Novel Microextraction Techniques for Bioanalysis of

Neurotransmitters and Biomarkers in Biological Fluids

Aziza El-Beqqali

Doctoral Thesis

Department of Environmental Science and Analytical Chemistry

Stockholm University, Stockholm, Sweden

2016

ii

Doctoral Thesis 2016 Aziza El-Beqqali. Novel Microextraction Techniques for Bioanalysis of Neurotransmitters and Biomarkers in Biological Fluids Department of Environmental Science and Analytical Chemistry (ACES) Geovetenskapens Hus, Stockholm University Svante Arrhenius väg 8, SE-11418 Stockholm, Sweden Copyright � Aziza El-Beqqali, Stockholm University 2016 Front cover: Photograph of MIPT (sol gel tablets) and illustration of molecularly imprinted polymer. ISBN 978-91-7649-581-0 ISBN 978-91-7649-582-7 Printed in Sweden by Universitetsservice US-AB, Stockholm 2016 Distributor: Department of Environmental Science and Analytical Chemistry

Stockholm University

iii

À mes parents, mon mari et mes enfants

iv

This thesis is a confirmation that, regardless of all the difficulties I have faced, there is always a motivation and inspiration to continue striving for my professional goals. I procure this encouraging spirit every day from Quran and the words from the prophet Mohammed.

حیم حمن الر بسم الله الر اقرأ باسم ربك الذي خلق

In the name of God, the Gracious, the Merciful.

Read: In the Name of your Lord who created.

I Guds namn, den Nåderike, den Barmhärtige. Läs: I din Herres namn, Han som har skapat.

QURAN 96 Clot (al-’Alaq)

قال رسول الله صلى الله علیھ و على آلھ و صحبھ و سلم من سلك طریقا یلتمس فیھ علما سھل الله لھ طریقا إلى الجنة

حدیث حسن رواه الترمذي

“For that one who treads a path in search of knowledge, Allah provides a road to Paradise”

"För den som beträder en väg för att söka kunskap,

underlättar Allah en väg till Paradiset"

The Prophet Mohammed (Al-Tirmidhi. Hasan)

v

Aziza El-Beqqali. Novel Microextraction Techniques for Bioanalysis of Neurotransmitters and Biomarkers in Biological Fluids

Department of Environmental Science and Analytical Chemistry Stockholm University, Stockholm, Sweden

Abstract

Sample preparation (sample pre-treatment) is the initial step and an essential part in bioanalysis procedure. The main role of sample preparation is to extract and transfer the analyte(s) of interest from a complex matrix to a purified media such as a pure solvent for analysis and quantification. Biological fluids are complex and contain, in addition to the target analyte(s), many different unwanted compounds from salts to proteins. Thus, the analysis of these samples requires an effective sample preparation method prior to the liquid chromatography-mass spectrometry (LC/MS) and gas chromatography-mass spectrometry (GC/MS) assays. The aim of the present work was to evaluate micro-extraction by packed sorbent (MEPS) and develop new sample preparation techniques for the extraction of neurotransmitters and biomarkers from biological samples. Moreover, two new sample preparation techniques were developed. The first developed technique is molecular imprinting on polysulfone membrane (MIPM), and the second one is molecularly imprinted polymer in tablet form (MIPT).

MEPS is a well-known sample preparation technique that can be used online with analytical instruments without any modifications. In this thesis, MEPS was used online with liquid chromatography-tandem mass spectrometry (LC/MS/MS) for the quantification of dopamine and serotonin in human urine samples (Study I) and with GC/MS for the analysis of methadone in humane urine (Study II). Polystyrene polymer and silica-C8 were used as sorbent in Study I and Study II, respectively. In both studies, small sample volumes (50 μL) were used and full method validation was performed. In both studies (I and II), MEPS enhanced the limit of detection (LOD) and reduced the extraction time compared to the previously published methods. In Study I, the mean accuracies of quality control (QC) samples of dopamine and serotonin were 99–101% while the precision values (RSD) were 6–11%. In Study II, the accuracy values of methadone were between 97 and 107% while the precisions (RSD) were between 11 and 15%.

MIP-sol-gel on polysulfone membrane (MIPM) was developed and used in combination with MEPS for the extraction of hippuric acid (HA) in plasma and urine samples (Study III). A good selectivity was obtained using plasma and urine samples. The precision of QC samples in plasma and urine samples were 2.2–4.8% and 1.1–6.7%, respectively. The method recovery was above 90%.

In Study IV, a new technique was developed using a tablet form of molecularly imprinted sol-gel (MIPT) for the extraction of methadone from human plasma samples. Methadone-d9 was selected as the template for accurate recovery, and 3-(propyl methacrylate) trimethoxysilane (3PMTMOS) was used as a precursor. The extraction recovery was higher than 80%. The LOD and LLOQ were 1.0 and 5.0 ng mL-1, respectively. The validation showed good selectivity, accuracy and precision.

vi

In Study I, III and IV, LC/MS/MS system was used while GC/MS was used in Study II for the separation and detection of target analytes.

There will always be a high demand for rapid, selective, reliable and sensitive techniques for sample preparation. It is convenient to use molecular dynamics simulations as a theoretical tool for the optimization of molecularly imprinted system. Suitable monomers and cross-linkers are crucial to synthesize the new membrane and tablet molecular imprinted polymer platforms for all kinds of molecules.

Keywords: Bioanalysis; Sample Preparation; Micro-Extraction; MEPS; Micro-Solid Phase Extraction; Molecularly Imprinted Polymer; Sol-Gel; Membrane; Tablet; LC/MS/MS; GC/MS; Neurotransmitter; Biomarker; Dopamine; Hippuric Acid; Methadone; Serotonin; Plasma; Urine

� 2016 Aziza El-Beqqali

vii

Sammanfattning

Bioanalys används för kvantitativ och kvalitativ mätning av droger och deras metaboliter i biologiska prover som blod, plasma och urin. Den används också i polislaboratorium för illegal droganalysis, inom dopingtest- och miljöanalyser. Bioanalys används också på kliniska laboratorier för att ta reda på vad som ligger bakom en sjukdom. Masspektrometer skiljer mellan olika molekyler utifrån deras masstal. Eftersom biologiska prover innehåller ett antal endogena föreningar måste provet extraheras från blod, urin eller vatten dvs. förbereda provet som kommer att injiceras i en masspektrometer. Denna process kallas provberedning. Denna forskning går ut på att hitta, utveckla och utvärdera innovativa provberedningsmetoder som ska vara billiga, snabba, effektiva, noggranna, selektiva och miljövänliga. Avhandlingen är indelad i evaluering av tre olika tekniker. Den första delen handlar om MEPS (Microextraction by Packed Sorbent) som är en teknik patenterad av AstraZeneca som Apparatur för Provberedning. MEPS är en miniatyrisering av konventionella Solid Phase Extraktion (SPE) och består av en spruta och en insats (BIN) som innehåller ca 2 mg Sorbent. MEPS kan användas 60-100 gånger. Huvuddragen om MEPS är kortare tid för förberedelser, samt färre steg mellan förberedelser och injicering i masspektrometern. Dessutom MEPS behöver mindre mängd buffert och lösningsmedel, olika sorters av sorbent samt den kan användas antigen manuellt eller ansluten till analytiskt system. I denna studie analyserades dopamin, metadon och serotonin i urin och plasma. Dessa är välkända neurotransmittorer. Den andra delen handlar om utvecklingen av en ny molekylärt avtryckt polysulfon membran (MIP-Membran eller MIPM) för att öka selektiviteten och hållbarheten hos det absorberande materialet (sorbent) för MEPS. I denna studie analyserades lungcancer biomarkör hippursyra i plasma. Den tredje delen omfattar skapande av en ny extraktion metod, MIP-Tablett (MIPT). MIPT är en kombination av MEPS teknik, molekylärt avtryckt polymer (MIP) som sorbent, och sol–gel koncept i form av tablett för kvantifiering av metadon, i plasma. Alla dessa metoder validerades i termer av linjäritet, selektivitet, noggrannhet och precision. Kriterierna för valideringsstudiers godkännande var väl i linje med de internationella kriterierna.

