of diclofenac sodium indifferent vehicles, noncoated liposome (Non-L), LCHL2 (0.25%Low molecular weight chitosan, w/v), and LCHL3 (0.5% lowmolecular weight chitosan, w/v), (with permission from [10])

the elimination of chitosan-coated liposomes was slowerthan non-coated liposomes. Authors suggested that mucinfilm, which primarily covers the surface of cornea andconjunctiva, is composed of negatively charged glycoprotein.Electrostatic alteration between positively charged LCH andmucin promotes prolonged retention. In addition, hydrogenbonding interactions of LCH with the ocular surface alsofavors precorneal retention. This study demonstrated the roleof LCH in improving the precorneal retention. However,previous studies with high molecular weight chitosan-coatedliposomes did not improve the precorneal retention dueto enhanced intramolecular interactions. Histopathologicalanalysis of the LCH-coated liposomes in rabbits after long-term irritation test revealed that the formulation was bio-compatible with the ocular tissues (Figure 3) [10].

Application of quaternized derivatives of chitosan thatis, N-trimethyl chitosan chloride (TMC), with significantlyhigher water solubility at physiological pH, was evaluatedfor surface modification of coenzyme Q10-loaded liposomes.Improved stability of the modified liposomes was reported.In addition, surface modification with cationic polymericfilm reduced particle aggregation through stearic stabi-lization and improved precorneal retention than uncoatedliposomes due to ionic interaction with negatively chargedcorneal surface. Investigators reported almost 4.8-foldincrease in precorneal residence time measured by gammascintigraphy after administration of 25 μL of formulation.Histological analysis and draize test performed on rabbitsrevealed that TMC was biocompatible with corneal epithe-lium. Moreover, higher molecular weight TMC exhibitedbetter anticataract activity in Sprague Dawley rats [43].To take the dual advantage of chitosan-based nanoparticlesand liposomes, Diebold et al. prepared liposome-chitosannanoparticle complexes [44]. As mentioned earlier, chitosannanocarriers were employed in topical drug delivery because


(a) (b) (c)

(d) (e) (f)

Figure 3: Histopathology microscopy of the ocular tissues after being treated with LCHL for 7 days. cornea of nontreated (a), treated withLCHL2 (b), and LCHL3 (c); conjunctiva of nontreated (d), treated with LCHL2 (e) and LCHL3 (f) (with permission from [10]).

of its mucoadhesive nature, whereas liposomes can incor-porate variety of drug molecules and improve ocular drugbioavailability [45–47]. These nanosystems were formulatedas eye drops, which possessed combined properties ofboth carriers and overcome the ocular mucosal barriers.These authors evaluated the nanosystems for toxicity onspontaneously immortalized epithelial cell line from nor-mal human conjunctiva (IOBA-NHC). Cells pre incubatedwith XTT (2,3-bis[2-methoxy-4-nitro-5-sulfophenyl]-2H-tetrazolium-5-carboxyalinide) solution (1 mg/mL XTT in100 mL of phenol red-free RPMI culture medium) wereexposed to different concentrations of chitosan nanoparticlesand liposome-chitosan nanoparticles complexes. Cytotoxic-ity was determined by measuring the production of yellowcolor due to cleavage of XTT by mitochondrial enzymes. Cellviability after exposure of liposome-chitosan nanoparticlecomplexes was higher in comparison to chitosan nanoparti-cles alone. They also performed in vivo acute tolerance testby administrating the formulations topically to the femalealbino New Zealand rabbits. The nanosystems did not showany evidence of toxicity to the both sham-controlled andtreated eyes. No sign of irritation on ocular surface wasconfirmed by clinical microscopic sign score. Also, in vivoexperiments have shown that nanosystems can enter theconjunctival cells without causing histological alteration tothe cornea, conjunctiva, and lid tissues in the rabbit model.In addition, the complexes did not release any inflammatorymediators in cornea, conjunctiva, and eyelids [44].

Vaccination approach can successfully overcome thelimitations of antiviral agents in the treatment of HSVinfections. However, delivery of vaccines is the majorhurdle facing by pharmaceutical scientists. Administrationby conventional parenteral route has several drawbackssuch as high cost, need of highly trained personnel, andneedle-stick injuries. Cationic liposomes containing herpes

simplex virus (HSV) antigens were proposed as potentialcarriers, in the form of a periocular vaccine, to protectanimals against subsequent HSV-1 ocular challenges. Twodifferent peptides, namely, DTK1 and DTK2 (DTKs), havingantiherpetic activity were synthesized. Cationic liposomescontaining both DTK and secretory HSV-1 glycoprotein Bwere formulated. Liposomal formulation showed effectiveresults in a rabbit model of HSV-1 infection [29].

Zhang et al. utilized cytochrome-C (Cyt-C) loadedcationic liposomes for the treatment of selenite-inducedcataract in rats. These liposomes were fabricated by thin-layer evaporation technique. Authors investigated the effectof composition on the encapsulation efficiency. This studyreported improvement in the entrapment efficiency (EE)with increasing phosphatidylcholine component, whereasEE was lowered by incorporating stearylamine. Cyt-Cloaded freeze-dried liposomes were stable for one year at4◦C. Furthermore, these liposomes exhibited remarkableefficacy (28% decrease in lens opacity) in minimizingthe cataract formation. Liposomal encapsulation of Cyt-C has significance, but the preparation method adaptedby these authors was similar to previous investigations[48].

In another study, fluconazole liposomal formulation wasevaluated in the candidal keratitis model in rabbits. Inthis investigation, comparative efficacy of the fluconazolesolution and fluconazole-loaded liposomes was determined.The purpose of developing liposomal formulation was toprolong the antifungal action by increasing the contacttime. In the rabbits treated with fluconazole solution, 50%healing was observed in 3 weeks, whereas 86.4% healing wasobserved in rabbits treated with fluconazole encapsulatedliposomes. Authors attributed enhanced pharmacologicalactivity to higher viscosity and lipid solubility of fluconazole-loaded liposomes [49].


Chronic ocular infectious diseases such as conjunctivitis,bacterial keratitis need high drug concentration at the siteof infection. Treatment of these diseases requires frequenteye drop administrations that may cause drug resistanceand also decrease patient compliance. In order to minimizeprecorneal drainage and increase bioavailability viscosityenhancers such as poly (vinyl alcohol) and polymethacrylicacid were blended with eye drop solution [50]. Many investi-gators evaluated the role of liposomal hydrogel formulationfor the delivery of fluoroquinolone antibiotics. For example,liposomal hydrogel formulation of ciprofloxacin (CPX)was reported to avoid tear-driven dilution in the cul desac. Lecithin (LEC) and α-L-dipalmitoylphosphatidylcholine(DPPC) were utilized as major ingredients in the preparationof multilamellar liposomes. Poly (vinyl alcohol) (PVA)and polymethacrylic acid (PMA) derivatives were utilizedfor gel formulation. Various formulation parameters suchas viscosity and rheological property of liposomes wasevaluated in relation to the in vitro release [51]. CPX becauseof its negative charge electrostatically interacts with lipidhead group of the phospholipid bilayers [52]. Therefore,majority of drug was entrapped inside the liposomes. Similarelectrostatic interaction between lipid bilayers and otherfluoroquinolones such as ofloxacin and lomefloxacin werereported in other studies [53]. The investigator observedthat use of viscosity enhancing agents in the formulationhad affected the drug release rate. The addition of gelforming agents PVA and PMA did not affect the rigidity ofliposomal membrane, instead these polymers were adsorbedon the surface of multilamellar liposomal surface becauseof method of formulation. Hydration of lipids with properconcentration of PVA and PMA results in the formation ofpolymer layer on the surface of the liposomes [54, 55]. Directcorrelation was observed between viscosity of hydrogel anddrug release rate. In addition, they found a remarkabledifference in drug release half-time between two differentlipids, that is, LEC and DPPC. The presence of unsaturatedlipid in LEC provides less rigid structure to the liposomeformulation that resulted in faster drug release in com-parison to DPPC. Hydrogel formulation has shown plasticproperties; that is, under higher shear stress condition, itremained in free flowing state, whereas it exhibited no flowstate at rest. Overall, the use of optimized formulation ofliposomal hydrogel can sustain the release of antibacterialagents in comparison to liposomes alone, and this approachcould be beneficial in the treatment of various chronicocular infectious diseases. In another study, CPX-loadedliposomal hydrogel formulation improved transcorneal per-meation in rabbit model. Liposomes were suspended in thehydrogel matrix composed of carbopol 940. The investigatorsreported that drug entrapment efficiency was enhanced withthe increase of cholesterol concentrations, which providedhigher stability and lower permeability of lipid bilayers.Furthermore, higher encapsulation efficiency with positivelycharged liposomes was observed due to favorable electro-static attraction between CPX and cationic stearylamine.Liposomes of higher size were obtained upon incorporationof charge inducing agents, which expand lipid bilayers dis-tance. Positively charged liposomes exhibited slower release

rate, and CPX release was more sustained from the liposomessuspended in the carbopol gel because of additional barriersfor diffusion. Liposomal hydrogel displayed fivefold higherin vitro transcorneal permeation across excised rabbit corneathan the aqueous solution. This approach was alreadyexplored by other investigators. Although authors observedenhanced in vitro transcorneal permeation, it would beinteresting to evaluate these formulations for in vivo studies,where tear dilution plays a major role [27]. In a similar studyby these researchers, transcorneal permeation of ofloxacin-loaded thermosensitive liposomal hydrogel was evaluated.Two different types of liposomes, MLV and reverse phaseevaporation vesicles (REV), were prepared. Authors observedsmaller particle size with REV relative to MLV due todifferences in the method of preparation. Splicing of thelipid monolayer in a more curved structure resulted inREV of smaller diameter. Authors evaluated chitosan/β glyc-erophosphate thermosensitive hydrogel system. Incorpora-tion of liposomes in thermosensitive gels reduced the gellingtime from 5 to 1 minute. The researchers suggested thathydrophobic interaction can reduce energy requirement forgelation. Transcorneal permeation studies across excised rab-bit cornea revealed sevenfold higher drug permeation fromthe liposomal formulation than ofloxacin aqueous solution.This effect was observed due to mucoadhesive nature of thehydrogel base which prolonged the retention of formulationacross the excised rabbit cornea. In addition, cationic natureof chitosan in the thermogelling system promoted cornealadherence and opened corneal epithelial tight junctions.Researchers concluded that ofloxacin liposomal formulationwill reduce the formation of crystalline deposit and alsofrequency of administration. Another investigation suggeststhreefold increase in corneal residence of ophthalmic for-mulation containing chitosan. The ocular irritation testsuggests excellent tolerance of chitosan formulation evalu-ated with confocal laser scanning ophthalmoscope [56, 57].A liposomal spray formulation was recently evaluated forchanges in preocular tear film. After application of the spray,liposomes traverse to the tear film. Liposomal formulationwas evaluated on human subjects, and effectiveness wascompared to normal saline at different time points. Authorsreported statistically significant improvement in tear filmstability and lipid layer stability in comparison to control.These studies suggest the potential of liposomal sprays in thetreatment of dry eye syndrome [58].

Liposomes were also investigated for the topical deliv-ery of intraocular pressure (IOP) lowering agents. Forexample, acetazolamide was encapsulated in liposomes toenhance the solubility and corneal permeation. Liposomeswere formulated by reversed phase evaporation and liquidhydration methods with and without the use of positiveor negative charge inducers to prepare REV and MLV.Liposomes of different compositions were evaluated forentrapment efficiency, stability, in vitro release, and IOPlowering efficacy in rabbit model. The entrapment efficiencyof acetazolamide was found highest with positively chargedliposomes followed by neutral and negatively charged lipo-somes because of ionic interaction between anionic drug andlipid bilayers. Cationic and neutral MLVs of acetazolamide


43210

Time (hours)

0

5

10

15

20

25

Cu

mu

lati

veam

oun

tof

GC

Vpe

rmea

ted

(μg)

GCV liposomesGCV solution

Figure 4: In vitro transcorneal permeation of GCV liposomes andsolution (with permission from [26]).

exhibited maximum effectiveness in terms of release profilefor the same reason [30]. Another IOP reducing agent, deme-clocycline (DEM), was encapsulated in liposomes whichenhanced ocular permeability. This formulation achievedlong-lasting IOP lowering effects relative to pilocarpineliquid formulation [59]. Monem et al. reported pilocarpineHCl loaded liposomes which were administered topically.This study reported two different liposomes with neutral andnegatively charged multilamellar surface. Neutral liposomeswere more effective in IOP lowering effect than negativelycharged liposomes or free drug. In addition, phase transitionand size distribution studies showed long term stability (15months) of the liposomal formulation [60].

Liposomal formulations were also developed for thedelivery of antiviral agents. Shen and Tu reported the appli-cation of liposomes for the delivery of ganciclovir (GCV)to the vitreous humor via topical administration in therabbits. GCV liposomes were prepared by the reversed phaseevaporation method utilizing PC/CH/sodium deoxycholatemixture. In vitro transcorneal permeability and in vivoocular pharmacokinetics of the liposomal formulation werecompared with the GCV solution. Transcorneal permeabilitywas 3.9-fold higher (Figure 4), and ocular bioavailabilityof GCV liposomes was 1.7-fold greater in comparison tosolution (Figure 5). GCV concentrations from liposomalformulation were 2 to 10 times higher in various oculartissues. In addition, in vivo experiments suggested thatthe scleral pathway contributed in the absorption of GCVliposomes, as the highest concentration of GCV was obtainedin the sclera. Concentrations of GCV attained in the corneaand the sclera were higher than IC50 value of GCV againstCMV. The author suggested that the particle size (i.e.,200 nm) and composition of the liposomes played a majorrole in transocular permeation [26].

Disposable contact lenses presoaked with medicationsolution have been utilized for continuous drug delivery.However, in presoaked contact lenses, drug molecules ran-domly disperse within the contact lenses and show burstrelease that can cause local tissue toxicity or other side

403020100

Time (hours)

0

100

200

300

400

500

600

Con

cen

trat

ion

ofG

CV

inaq

ueo

us

hum

or(n

g/m

L)

GCV liposomes preparationGCV solution

Figure 5: Concentration-time profile of GCV in aqueous humorafter instillation of 1 mg/mL GCV liposome preparation and GCVsolution in rabbits (with permission from [26]).

effects [61]. To avoid rapid drug release and to provide site-specific delivery, another novel strategy, liposomes loadedsoft contact lenses, was proposed for the antibiotics inthe treatment of ocular infections such as bacterial kerati-tis. Multilamellar liposomal surface of soft contact lenseswas decorated with PEG-Biotin linkage. Contact lenseswith surface-immobilized levofloxacin-loaded liposomes fol-lowed first-order release kinetics and released the drug overmore than 6 days. In addition, the liposomal formulationhas shown antibacterial activity against S. aureus [28, 62].In another study, chloramphenicol (CAP) was encapsulatedin dimyristoylphosphatidylcholine (DMPC) liposomes andformulated in the form of eye drops. Three methods, thatis, CAP-PART (partitioning of CAP in the vesicle bilay-ers), CAP-EN (entrapment of CAP via normal hydrationmethod), and CAP-ADS (adsorption of CAP on the vesiclesurface) were employed for the preparation of liposomes.The formulation was evaluated for interaction of the drugwith the phospholipid bilayers resulting in optimum efficacyagainst S. aureus. CAP was localized in the interfacial lipidbilayers in the case of CAP-EN whereas entrapped deeper inthe bilayers in the case of CAP-PART. These results showedthat CAP located near the interfacial region within thehydrophobic core of the liposomes had shown highest anti-bacterial activity against S. aureus for almost 5 hrs [63].

Chetoni et al. reported acyclovir (ACV) containing posi-tively charged unilamellar liposomes (LIPO-ACV), adminis-tered topically into rabbit eyes. The bioavailability of LIPO-ACV was compared with free ACV in solution (SOL), ACVencapsulated in “empty” liposomes (LIPO-EMPTY), anda diluted dose of commercially available ACV ointment,containing same ACV concentrations (0.12%). The phar-macokinetic profile of the drug in the aqueous humor ofrabbits showed highest drug concentration profile for LIPO-ACV system with 90 minutes plateau. LIPO-ACV exhibitedaqueous humor ACV concentration in the upper range of theID50 (0.01 to 0.7 μg/mL). In a separate study concentrated


ACV ointment (containing 8-fold greater dose of ACV)was compared with LIPO-ACV. Only 1.6 times higherbioavailability was observed with ACV ointment. Theseresults indicate a significant advantage of LIPO-ACV as analternative to ACV ointment [47]. In order to give an insighton release mechanism of ACV from liposomal vehicle, invitro release experiments through a cellophane membranewas performed which showed lower drug release from theliposomal vehicle through cellophane membrane comparedto that of SOL and LIPO-EMPTY. These results sustained theconcept that negatively charged corneal epithelium enhancesthe efficacy of positively charged liposomal formulation.

Pleyer et al. formulated different cationic liposomesby changing their lipid compositions in order to improvegene expressions in corneal endothelial cells. The authorsreported six formulations with different cationic lipids3β[N-(N,N′-dimethylaminoethane)-carbamoyl] (DAC), di-carbobenzoxyspermin-carbamoyl (SP), N-Amidino-β-al-anin-[2-(1,3-dioleoyloxy)propyl]amid-hydrochlorid (DOSGA),and 1,2-dimyristyloxypropyl-3-methylhydroxethylammoni-umbromide (DMRIE) which were coupled in varying con-centrations with neutral lipid dioleoylphosphatidylethanol-amine (DOPE). Fixed amount of DNA was entrapped ineach liposome which expressed E. coli beta-galactosidase.Transfection experiments on bovine corneal endothelial cells(BCEC) indicated that SP20 (SP/DOPE 20/80) generatedhighest efficiency followed by DMRIE 50 (DMRIE/DOPE50/50) ranging at approximately 3 mU per β-gal per well.The researchers observed low gene expressions with DAC30 (DAC/DOPE 30/70), and DOSGA 30 (DOSGA/DOPE30/70), DOSGA 100 (DOSGA 100) and no gene expressionsfor free DNA. At a fixed DNA concentration, the relativeβ-galactosidase expressions were decreased with increasingthe cationic lipid dose, which might be due to either toxiceffects of cationic lipids at higher concentrations to the cellsor non-optimal lipid/DNA ratios. The highest efficiencyof SP20 liposomes in delivering DNA into BCEC can berationalized by considering its rapid and stable complexationwith DNA due to result of ionic interactions between themultivalent lipid and negatively charged phosphate groupsof DNA. SP20 was completely biodegradable compared tomany synthesized lipids as it was derived from naturallyoccurring compounds resulting in least toxicity compared toother liposomal formulations [64].

Teshima et al. studied prednisolone- (PLS-) incorporatedliposomes to improve retention property of prednisolone.Introduction of a lipophilic moiety (palmitoyl) to pred-nisolone (Pal-PLS) greatly enhanced drug retention inliposomes as lipophilic moiety increased its affinity to lipo-somal lipid bilayer. The investigators studied two liposomescontaining two different lipids, egg phosphatidylcholine(EggPC) and distearoyl phosphatidylcholine (DSPC). Ultra-filtration and gel filtration techniques were used to investi-gate retention properties of PLS and pal-PLS in liposomes.While ultrafiltration method showed high incorporation effi-ciency of PLS into the liposomes, a significant decrease of itsincorporation efficiency was observed in gel filtration. Thisresult indicated that elution medium in gel filtration studiesreleased incorporated PLS from liposomes. Pal-PLS showed

higher incorporation efficiency in both ultrafiltration and gelfiltration studies. However, incubation of liposomes with ratplasma for 1 min effectively decreased Pal-PLS incorporationinto EggPC/Chol liposomes as detected by gel filtration. Thereducing effect of Pal-PLS incorporation into liposomes byrat-plasma was overcome by using DSPC lipid in liposomalformulation. Further surface modification of liposomes witha hydrophilic polymer PEG resulted in the protection ofthe entrapped palmitoyl-PLS and thus generated a stableretention property of the drug [65].

Law et al. reported topical administration of acyclovir-(ACV)- encapsulated liposomes, where in vitro cornealpenetration and in vivo corneal absorption (using malerabbits) of acyclovir from ACV-encapsulated liposomes werestudied. This study reported the effect of liposomal surfacecharge on their corneal penetration and absorption. Surfacecharge of liposomes plays a significant role in improvingthe efficiency of ocular drug delivery system. Positivelycharged liposomes exhibited higher drug loading efficienciesas well as faster drug release rates compared to negativelycharged liposomes. Prolonged penetration across the corneawas observed for ACV-encapsulated liposomes. This phe-nomenon was more evident in case of positively chargedliposomes. The penetration rate for positively chargedliposomes was found to be approximately 3.6-fold lower thanfree ACV and approximately 2-fold lower than negativelycharged liposomes. Similarly, ACV concentration profile inaqueous humor indicated higher corneal absorption andgreater corneal deposition of ACV for positively chargedliposomes relative to negatively charged ACV and free ACV.The researchers suggested that positively charged liposomescan interact electrostatically with the negatively chargedsurface of cornea. This interaction can result in strongerbinding which leads to formation of a completely coatedlayer on the corneal surface. This layer may cause an increasein residence time on the cornea surface resulting in higherACV absorption and greater extent of ACV deposition inthe cornea compared to that of negatively charged liposomes[46].

Kawakami et al. reported O-palmitoyl prodrug oftilisolol-encapsulated liposome to improve the retentiontime of tilisolol in the precorneal area and vitreous body.The liposomes were administered topically, as well asintravitreally to the rabbit eye. Following topical adminis-tration, the researchers observed very low retention of O-palmitoyl tilisolol in the tear fluid even when it was appliedas liposomal formulation. The investigators significantlyincreased the retention property of liposomes by adding2% of carmellose sodium which acted as a reservoir forliposomes. In case of intravitreal administration, o-palmitoyltilisolol-encapsulated liposomes responded well resulting inhigher drug concentration in the vitreous body compared tofree tilisolol [66].

In the last decade numerous researchers addressed thechallenge of minimizing rapid clearance from precornealsite and enhancing the corneal permeation through vari-ous approaches. Utilization of chitosan in the preparationof mucoadhesive and cationic formulations was widelyexplored for the delivery of small therapeutic molecules from


different categories. Other mucoadhesive polymers were alsoapplied in the formulation of hydrogels that can regulate thedrug release rate at the ocular surface.

4. Intravitreal Applications

Liposomes represent the first injectable systems for intrav-itreal administrations. Liposomes can provide sustainedrelease for prolonged period. In addition, liposomal for-mulation can minimize the tissue toxicity and enhance theintravitreal half-life of drugs by decreasing rapid clearancefrom vitreous cavity [67, 68].

Barza et al. delineated the effect of liposome size andpathological state of eye on the intravitreal eliminationkinetics of carriers. Investigators observed that the clearancerate of SUVs was faster than LUVs. Moreover, intraocularinflammation also increases the intravitreal clearance rate[69]. Recently ocular pharmacologists have utilized liposo-mal hydrogel and sterically (pegylated) stabilized liposomesto address the drawbacks associated with intravitreal admin-istrations of liposomes [70].

In an application, rhodamine-conjugated liposomesloaded with vasoactive intestinal peptide (VIP) were givenintravenously to healthy rats to examine efficacy in the treat-ment of ocular inflammation. VIP is an immunomodulatoryneuropeptide involved in the regulation of ocular immuneresponse by modulating the activities of macrophages, Tlymphocytes, and dendritic cells [19, 71]. Intravitreal appli-cation of VIP-loaded liposomes was proposed for the treat-ment of endotoxin-induced uveitis [72]. Internalization ofrhodamine-conjugated liposomes (Rh-Lip) alone and loadedwith VIP (VIP-Rh-Lip) was examined in male Lewis rats.Intraocular and systemic biodistributions of the liposomeswere also determined. The authors reported that, after singleintravitreal injection, liposomes were internalized by retinalMuller glial cells, resident macrophages, and rare infiltratingactivated macrophages. Majority of the liposomes reachedthe cervical lymph nodes via conjunctival lymphatics. VIP-Rh-Lip internalized via macrophages resulted in slowerrelease and long-term expression inside the ocular tissuesand cervical lymph nodes. Thus, intravenous delivery of VIPby liposomes would be helpful in the treatment of uveitisand other immune-mediated eye diseases by modulating theimmune microenvironment of the ocular region [32].

Camelo et al. evaluated the liposomal formulation dis-persed in hyaluronic acid (HA) gel for the delivery of VIPin the treatment of uveitis and uveoretinitis in Lewis rats.Major limitation with the VIP-LP was shorter residence timein the vitreous cavity due to rapid elimination through thelymphatic circulation. Investigators attempted to increase thehalf-life of VIP-loaded liposomes (VIP-LP) after intravitrealadministration by suspending them in the hydrogel. HAwhich is the major component of vitreous was utilizedfor the studies. The researchers incorporated liposomes inHA gel in order to attain sustained release of VIP fromthe liposomes. VIP-LP suspended in HA gel was retainedin the vitreous cavity for 8 days after single intravitrealinjection. Authors reported that incorporation of liposomesin the gel had increased the viscosity of the gel due to

the enhanced interaction between HA gel and phospholipids.Moreover, it was reported that formulation was effectivein the treatment as evident by reduced clinical score andnumber of polymorphonuclear cells [73].

In a study tacrolimus-loaded liposomes were utilized forthe treatment of uveoretinitis. The vesicles were preparedby reverse phase evaporation technique and subsequentlyevaluated for efficacy and safety following intravitreal injec-tion in rats. No change in the retinal function was observedin the liposome-treated rats. Histopathological examinationrevealed reduced inflammatory response in comparison tofree drug. Liposomes were able to maintain the vitreousconcentration of more than 50 ng/mL for 2 weeks after singleadministration. Investigators concluded that tacrolimus-loaded liposomes were more effective than the drug alone.The formulation also reduced drug-related toxicity to innerretinal cells [31].

In another study, Abrishami et al. prepared nano-liposomes of bevacizumab. The investigators utilizeddehydration-rehydration method for achieving highestencapsulation efficiency. Researchers attempted to reducethe clearance of bevacizumab liposomes by incorporationof cholesterol. In comparison to free drug, concentration ofliposomal formulation was 5 times higher at 42 days. Thisstudy revealed that liposomal formulation of bevacizumabwas proven effective in the controlled release of bevacizumabfor more than 6 weeks in rabbit model [74].

Fluconazole liposomes were evaluated for the treatmentof candidal endophthalmitis. In the comparative study,intravitreal injections of fluconazole solution or liposomalformulation were given at different dose levels in therabbit eyes. Administration of fluconazole solution causedphotoreceptor disorientation and ultrastructural changes ofthe retina at the concentration of 100 μg in 0.1 mL or above.In contrast, liposomal formulation of fluconazole did notshow any retinal alteration up to concentration of 200 μg in0.1 mL [75].

Cheng et al. Formulated lipid prodrug of ganci-clovir (GCV), 1-O-hexadecylpropanediol-3-phospho-GCVinto liposomes which were injected intravitreally in rabbits.The researchers used this liposomal formulation for antiviraltreatment against herpes simplex virus type 1 (HSV-1)and human cytomegalovirus (HCMV). Intravitreal injectionwith 0.2 nM intravitreal concentration was the most effectivewithout causing any side effects of vitreous clarity orcataracts development in the eye. Moreover, this formulationprovided complete retinal protection even after simultaneousintravitreal injection [76].