viii

Abbreviations

3PMTMOS 3-(Propylmethacrylate) Trimethoxysilane 5-HIAA 5-Hydroxyindoleacetic Acid 5-HT Serotonin; 5-Hydroxytryptamine; 3-(2-Aminoethyl)indol-5-ol) 5-HTP L-Tryptophan and 5-Hydroxytryptophan ACN Acetonitrile ADHD Attention Deficit Hyperactive Disorder API Atmospheric Pressure Ionization BIN Barrel Insert-Needle Assembly C2 Ethyl silica C8 Octyl silica C18 Octadecyl silica CE Capillary Electrophoresis CID Collision-Induced Dissociation CTC-PAL Auto sampler for Online or Offline Analysis DA Dopamine; 4-(2-Aminoethyl)benzene-1,2-diol DHBA 3,4-Dihydroxybenzylamine ECD Electrochemical Detection EDDP 2-Ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine EF Extraction Recovery EI Electron Impact Ionization EMA European Medicines Agency (European Union) ESI Electrospray Ionization FA Formic Acid FDA Food & Drug Administration (USA) GC Gas Chromatography HA Hippuric Acid; Benzoylaminoethanoic Acid or N-Benzoylglycine HPLC High Performance Liquid Chromatography HQC High Quality Control I.S. Internal Standard LC Liquid Chromatography LLE Liquid-Liquid Extraction LLOQ Lower Limit of Quantification LOD Limit of Detection LPME Liquid Phase Microextraction LQC Low Quality Control MEPS Microextraction by Packed Sorbent MeOH Methanol MIP Molecularly Imprinted Polymer MIPM Molecularly Imprinted Sol-Gel Membrane

ix

MIPT Molecularly Imprinted Sol-Gel Tablet MISPE Molecularly Imprinted Polymer for Solid-Phase Extraction MRM Multiple Reaction Monitoring MS Mass Spectrometry MS/MS Tandem Mass Spectrometry MT Methadone; RS-6-Dimethylamino-4,4-diphenylheptan-3-one MW Molecular Weight mp

+ precursor (parent) ions md

+ product (daughter) ions m/z Mass-to-Charge Ratio NIP Non-Molecularly Imprinted Polymer NIPM Non-Molecularly Imprinted Sol-Gel Polymer Membrane NIPT Non-Molecularly Imprinted Sol-Gel Polymer Tablet PCB Polychlorinated Biphenyls PE Polyethylene PP Protein Precipitation PS-DVB Polystyrene-Divinylbenzene Copolymer PSM Polysulfone Membrane QC Quality Control RAM Restricted Access Material RSD Relative Standard Deviation SBSE Stir-Bar Sorptive Extraction SEM Scanning Electron Micrographs SIM Selected Ion Monitoring Sol-Gel Solution (Sol) and Gelation (Gel) SPE Solid-Phase Extraction μ-SPE Micro-Solid Phase Extraction SPME Solid-Phase Microextraction TDM Therapeutic Drug Monitoring TEOS Tetraethyl Orthosilicate TFA Trifluoroacetic Acid tR Retention Time

x

List of Publications

This thesis is based on the following scientific papers:

I. Determination of dopamine and serotonin in human urine samples utilizing

microextraction online with liquid chromatography/electrospray tandem mass

spectrometry. Aziza El-Beqqali, Anders Kussak and Mohamed Abdel-Rehim. Journal

of Separation Science (2007) 30 (3), 421-424.

II. Quantitative analysis of methadone in human urine samples by microextraction in

packed syringe-gas chromatography-mass spectrometry (MEPS-GC-MS). Aziza El-

Beqqali and Mohamed Abdel-Rehim. Journal of Separation Science (2007) 30 (15),

2501-2505.

III. On-line detection of hippuric acid by microextraction with a molecularly-imprinted

polysulfone membrane sorbent and liquid chromatography-tandem spectrometry.

Mohammad Mahdi Moein, Aziza El-Beqqali, Mehran Javanbakht, Mohammad

Karimi, Behrouz Akbari-adergani and Mohamed Abdel-Rehim. Journal of

Chromatography A (2014) 1372, 55-62.

IV. Molecularly imprinted polymer-sol-gel tablet toward micro-solid phase extraction: I.

Determination of methadone in human plasma utilizing liquid chromatography-tandem

mass spectrometry. Aziza El-Beqqali and Mohamed Abdel-Rehim. Analytica Chimica

Acta (2016) 936, 116-122.

Papers I-IV: The author of this thesis was responsible for performance of the research work,

experimental work, all data validation and collaboration in writing the scientific papers.

Publications not copyrighted by the author have been reprinted with kind permission: Papers I & II: � 2007 Wiley and Sons Paper III: � 2014 Elsevier Paper IV: � 2016 Elsevier

xi

Related scientific papers not included in the thesis:

V. Determination of amphetamine in human urine samples utilizing molecularly imprinted

polymer-sol-gel tablet. Aziza El-Beqqali and Mohamed Abdel-Rehim. In preparation.

VI. Bioanalytical method development and validation: Critical concepts and strategies.

Mohammad Mahdi Moein, Aziza El-Beqqali and Mohamed Abdel-Rehim. J.

Chromatography B (2016). In Press.

VII. Preparation of monolithic molecularly imprinted polymer sol–gel packed tips for high-

throughput bioanalysis: Extraction and quantification of L-tyrosine in human plasma

and urine samples utilizing liquid chromatography and tandem mass spectrometry.

Mohammad Mahdi Moein, Aziza El-Beqqali, Abbi Abdel-Rehim, Amin Jeppsson-

Dadoun and Mohamed Abdel-Rehim. J. Chromatography B (2014) 967, 168-173.

VIII. Determination of amphetamine and methadone in human urine by microextraction by

packed sorbent coupled directly to mass spectrometry: An alternative for rapid clinical

and forensic analysis. Hana Vlčková, Aziza El-Beqqali, Lucie Nováková, Petr Solich

and Mohamed Abdel-Rehim. J. Separation Science (2014) 37(22), 3306-3313.

IX. Microextraction by packed sorbent for LC-MS/MS determination of drugs in whole

blood samples. Rana Said, Mohamed Kamel, Aziza El-Beqqali and Mohamed Abdel-

Rehim. J. Bioanalysis (2010) 2 (2), 197-205.

X. Microextraction in packed syringe/liquid chromatography/electrospray tandem mass

spectrometry for quantification of acebutolol and metoprolol in human plasma and urine

samples. Aziza E-Beqqali, Anders Kussak, Lars Blomberg and Mohamed Abdel-

Rehim. J. Liquid Chromatography and Related Technologies (2007) 30 (4), 575-586.

XI. Fast and sensitive environmental analysis utilizing microextraction in packed syringe

online with gas chromatography-mass spectrometry. Determination of polycyclic

aromatic hydrocarbons in water. Aziza El-Beqqali, Anders Kussak and Mohamed

Abdel-Rehim. J. Chromatography A (2006) 1114, 234-238.

XII. Use of carbon dioxide as collision gas in tandem mass spectrometry. Aziza El-Beqqali

and Mohamed Abdel-Rehim. Rapid Communications in Mass Spectrometry (2005) 19,

2099-2102.

xii

Table of Contents

Abstract ................................ ................................ ....... v

Sammanfattning ................................ .......................... vii

Abbreviations ................................ ............................ viii

List of Publications ................................ ....................... x

1 Introduction ................................ ............................. 1 1.1 Sample Preparation in Bioanalysis 2 1.2 Solid-Phase Extraction (SPE) 2 1.3 Solid-Phase Microextraction (SPME) 4 1.4 Microextraction by Packed Sorbent (MEPS) 5 1.5 Extraction Sorbents 6

Silica-Based Sorbents and Molecular Interactions 7 Molecularly Imprinted Polymers 8 Sol-Gel Concept 10

1.6 Neurotransmitters and Biomarkers 11 1.7 Separation and Detection Techniques 15

Liquid Chromatography (LC) and Gas Chromatography (GC) 15 Mass Spectrometry (MS) 17 Electrospray Ionization (ESI) 18 Tandem Mass Spectrometry (MS/MS) 19 Role of LC/MS/MS and GC/MS in Bioanalysis 20

2 Aim of this Thesis ................................ ................... 21

3 Results and Discussion ................................ ............ 22 3.1 MEPS Procedure and Validation (Papers I-II) 22

3.1.1 Sorbent Selection 22 3.1.2 Washing and Elution 22 3.1.3 Selectivity 23 3.1.4 Calibration Data 25 3.1.5 Accuracy and Precision 26 3.1.6 LOD, LOQ and Carry-Over 27 3.1.7 Comparison with Other Extraction Techniques 29

3.2 Development of a New Sorbent “MIPM” for MEPS (Paper III) 29 3.2.1 Sol-Gel MIP Sorbent Preparation Conditions 29 3.2.2 Preparation of Sol-Gel MIP Coated Membrane (MIPM) 30 3.2.3 MIPM Procedure and Validation 31 Conditioning, Washing and Elution Solutions 31 Effect of Added Salt, pH and Extraction Time 33 Method Validation 35 Comparison with Other Extraction Techniques 36