Bochot et al. reported that phosphodiester oligonu-cleotide encapsulated sterically (pegylated) stabilized lipo-somes which were administered intravitreally in rabbits.It was the first reported use of liposomes as vehicle forintravitreal delivery of phosphodiester oligonucleotides. Theinvestigators tried to overcome the problem of short intrav-itreal half-life of oligonucleotide by encapsulating [33P]labeled 16-mer oligothymidylate (PdT16) within liposome.After intravitreal injection liposomal formulations yieldedsignificantly higher concentration of radiolabeled 33P withinthe posterior segment of the eye (vitreous, retina, choroid,


and sclera) than the solution. A heterogeneous competi-tive hybridization assay revealed a significantly improvedintraocular stability of PdT16 when it was administered in aliposomal formulation. The sterically stabilized hydrophilicpolyethylene glycol (PEG) chains on the liposome’s surfaceprotected them from degradation, resulting in prolongedresidence time in vitreous and sustained release of encap-sulated oligonucleotide into the vitreous and the retina-choroid. Controlled release of [33P] PdT16 from liposomesalso inhibited unwanted distribution of oligonucleotide inthe nontargeted tissues (sclera, lens) and thus reduced overallocular toxicity [77].

Peeters et al. reported cationic liposomes as nonviralgene carriers which were complexed with therapeutic DNA,called lipoplexes (LPXs). The authors investigated the factorsresponsible for inefficient vitreous diffusion of nonviral genecomplexes and addressed the problems to overcome vitreousbarrier for lipoplexes. FITC-dextran, fluorescent polystyrenenanospheres as models for LPXs and LPXs were mixed withvitreous gel obtained from bovine eyes, and their mobilityin vitreous was studied by fluorescence recovery afterphotobleaching (FRAP) technique. Polystyrene nanospherescan bind to collagen fibers within the vitreous due tohydrophobic interactions resulting in restricted mobilityin the vitreous. To overcome this problem, hydrophilicpolyethylene glycol (PEG) chains were grafted on the surfaceof nanoparticles that had prevented adsorption to the colla-gen fibers and thus increased their mobility in the vitreous.They reported that the size of the nanospheres should beless than 500 nm to obtain good vitreous mobility; otherwiseit would be sterically hindered by vitreous network andspread nonhomogeneously throughout the vitreous result-ing in accumulation near the injection site. Nonpegylatedcationic liposomes aggregated in the vitreous as negativelycharged glycosaminoglycans (GAGs) strongly bind to thecationic lipoplexes, which neutralize positive zeta potentialof lipoplexes, and thus favor aggregation. Low to moderatepegylation (1.9 mol% DSPE-PEG to 9.1 mol% DSPE-PEG)on cationic lipoplexes prevented their aggregation but,binding to biopolymers in the vitreous still occurred. Furtherincrease of DSPE-PEG to 16.7 mol% prevented both vitreousaggregation as well as binding to vitreous fibrils, resultingin homogeneous vitreous distribution and vitreous mobility[35]. The size and zeta potential of pegylated LPXs decreasedwith increasing the amount of pegylated lipids (DSPE-PEG)in LPXs. Gel electrophoresis experiments indicated that LPXsin vitreous remain stable and do not disassemble. The dataon mobility, aggregation, and stability of lipoplexes openedup a new direction to nonviral ocular gene therapy, butsome factors need to take into consideration. Here transportof drugs in vitreous was assumed by diffusion mechanismonly but in case of larger animal species like humansdrug transport through convection plays a significant role.Moreover here transport was focused in the central parts ofvitreous samples. Cortical vitreous zone containing denselypacked collagen and inner limiting lamina may produceadditional barriers to the diffusion of LPXs into the retina.

Gupta et al. evaluated fluconazole-encapsulated lipo-somes which were administered intravitreally in rabbit

eyes. Entrapping of fluconazole into liposomes significantlyslowed down clearance of free fluconazole after intravitrealinjection and thereby achieved higher fluconazole concen-tration in the vitreous. The liposomes showed longer half-life (23.40 h) in comparison to free fluconazole (3.08 h)[78]. Among all these investigations performed by numerousresearchers, approach of entrapping bevacizumab will beadvantageous for designing controlled release system fortherapeutic macromolecules. Another approach of usingsterically stabilized liposomes for oligonucleotide deliverycan be further explored for resolving the challenges inocular gene therapy. This approach will be advantageousin minimizing the intravitreal clearance of liposomes anddistribution of oligonucleotide in the non-targeted tissues.

5. Subconjunctival Applications

Subconjunctival mode of administration has gained newmomentum in delivering the drugs to both the anteriorand posterior segments [79]. Subconjunctival injection ofliposomes can provide retentive effect and steady-staterelease at the site of application. Therefore, higher drugconcentrations can be achieved at the target site. In addi-tion, subconjunctival injection is better in comparison totopical application as it can improve patience complianceby avoiding repeated administrations and provide directaccess of the drug to the target site [80, 81]. Absorptionrate of liposome-bound low molecular weight heparin(LMWH) was investigated after subconjunctival injectionin the treatment of subconjunctival hemorrhage (SH) inrabbits. Low concentration of liposome-bound LMWH wasobserved as compared to the free LMWH in the intraocularregions (aqueous and vitreous). Moreover, lower systemiclevel of LMWH was noted after subconjunctival injection.The paper suggested that, due to larger size (approx.550 nm in size), liposomes remained at the site of injectionand avoided lymphatic drainage. Also, positively chargedliposomes encapsulated higher amounts of LMWH andreleased the drug in a sustained manner, thus providinglonger residence time and increased concentration at thetargeted site. Thus, subconjunctival application of liposomesis a possible strategy to avoid systemic side effects of LMWH[81]. In a similar study, Baek et al. attempted subconjunctivaladministration of streptokinase- (SK-) loaded liposomes forthe treatment of SH in rabbits. Freeze thaw method wasutilized for the production of liposomes. The study reportedthat 81% of the drug was released in 48 h. Higher absorptionefficiency of liposomes in comparison to free drug wasobserved. SK-encapsulated liposomes in the early phase ofSH need to be assessed [82].

Fukushima et al. reported clodronate liposomes(CL2MDP-lip), which were used to inhibit infiltration ofmacrophages in the conjunctiva in the case of blepharoconjunctivitis (EC) developed in Brown Norway rats.The liposomes were administered by subconjunctivalinjection as well as by intravenous injection. They foundthat CL2MDP-lip effectively decreased the number ofED2-positive macrophages in the conjunctivas, where ED1-positive macrophages infiltration could only be controlled if


the injection was administered just prior to OVA challenge[33]. Limited investigations on subconjunctival deliveryof liposomes were performed in the last decade. However,approach of utilizing liposomes of size greater than 550 nmcan be explored in future for long-term delivery byminimizing the systemic clearance of liposomes throughconjunctival capillaries. It would be interesting to investigatethe subconjunctival clearance of liposomes of various sizes.

6. Conclusion

Numerous applications of liposomes in ophthalmic drugdelivery were extensively studied. These carriers have suc-cessfully improved the drug bioavailability by controlled andtargeted delivery. In the case of topical application improve-ment in the precorneal retention, transcorneal permeation,and therapeutic efficacy was achieved by utilizing liposomalformulations. In addition, effects of charge and compositionof liposomes were explored in detail, which have providedcomprehensive understanding of the interaction betweenliposome and ocular tissues. The applications of chitosanand hydrogel for improving the precorneal retention ofliposomes were explored and shown potential for furtherinvestigation. Liposomal formulations have been evaluatedfor encapsulation of various drug molecules of differenttherapeutic classes. In particular, liposomal formulation ofsmall molecules for the treatment of bacterial conjunctivitisand glaucoma was developed. Moreover, posterior segmentdelivery of liposomes was proven successful in enhancing theintravitreal half-life and targeted delivery to the inner retinalcells. In the case of posterior segment disorders liposomalformulation of therapeutic macromolecules was examined.However, research on targeted delivery of liposomes waslimited. Receptors expressed on the cornea and retina couldbe explored in future for targeted drug delivery utilizingsurface-modified liposomes.

Acknowledgments

This study was supported by NIH R01EY09171-14 and NIHRO1 EY10659-12.

References

[1] D. H. Geroski and H. F. Edelhauser, “Drug delivery forposterior segment eye disease,” Investigative Ophthalmologyand Visual Science, vol. 41, no. 5, pp. 961–964, 2000.

[2] A. Urtti, “Challenges and obstacles of ocular pharmacokineticsand drug delivery,” Advanced Drug Delivery Reviews, vol. 58,no. 11, pp. 1131–1135, 2006.

[3] A. K. Mitra, “Role of transporters in ocular drug deliverysystem,” Pharmaceutical Research, vol. 26, no. 5, pp. 1192–1196, 2009.

[4] I. P. Kaur and M. Kanwar, “Ocular preparations: the formu-lation approach,” Drug Development and Industrial Pharmacy,vol. 28, no. 5, pp. 473–493, 2002.

[5] S. Wadhwa, R. Paliwal, and S. P. Vyas, “Nanocarriers in oculardrug delivery: an update review,” Current PharmaceuticalDesign, vol. 15, no. 23, pp. 2724–2750, 2009.


[7] I. P. Kaur, A. Garg, A. K. Singla, and D. Aggarwal, “Vesicularsystems in ocular drug delivery: an overview,” InternationalJournal of Pharmaceutics, vol. 269, no. 1, pp. 1–14, 2004.

[8] R. M. Mainardes, M. C. C. Urban, P. O. Cinto et al., “Colloidalcarriers for ophthalmic drug delivery,” Current Drug Targets,vol. 6, no. 3, pp. 363–371, 2005.

[9] T. A. Elbayoumi and V. P. Torchilin, “Current trends inliposome research,” Methods in Molecular Biology, vol. 605, pp.1–27, 2010.

[10] N. Li, C. Zhuang, M. Wang, X. Sun, S. Nie, and W. Pan,“Liposome coated with low molecular weight chitosan and itspotential use in ocular drug delivery,” International Journal ofPharmaceutics, vol. 379, no. 1-2, pp. 131–138, 2009.

[11] X. Ding, W. G. Alani, and J. R. Robinson, “extended-releaseand targeted drug delivery systems,” in Remington: The Scienceand Practice of Pharmacy, D. B. Troy, Ed., Lippincott Williamsand wilkins, Philadelphia, Pa, USA, 2005.

[12] A. Jesorka and O. Orwar, “Liposomes: technologies and ana-lytical applications,” Annual Review of Analytical Chemistry,vol. 1, no. 1, pp. 801–832, 2008.

[13] T. Lian and R. J. Y. Ho, “Trends and developments in liposomedrug delivery systems,” Journal of Pharmaceutical Sciences, vol.90, no. 6, pp. 667–680, 2001.

[14] M. R. Niesman, “The use of liposomes as drug carriers inophthalmology,” Critical Reviews in Therapeutic Drug CarrierSystems, vol. 9, no. 1, pp. 1–38, 1992.

[15] D. Papahadjopoulos, K. Jacobson, S. Nir, and T. Isac, “Phasetransitions in phospholipid vesicles. Fluorescence polarizationand permeability measurements concerning the effect oftemperature and cholesterol,” Biochimica et Biophysica Acta,vol. 311, no. 3, pp. 330–348, 1973.

[16] R. H. Guy, J. Hadgraft, M. J. Taylor, and I. W. Kellaway,“Release of non-electrolytes from liposomes,” Journal ofPharmacy and Pharmacology, vol. 35, no. 1, pp. 12–14, 1983.

[17] K. Tsukada, S. Ueda, and R. Okada, “Preparation of liposome-encapsulated anti-tumor drugs: relationship betweenlipophilicity of drugs and in vitro drug release,” Chemical andPharmaceutical Bulletin, vol. 32, no. 5, pp. 1929–1935, 1984.

[18] L. Zeng and X. Wu, “Modeling the sustained release oflipophilic drugs from liposomes,” Applied Physics Letters, vol.97, no. 7, Article ID 073701, 3 pages, 2010.

[19] V. P. Torchilin, “Recent advances with liposomes as pharma-ceutical carriers,” Nature Reviews Drug Discovery, vol. 4, no. 2,pp. 145–160, 2005.

[20] G. Bonacucina, M. Cespi, M. Misici-Falzi, and G. F. Palmieri,“Colloidal soft matter as drug delivery system,” Journal ofPharmaceutical Sciences, vol. 98, no. 1, pp. 1–42, 2009.

[21] J. M. Ruiz-Moreno, J. A. Montero, L. Arias et al., “Pho-todynamic therapy in subfoveal and juxtafoveal idiopathicand postinflammatory choroidal neovascularization,” ActaOphthalmologica Scandinavica, vol. 84, no. 6, pp. 743–748,2006.

[22] S. Lazzeri, M. Figus, E. Di Bartolo, S. Rizzo, and M. Nardi,“Verteporfin photodynamic therapy for retinal hemangioblas-toma associated with Von Hippel-Lindau Disease in a 9-year-old child,” Clinical & Experimental Ophthalmology. In press.

[23] R. Sachdeva, H. Dadgostar, P. K. Kaiser, J. E. Sears, and A.D. Singh, “Verteporfin photodynamic therapy of six eyes withretinal capillary haemangioma,” Acta Ophthalmologica, vol.88, no. 8, pp. e334–e340, 2010.