3.3 Development of a New Extraction Technique “MIPT” (Paper IV) 36 3.3.1 Preparation of MIP Sol-Gel Tablet (MIPT) 36 Adsorption Capacity, Extraction Time and Desorption Time 39 3.3.2 MIPT Development and Validation 39 Effect of Added Salt and pH 41 Selectivity of MIPT, NIPT and PE Tablets 42 Method Validation 43 Comparison with Other Extraction Techniques 45

xiii

4 Conclusions and Future Aspects ................................ 46

Acknowledgements ................................ ...................... 48

References ................................ ................................ . 50

Original Papers (I-IV) ................................ ................. 55

1

1 Introduction

Bioanalysis is a part of analytical chemistry that deals with analysis (identification and quantification) of analytes of interest in different biological specimens. Examples of the application of bioanalysis include drugs, metabolites, genes and biomarkers, including many other biological molecules that are present in biological matrices as for example: whole blood, plasma, serum, urine, saliva, sweat, as well as tissue and cerebral spinal fluid. Samples from both human and animal are used in bioanalysis. Bioanalysis is widely used in areas such as medical research and development projects, mostly in pharmaceutical and biotechnology industries for studying pharmacokinetics, toxicokinetics and bioequivalence of medical drugs. In addition, bioanalysis is important in therapeutic drug monitoring (TDM), forensic toxicology investigations, environmental matters, and finally in the determination of illicitly used drugs in sports’ anti-doping tests. The implementation of new sample preparation techniques in bioanalysis is of great interest and value to the bioanalytical community. In the present thesis, the focus will be on two important groups of bioanalytes: neurotransmitters and biomarkers. Typical neurotransmitters of relevance for this thesis are dopamine, serotonin and methadone, while hippuric acid is proposed as a biomarker for lung cancer [An et al., 2010], and all of them are determined in plasma and urine samples. More information about neurotransmitters and biomarkers is described further in Section (1.6). Apart from the discovery and qualification of neurotransmitters and biomarkers, there is a great demand and need to measure the concentration of these substances in biological fluids. The neurotransmitters have great effect on our bodies and our health in terms of thinking, feeling and moving. Biomarkers help in the diagnosis and/or treatment of several diseases. Good characterization, full validation and documentation of the applied bioanalytical methods are needed. Due to the low concentration levels of these analytes in biological fluids, the analytical estimation of these substances is challenging. Normally those analytes incorporate a variety of chemicals and metabolites that can have a negative impact on the accuracy, precision and reliability when determining these entities. A powerful bioanalytical method consists of two integral stages: 1) Sample preparation (treatment/clean-up/extraction) of the analyte from its bulk

biological matrix. 2) Separation, detection and quantification of the target analyte. Among modern separation-detection techniques, liquid chromatography (LC), gas chromatography (GC), liquid chromatography together with spectrometric tandem mass (LC/MS/MS) and gas chromatography coupled with mass spectrometry (GC/MS) are the benchmark for qualitative and quantitative bioanalytical studies as summarized in Section (1.7).

2

1.1 Sample Preparation in Bioanalysis

Efficient pre-treatment of samples is among the key procedures within bioanalysis requiring huge time of the work activity and operating costs to make the sample ready for introduction into a chromatographic system [Vaghela et al., 2016]. The aim of sample preparation is the removal of extraneous material from biological samples before analysis. Furthermore, sample preparation is required to enhance the selectivity, to isolate the analyte from the complex matrix and pre-concentrate the analyte if it is needed. Thus, a highly selective and specific sample preparation is a prerequisite for reasonable and sensitive analyses. Current progress in sample preparation techniques is oriented toward fully automation, creation of liquid-handling robots, miniaturization to decrease sample sizes. The development in sample preparation should be measured by means of improved sensitivity, selectivity, accuracy, precision, and data quality. Sample preparation advancement should be directed to increase efficiency, to give a high recuperation of the target analytes including less number of working steps, to increase assay throughput, as well as to reduce costs and environmental impact. In bioanalysis, normal sample-preparation systems include solid phase extraction (SPE), liquid-liquid extraction (LLE) as well as protein precipitation (PP) [Ashri and Abdel-Rehim, 2011]. LLE and PP have several limitations that are associated with the need of large sample volume, low recovery, poor selectivity, time-consuming and matrix effects in LC/MS/MS analyses. Recent approaches have improved their disadvantages [Kole et al., 2011]. However, the last decades have seen a rapid introduction of more effective sample preparation strategies including SPE and newly developed microextraction techniques that have contributed to master the restrictions of LLE and PP. In the next Sections (1.2) – (1.4), a summary of the development from the conventional SPE to novel sample preparation techniques has been provided with emphasis on solid-phase microextraction (SPME) and microextraction by packed sorbent (MEPS).

1.2 Solid-Phase Extraction (SPE)

Solid-phase extraction (SPE) is a highly reliable technique used for sample preparation allowing a good purification and concentration of trace elements in biological fluids. The method has been used for more than fifty years and applied broadly in many fields, e.g., in environmental, pharmaceutical, biological, organic chemistry and food analyses. SPE process is very simple, and consists of using a solid phase (sorbent) to extract the desired analytes from a sample (Figure 1). SPE can be performed off-line or coupled on-line directly to a chromatograph. The purpose of SPE is to obtain extracts

3

of the target analyte free from matrix interferences in a few process steps: conditioning the sorbent, applying the sample, washing and eluting. The most important parameters in SPE are first, the selection and the amount of the sorbent; second, the washing solution’s composition and volume which should be carefully selected to guarantee no leakage of the analytes; and third, the elution solution’s composition and volume. Nowadays, a great variety of highly reproducible and cleaner sorbents is available, leading to an increased acceptance and interest towards SPE. The benefits of SPE are, when compared to classical LLE, that it can easily be automated, it can save time and the interferences can be separated more efficiently from analytes. Furthermore, SPE demands less consumption of organic solvents and results in higher method selectivity. Since the sorbent and the analyte(s) participate in different intermolecular interactions, sample preparation task using SPE has been more selective, and the optimization of the technique can be performed by adjusting the conditions of chromatographic analysis. The interactions occurring in SPE can be of polar, hydrophobic or ionic character, while in LLE occurs only a partition equilibrium in the fluid phase. LLE utilize relatively large amounts of solvents with compromised purity of extracted analytes while SPE requires small amounts of solvents and produce better recovery. The main drawback of SPE is the cartridge variations stem from lot to lot. Moreover, during elution of highly polar analytes using strong bases and acids to guarantee a high recovery is not compatible with mass spectrometry. Further advantages and disadvantages of SPE during drug analysis are reviewed in the literature [de Zeeuw, 1997].

Figure 1. Schematic diagram of solid-phase extraction (SPE).

4

1.3 Solid-Phase Microextraction (SPME)

The studies on solid-phase microextraction (SPME) were initiated 30 years ago [Arthur and Pawliszyn, 1990; Pawliszyn, 1997]. This technique has been adopted as an option to the existing sample preparation methods, and today it has various fields of applications for the extraction of a diversity of analytes. SPME is a rapid sample preparation method for GC and LC (Figure 2) that does not require any solvent. The extraction technique SPME is based on the partitioning of the analyte between the matrix and the coated silica fiber. Several factors, including temperature, pH, stirring, and salt concentration may influence the equilibration time and the equilibrium constant [Pawliszyn, 1997]. With the adjustment of pH, the neutrality of the target analyte can be reached. Basic pH improves the recovery of basic analytes, while acidic analytes are better recovered at acidic pH. Recovery is also enhanced by adding salt, e.g., NaCl, which is a source of strong ionic effects. Many target drugs are polar and strongly attached to the sample matrix. In order to make these drugs disposed to thermal desorption from the SPME fiber to GC analysis a derivatization may be necessary. Several techniques for derivatization have been used and described in the literature [Pan and Pawliszyn, 1997]. Derivatization process comprises first of the addition of appropriate reagents to the sample matrix, and second the extraction or derivatization on the fiber. There is a large potential for derivatization within SPME that allows enhancing drug analysis but this potential has not been explored sufficiently. Although several studies on SPME using drugs and their metabolites in plasma or human urine have been published [Fritz, 1999], only few of them involve SPME optimization for plasma analysis. In most applications, SPME is linked on-line with GC/MS. On the other hand, LC/MS has been applied very little.

Figure 2. Depiction of solid-phase micro-extraction (SPME).

5

1.4 Microextraction by Packed Sorbent (MEPS)

A novel development in the field of sample preparation is the technique microextraction by packed sorbent (MEPS) originally designed and patented by AstraZeneca, Sweden [Abdel-Rehim, 2003; Abdel-Rehim, 2004]. MEPS resembles a miniaturized SPE packed bed device with different range of volumes. Therefore, its function is the same as SPE, i.e., cleaning-up interferences from the biological matrix, separation and enrichment of the sample with the target analytes. Typical MEPS setup (Figure 3) consists of two parts: a liquid-handling microliter syringe of (100-250 μL) and a MEPS barrel inserted-needle assembly (BIN) containing circa 1-2 mg of the solid phase material (packed sorbent) loaded as a plug or as a cartridge into the syringe [Abdel-Rehim, 2010; Abdel-Rehim, 2011]. Different types of packed sorbent material may be used in the MEPS cartridge depending on the conditions: normal phase, reverse phase, mixed mode or ion exchange. Most common materials are silica-C2, silica-C8, or silica-C18, carbon, polystyrene-divinylbenzene copolymer (PS-DVB), restricted access material (RAM), molecular imprinted polymers (MIP) and including coated sorbents to enhance the target analyte’s selectivity [Abdel-Rehim, 2010]. In the MEPS technique, the biological fluid sample is pulled up and down throughout the syringe with the help of an auto sampler pump. Passing through the solid support, the analytes are then adsorbed into the sorbent material. Together with the analytes some interfering materials may eventually also be adsorbed. Thus, a washing solution has to be applied to the solid phase material to wash it and thus to remove the interferences. After that, the elution of the analytes is performed with help of an organic solvent, e.g., methanol or a mobile phase that is injected directly into the LC or GC instrument.

Figure 3. Setup of microextraction by packed sorbent (MEPS).

6

Unlike conventional SPE that requires a separate column to upload the sorbent, in MEPS technique the solid phase is packed right away in the syringe. That differs MEPS from SPE where a separate robot to administer the sample into the solid phase is needed. This allows washing the sorbent effectively between different assays and thus reducing any chance of analyte(s) for carry-over. The small amount of packed sorbent material used in MEPS makes it easier to be cleaned, and hence several cycles can be applied. For instance, for plasma or urine samples, the sorbent may be used over 100 times while for aqueous samples it may be used as much as 400 times. However, commercial SPE columns can be used only once, even when the same sorbent material as in MEPS is used. In MEPS, small sample volumes can be used. In addition, instead of solvent volumes in the order of mL required to elute a standard SPE cartridge, MEPS technique can handle small elution volumes as 10-100 μL. This elution volume is such that it can easily be injected directly into a GC or LC without further steps. This gives the possibility to perform the process manually or to connect MEPS automatically into LC or GC with help of a CTC-PAL auto sampler for offline or online analysis without any modifications [Abdel-Rehim, 2004]. Another advantage of MEPS technique before conventional SPE or even LLE, is that with MEPS the volume of sorbent, the consumption of solvents and the sample preparation time are significantly reduced. In SPME, the sampling on the coated fiber may not be always used due to its vulnerability to the nature of different complex matrices, e.g., urine, blood or plasma. In that sense, MEPS is a more robust technique than SPME because MEPS is less sensitive as normal phase or reversed phase sorbents can be used which acquire high stability towards different solvents and temperatures and consequently can be used without problems. Regarding extraction recoveries, MEPS shows much higher values (60-90%) than SPME in which hardly 1 to 10% can be obtained. As little as 10 μL of sample volume can be treated with MEPS, while SPME favors higher sample volumes to improve the recovery.

1.5 Extraction Sorbents

Sorbents are characterized in terms of capacity and selectivity. Selectivity is the capability of a solid phase sorbent to adsorb the analyte present in a sample matrix in the presence of other components in the sample, thus not paying attention to the other undesirable constituents. Selectivity is influenced by the sample matrix’ composition, the analyte’s chemical structure, the sorbent’s properties, the washing solution and the eluent solution used. The best way to obtain a high selectivity is by selecting such functional groups of the analyte that are “missing“ in the sample matrix and other interferences.

7

Capacity is defined as the total amount of the analyte retained in a defined mass of sorbent according to the process optimal conditions. Hence, it is easy to estimate approximately the required amount of sorbent when we know the capacity and the quantity of analyte that we wish to extract. It has been reported that the sample capacity of a SPE cartridge is about 1-5% of the sorbent mass [Poole, 2003]. Silica-Based Sorbents and Molecular Interactions

Between functional groups of sorbent and analyte occur molecular interactions that can be polar, non-polar and ionic interactions. Most organic compounds possess a non-polar structure, so they can easily be adsorbed by non-polar sorbents where Van-der-Waals forces and non-polar interactions dominate. Typical sorbents with a pronounced degree of non-polar interactions are modified silicas in which a hydrocarbon spacer attaches the functional groups to the silica surface. Their selectivity is apparently low because of the ability of their functional groups, e.g., the alkyl substituents, to interact with practically all non-polar analytes. Therefore, a series of substances of different structure can be isolated. Surface-modified silicas, such as octyl-modified silica (C8) and octadecyl-modified silica (C18) are stable in almost all organic solvents and show pH stability in the range 2 – 8. Their use in a wider pH range is guaranteed because functional groups cleavage is time-depending, and the time contact between sorbent and solvent is normally very short. Silicas being solid and rigid materials, they never show disintegration or shrinking. Octadecyl-modified silicas (C18) are very non-polar sorbents allowing hydrophobic interactions with a large variety of organic substances. Thus, C18 is recommended for extraction of drugs, pharmaceuticals, antibiotics, amphetamines, antiepileptics, caffeine, barbiturates, fatty acids, preservatives, nicotine, vitamins and pesticides among many other non-polar compounds [Montesano et al., 2014]. On the other hand, octyl-modified silica (C8) is somewhat more polar than C18 due to the presence of shorter alkyl sequences that allow secondary interactions with polar compounds. C8 is applied for extraction of pesticides, polychlorinated biphenyls (PCB), etc. Among polar interactions that occur between many different sorbents and functional groups of the analytes, we can find hydrogen bonds, dipole-dipole and π-π interactions. Polar interactions are typical for sorbents like unmodified silica, NH2, CN and OH (diol) modified silicas. Commonly, polar sorbent is used effectively to adsorb polar substances from a non-polar environment, and as elution solution, a polar solvent is applied. Contrarily, non-polar sorbent is used to adsorb non-polar compounds from a polar environment. Low polarity solvents are then used for elution. Charged analytes interacting with a sorbent with functional groups of opposite charge undergo ion exchange interactions.

8

Positively charged groups may be found in amines (from primary to quaternary) and many inorganic cations such as Ca2+, Na+ and Mg2+. Negatively charged groups may exist in phosphates, anions of sulphonic and carboxylic acids, and similar anionic units. The adsorption via these ionic interactions is intensified in a low-ionic-strength matrix and a low-selectivity counter ion, for instance, Na+ or acetate. Solvents having a high ionic strength and high selectivity, e.g., citrate or Ca2+, are usually chosen as elution solution. Nevertheless, pure ionic interactions rarely occur. Alternatively, they take part in association with others as secondary interactions. Molecularly Imprinted Polymers

Molecular imprinting is a robust technique for the synthesis of materials with the capability of selecting effectively some specific chemical compounds, e.g., a target analyte. Polyakov in 1931 demonstrated the idea of the molecular imprinting [Alexander et al., 2006]. He demonstrated that molecular specificity could be imprinted into silica gel by pre-treating silicic acid with organic adsorbates before polycondensation. Under last years, many developments and applications of imprinting polymers in different areas have carried out by several research groups [Alexander et al., 2006; Chen et al., 2011]. As illustrated in Figure 4, in order to generate a molecularly imprinted polymer (MIP), a number of functional monomers are arranged around a template molecule via covalent and non-covalent interactions. Subsequent co-polymerization of the monomer molecules with a cross-linker leads to a trapping of the template molecule in a highly cross-linked amorphous polymeric structure. After extracting the trapped template, a porous polymer can be obtained with an imprinted arrangement of ligands or binding sites located at the surface. The resulting molecularly imprinted polymer (MIP) holds a predetermined shape, size and functional groups for a specific target analyte. Due to its high binding affinity and high ligand selectivity, MIP can recognize and rebind with the target analyte even in presence of some structurally related compounds [Khorrami and Rashidpur, 2012]. Therefore, MIP can be used as a feasible alternative to the more conventional sorbents for the extraction of analytes during sample preparation [Andersson et al., 2004; Widstrand, et al., 2006]. MIP has been implemented during the last twenty years as extracting media in different bioanalytical extraction techniques. Comprehensive studies include off-line SPE [Ali et al., 2010; Chen et al., 2013], on-line SPE [Moein et al., 2011; Moein, Javanbakht et al., 2014], monolithic columns [Huang et al., 2011; Javanbakht et al., 2012; Lin et al., 2013], SPME [Hu et al., 2012; Prasad et al., 2009; Prasad et al., 2013b; Qiu et al., 2010], stir-bar sorptive extraction (SBSE) [Prasad et al., 2013a; Zhu et al., 2006] and MEPS [Ashri et al., 2013; Daryanavard et al., 2013].