[24] U. Chakravdrthy, G. Soubrane, F. Bandello et al., “EvolvingEuropean guidance on the medical management of neovas-cular age related macular degeneration,” British Journal ofOphthalmology, vol. 90, no. 9, pp. 1188–1196, 2006.

[25] “Rostaporfin: PhotoPoint SnET2, purlytin, Sn(IV) etiopur-purin, SnET2, tin ethyl etiopurpurin,” Drugs in R and D, vol.5, no. 1, pp. 58–61, 2004.

[26] Y. Shen and J. Tu, “Preparation and ocular pharmacokineticsof ganciclovir liposomes,” AAPS Journal, vol. 9, no. 3, pp.E371–E377, 2007.

[27] K. M. Hosny, “Ciprofloxacin as ocular liposomal hydrogel,”AAPS PharmSciTech, vol. 11, no. 1, pp. 241–246, 2010.

[28] A. Danion, I. Arsenault, and P. Vermette, “Antibacterialactivity of contact lenses bearing surface-immobilized layersof intact liposomes loaded with levofloxacin,” Journal ofPharmaceutical Sciences, vol. 96, no. 9, pp. 2350–2363, 2007.

[29] R. Cortesi, R. Argnani, E. Esposito et al., “Cationic liposomesas potential carriers for ocular administration of peptides withanti-herpetic activity,” International Journal of Pharmaceutics,vol. 317, no. 1, pp. 90–100, 2006.

[30] R. M. Hathout, S. Mansour, N. D. Mortada, and A. S. Guinedi,“Liposomes as an ocular delivery system for acetazolamide: invitro and in vivo studies,” AAPS PharmSciTech, vol. 8, no. 1,article 1, 2007.

[31] R. Zhang, R. He, J. Qian, J. Guo, K. Xue, and Y. F. Yuan,“Treatment of experimental autoimmune uveoretinitis withintravitreal injection of tacrolimus (FK506) encapsulated inliposomes,” Investigative Ophthalmology & Visual Science, vol.51, no. 7, pp. 3575–3582, 2010.

[32] S. Camelo, L. Lajavardi, A. Bochot et al., “Ocular and systemicbio-distribution of rhodamine-conjugated liposomes loadedwith VIP injected into the vitreous of Lewis rats,” MolecularVision, vol. 13, pp. 2263–2274, 2007.

[33] A. Fukushima, A. Ozaki, W. Ishida, N. van Rooijen, K. Fukata,and H. Ueno, “Suppression of macrophage infiltration intothe conjunctiva by clodronate liposomes in experimentalimmune-mediated blepharoconjunctivitis,” Cell Biology Inter-national, vol. 29, no. 4, pp. 277–286, 2005.

[34] S. Kawakami, A. Harada, K. Sakanaka et al., “In vivogene transfection via intravitreal injection of cationic lipo-some/plasmid DNA complexes in rabbits,” International Jour-nal of Pharmaceutics, vol. 278, no. 2, pp. 255–262, 2004.

[35] L. Peeters, N. N. Sanders, K. Braeckmans et al., “Vitreous:a barrier to nonviral ocular gene therapy,” InvestigativeOphthalmology and Visual Science, vol. 46, no. 10, pp. 3553–3561, 2005.

[36] H. E. Schaeffer and D. L. Krohn, “Liposomes in topical drugdelivery,” Investigative Ophthalmology and Visual Science, vol.22, no. 2, pp. 220–227, 1982.

[37] K. Singh and M. Mezei, “Liposomal ophthalmic drug deliverysystem. I. Triamcinolone acetonide,” International Journal ofPharmaceutics, vol. 16, no. 3, pp. 339–344, 1983.

[38] K. Singh and M. Mezei, “Liposomal ophthalmic drug deliverysystem. II. Dihydrostreptomycin sulfate,” International Journalof Pharmaceutics, vol. 19, no. 3, pp. 263–269, 1984.

[39] P. Fitzgerald, J. Hadgraft, and C. G. Wilson, “A gammascintigraphic evaluation of the precorneal residence of lipo-somal formulations in the rabbit,” Journal of Pharmacy andPharmacology, vol. 39, no. 6, pp. 487–490, 1987.

[40] S. Aiba, “Studies on chitosan: 4. Lysozymic hydrolysis ofpartially N-acetylated chitosans,” International Journal ofBiological Macromolecules, vol. 14, no. 4, pp. 225–228, 1992.

[41] K. Tomihata and Y. Ikada, “In vitro and in vivo degradation offilms of chitin and its deacetylated derivatives,” Biomaterials,vol. 18, no. 7, pp. 567–575, 1997.

[42] M. M. Mehanna, H. A. Elmaradny, and M. W. Samaha,“Mucoadhesive liposomes as ocular delivery system: physical,microbiological, and in vivo assessment,” Drug Developmentand Industrial Pharmacy, vol. 36, no. 1, pp. 108–118, 2010.

[43] J. Zhang and S. Wang, “Topical use of Coenzyme Q10-loaded liposomes coated with trimethyl chitosan: tolerance,precorneal retention and anti-cataract effect,” InternationalJournal of Pharmaceutics, vol. 372, no. 1-2, pp. 66–75, 2009.

[44] Y. Diebold, M. Jarrın, V. Saez et al., “Ocular drug deliveryby liposome-chitosan nanoparticle complexes (LCS-NP),”Biomaterials, vol. 28, no. 8, pp. 1553–1564, 2007.

[45] A. M. de Campos, A. Sanchez, and M. J. Alonso, “Chitosannanoparticles: a new vehicle for the improvement of the deliv-ery of drugs to the ocular surface. Application to cyclosporinA,” International Journal of Pharmaceutics, vol. 224, no. 1-2,pp. 159–168, 2001.

[46] S. L. Law, K. J. Huang, and C. H. Chiang, “Acyclovir-containing liposomes for potential ocular delivery Cornealpenetration and absorption,” Journal of Controlled Release, vol.63, no. 1-2, pp. 135–140, 2000.

[47] P. Chetoni, S. Rossi, S. Burgalassi, D. Monti, S. Mariotti,and M. F. Saettone, “Comparison of liposome-encapsulatedacyclovir with acyclovir ointment: ocular pharmacokinetics inrabbits,” Journal of Ocular Pharmacology and Therapeutics, vol.20, no. 2, pp. 169–177, 2004.

[48] J. Zhang, P. Guan, T. Wang, D. Chang, T. Jiang, and S.Wang, “Freeze-dried liposomes as potential carriers for ocularadministration of cytochrome c against selenite cataractformation,” Journal of Pharmacy and Pharmacology, vol. 61,no. 9, pp. 1171–1178, 2009.

[49] F. S. Habib, E. A. Fouad, M. S. Abdel-Rhaman, and D. Fathalla,“Liposomes as an ocular delivery system of fluconazole: in-vitro studies,” Acta Ophthalmologica, vol. 88, no. 8, pp. 901–904, 2010.

[50] J. P. Guzek, J. M. Roosenberg, D. L. Gano, and I. F.Wessels, “The effect of vehicle on corneal penetration oftriturated ketoconazole and itraconazole,” Ophthalmic Surgeryand Lasers, vol. 29, no. 11, pp. 926–929, 1998.

[51] L. Budai, M. Hajdu, M. Budai et al., “Gels and liposomesin optimized ocular drug delivery: studies on ciprofloxacinformulations,” International Journal of Pharmaceutics, vol. 343,no. 1-2, pp. 34–40, 2007.

[52] S. Merino, J. L. Vazquez, J. L. Domenech et al.,“Fluoroquinolone-biomembrane interaction at the DPPC/PGlipid-bilayer interface,” Langmuir, vol. 18, no. 8, pp. 3288–3292, 2002.

[53] M. Budai, R. Pallaghy, Zs Szabo, A. Zimmer, and P. Grof,“Molecular interactions in lomefloxacin—liposome systems,”Cellular and Molecular Biology Letters, vol. 10, no. 70, 2005.

[54] X. Mu and Z. Zhong, “Preparation and properties ofpoly(vinyl alcohol)-stabilized liposomes,” International Jour-nal of Pharmaceutics, vol. 318, no. 1-2, pp. 55–61, 2006.

[55] A. F. Xie and S. Granick, “Phospholipid membranes assubstrates for polymer adsorption,” Nature Materials, vol. 1,no. 2, pp. 129–133, 2002.

[56] G. Abdelbary, “Ocular ciprofloxacin hydrochloride mucoad-hesive chitosan-coated liposomes,” Pharmaceutical Develop-ment and Technology, vol. 16, no. 1, pp. 44–56, 2009.


[57] O. Felt, P. Furrer, J. M. Mayer, B. Plazonnet, P. Buri,and R. Gurny, “Topical use of chitosan in ophthalmology:tolerance assessment and evaluation of precorneal retention,”International Journal of Pharmaceutics, vol. 180, no. 2, pp. 185–193, 1999.

[58] J. P. Craig, C. Purslow, P. J. Murphy, and J. S. W. Wolffsohn,“Effect of a liposomal spray on the pre-ocular tear film,”Contact Lens and Anterior Eye, vol. 33, no. 2, pp. 83–87, 2010.

[59] M. I. Afouna, I. S. Khattab, and I. K. Reddy, “Preparation andcharacterization of demeclocycline liposomal formulationsand assessment of their intraocular pressure-lowering effects,”Cutaneous and Ocular Toxicology, vol. 24, no. 2, pp. 111–124,2005.

[60] A. S. Monem, F. M. Ali, and M. W. Ismail, “Prolonged effectof liposomes encapsulating pilocarpine HCl in normal andglaucomatous rabbits,” International Journal of Pharmaceutics,vol. 198, no. 1, pp. 29–38, 2000.

[61] G. A. Lesher and G. G. Gunderson, “Continuous drug deliverythrough the use of disposable contact lenses,” Optometry andVision Science, vol. 70, no. 12, pp. 1012–1018, 1993.

[62] A. Danion, H. Brochu, Y. Martin, and P. Vermette, “Fabrica-tion and characterization of contact lenses bearing surface-immobilized layers of intact liposomes,” Journal of BiomedicalMaterials Research A, vol. 82, no. 1, pp. 41–51, 2007.

[63] S. S. Mahmoud, J. D. Gehmam, K. Azzopardi, R. M. Robins-Browne, and F. Separovic, “Liposomal phospholipid prepa-rations of chloramphenicol for ophthalmic applications,”Journal of Pharmaceutical Sciences, vol. 97, no. 7, pp. 2691–2701, 2008.

[64] U. Pleyer, D. Groth, B. Hinz et al., “Efficiency and toxicity ofliposome-mediated gene transfer to corneal endothelial cells,”Experimental Eye Research, vol. 73, no. 1, pp. 1–7, 2001.

[65] M. Teshima, S. Kawakami, K. Nishida et al., “Prednisoloneretention in integrated liposomes by chemical approach andpharmaceutical approach,” Journal of Controlled Release, vol.97, no. 2, pp. 211–218, 2004.

[66] S. Kawakami, K. Yamamura, T. Mukai et al., “Sustained oculardelivery of tilisolol to rabbits after topical administration orintravitreal injection of lipophilic prodrug incorporated inliposomes,” Journal of Pharmacy and Pharmacology, vol. 53,no. 8, pp. 1157–1161, 2001.

[67] G. A. Peyman, J. A. Schulman, B. Khoobehi, H. M. Alkan,M. E. Tawakol, and H. Mani, “Toxicity and clearance ofa combination of liposome-encapsulated ganciclovir andtrifluridine,” Retina, vol. 9, no. 3, pp. 232–236, 1989.

[68] A. A. Alghadyan, G. A. Peyman, B. Khoobehi, S. Milner,and K. R. Liu, “Liposome-bound cyclosporine: clearance afterintravitreal injection,” International Ophthalmology, vol. 12,no. 2, pp. 109–112, 1988.

[69] M. Barza, M. Stuart, and F. Szoka Jr., “Effect of size andlipid composition on the pharmacokinetics of intravitrealliposomes,” Investigative Ophthalmology and Visual Science,vol. 28, no. 5, pp. 893–900, 1987.

[70] S. Ebrahim, G. A. Peyman, and P. J. Lee, “Applications ofliposomes in ophthalmology,” Survey of Ophthalmology, vol.50, no. 2, pp. 167–182, 2005.

[71] A. W. Taylor, J. W. Streilein, and S. W. Cousins, “Immunoreac-tive vasoactive intestinal peptide contributes to the immuno-suppressive activity of normal aqueous humor,” Journal ofImmunology, vol. 153, no. 3, pp. 1080–1086, 1994.

[72] L. Lajavardi, A. Bochot, S. Camelo et al., “Downregulation ofendotoxin-induced uveitis by intravitreal injection of vasoac-tive intestinal peptide encapsulated in liposomes,” InvestigativeOphthalmology and Visual Science, vol. 48, no. 7, pp. 3230–3238, 2007.

[73] S. Camelo, L. Lajavardi, A. Bochot et al., “Drainage offluorescent liposomes from the vitreous to cervical lymphnodes via conjunctival lymphatics,” Ophthalmic Research, vol.40, no. 3-4, pp. 145–150, 2008.

[74] M. Abrishami, S. Z. Ganavati, D. Soroush, M. Rouhbakhsh, M.R. Jaafari, and B. Malaekeh-Nikouei, “Preparation, characteri-zation, and in vivo evaluation of nanoliposomes-encapsulatedbevacizumab (avastin) for intravitreal administration,” Retina,vol. 29, no. 5, pp. 699–703, 2009.

[75] T. Velpandian, K. Narayanan, T. C. Nag, A. K. Ravi, and S.K. Gupta, “Retinal toxicity of intravitreally injected plain andliposome formulation of fluconazole in rabbit eye,” IndianJournal of Ophthalmology, vol. 54, no. 4, pp. 237–240, 2006.