9

Figure 4. Synthesis of molecularly imprinted polymer (MIP).

10

Figure 5. Hydrolysis and condensation during sol-gel production of silica oligomers from

an alkoxy precursor, tetraethyl orthosilicate (TEOS). Reproduced with permission from Ref. [Owens et al., 2016] Copyright 2016, Elsevier.

Molecularly imprinted materials prepared with the “sol-gel” method have recently received special consideration as a sorbent in solid-phase extraction and other micro-extraction sorptive techniques [Moein, El-Beqqali, et al., 2014]. Sol-Gel Concept

Sol-gel is a technique for preparing advanced inorganic polymers and organic-inorganic hybrid materials under mild temperature and reaction conditions [Hench and West, 1990; Krasia-Christoforou, 2015]. Sol-gel process implies two different phases: a solution (sol) and a gelation (gel). A sol, in which a solid matter is dispersed through the bulk of a colloidal fluid, is exchanged into a gel consisting of a solid phase interconnected matrix, i.e., a continuous three-dimensional network [Brinker and Scherer, 1990]. Sol-gel process occurs via two main reactions: hydrolysis of precursor(s) and poly-condensation of sol-gel entities that are active in the bulk of the fluid. An example of sol-gel generation is illustrated in Figure 5. The combination of the MIP technique with the sol-gel methodology can be used to develop novel sorbent materials that possess unique properties: a molecular-level composition with a strong chemical bonding and a structure showing high specific surface area and controllable pore size. In addition to high selectivity and specificity, the resulting MIP sol-gel materials offer many advantages including high chemical and thermal stability, excellent mechanical properties giving a long shelf lifetime, fast mass transfer, fast adsorption/desorption kinetics resulting in higher extraction efficiency [Corriu, 2000; Shea and Loy, 2001]. Through sol-gel MIP technique, a wide range of solid materials (Figure 6) may be obtained at a very low cost and quickly. Example of manufactured devices by this technique include solid-phase adsorbents, catalysts, stationary phases and sensors.

11

Figure 6. Sol-gel technologies and applications. Reproduced with permission from Ref. [Owens et al., 2016] Copyright 2016, Elsevier.

1.6 Neurotransmitters and Biomarkers

Neurotransmitters are chemical compounds released by nerve cells, i.e., brain neurons, which allow sending signals to other neighboring nerve cells, muscle cells, or gland cells [Lodish et al., 2000]. In other words, they act as chemical messengers facilitating communication between the brain and the body organs. For example, neurotransmitters make it possible that the brain commands the heart to beat, the stomach to digest, the lungs to breathe and to perform other body functions. Neurotransmitters can act in an inhibitory or excitatory way. Inhibitory neurotransmitters relax the brain and are supportive in keeping the balance, while those that stimulate the brain are excitatory. Neurotransmitter imbalance can be caused by factors like stress, poor diet, drugs, alcohol, caffeine, neurotoxins and genetic predisposition. Certain types of neurotransmitters are synthesized and activated in their own systems of neurons and cells. Major neurotransmitter systems include the dopamine system, the serotonin system, among others. Other types of neurotransmitters are produced artificially by chemical synthesis, e.g., methadone. Dopamine (DA; 4-(2-aminoethyl) benzene-1,2-diol) is a monoamine neuro-transmitter that belongs to the catecholamine and phenethylamine families of psychoactive drugs, and its chemical structure consists of a benzene ring with one ethyl section ending in an amine group, and in addition there are two hydroxyl groups attached to the benzene ring (Table 1).

12

Table 1. Chemical structures of analytes and techniques used in this thesis.

Compound Formula [CAS No]

Chemical Structure Extraction Method

Analytical Method

Publication

Neurotransmitters

Dopamine C8H11NO2 [51-61-6]

MEPS

LC/MS-

MS

Paper I

Methadone C21H27NO [76-99-3]

MEPS

MIPT

GC/MS-

MS

LC/MS-MS

Paper II

Paper IV

Serotonin

C10H12N2O [50-67-9]

MEPS

LC/MS-

MS

Paper I

Biomarkers

Hippuric acid

C9H9NO3 [495-69-2]

MIPM

LC/MS-

MS

Paper III

13

DA is a neurotransmitter with a peculiar property that allows it to act as both excitatory and inhibitory. Several distinct pathways and cell groups make up a DA system. In the brain, the most important dopamine pathways are those involved in executive functions such as motivation, motor behavior, alertness, emotional and sexual arousal, orgasm, and refractory period. If the level of DA activity is high, then it is required less impetus to initiate a given conduct or action. DA plays a critical role as a reward-learning signal, a form of operant conditioning, which establishes a strong relationship provided by pleasurable experiences, such as playing video games or taking addictive drugs, followed by an increase in DA activity. A typical example of increased anticipatory behavior is the process of freeing DA to increase the appetite after we smell some food. Other lower-level functions of DA are those related with controlling the production of several hormones, including lactation, nausea and vomiting. Outside the brain, DA acts as a local chemical messenger in several sections of the peripheral nervous system. In addition, DA functions as vasodilator in blood vessels; it regulates insulin release in the pancreas; and in the immune system, it decreases the lymphocyte activity. DA also stimulates the elimination of sodium and delivery of urine in the kidneys; while the gastrointestinal system is favored by DA because it controls the movement of food through the intestines and protects intestinal mucosa from damage. Alterations in DA neurotransmission can lead to serious dysfunctions in the nervous system. For instance, Parkinson’s disease, which is distinguished by rigidity of the body, shaking, and difficulty initiating movement, is caused by damage and loss of DA-secreting neurons, while schizophrenia is due to an increased DA activity. Low levels of DA activity are related to disorders like ADHD, fibromyalgia, and other syndromes, e.g., burning mouth and restless legs. DA is found in many food plants (fruit pulp of bananas, potatoes, avocados, broccoli, Brussels sprouts, oranges, tomatoes, spinach and beans) [Kulma and Szopa, 2007] but when consumed they have no effect on the brain due to their difficulty to cross the blood-brain barrier. Fortunately, many plants such as velvet beans, fava beans and marine green algae contain L-DOPA, which is the metabolic precursor of DA and capable of crossing the blood-brain barrier [Ingle, 2003]. The determination of dopamine in urine is presented in Paper I. Methadone (MT; RS-6-dimethylamino-4,4-diphenylheptan-3-one) (Table 1) is a synthetic opioid with analgesic effect that is widely used to relief chronic pain. Racemic MT is found to be effective during maintenance treatment of opioid dependence and in the management of detoxification in heroin-addicted individuals [Bell, 2012; Isaza et al., 2014]. MT and its metabolite EDDP are often determined in biological fluids to respond to drug abuse detection programs (in which urine is tested), to attest intoxication cases (plasma or serum) or to support forensic investigations of sudden deceases, traffic or other criminal violations (whole blood). By reviewing MT usage history, the results