[76] L. Cheng, K. Y. Hostetler, S. Chaidhawangul et al., “Intravitrealtoxicology and duration of efficacy of a novel antiviral lipidprodrug of ganciclovir in liposome formulation,” InvestigativeOphthalmology and Visual Science, vol. 41, no. 6, pp. 1523–1532, 2000.

[77] A. Bochot, E. Fattal, V. Boutet et al., “Intravitreal delivery ofoligonucleotides by sterically stabilized liposomes,” Investiga-tive Ophthalmology and Visual Science, vol. 43, no. 1, pp. 253–259, 2002.

[78] S. K. Gupta, T. Velpandian, N. Dhingra, and J. Jaiswal, “Intrav-itreal pharmacokinetics of plain and liposome-entrappedfluconazole in rabbit eyes,” Journal of Ocular Pharmacologyand Therapeutics, vol. 16, no. 6, pp. 511–518, 2000.

[79] S. Raghava, M. Hammond, and U. B. Kompella, “Periocularroutes for retinal drug delivery,” Expert Opinion on DrugDelivery, vol. 1, no. 1, pp. 99–114, 2004.

[80] K. K. Assil, J. Frucht-Perry, E. Ziegler, D. J. Schanzlin, T.Schneiderman, and R. N. Weinreb, “Tobramycin liposomes:single subconjunctival therapy of pseudomonal keratitis,”Investigative Ophthalmology and Visual Science, vol. 32, no. 13,pp. 3216–3220, 1991.

[81] J. W. Moon, Y. K. Song, J. P. Jee, C. K. Kim, H. K. Choung, andJ. M. Hwang, “Effect of subconjunctivally injected, liposome-bound, low-molecular-weight heparin on the absorptionrate of subconjunctival hemorrhage in rabbits,” InvestigativeOphthalmology and Visual Science, vol. 47, no. 9, pp. 3968–3974, 2006.

[82] S. H. Baek, S. J. Park, S. E. Jin, J. K. Kim, C. K. Kim, and J. M.Hwang, “Subconjunctivally injected, liposome-encapsulatedstreptokinase enhances the absorption rate of subconjunctivalhemorrhages in rabbits,” European Journal of Pharmaceuticsand Biopharmaceutics, vol. 72, no. 3, pp. 546–551, 2009.


Research Article

Preparation and Characterization of Stealth Archaeosomes Basedon a Synthetic PEGylated Archaeal Tetraether Lipid

Julie Barbeau,1, 2 Sandrine Cammas-Marion,1, 2 Pierrick Auvray,3 and Thierry Benvegnu1, 2

1 Ecole Nationale Superieure de Chimie de Rennes, UMR 6226 CNRS, Avenue du General Leclerc, CS 50837,35708 Rennes Cedex 7, France

2 Universite europeenne de Bretagne, France3 C-RIS Pharma, Parc Technopolitain, Atalante Saint-Malo, 35400 Saint-Malo, France

Correspondence should be addressed to Sandrine Cammas-Marion, [email protected] andThierry Benvegnu, [email protected]

Received 1 July 2010; Revised 4 January 2011; Accepted 20 January 2011

Academic Editor: Adrian Williams

Copyright © 2011 Julie Barbeau et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

The present studies were focused on the formation and characterization of sterically stabilized archaeosomes made from a syntheticPEGylated archaeal lipid. In a first step, a synthetic archaeal tetraether bipolar lipid was functionalized with a poly(ethylene glycol),PEG, and (PEG45-Tetraether) with the aim of coating the archaeosome surface with a sterically stabilizing hydrophilic polymer.In a second step, Egg-PC/PEG45-Tetraether (90/10 wt%) archaeosomes were prepared, and their physicochemical characteristicswere determined by dynamic light scattering (size, polydispersity), cryo-TEM (morphology), and by high-performance thin layerchromatography (lipid composition), in comparison with standard Egg-PC/PEG45-DSPE formulations. Further, a fluorescent dye,the carboxyfluorescein, was encapsulated into the prepared archaeosomes in order to evaluate the potential of such nanostructuresas drug carriers. Release studies have shown that the stability of Egg-PC/PEG45-Tetraether-based archaeosomes is significantlyhigher at 37◦C than the one of Egg-PC/PEG45-DSPE-based liposomes, as evidenced by the slower release of the dye encapsulatedinto PEGylated archaeosomes. This enhanced stability could be related to the membrane spanning properties of the archaealbipolar lipid as already described with natural or synthetic tetraether lipids.

1. Introduction

In the drug-delivery field, several nanocarriers have beenproposed to improve the therapeutic index of variousbiologically active molecules such as peptides. Indeed, in vivoadministration of peptides is still limited by their poorbioavailability and susceptibility to cleavage by proteases.In order to obtain a satisfactory therapeutic effect, thepeptide has to be frequently administrated at high dosesleading to unwanted toxic effects, such as induction ofimmune response. Consequently, peptide encapsulation intosite-specific delivery systems can offer solutions to theabove-mentioned problems. Indeed, the nanocarriers can (i)enhance drug solubility, (ii) control drug release thus avoid-ing toxic side effects, (iii) improve drug biodistribution, (iv)and, if appropriate molecule is grafted on the nanocarriersurface, target a specific site of action. Several nanovectorshave been used to encapsulate various therapeutic peptides

such as liposomes, nanoparticles, and nano- or microgels[1–8]. Among these nanocarriers, liposomes are of greatimportance because of their relatively large carrying capacityand the possibility to entrap either hydrophilic, hydrophobic,or amphiphilic drugs. Moreover, a good knowledge of suchvectors has been acquired since the first discovery of lipo-somes by Bangham and Horne [9] attested by commerciallyavailable anticancer liposomial formulations such as Doxil[10, 11]. However, despite encouraging results, a majorlimitation to the development of liposomes as drug carriersis their instability, especially during their transit to the siteof action [12]. Attempts to improve their stability, either byincorporation of high amount of cholesterol or by coatingthe liposome surface with poly(ethylene glycol), have led tolimited success.

Within this context, archaeosomes, made with one ormore of either the ether lipids found in Archaea bacteriaor synthetic archaeal lipids, constitute a novel family of


liposomes exhibiting higher stabilities in several conditions,such as high temperature, alkaline or acidic pH, presenceof phospholipases, bile salts, and serum media [13, 14].Therefore, because of their biocompatibility and higherstability, archaeosomes have been extensively studied forpotential applications as drug/gene and vaccine deliverysystems [14, 15].

Over the last decade, our research group has developedsynthetic analogues of natural archaeal tetraether lipids andstudied their uses in cationic archaeosome formulations asefficient gene delivery systems [16–18]. Our next objectivewas to evaluate the potential applications of archaeosometechnology for the delivery of additional hydrophilic sub-strates such as antitumoral peptides (Project Sealacian:encapsulation of natural marine peptides, extracted fromScyliorhinus canicula, for their site-specific delivery). Ourattention was then directed towards the preparation and theformulation of a PEGylated archaeal tetraether lipid (PEG45-Tetraether) to provide neutral coated archaeosomes valuableas peptide nanocarriers. In order to assess the value of thisnew family of stealth liposomes, physicochemical charac-teristics (DLS, cryo-TEM, and HPTLC), dye encapsulationand release profile for a PEGylated archaeosome formulationwere determined and compared to those measured from aconventional PEGylated liposome formulation.


2.1. Materials. Egg-PC was purchased from Sigma. 1,2-distearoyl-sn-glycero-3-phosphatidylethanolamine-N -[methoxy-poly(ethylene glycol)-2000], ammonium salt,(PEG45-DSPE) was purchased from Aventi Polar. PEG45-Tetraether was synthesized according to a four-step pro-cedure from the tetraether diol 1 available in our labora-tory [13]. All reactions were carried out under nitrogenatmosphere with dry, freshly distilled solvents under anhy-drous conditions. Dichloromethane (CH2Cl2) and methanol(MeOH) were distilled over calcium hydride. All otherreagents were used directly from the supplier without furtherpurification unless noted. Analytical thin-layer chromatogra-phy (TLC) was performed on Merck 60 F254 silica gel nonac-tivated plates. A solution of 5% H2SO4 in EtOH or ultravioletfluorescence was used to develop the plates. Column chro-matography was performed on silica gel MERCK 60 H (5–40 μm). Nuclear magnetic resonance spectra (1H NMR and13C NMR) were recorded on a Brucker ARX 400 instrument(1H at 400 MHz, 13C at 100 MHz). Data are reported asfollows: chemical shift (number of hydrogen, multiplicity,and coupling constants if applicable). The chemical shifts(δ) are reported as parts per million (ppm) referenced tothe appropriate residual solvent peak. Coupling constantsare reported in Hertz (Hz). Abbreviations are as follows: s(singlet), d (doublet), t (triplet), q (quartet), dd (doublet ofdoublet), and m (multiplet). High-resolution mass spectra(HRMS) were performed by CRMPO (Universite de Rennes1) on a MS/MS ZabSpec TOF Micromass. Accurate massesare reported for the molecular ions [M+H]+, [M+Na]+,[M+K]+, or [M−H]−. Optical rotations were measured on

a Perkin-Elmer 341 polarimeter. IR spectra were recorded ona Nicolet 250 FT-IR spectrometer.

HPTLC plates (20∗10 cm, silica gel 60, 0.2 mm layerthickness, Nano-Adamant UV254) were purchased fromMacherey-Nagel. Before use, the HPTLC plates were pre-washed with methanol, dried on a CAMAG TLC plate heaterIII at 120◦C for 20 min, and kept in an aluminum foil in adesiccator at room temperature. All solvents were of HPTLCgrade.

2.2. Synthesis of PEG45-Tetraether

1-O-acetyl-2,2′-di-O-(3,7,11,15-tetramethylhexadecyl)-3,3′-O-(1,32-(13,20-dioxa)-dotriacontane-(cis-15,18-methylid-en))diyl-di-sn-glycerol 2. A mixture of tetraether diol 1(600 mg, 0.495 mmol, 1 equiv.), acetic anhydride (151 μL,3.5 equiv.) and sodium acetate (41 mg, 1 equiv.) was stirredunder reflux for 24 h. Water was added and the aqueousphase was extracted twice with CH2Cl2. The combinedorganic phases were dried (MgSO4) and concentrated underreduced pressure. The residue was purified by flash chro-matography on silica gel (petroleum ether (PE)/AcOEt:98 : 2) to yield the monoacetate derivative 2 (305 mg, 49%)as a colorless oil. Rf = 0.15 (PE/AcOEt: 9 : 1). [α]20

D : +9◦

(c 1.0, CHCl3). FT-IR υ (cm−1) 2924 (CH3), 2853 (CH2),1746 (CO), 1463 (CH2), 1377 (CH3), 1115 (COC); 1H NMR(CDCl3, 400 MHz) δ 0.80–0.89 (31H, m), 1.02–1.81 (92H,m), 1.91–1.98 (1H, m), 2.07 (3H, s), 2.13–2.23 (2H, m), 3.29(4H, d, J = 6.9 Hz), 3.39 (4H, t, J = 6.7 Hz), 3.43 (4H,t, J = 6.6 Hz), 3.44–3.74 (m, 8H), 4.11 (1H, dd, J = 5.7,11.6 Hz), 4.22 (1H, dd, J = 4.1, 11.6). 13C NMR (CDCl3,100 MHz) δ 19.61, 19.68, 19.75, 20.93, 22.63, 22.72, 24.32,24.46, 24.48, 24.81, 26.13, 28.02, 29.53, 29.62, 29.71, 29.79,30.03, 31.61, 32.81, 33.01, 36.73, 37.22, 37.33, 37.38, 37.43,37.51, 38.79, 39.38, 40.12, 40.68, 63.12, 64.13, 68.61, 68.89,68.91, 70.16, 70.19, 70.6, 70.9, 71.7, 71.9, 75.6, 76.5, 78.6,170.9. HRMS (ESI) calcd. for C79H157O9 (M+H)+ 1250.1827,found 1250.1823; HRMS (ESI) calcd. for C79H156O9Na[M+Na]+ 1272.1647, found 1272.1650; HRMS (ESI) calcd.for C79H156O9K [M+K]+ 1288.1386, found 1288.1381.

1-O-acetyl-1′-carboxy-2,2′-di-O-(3,7,11,15-tetramethylhex-adecyl)-3,3′-O-(1,32-(13,20-dioxa)-dotriacontane-(cis-15,18-methyliden))-diyl-di-sn-glycerol 3. To a solution of alco-hol 2 (50 mg, 0.04 mmol, 1 equiv.) in AcOEt (1 mL), a 0.5 Maqueous solution of KBr (8 μL, 0.1 equiv.) and TEMPO(1 mg, 0.2 equiv.) were added. At 0◦C, a 5% aqueous solutionof NaOCl (69 μL) was then added dropwise. The reactionmixture was stirred at room temperature for 2 h, the solutionwas acidified until pH 3-4 using 5% HCl and a 25% aqueoussolution of NaO2Cl (17 μL) was added slowly. After stirringfor 3 h at room temperature, the mixture was extracted withAcOEt, washed with a saturated aqueous solution of NaCl,dried (MgSO4), and concentrated under reduced pressureto give the carboxylic acid derivative 3 (45 mg, 90%) asa colorless oil. Rf = 0.28 (CH2Cl2/CH3OH: 9 : 1). FT-IRυ (cm−1) 2924 (CH3), 2853 (CH2), 1746 (COCH3), 1733(COOH), 1463 (CH2), 1377 (CH3), 1115 (COC); 1H NMR


(CDCl3, 400 MHz) δ 0.80–0.89 (31H, m), 1.02–1.81 (92H,m), 1.91–1.98 (1H, m), 2.07 (3H, s), 2.13–2.23 (2H, m), 3.29(4H, d, J = 6.9 Hz), 3.39 (4H, t, J = 6.7 Hz), 3.41–3.72 (m,12H), 3.79 (1H, ddd, J = 1.0, 3.3, 10.5 Hz), 4.03 (1H, dd,J = 4.1, 11.6 Hz), 4.11 (1H, d, J = 5.7, 11.6 Hz), 4.22 (1H,dd, J = 4.1, 11.6 Hz). 13C NMR (CDCl3, 100 MHz) δ 19.61,19.68, 19.75, 20.91, 22.63, 22.72, 24.3, 24.46, 24.48, 24.81,26.11, 28.02, 28.79, 29.51, 29.62, 29.73, 29.82, 30.02, 31.59,32.82, 33.01, 36.68, 36.81, 36.93, 37.04, 37.12, 37.19, 37.33,37.38, 37.41, 37.52, 38.84, 39.37, 40.12, 40.66 63.08, 63.12,64.15, 68.65, 68.89, 68.91, 70.16, 70.19, 70.63, 70.91, 71.72,71.91, 75.57, 76.53, 78.59, 170.91, 171.88. HRMS (ESI) calcd.for C79H153O10 [M−H]− 1262.1463, found 1262.1447.

PEG45-Tetraether. To a solution of carboxylic acid 3(16.6 mg, 0.015 mmol, 1 equiv.) and TBTU (4.6 mg, 1 equiv.)in dry CH2Cl2 (1 mL) was added DIEA (3.4 μL, 1.3 equiv.)under a nitrogen atmosphere. After 20 min at room temper-ature, a solution of H2N-PEG45-OMe 4 (24.4 mg, 1 equiv.)in dry CH2Cl2 (2 mL) was added and the reaction mixturewas stirred under reflux for 12 h. A few drops of a 5%HCl aqueous solution were then added and the solventswere removed under reduced pressure. The residue wasdissolved in CHCl3 (1 mL) and purified on a SephadexLH-20 column eluting with a mixture of CHCl3/CH3OH(9 : 1) to give a white solid (41 mg, 80%) composed of theexpected monoacetate derivative 5 and the starting H2N-PEG45-OMe 4 in a 80 : 20 ratio. FT-IR υ (cm−1) 2924 (CH3),2855 (CH2), 1746 (COCH3), 1651 (CONH), 1103 (COC);1H NMR (CDCl3, 400 MHz) δ 0.82–0.86 (31H, m, 10 CH3),1.00–1.80 (92H, m), 1.91–1.98 (1H, m), 2.06 (3H, s), 2.13–2.23 (2H, m), 3.27 (4H, d, J = 6.9 Hz), 3.36–3.58 (23H, m),3.37 (3H, s), 3.59–3.68 (169H, m), 3.73–3.77 (1H, m), 3.81(1H, dd, J = 4.1, 5.6 Hz), 3.88 (1H, dd, J = 2.5, 6 Hz), 4.08(1H, dd, J = 5.6, 11.6 Hz), 4.21 (1H, dd, J = 4.1, 11.6 Hz),7.03 (1H, m). 13C NMR (CDCl3, 100 MHz) δ 14.08, 19.58,19.65, 19.72, 20.91, 22.60, 22.69, 24.32, 24.45, 24.77, 26.03,26.07, 26.15, 27.93, 28.82, 29.32–29.84, 32.76, 33.9, 36.84–37.50, 38.57, 39.33, 39.70 59.00, 62.97, 68.60, 69.74, 69.83,70.29, 70,53, 70.91, 71.53, 71.56, 71.69, 71.83, 71.89, 75.60,77.20, 78.21, 80.50, 170.53, 170.72. To a solution of thiswhite solid (41 mg) in a CH2Cl2/CH3OH (1 : 1) mixture, wasadded a freshly prepared solution of CH3ONa in CH3OH(0.1 M, 1 equiv.). The reaction mixture was stirred at roomtemperature for 4 h. Amberlite resin (IR120) was added,the reaction mixture was filtered, and the solvents wereevaporated under reduced pressure. A white powder wasisolated (41 mg) composed of the desired PEG45-Tetraetherand the starting H2N-PEG45-OMe 4 in a 80 : 20 ratio. Rf

= 0.28 (CHCl3/CH3OH/H2O: 9 : 1). FT-IR υ (cm−1) 2927(CH3), 2855 (CH2), 1652 (CONH), 1103 (COC); 1H NMR(CDCl3, 400 MHz) δ 0.82–0.86 (31H, m, 10 CH3), 1.00–1.80(92H, m), 1.91–1.98 (1H, m), 2.13–2.23 (2H, m), 3.27 (4H,d, J = 6.9 Hz), 3.36–3.58 (23H, m), 3.37 (3H, s), 3.59–3.68(169H, m), 3.73–3.77 (1H, m), 3.81 (1H, dd, J = 4.1, 5.6 Hz),3.88 (1H, dd, J = 2.5, 6 Hz), 7.03 (1H, m). 13C NMR (CDCl3,100 MHz) δ 14.09, 19.58, 19.65, 19.72, 22.60, 22.69, 24.32,24.45, 24.77, 26.03, 26.07, 26.15, 27.93, 28.82, 29.32–29.84,

32.76, 33.9, 36.84–37.50, 38.57, 39.33, 39.70 59.00, 62.97,68.60, 69.74, 69.83, 70.29, 70,53, 70.91, 71.53, 71.56, 71.69,71.83, 71.89, 75.60, 77.20, 78.21, 80.50, 170.52.

2.3. Preparation of PEGylated Archaeosomes and PEGy-lated Liposomes. Stock solutions of Egg-PC (1 mg/mL) andPEG45-DSPE (1 mg/mL) were prepared in CHCl3 : CH3OH(2 : 1, v/v), while stock solutions of PEG45-Tetraether(1 mg/mL) were prepared in CHCl3.

Liposomes and archaeosomes were obtained by thehydration method as already described elsewhere [16–18].Briefly, the selected lipid solutions were mixed to yield eithera mixture of Egg-PC and PEG45-DSPE (90 : 10 wt%) or amixture of Egg-PC and PEG45-Tetraether (90 : 10 wt%) witha total lipid concentration of 1 mg/mL. The organic solventswere then evaporated using a rotary evaporator, and thelipid films thus obtained were dried under high vacuum for2 hours at room temperature. The dried lipid films werethen hydrated with 1 mL of milliQ water. The solutions werevortexed and left at 4◦C overnight. Archaeosome or liposomeformulations were sonicated at room temperature for twotimes 5 min with interval of 5 min using a Fischer scientificsonication bath (FB 15051) at 80 KHz. Each formulation wasrealized in duplicate.

2.4. Encapsulation of Carboxyfluorescein into PEGylated Ar-chaeosomes. PEG45-Tetraether (90 : 10 wt%) based archaeo-somes and Egg-PC/PEG45-DSPE (90 : 10 wt) PEGylated Li-posomes: Carboxyfluorescein (CF) was encapsulated inEgg-PC based liposomes during the hydration phase asdescribed elsewhere [19]. Briefly, Egg-PC/PEG45-Tetraether(90 : 10 wt%) and Egg-PC/PEG45-DSPE (90 : 1 wt%) lipidfilms were prepared as described above. After drying, bothlipid films were hydrated with 1 mL of a tris(hydroxylmethyl) methylamine buffer (Tris buffer) at pH 7.4 con-taining CF at a concentration of 100 mM. The solutionswere vortexed and left at 4◦C overnight. Both PEGylatedarchaeosomes and PEGylated liposomes containing CF weresonicated (Fischer scientific sonication bath FB 15051-80 KHz) at room temperature for two times 5 min withinterval of 5 min. Nonencapsulated CF was eliminated by sizeexclusion column chromatography on the Sephadex G-50 gelwith the Tris buffer as eluent. Both PEGylated archaeosomesand PEGylated liposomes containing CF were analyzed byDLS and by fluorescence using a Fluoromax-3 (Horiba)spectrofluorimeter with excitation and emission wavelengthsof 490 and 515 nm, respectively.

2.5. Size, Polydispersity, and Zeta Potential Measurements.The size (average diameter obtained by the cumulant resultmethod), polydispersity and zeta potential of the formu-lations were measured by dynamic light scattering usinga Delsa Nano Beckman Coulter apparatus at 25◦C. Thesamples were diluted 2 times with milliQ water.

2.6. Cryo-TEM Measurements. The cryo-TEM analysis ofPEGylated liposomes and PEGylated archaeosomes was real-ized by Dr. Olivier LAMBERT at the University of Bordeaux


(Group “Chimie et Biologie des Membranes et Nano-objets”,UMR 5248 CNRS).

Each sample (5 μL) was deposited on a grid coveredwith a carbon film having 2 μm diameter holes previouslyexposed to treatment with UV-ozone. The excess of waterwas removed by absorption with filter paper to form a thinlayer of water suspended inside the holes. This grid was thenplunged quickly (EM CPC, Leica) in liquid ethane (−178◦C).Rapid freezing of the thin layer of liquid water in vitreous ice(absence of crystals) preserved biological structures. Gridswere then placed in a suitable object carrier for observing thesamples at −170◦C. Observation under a microscope (FEITecna F20) was carried out in the mode low dose, limitingthe effects of beam irradiation on the lipid material. Imageswere recorded using an ultrasensitive camera (Gatan, USC1000) 2K∗2K with pixel size of 14 μm. The electron doseused was 10–20 electrons/A2. The image resolution underthese conditions was about 2 nm.

2.7. Lipid Composition of Liposomes and Archaeosomes byHPTLC. The lipid compositions of formulations were deter-mined after ultrafiltration. The samples were filtered through10 000 NMWL pore filters (Micron YM-10, Millipore Corpo-ration) by ultracentrifugation at 15 000 g for 1 hour at 15◦C.The supernatants were recovered, lyophilized, dissolved in1 mL of methanol, and analyzed by HPTLC using the auto-mated HPTLC system from CAMAG (Muttenz, Switzerland).The samples, the appropriate lipid standard solutions anda blank solution composed by pure methanol were spottedon 20 × 10 cm HPTLC plates using the Automatic TLCSampler 4 from CAMAG (Muttenz, Switzerland). Each lanewas spotted 10 mm above the bottom edge of the plate andwas 6 mm length with 17 mm spacing between lanes. Thespotting volume was 10 μL or 20 μL. A maximum of 20 laneswas spotted on a single plate. After evaporation of the samplesolvent, the plates were developed in a closed twin troughchamber for 20∗10 cm plates (CAMAG) containing 10 mLof the mobile phase (CHCl3/MeOH/H2O, 18/4/0.5) in eachtrough. The chamber was pre-equilibrated at least 20 minbefore the development. The development was stoppedwhen the solvent had migrated 80 mm. The plates weredried on a CAMAG TLC plate heater III at either 60◦Cfor 30 min. The HPTLC plates were postchromatographicderivatizated by dipping 5 s into a primuline solution (5 mgof primuline in 100 mL of acetone/H2O (80/20) mixture).HPTLC plates were then dried at room temperature for10 min and at 60◦C for 30 min on a CAMAG TLC plateheater III. Plates were then scanned from 6 mm abovethe bottom edge of the plate to the solvent front, usinga CAMAG TLC scanning densitometer. The measurementswere performed in fluorescence mode at λ = 366 nmwith a scanning speed of 20 mm/s, a slit dimension of4∗0.2 mm (Micro) and deuterium and tungsten lamps. Datawere stored online on a personal computer, and integrationas well as quantification was performed with the softwarepackage CATS from CAMAG. Calibration was performedby applying standard solutions in concentration givenbelow:

Egg-PC (Rf = 0.04): 10 μg, 7.5 μg, 5 μg, and 2.5 μg,

PEG45-DSPE (Rf = 0.46): 2 μg, 1 μg, 0.5 μg, and0.25 μg,

PEG45-Tetraether (Rf = 0.79): 2 μg, 1 μg, 0.5 μg, and0.25 μg.

Peak heights and peak areas were used for quantification.Calibration curves were calculated for each lipid or archaeallipid, with a linear regression mode. In order to reduce exper-imental errors, individual calibration curves were obtainedfor every HPTLC plate. The amount of Egg-PC and PEG45-DSPE in liposomes, after ultrafiltration, and of Egg-PC andPEG45-Tetraether in archaeosomes, after ultrafiltration, werecalculated from the calibration curves.

2.8. Carboxyfluorescein Release Profile. CF release profilefrom both PEGylated archaeosomes and PEGylated lipo-somes was measured by fluorescence using a Fluoromax-3 (Horiba) spectrofluorimeter with excitation and emissionwavelengths of 490 and 515 nm, respectively. Release wasstudied at 4◦C and 37◦C. The fluorescence of both formu-lations was measured at T0, before (I0) and after (Imax)Triton-X-100 (2 v%) addition (total disruption of liposomialmembranes) and at various times (It) until almost completeCF release at 4◦C and at 37◦C. Release of the incorporateddye was calculated using the following equation:

Release (%) = It− I0Imax− I0

∗ 100. (1)


Archaeosomes made with one or more of the ether lipidsfound in Archaea represent an innovative family of liposomesthat demonstrate higher stabilities to several conditions incomparison with conventional liposomes. The definition ofarchaeosomes also includes the use of synthetically derivedlipids that have the unique structure characteristics ofarchaeobacterial ether lipids, that is, regularly branchedphytanyl chains attached via ether bonds at sn-2,3 glycerolcarbons [15]. The lipid membrane of archaeosomes maybe entirely of the bilayer form if made exclusively frommonopolar archaeol (diether) lipids or a monolayer if madeexclusively from bipolar caldarchaeol (tetraether) lipids, ora combination of monolayers and bilayers if made fromcaldarchaeol lipids in addition to archaeol lipids or standardbilayer-forming phospholipids. The large variety of lipidstructures reflects the need for Archaea to adjust their corelipid structures in order to be able to ensure membranefunctions despite harsh destabilizing environmental condi-tions (high or low temperatures, high salinity, acidic media,anaerobic atmosphere, and high pressure) [20].