14

obtained from chronic users can be interpreted better as they can develop tolerance to doses that would incapacitate an opioid-naive individual. The determination of methadone in urine and human plasma is presented, respectively, in Papers II and IV. Serotonin (5-HT; 5-hydroxytryptamine or 3-(2-Aminoethyl) indol-5-ol)) (Table 1) is another monoamine neurotransmitter that along with DA has influence on various functions of the central nervous system and on the rest of the body. It is popularly associated to feelings of happiness and well-being. Serotonin is necessary for a stable mood and control of emotions, sleep cycle, pain control, regulation of body temperature and blood pressure, appetite satiety, smooth muscle contraction and appropriate digestion. It can also act as a growth factor, playing an important role in regeneration of liver and aiding healing the body. Serotonin is a natural regulator of anxiety, weight, and is speculated to be an effective antidepressant. It is known for some cognitive functions like concentration, memory, sensory perception and learning. Low levels of 5-HT can lead to decreased immune system function, obsessive-compulsive disorder, depression, social phobia, migraine, bipolar disorder, and emesis. Serotonin, along with DA, is also involved Alzheimer’s disease, schizophrenia, and ADHD [Moriarty et al., 2011]. Extensive studies performed over 30 years ago [Feldman and Lee, 1985], evaluating the 5-HT content of fruits and vegetables, determined that 5-HT is found in high concentrations in 80 types of food such as nuts of the walnut and hickory (Caraya) families, fruits as banana, plantain, pineapple, Kiwi fruit, plums, avocado, tomatoes, dates, mushrooms, and in many vegetables. Eating foods containing 5-HT does not increase brain serotonin, because it cannot pass through the blood–brain barrier, a similar mechanism to DA as explained above. However, its metabolic precursors L-tryptophan and 5-hydroxytryptophan (5-HTP) does traverse the blood–brain barrier. These substances are effective serotonergic agents than can be found as dietary supplements [Young, 2007]. 5-HT and its derivative 5-hydroxyindoleacetic acid (5-HIAA) are excreted in the urine. Therefore, they can be used as cancer biomarkers because their overbalance in urine is a sign of excess production due to the presence of certain tumors or cancers. The determination of serotonin in urine is presented in Paper I. Biomarker or biological marker is defined by regulatory agencies worldwide [EMA, 2011; US FDA, 2014] as a physiological, pathological, or anatomic property that is determined impartially and evaluated as evidence of a normal biological state or condition, a disease process, or a response to a treatment in humans and animals. It also refers to specific chemicals that living organisms shed into the environment, thus used as indicator of their existence or their presence in a particular place.

15

Recently, there has been a rapid development in the use of biomarkers in medical science, and consequently, an increased interest in finding and characterizing new biomarkers. Hippuric acid (HA; Benzoylaminoethanoic acid or N-Benzoylglycine) is a carboxylic acid (Table 1) excreted in the urine of horses and other herbivores. In some occupations, humans take many aromatic compounds such as benzoic acid and toluene that are converted in the interior of the body into hippuric acid by reaction with amino acids, e.g., glycine. Therefore, HA is considered a lung cancer biomarker in the biological monitoring of human exposure to toxic organic solvents as toluene [An et al., 2010; Bahrami et al., 2005]. The determination of hippuric acid in urine and plasma is presented in Paper III.

1.7 Separation and Detection Techniques

Among the existing separation and detection techniques, the combination of high performance liquid chromatography (HPLC or LC) and mass spectrometry (MS) as well as the combination of gas chromatography (GC) and MS, are the most effective and powerful techniques in bioanalysis with a high sensitivity and a good selectivity. They are useful in many applications like pharmaceutical analysis, environmental analysis and other related fields of chemistry, food and life sciences [Hansen and Pedersen-Bjergaard, 2015; Hsieh and Korfmacher, 2006; Inoue et al., 2011; Koeber et al., 2001; Smeraglia et al., 2002]. Liquid Chromatography (LC) and Gas Chromatography (GC)

Liquid chromatography (LC) is a fundamental technique for separation of a wide range of bioanalytes with different molecular weights, polarities, thermal stability or tendency to ionize in solution. In LC, the separation of the analytes is the result of partitioning between the mobile phase and the stationary phase. The mobile phase consists of solvents of different compositions, while particles (usually silica spheres) make up the stationary phase. LC can be run in various separation modes, for instance, normal-phase, reverse phase, ion-exchange, ion-pairing and size-exclusion. The most common technique in LC is the reverse-phase, in which the two phases, silica particles with bonded hydrophobic (Silica-C4 to C18) stationary phase and a hydrophilic mobile phase, have a strong influence on the separation of analytes on the LC column. A setup of LC system is show in Figure 7. A sample is introduced to the column using an injector while a flow of mobile phase is pumped through the column. The analytes follow the direction of the mobile phase flow through the column at different velocities and elute from the column one at a time. After leaving the column, the analytes are introduced into the MS detector and measured one by one.

16

Figure 7. Setup of a liquid chromatography (LC) apparatus.

The mean velocity at which the analytes move through the column relies on the hydrophilicity and the hydrophobicity of the analyte. Analytes with higher affinity to the stationary phase will migrate slowly in the column, and thus will show longer retention time (tR) than those analytes with lower affinity. Gas chromatography (GC) is a separation and detection technique applied mainly to volatile chemicals or compounds that can be derivatized into volatile substances. Figure 8 illustrates a setup of GC/MS system. In GC, a small sample (in the order of μL) is injected into a heated gaseous mobile phase (carrier gas) where the sample component evaporates instantaneously. The vapor sample is then transported with the carrier gas into the packed or capillary separation column (GC column).

Figure 8. Setup of gas chromatography-mass spectrometry (GC/MS) apparatus.

17

The different sample elements interact with the stationary phase as they pass through the GC column, and different components are physically separated based on differences in their volatility and ability to partition between the mobile phase and the stationary phase. Consequently, constituents with low volatility and strong interaction with the stationary phase will be effectively retained in the GC column, and move gradually through the column. After leaving the GC column, the sample elements are introduced into a MS detector where each individual sample element is detected and quantified. In GC, analytes should be volatile or semi-volatile to be carried with the gas to the detector. Temperature plays an important role because both volatility and vapor pressure of the sample constituents raises with temperature. For very volatile compounds, lower temperatures are applied; while for poor volatile higher temperatures are required. Owing to the volatility criterion, bioanalytical GC may be restricted to small-molecule drug substances under approximately 500 molecular weights. Applicability of GC is also limited by the analyte thermal stability and polarity. Thermally fragile molecules are not suitable for GC if they cannot withstand elevated temperatures up to 250–300 oC. The use of GC to separate polar and large-molecule compounds is troublesome because these substances tend to demonstrate severe peak tailing. Therefore, compounds with high degree of polarity and with high molecular weights are preferably analyzed by LC, while non-polar and small-size molecules can ideally be analyzed by GC. Mass Spectrometry (MS)

Mass spectrometry (MS) plays an important role as a detection techniques used today in molecular analysis [De Hoffman and Stroobant, 2007]. MS can generate three-dimensional data; i.e., data representing signal strength as a function of time and mass spectral data for each point of time. MS has the potential to yield valuable information on the molecular weight, molecular structure, quantity, identity and purity of a bioanalyte. MS has a high sensitivity and specificity that provide good results in both qualitative and quantitative bioanalysis. The MS process consists of an ionization of the bioanalyte molecules followed by a separation, fragmentation and identification of the resulting ions according to their mass-to-charge (m/z) ratios. The generation of ions occurs in the ionization source of the mass spectrometer prior to their characterization in the mass analyzer. Modern ion sources are based on the atmospheric pressure ionization (API) technique [Takats et al., 2004; Cody et al., 2005] in which the bioanalyte molecules are ionized at atmospheric pressure and then the product ions are separated from neutral molecules either mechanically or electrostatically. Common API techniques include electrospray ionization, chemical ionization and photoionization.

18

Electrospray Ionization (ESI)

Electrospray ionization (ESI) relies on the generation of bioanalyte ions in liquid-phase (solution) before they reach the mass spectrometer analyzer. Several books and extended reviews have been published on ESI principles and bioanalytical applications [Cole, 1997; Cech and Enke, 2001; De la Mora et al., 2000; Rohner et al., 2004]. As illustrated in Figure 9, a slow flux of the ESI solution (effluent) passing through a capillary tube is sprayed (nebulized) by nitrogen gas at atmospheric pressure in the presence of a strong electrostatic field (2-5 kV) and heated drying gas (N2) at a specific temperature (200-500 oC) forming highly charged droplets. As the solvent evaporates from the droplets, the charge density in the droplet surface increases continuously until ions with like charges are ejected (desorbed) from the droplets [Kebarle and Tang, 1993]. By this time, solvent molecules having lower molecular mass than the sample ions are quickly diffused away. Then, the remaining bioanalyte ions are focused into the mass analyzer through some vacuum pumping stages. Effective generation of bioanalyte ions in solution is very important to achieving the desired electrospray and good MS results. During LC/MS separations, and in ESI mode, it is important to take in consideration a few parameters related to sample preparation and solution chemistry. For instance, the use of adequate bioanalyte concentration in the ESI solution, the adjustment of solvent pH accordant to the polarity of ions desired and the pH of the sample, and the selection of solvents with low heat of vaporization and low surface tension to enhance ion desorption.

Figure 9. Process of ion generation during electrospray ionization (ESI).