These atypical characteristics should be particularlyuseful for the preparation of highly stable archaeosomes. Inparticular, specific archaeal lipid membrane properties haveto be considered in view to optimize the performance ofarchaeosomes: (1) the ether linkages are more stable thanesters over a wide range of pH, and the branching methylgroups help both to reduce crystallization (membrane lipids


O

OO

OP O

O

N

OO

O P

O

O NH

O O

O

O

O

O O O

O

NH

O O

OOO

OHO

O

O

R1 et R2 = fatty acids

NH3

PEG45-DSPE

PEG45-tetraether

Egg-PC

45

45

3

33

3

R1

R2

Figure 1: Structure of Egg-PC, PEG45-DSPE, and PEG45-Tetraether.

in the liquid crystalline state at ambient temperature) andmembrane permeability (steric hindrance of the methyl sidegroups); (2) the saturated alkyl chains would impart stabilitytowards oxidative degradation; (3) the unusual stereochem-istry of the glycerol backbone (opposite to mesophilic organ-isms) would ensure resistance to attack by phospholipasesreleased by other organisms; (4) the bipolar lipids span themembranes and enhance their stability properties and (5)the addition of cyclic structures (in particular five-memberedrings) in the transmembrane portion of the lipids appearsto be a thermoadaptive response, resulting in enhancedmembrane packing and reduced membrane fluidity.

Consequently, formulations including archaeal lipidsdemonstrate relatively higher stabilities to oxidative stress,high temperature, alkaline or acidic pH, action of phos-pholipases, bile salts, and serum media. Archaeosomescan be formed using standard procedures (hydrated filmsubmitted to sonication, extrusion or detergent dialysis) atany temperature in the physiological range or lower, thusmaking it possible to encapsulate thermally labile com-pounds. Moreover, they can be prepared and stored in thepresence of air/oxygen without any degradation. The in vitroand in vivo studies indicate that archaeosomes are safe and donot elicit toxicity in mice. Thus, the biocompatibility and thesuperior stability properties of archaeosomes in numerousconditions offer advantages over conventional liposomes inthe manufacture and the use in biotechnology includingvaccine and drug/gene delivery.

However, to study in depth archaeolipid structure-archaeosome property relationships with a view of optimiz-ing the performance of these unusual liposomes as gene/drug

nanocarriers, sufficient amounts of pure natural lipids arerequired. Well-defined lipids are difficult to isolate fromnatural extracts, and chemical synthesis appears, therefore,as an attractive means of producing model lipids thatmimic the natural lipids. Within this context, our groupfocused on the synthesis and the evaluation of chemicallypure archaeal diether and tetraether lipids that retain someof the essential structural features of archaeal membranelipids. These studies clearly showed the interest in developingarchaeosome technology from synthetic tetraether lipidspossessing neutral, zwitterionic, or cationic polar headsgroups for in vitro and in vivo delivery applications of nucleicacids and drugs [13, 16–18].

In order to propose a stealth version of syntheticarchaeosomes that could increase blood circulation longevityby reducing or preventing protein binding and/or byinhibiting cell binding/uptake, an additional archaeosomeformulation based on a novel synthetic tetraether lipid wasdeveloped. These stealth archaeosomes could be suitablefor the encapsulation and the in vivo delivery of variousbioactive molecules including peptides which are known tobe highly sensitive to enzymatic or chemical degradations.Comparative studies in terms of drug-encapsulation efficacyand formulation stability between standard PEGylated lipo-somes and PEGylated archaeosomes were then investigatedby following the leakage of the encapsulated aqueous dye5(6)-carboxyfluorescein as a marker.

For that purpose, an archaeosome formulation com-posed by 90 wt% of a classical lipid, Egg-PC, and 10 wt%of a PEGylated tetraether archaeal lipid, PEG45-Tetraether(Figure 1) was selected. Indeed, previous studies relative to


OOO

O

O

O

OOO

OO

O

O

O

OOO

OO

O

O

O O

O O

NH

O O

O

OOO

O

O

O

(CH3CO)2O, CH3COONa,CH2Cl2, reflux 49%

(i) NaOCl, KBr, TEMPO, AcOEt, r.t.(ii) NaO2Cl, AcOEt, r.t.

(1) TBTU, DIEA,CH2Cl2, r.t.

( ) MeONaMeOH/CH2Cl2 (1/1)

quant

2

PEG45-tetraether

PEG45-tetraether/4: 80/20

+ 4

4

90%

80%

H2N

HO

HO

OH

OH

OH3 3

3

3

33

3

3

3

3

3

33

3

3

3

45

45

1

2

3

Scheme 1: Synthesis of PEG45-Tetraether lipid.

the use of archaeosomes as gene nanocarriers showed thatthe incorporation of 5 wt% to 10 wt% of tetraether archaeallipids into bilayered vesicles led to the best efficient in vitrogene transfection properties [16]. In parallel, a classicalliposomal formulation composed by 90 wt% of Egg-PC and10 wt% of PEG45-DSPE, was prepared in order to evaluatethe influence of the tetraether structure on the formulationproperties in terms of stability, drug-encapsulation effi-ciency, and further on the in vivo formulation efficacy. In thepresent approach, the vesicle formulations were studied froma fundamental point of view, that is, through DLS and cryo-TEM measurements (size, polydispersity, and morphology),HPTLC (lipid composition), and CF release (formulationstability) in order to assess the potentiality of PEGylatedarchaeosomes as in vivo nanocarriers.

3.1. Synthesis of PEG45-Tetraether Lipid. The novel PEGy-lated archaeal lipid (PEG45-Tetraether) was synthesizedthrough the functionalization of the tetraether backboneat one terminal end. The synthesis of this unsymmetricalPEGylated lipid involved the monoprotection of the startingtetraether diol 1 [13] followed by the introduction of thepoly(ethylene glycol) chain (Scheme 1). The first step wascarried out by an easy monoacetylation of diol 1 withsodium acetate (1 equiv.) and acetic anhydride (3.5 equiv.)to give monoacetate 2 in a 49% yield. Alcohol 2 wasthen oxidized in a one-pot two-step procedure underTEMPO catalysis conditions with NaOCl and NaClO2 asthe oxidizing agents. Fine tuning of the pH during thereaction led to a clean oxidation of 2 to carboxylic acid3 in a yield of 90%. With acid 3 in hand, we introduced


Table 1: Size (cumulant results), polydispersity (Ip), and zeta potential of prepared formulations. (ND = nondetermined).

Formulation Size (nm), (Std Dev) Ip Zeta potential (mV)

Egg-PC/PEG45-DSPE 70 (40) 0.30 −20.0 ± 9

Egg-PC/PEG45-Tetraether 80 (30) 0.26 −13.0 ± 6

CF-encapsulated Egg-PC/PEG45-DSPE 90 (37) 0.21 Nd

CF-encapsulated Egg-PC/PEG45-Tetraether 100 (45) 0.26 Nd

50 nm

(a)

50 nm

(b)

Figure 2: Cryo-TEM photos of (a) Egg-PC/PEG45-Tetraether (90 : 10 wt%) archaeosomes and (b) Egg-PC/PEG45-DSPE (90 : 10 wt%)liposomes. Bar is 50 nm.

a 45-unit PEG chain using commercially available H2N-PEG45-OMe 4. After optimization of the coupling reactionconditions, the use of the uronium salt (O-(benzotriazol-1-yl)1,1,3,3-tetramethyluronium tetrafluoroborate (TBTU)/N,N′-diisopropylethylamine (DIEA) system furnished theexpected PEGylated tetraether (80% yield) in addition tothe starting H2N-PEG45-OMe chain (ratio: 80 : 20). It isnoteworthy that the purification of the crude reactionmixture on a Sephadex LH-20 column allowed the totalremoval of the starting carboxylic acid 3. The final dea-cylation of the hydroxyl group under Zemplen conditions(MeONa, MeOH) gave the targeted PEG45-Tetraether lipidin a quantitative yield.

3.2. Physicochemical Characteristics of PEGylated Archaeo-somes and PEGylated Liposomes. As described in the exper-imental part, formulations have been prepared using theclassical lipid film hydration method followed by vesicle sizereduction under sonication. The mean particle size and zetapotential of archaeosomes and liposomes were measuredby dynamic light scattering. Particle mean diameters andpolydispersity index are gathered in Table 1 and show thatboth liposomes and archaeosomes are similar in size, lowerthan 100 nm, with a quite narrow dispersity (around 0.30).In the same way, the mean surface potential of archaeosomes

and liposomes were comparable with slightly negative values.These results are in good agreement with several reports[21, 22] that pointed out the impact of the PEG chains onliposomal size decrease and on zeta potential values close toneutrality. Most importantly, these studies revealed that theatypical structure of the tetraether did not modify the maincharacteristics of the resulting PEG-grafted vesicle structures(shape, size).

Cryo-TEM was employed to investigate the morphologyof the vesicles composed of PEGylated lipids. The imagesin Figure 2 show that PEG-bearing archaeosomes weredispersed and spherical as for classical PEGylated liposomes.The presence of an external dark circle evidenced thelipid layer surrounding the internal aqueous volume of thevesicles. It is noteworthy that no phase segregation has beenevidenced meaning that the prepared formulations are quitehomogenous. The sizes of the vesicles were under 100 nmand the diameter was comprised between 20 to 100 nm,which was in relatively good agreement with data obtainedby DLS. Indeed, DLS measurements gave average diameters(cumulant results) lower than 100 nm with objects havingdiameters ranging from around 20 nm to around 200 nm.

Besides these characteristics, it is of great interest todetermine the lipid composition after formulation. Forthat purpose, we have used an innovative method basedon quantitative thin layer chromatography, named high


Egg-PC

PEG45-DSPE

PEG45-tetraether

Standard

s

Egg-PC/PEG45-tetraether

Egg-PC/PEG45-DSPE

Blank

900800700600500400300200100

0

−0.20

0.20.4

0.60.8

1

All tracks at wavelength

(a.u

.)

0

50

100

150

200

(mm

)

(Rf)

(a) Scan of a plate at 366 nm (fluorescence mode)

y = 26x+ 371.5

R2 = 0.9904

0100200300400500600700800

0 2 4 6 8 10 12

Egg-PC

Amount (μg)

Amount (μg)

(a.u

.)(a

.u.)

PEG45-DSPE

y = 53.635x+ 67.217

R2 = 0.9946

0

50

100

150

200

250

0 0.5 1 1.5 2 2.5

(b) EggPC/PEG45-DSPE (90 : 10 wt%) liposomes

0100200300400500600700

0 2 4 6 8 10 12

Egg-PC

Amount (μg)

Amount (μg)

(a.u

.)(a

.u.)

0

50

100

150

200

250

0 0.5 1 1.5 2 2.5

y = 19.12x + 404.5

R2 = 0.9191

PEG45-tetraether

y = 64.035x + 74.217

R2 = 0.994

(c) EggPC/PEG45-Tetraether (90 : 10 wt%) archaeosomes

Figure 3: HPTLC measurements: (a) Scan of a plate at 366 nm (fluorescence mode); (b and c) standard curves, based on peak height, foreach lipid composing the prepared liposomes and archaeosomes. (AU = arbitrary unit).

performance thin-layer chromatography (HPTLC). TheHPTLC is a qualitative and quantitative analytical methodallowing obtaining reproducible and reliable results [23].This method is used, since several years, for analysis andquantification of lipids extracted from various sources [23–29]. More recently, the use of HPTLC has been developedfor the determination of lipid compositions of liposomes[30–34] and for peptide analysis in liposomes [35]. We

have, therefore, studied possibilities to use HPTLC for thedetermination of lipid compositions of the studied liposomesand archaeosomes. We have found conditions, describedin experimental part, which allowed us to measure lipidcomposition. After removal of nonaggregated lipids, thesupernatants were lyophilized and solubilized in methanol inorder to disrupt the nanostructure leading to the recoveringof nonaggregated lipids which can be further analyzed


Table 2: Amounts of lipids contained in liposomes and archaeosomes calculated from HPTLC data. The given values are an average betweenpeak height and peak area values. The values are reported to a volume of 1 mL.

Liposome formulations Archaeosome formulations

Egg-PCInitial amount (μg) 0.900 (90 wt%) 0.900 (90 wt%)

Amount in formulation (μg) 0.614 (88 wt%) 0.589 (86 wt%)

PEG45-DSPEInitial amount (μg) 0.100 (10 wt%) —

Amount in formulation (μg) 0.087 (12 wt%) —

PEG45-TetraetherInitial amount (μg) — 0.100 (10 wt%)

Amount in formulation (μg) — 0.096 (14 wt%)

0

10

20

30

40

50

60

70

80

90

100

0 40 80 120 160 200 240 280 320 360

Time (hours)

CF

rele

ase

(%)

Egg-PC/PEG42-DSPE (90 : 10 wt(%))

Egg-PC/PEG45-tetraether (90 : 10 wt(%))

(a) CF release at 4◦C

Time (hours)

0

10

20

30

40

50

60

70

80

90

100

CF

rele

ase

(%)

0 5 10 15 20

Egg-PC/PEG42-DSPE (90 : 10 wt(%))

Egg-PC/PEG45-tetraether (90 : 10 wt(%))

(b) CF release at 37◦C

Figure 4: Release (%) of CF from Egg-PC/PEG45-Tetraether (90 : 10 wt%) archaeosomes and from Egg-PC/PEG45-DSPE (90 : 10 wt%)liposomes at (a) 4◦C and (b) 37◦C.

by HPTLC as described in the experimental part. It isworth to note that no peak has been observed on the lanecorresponding to the blank solution. Such result allowed usto conclude that peaks corresponding to the analyzed lipids(Egg-PC: Rf = 0.04, PEG45-DSPE: Rf = 0.46 and PEG45-Tetraether: Rf = 0.79) were not overestimated because ofthe presence of other peaks having similar Rf values (Figure3(a)). Calibration curves, based on either peak height orpeak area, were plotted for each lipid (Figures 3(b) and3(c)). From these calibration curves, amounts of lipidscontained in each formulation studied were calculated (Table2) and compared to initial amount of lipids used to prepareliposomes and archaeosomes (Table 2). Results given inTable 2 demonstrated that lipid composition of the preparedliposomes and archaeosomes are very similar to the initiallipid compositions: 88/12 wt% for Egg-PC/PEG45-DSPEliposomes instead of an initial composition of 90/10 wt%and 86/14 wt% for Egg-PC/PEG45-Tetraether archaeosomesinstead of an initial composition of 90/10 wt%.