19

Furthermore, volatile additives and acid-base buffering species are preferred to prevent partial ionization, signal suppression and contamination or corrosion in the MS ion source [De Hoffman and Stroobant, 2007]. Volatile buffers like acetate and formate are frequently used thanks to compatibility with MS. Acetate and formate are commonly useful for positive-ion mode detection. The use of triflouroacetic acid (TFA), which is an efficient ion pairing agent, can however strongly suppress signal with ESI mode [Temesi and Law, 1999; Apffel et al., 1995]. ESI can be applied to a wide spectrum of small and large biomolecules [Cole, 1997; Kushnir et al., 2002]. Biological macromolecules can acquire multiple charges, and for this reason, ESI can be used successfully to analyze compounds with high molecular weights such as proteins. In addition to molecular weight information obtained from MS, complementary structural elucidation is very valuable. Structural information is acquired by fragmentation of ion bioanalytes during a so-called collision-induced dissociation (CID) process.

Tandem Mass Spectrometry (MS/MS)

Tandem mass spectrometry (MS/MS) is one of the most powerful ways to obtain structural information and it consists of a multiple-stage MS; commonly two mass analyzers are connected. The basic approach of MS/MS (Figure 10) is the measurement of m/z of bioanalyte ions before and after a CID reaction within the MS. Precursor ions with a specific m/z-value are produced in the ionization source and isolated by the first mass analyzer (MS1). Then, these precursor (parent) ions (mp+) undergo a fragmentation or dissociation in the collision cell into the product (daughter) ions (md+) and residual neutral fragments (mn) according to the following acceleration-collision reaction: mp+ → md+ + mn

Figure 10. Principle of tandem mass spectrometry (MS/MS).

20

md+ can be sampled and analyzed by the second mass spectrometer (MS2) but the latter species, mn, cannot be detected. However, from the mass difference between mp+ and md+, the mass of mn can be calculated. Interesting in MS/MS is that already in the first mass analyzer non-precursor ions can be discarded. Accordingly, previous steps like sample preparation and chromatographic separation become much less critical. The major disadvantage of MS/MS is the high cost of the equipment. Both LC/MS/MS and GC/MS employ a mass spectrometer as a detector, but they differ in the way the ionization is performed. In LC/MS/MS, ionization takes place outside the mass spectrometer at atmospheric pressure, while in GC/MS it occurs under vacuum inside the mass spectrometer. Role of LC/MS/MS and GC/MS in Bioanalysis

Several analytical techniques have been developed for the quantification of neurotransmitters and biomarkers of interest in this thesis. For the detection of dopamine and serotonin in plasma or urine, a few researchers have used either LC/MS/MS [De Jong et al. 2010; Kushnir et al., 2002; Monaghan et al., 2009; Moriarty et al., 2011], LC with electrochemical detection (LC/ECD) [Vuorensola and Karjalainen, 2003] or capillary electrophoresis in combination with mass spectrometry (CE/MS) [Peterson et al., 2004; Vuorensola and Karjalainen, 2003]. However, they often confronted difficulties during sample preparation that, consequently, resulted in time-consuming assays, low extraction recoveries and loss of sensitivity. In some cases [Kushnir et al., 2002; Vuorensola and Karjalainen, 2003], the sample preparation method was designed specifically for catecholamines (e.g., dopamine) and may not be effective for serotonin. Determination of methadone enantiomers has been performed using LC/MS [He et al., 2005; Kintz et al., 1997; Quintela et al., 2006], GC/MS [Myung et al., 1999; Sporkert and Pragst, 2000; Wilkins et al., 1996] and CE [Cherkaoui et al., 2001; Rudaz et al., 2003]. Besides improved sensitivity and selectivity, LC/MS/MS gives specificity, speed and cost-effectivity. However, this technique has also limitations related to matrix effect, decrease of sensitivity of the analyte of interest in the processed matrix and compromised selectivity [Smeraglia et al., 2002]. The matrix effect (ion suppression) occurs when interferences from biological matrix, for example salts and phospholipids, are co-eluting with the analyte from LC and disturb the ionization process in the ESI interface resulting in a reduced or enhanced analyte signal. To reduce or eliminate matrix effect some modifications related to instrumentation and methodological issues include modified ionization, ionization switching, extraction mechanism modification and gradient HPLC techniques, which have demonstrated significantly improved robustness of complex bioanalytical methods to avoid matrix-related issues.

21

2 Aim of this Thesis

The aim of the present work can be divided into three parts. The first part deals with the development and validation of the Microextraction by Packed Sorbent (MEPS) technique as a new sample preparation method for online quantification of drugs and metabolites in biological samples. The target analytes used in this study are dopamine, methadone and serotonin, which are well-known as neurotransmitters. Human urine is used as biological fluid. The analytical screening and detection is carried out by connecting MEPS online directly to high performance liquid chromatography-tandem mass spectrometry (LC/MS/MS) or to gas chromatography-mass spectrometry (GC/MS). MEPS results and comparison with previously published data using other extraction techniques will be presented and discussed in Section 3.1 of this thesis (Paper I and Paper II). The second part is about the evaluation and application of a new molecularly imprinted polysulfone membrane (MIPM) to increase selectivity and durability of the sorbent material for MEPS. High performance liquid chromatography-tandem mass spectrometry (LC/MS/MS) is used to analyze hippuric acid (lung cancer biomarker) in human plasma samples. Section 3.1.7 of this thesis (Paper III) summarizes the evaluation of different factors in the extraction process. The third part embraces the development of a novel micro-solid phase extraction method, molecularly imprinted tablet (MIPT) by combination of MEPS technique, Molecularly Imprinted Polymer (MIP) as sorbent, and Sol-Gel as the sorbent synthesis approach. Online quantification of neurotransmitters (methadone) in plasma is accomplished by high performance liquid chromatography-tandem mass spectrometry (LC/MS/MS). The MIPT results including the evaluation and optimization of this novel technique will be presented and discussed in Section 3.3 (Paper IV).

22

3 Results and Discussion

3.1 MEPS Procedure and Validation (Papers I-II)

During sample preparation using MEPS technique (Figure 11), 50-250 μL of the biological fluid sample (human urine) were pumped up through the syringe by means of an auto sampler (Phase 1) and then the analytes were retained in the sorption bed. In Phase 2, the interfering materials present in the solid phase were removed with 50-100 μL of washing solution (pure water or water/methanol mixture). Afterwards, the elution of the analytes (Phase 3) was evaluated using solutions with different compositions of water, methanol, ammonium hydroxide and formic acid, making the sample ready to introduce it straight into a GC or LC instrument's injector (Phase 4). In MEPS, the recovery was defined as the relation of the mean peak area resulting from three trials using spiked urine samples to the peak area obtained from pure standard solution at two levels of concentration, i.e., 100 and 1000 ng mL-1. 3.1.1 Sorbent Selection

For MEPS sorbent, the recovery was measured for different solid–phase silica sorbent materials (1 mg of C2, C8 or C18). The method sensitivity was enhanced by allowing the auto sampler (set to a flow rate of 20 μL s-1) to pump the urine sample up and down through the syringe five times. During the elution step, a warmed syringe (at 40°C) was used, and the elution solution consisted of pure methanol. A good recovery (around 60%) was obtained for silica–based C8 resulting in an extraction time of 2 min. After each trial, the sorbent was rinsed five times with methanol (100 μL) in order to avoid carry-overs. 3.1.2 Washing and Elution

Various washing and elution solutions were used to study their effect on the recovery of the analytes. In general, the presence of methanol in the washing solution instead of pure water contributed to a high level of cleanliness (no interferences detected) in the extract. However, the use of large amount of methanol led to a weaker recovery and to an increased leakage. The optimal composition of water/methanol in the washing solution was 90:10 (v/v) resulting in the highest recovery of the analytes, the lowest amount of leakage and a very clean extract. To evaluate the elution efficiency, different elution solutions were studied. The extract of 50 μL of the urine sample in the packed sorbent was washed with 100 μL of water/methanol mixture (90:10, v/v), then different compositions of water, methanol and ammonium hydroxide were tested during the elution step, and its

23

Figure 11. Steps involved in the sample preparation by MEPS.

recovery was compared with the results obtained for pure standard solution with a concentration of 100 ng mL-1. The increment of methanol content in the elution solution contributed to an increase in the elution efficiency. Admissible recovery of about 50% resulted when an elution solution with a composition of water/methanol 5:95 (v/v) containing 0.2% ammonium hydroxide was employed. 3.1.3 Selectivity

To define selectivity of MEPS, a non-interference with the endogenous material present in the areas of interest was considered. When blank human urine samples were analyzed with LC/MS/MS, no endogenous interference peaks were detected at or around the time when analytes or the IS elute. Typical mass chromatograms obtained for blank urine, internal standard, as well as for urine samples spiked with dopamine or serotonin (300 μg L-1), and also for urine spiked with methadone at low quality control (QC) sample (62 ng mL-1) are presented in Figure 12.