3.3. Carboxyfluorescein Encapsulation and Release Profile.To assess vesicle stability, the kinetics of encapsulated CFrelease from PEG-bearing liposomes and archaeosomes wasstudied at 4◦C (standard storage temperature of liposomalformulations) and 37◦C (human physiological temperature).The percent release of CF was calculated from the formula

described in the experimental part after evaluating theinitial amount of encapsulated CF. Thus, a part of thesample containing the vesicle dispersion was treated withtriton X-100 [36] for lipid membrane disruption. Then,the fluorescence analysis of the resulting sample allowedus to determine the CF concentration initially entrappedin the nanocarrier using a calibration curve beforehandestablished.

The release profile of CF from vesicles at 4◦C (Figure4(a)) showed different rates of leakage between liposomeand archaeosome formulations. Indeed, 45% CF release wasfound to be approximately 20 h for the liposome sample and100 h for the archaeosome sample. This different behaviorwas dramatically increased when the formulations werestudied at 37◦C. As shown in Figure 4(b), there was a rapidleakage of CF from conventional liposomes, where almost70% of the encapsulated marker was lost within 3 hours.On the contrary, a significant improvement in stability wasnoted with archaeosomes, which released only 20% duringthe same period.

Despite their apparent identical characteristics in termsof morphology and surface potential, PEGylated liposomesand archaeosomes exhibited different vesicle stabilities. Thepresence of only 10 wt% of archaeal tetraether lipid in theliposomal formulations increased significantly the nano-object stability and allowed a slow release of the encapsulated


dye at 37◦C. This enhanced stability could result fromthe membrane spanning organization of the PEGylatedtetraether lipids within the Egg-PC bilayer membrane,forming a monolayer as previously shown with syntheticcationic tetraethers [13].

4. Conclusions

In conclusion, we have demonstrated that small proportionsof a novel synthetic PEGylated archaeolipid added to aliposomal formulation increase significantly the nanovectorstability and slow down the constant dye release at 37◦C.This result is quite promising in so far as a similar behaviorcould be expected for in vivo applications. This study has alsoshown that HPTLC is a powerful method for analyzing lipidcomposition. Following such a fundamental work, we haverecently evaluated the encapsulation of a therapeutic peptide(anticancer) extracted from marine resources into PEGylatedarchaeosomes and the in vivo efficiency of this peptide-loaded formulation. The first results are very promising andwill be published elsewhere.

Acknowledgments

The authors would like to thank the partners of theproject Sealacian for valuable discussion. They also wouldlike to thank the CNRS, the Direction Generale desEntreprises (DGE), the Region Bretagne, and the Ministerede l’Enseignement Superieur et de la Recherche for financialsupport. Finally, the authors thank Dr. Olivier Lambert forcryo-TEM analysis. J. Barbeau would like to thank the DGEfor the financial support, which enabled her to achieve thisstudy.

References

[1] C. R. Dass and P. F. M. Choong, “Carrier-mediated delivery ofpeptidic drugs for cancer therapy,” Peptides, vol. 27, no. 11, pp.3020–3028, 2006.

[2] R. A. Schwendener and H. Schott, “Delivery systems forpeptides/oligonucleotides and lipophilic nucleoside analogs,”in Handbook of Pharmaceutical Biotechnology, S. C. Gad, Ed.,Pharmaceutical Development Series, pp. 1149–1172, Wiley-Blackwell, 2007.

[3] A. Graf, E. Ablinger, S. Peters, A. Zimmer, S. Hook, and T.Rades, “Microemulsions containing lecithin and sugar-basedsurfactants: nanoparticle templates for delivery of proteinsand peptides,” International Journal of Pharmaceutics, vol. 350,no. 1-2, pp. 351–360, 2008.

[4] B. Stark, F. Andreae, W. Mosgoeller et al., “Liposomal vasoac-tive intestinal peptide for lung application: protection fromproteolytic degradation,” European Journal of Pharmaceuticsand Biopharmaceutics, vol. 70, no. 1, pp. 153–164, 2008.

[5] I. Brasnjevic, H. W. M. Steinbusch, C. Schmitz, and P.Martinez-Martinez, “Delivery of peptide and protein drugsover the blood-brain barrier,” Progress in Neurobiology, vol. 87,no. 4, pp. 212–251, 2009.

[6] S. Chono, R. Fukuchi, T. Seki, and K. Morimoto, “Aerosolizedliposomes with dipalmitoyl phosphatidylcholine enhance pul-monary insulin delivery,” Journal of Controlled Release, vol.137, no. 2, pp. 104–109, 2009.

[7] C. He, F. Cui, L. Yin, F. Qian, C. Tang, and C. Yin, “Apolymeric composite carrier for oral delivery of peptidedrugs: bilaminated hydrogel film loaded with nanoparticles,”European Polymer Journal, vol. 45, no. 2, pp. 368–376, 2009.

[8] L. Lajavardi, S. Camelo, F. Agnely et al., “New formulationof vasoactive intestinal peptide using liposomes in hyaluronicacid gel for uveitis,” Journal of Controlled Release, vol. 139, no.1, pp. 22–30, 2009.

[9] A. D. Bangham and R. W. Horne, “Negative staining ofphospholipids and their structural modification by surface-active agents as observed in the electron microscope,” Journalof Molecular Biology, vol. 8, no. 5, pp. 660–668, 1964.

[10] A. Gabizon, H. Shmeeda, and Y. Barenholz, “Pharmacokinet-ics of pegylated liposomal doxorubicin: review of animal andhuman studies,” Clinical Pharmacokinetics, vol. 42, no. 5, pp.419–436, 2003.

[11] A. A. Gabizon, H. Shmeeda, and S. Zalipsky, “Pros and consof the liposome platform in cancer drug targeting,” Journal ofLiposome Research, vol. 16, no. 3, pp. 175–183, 2006.

[12] V. P. Torchilin, “Recent advances with liposomes as pharma-ceutical carriers,” Nature Reviews Drug Discovery, vol. 4, no. 2,pp. 145–160, 2005.

[13] M. Brard, C. Laine, G. Rethore et al., “Synthesis of archaealbipolar lipid analogues: a way to versatile drug/gene deliverysystems,” Journal of Organic Chemistry, vol. 72, no. 22, pp.8267–8279, 2007.

[14] T. Benvegnu, L. Lemiegre, and S. Cammas-Marion, “Newgeneration of liposomes called archaeosomes based on naturalor synthetic archaeal lipids as innovative formulations for drugdelivery,” Recent Patents on Drug Delivery & Formulation, vol.3, no. 3, pp. 206–220, 2009.

[15] G. B. Patel and G. D. Sprott, “Archaeobacterial ether lipidliposomes (archaeosomes) as novel vaccine and drug deliverysystems,” Critical Reviews in Biotechnology, vol. 19, no. 4, pp.317–357, 1999.

[16] G. Rethore, T. Montier, T. Le Gall et al., “Archaeosomesbased on synthetic tetraether-like lipids as novel versatilegene delivery systems,” Chemical Communications, no. 20, pp.2054–2056, 2007.

[17] T. Benvegnu, L. Lemiegre, and S. Cammas-Marion, “Archaeallipids: innovative materials for biotechnological applications,”European Journal of Organic Chemistry, no. 28, pp. 4725–4744,2008.

[18] C. Laine, E. Mornet, L. Lemiegre et al., “Folate-equippedpegylated archaeal lipid derivatives: synthesis and transfectionproperties,” Chemistry: A European Journal, vol. 14, no. 27, pp.8330–8340, 2008.

[19] P. Goldbach, H. Brochart, P. Wehrle, and A. Stamm, “Sterilefiltration of liposomes: retention of encapsulated carboxyflu-orescein,” International Journal of Pharmaceutics, vol. 117, no.2, pp. 225–230, 1995.

[20] C. R. Woese and G. E. Fox, “Phylogenetic structure of theprokaryotic domain: the primary kingdoms,” Proceedings ofthe National Academy of Sciences of the United States ofAmerica, vol. 74, no. 11, pp. 5088–5090, 1977.

[21] D. Lasis and F. J. Martin, “Stealth Liposomes,” in Handbooks inPharmacology and Toxicology, M. A. Hollinger, Ed., pp. 1–320,CRC Press, 1995.

[22] J. Thevenot, A. L. Troutier, L. David, T. Delair, andC. Ladaviere, “Steric stabilization of lipid/polymer parti-cle assemblies by poly(ethylene glycol)-lipids,” Biomacro-molecules, vol. 8, no. 11, pp. 3651–3660, 2007.

[23] J. Sherma, “Planar chromatography,” Analytical Chemistry,vol. 74, no. 12, pp. 2653–2662, 2002.


[24] L. J. Macal, R. K. Yu, and S. Ando, “Analysis of brainlipids by high performance thinlayer chromatography anddensitometry,” Journal of Lipid Research, vol. 24, no. 9, pp.1243–1250, 1983.

[25] S. Zellmer and J. Lasch, “Individual variation of humanplantar stratum corneum lipids, determined by automatedmultiple development of high-performance thin-layer chro-matography plates,” Journal of Chromatography B, vol. 691, no.2, pp. 321–329, 1997.

[26] G. Helmerich and P. Koehler, “Comparison of methods for thequantitative determination of phospholipids in lecithins andflour improvers,” Journal of Agricultural and Food Chemistry,vol. 51, no. 23, pp. 6645–6651, 2003.

[27] I. Meisen, J. Pater-Katalinic, and J. Muthing, “Direct analysisof silica gel extracts from immunostained glycosphingolipidsby nanoelectrospray ionization quadrupole time-of-flightmass spectrometry,” Analytical Chemistry, vol. 76, no. 8, pp.2248–2255, 2004.

[28] T. Kriska, W. Korytowski, and A. W. Girotti, “Role ofmitochondrial cardiolipin peroxidation in apoptotic pho-tokilling of 5-aminolevulinate-treated tumor cells,” Archives ofBiochemistry and Biophysics, vol. 433, no. 2, pp. 435–446, 2005.

[29] R. Usami, T. Fukushima, T. Mizuki, Y. Yoshida, A. Inoue,and K. Horikoshi, “Organic solvent tolerance of halophilicarchaea, Haloarcula strains: effects of NaCl concentrationon the tolerance and polar lipid composition,” Journal ofBioscience and Bioengineering, vol. 99, no. 2, pp. 169–174,2005.

[30] J. Lasch, U. Schonfelder, M. Walke, S. Zellmer, and D.Beckert, “Photooxydative damage of skin lipids in liposomes(hSCLLs)-interference of sterols with the lipid peroxidationchain,” International Journal of Pharmaceutics, vol. 162, no. 1-2, pp. 129–136, 1998.

[31] S. Moffett, D. A. Brown, and M. E. Linder, “Lipid-dependenttargeting of G proteins into rafts,” Journal of BiologicalChemistry, vol. 275, no. 3, pp. 2191–2198, 2000.

[32] R. Zeisig, R. Stahn, K. Wenzel, D. Behrens, and I. Fichtner,“Effect of sialyl Lewis X-glycoliposomes on the inhibition of E-selectin-mediated tumour cell adhesion in vitro,” Biochimica etBiophysica Acta, vol. 1660, no. 1, pp. 31–40, 2004.

[33] A. M. Sætern, M. Skar, A. Braaten, and M. Brandl,“Camptothecin-catalyzed phospholipid hydrolysis in lipo-somes,” International Journal of Pharmaceutics, vol. 288, no.1, pp. 73–80, 2005.

[34] M. Zareba, M. Niziolek, W. Korytowski, and A. W. Girotti,“Merocyanine 540-sensitized photokilling of leukemia cells:role of post-irradiation chain peroxidation of plasma mem-brane lipids as revealed by nitric oxide protection,” Biochimicaet Biophysica Acta, vol. 1722, no. 1, pp. 51–59, 2005.

[35] M. Ricci, L. Tuttobello, G. Luca, and C. Rossi, “Microbiologi-cal and densitometric TLC analyses for peptides in liposomes,”Journal of Pharmaceutical and Biomedical Analysis, vol. 25, no.5-6, pp. 903–912, 2001.

[36] A. Hollmann, L. Delfederico, G. Glikmann, G. De Antoni, L.Semorile, and E. A. Disalvo, “Characterization of liposomescoated with S-layer proteins from lactobacilli,” Biochimica etBiophysica Acta, vol. 1768, no. 3, pp. 393–400, 2007.

Liposomes - Hindawi Publishing Corporationdownloads.hindawi.com/journals/focusissues/820732.pdf ·...

Documents

Transcript of Liposomes - Hindawi Publishing Corporationdownloads.hindawi.com/journals/focusissues/820732.pdf ·...