25

3.1.4 Calibration Data

The peak area ratio of each analyte (dopamine, methadone and serotonin) and the internal standard (analyte/I.S.) was determined and a standard calibration curve with non–zero concentration was drawn. The standard calibration curve was depicted by the quadratic regression:

c 2 ��� bxaxy

where y defines the peak/area ratio, x the concentration, a the curvature, b the slope, and c the intercept. The weight of the quadratic calibration curve was 1/x. The precision and accuracy were calculated for the quality control (QC) samples at 3 different trials. Table 2. Back calculated results for the calibration of urine samples.

Analyte Concentration Mean Concentration (n=6)

Mean Accuracy (%)

RSD (%)

Dopamine (μg L-1) 50 50 100 0.0 100 104 104 2.9 500 478 96 2.1 1000 1091 109 4.2 2000 2019 101 5.5 4000 3964 99 11.0 Methadone (ng mL-1) 2.3 3 120 1.0 23 25 107 8.4 78 82 106 7.2 155 153 99 11.0 310 300 97 8.9 465 441 95 9.7 1550 1572 101 3.4 3100 3079 99 1.7 Serotonin (μg L-1) 50 51 102 11.0 100 93 93 7.7 500 494 99 8.5 1000 1055 106 7.8 2000 1977 99 5.4 4000 4011 100 5.3

26

Optimized conditions were used for method validation. The calibration curves consisted of six-eight concentration levels of human urine spiked with the analytes in the concentration interval 2.3–3100 ng (mL)-1 for methadone, and 50–4000 μgL-1 for dopamine and serotonin. Simultaneously, several blank samples were tested. In all assays, the correlation coefficients of the regression r2> 0.99 were obtained. In Table 2, we can observe that the back calculated concentrations of the calibration points are in good agreement with the nominal concentrations. For n=6, the mean accuracy values were in the range between –5% and +20% of the nominal concentrations for methadone, while for dopamine and serotonin the values ranged from –1 to +9%. 3.1.5 Accuracy and Precision

Accuracy is a representation of systematic error that can be determined as the percentage of deviation between the measured and the nominal value at different concentration levels.

Accuracy (%) = × 100

Precision (RSD%) is a measure of the random error and is determined as the percentage of standard deviation of the observed values related to the mean value observed at the different levels of concentration.

RSD (%) = × 100

The accuracy and the precision of the method validation for quality control (QC) samples were calculated according to internationally accepted guidelines [EU EMA, 2011; US FDA, 2014; Shah et al., 2000]. Table 3 summarizes the accuracy and precision data. Three levels of concentrations: 30, 300 and 1500 μg L-1 were used to determine the precision and the accuracy, intra- and inter day experiments for dopamine and serotonin. At all calibration curves, two different levels for QC samples were used. For n=12, the intra-assay precisions (RSD) at three different concentrations for QC samples were about 3.2–6.4% for dopamine and 2.7–9.6% for serotonin in urine samples. For n=24, the inter-assay precisions (RSD) (n= 24) were 6–9% and 6.1–11% for dopamine and serotonin, respectively. For both dopamine and serotonin, the accuracy for intra-assay and inter-assay varied between 99 and 108%. For methadone, three levels of concentrations: 62, 186 and 1860 ng mL-1 were used to calculate precision and accuracy for intra– and inter-day assays.

27

Table 3. Accuracy and precision (intra- and inter day assay) of QC samples in urine.

Intra-assay Inter-assay (n=3) Analyte Conc. Mean

Conc. Mean

Accuracy RSD Mean

Conc. Mean

Accuracy RSD

(%) (%) (%) (%)

Dopamine (μg L-1) (n=12) (n=24)

30 33 108 6.0 32 106 9.0 300 301 100 6.4 302 101 7.7 1500 1481 99 3.2 1493 100 6.0

Methadone (ng mL-1) (n=6) (n=18)

62 67 109 11.0 67 107 11.0 186 185 99 12.0 186 100 11.0 1860 1652 89 14.0 1797 97 15.0

Serotonin (μg L-1) (n=12) (n=24) 30 32 107 9.6 32 107 8.8 300 296 99 8.8 296 99 11.0 1500 1514 101 2.7 1503 100 6.1

Six QC samples were prepared in human urine at each level. For n=6, the intra-assay precisions (RSD) at three different concentrations for QC samples were in the range 11–14%. For n=18, the inter-assay precisions (RSD) were between 11 and 15%. The accuracy was 89-109% for intra-assay (n=6), while for inter-assay (n=18), it varied between 97 and 107%. 3.1.6 LOD, LOQ and Carry-Over

The determination method for the limit of detection (LOD) and the limit of quantification (LOQ) was set in the line of FDA-guidelines [US FDA, 2014]. The LOD (S/N ≥ 3) was 1.0 μg L-1 for dopamine and serotonin, while for methadone it was 0.4 ng mL-1. Figure 3 of Paper II shows a chromatogram of a urine sample at the LOD. The LOQ for dopamine and serotonin was set to 50 μg L-1; and for methadone, it was set to 2.3 ng mL-1. The carry-over effect was investigated by injecting solution after extraction using the highest standard concentration followed by injection of blank solution. The memory effect was minimized by washing the MEPS five times with methanol and five times with water after every injection. The carry-over was lower than 0.2% in all the cases.

28

T

able

4.

Com

pari

son

of a

ccur

acy,

pre

cisi

on, L

OD

and

ext

ract

ion

time

betw

een

ME

PS a

nd o

ther

ext

ract

ion

tech

niqu

es.

Ana

lyte

Te

chni

que

Con

cent

ratio

n

Acc

urac

y (%

) R

SD

(%)

LOD

Extra

ctio

n Ti

me

(m

in)

Ref

eren

ces

Dop

amin

e

(μg

L-1)

(μg

L-1)

M

EPS-

LC/M

S/M

S 50

10

0 0.

0 1.

0 1

Pape

r I

LL

E-LC

/MS/

MS

66

97.7

8.

6 2.

5 12

[K

ushn

ir, 2

002]

Met

hado

ne

(n

g m

L-1)

(ng

mL-1

)

M

EPS-

GC

/MS

62

109

11.0

0.

4 2

Pape

r II

LL

E-LC

/UV

50

10

9 12

.0

2.3

15

[He

et a

l., 2

005]

SP

E (C

18)-

LC/U

V

20

108

6.7

1.0

26

[Qui

ntel

a et

al.,

200

6]

SP

ME-

GC

/NPD

50

10

8 7.

3 50

.0

30

[Myu

ng e

t al.,

199

9]

Sero

toni

n

(μg

L-1)

(μg

L-1)

M

EPS-

LC/M

S/M

S 30

10

7 9.

6 0.

1 1

H

PLC

15

8.4

104

2.1

1.76

15

[X

iao

et a

l., 1

998]

29

3.1.7 Comparison with Other Extraction Techniques

The results of the microextraction technique (MEPS) studies presented in Paper I and Paper II were compared with the results published in the literature for other sample preparation techniques (LLE, SPE, SPME) for bioanalysis of dopamine, methadone and serotonin in urine or plasma [Kushnir, 2002; He et al., 2005; Quintela et al., 2006; Myung et al., 1999; Xiao et al., 1998]. Table 4 shows that MEPS technique is as accurate and precise as the other methods. However, MEPS contributed to an improvement in the limit of detection (LOD) by two- and up to 125-fold. A significant extraction time reduction (7-15 times) was also obtained. MEPS reduced the consuming of solvent significantly. In addition, small sample volumes could be handled.

3.2 Development of a New Sorbent “MIPM” for MEPS (Paper III)

3.2.1 Sol-Gel MIP Sorbent Preparation Conditions

Figure 13. Modification process for MIP Sol-Gel membrane (MIPM) (Paper III) and MIP Sol-Gel Tablet (MIPT) (Paper IV).

30

In the course of preparation of the sol-gel MIP, and to achieve the most favorable extraction conditions, various key parameters have to be taken into account. The first one is the selection of an appropriate ratio of concentrations between the template (HA) and the precursor (3PMTMOS) [Schweitz et al., 1997], as well as the effect of different quantities of catalyst (TFA) during the sol-gel preparation. A proper ratio is necessary in order to build the optimum cavity density in the MIPM system after the template is removed. The results of these investigations are listed in Table 1 of Paper III. For 0.1 mmol of the template, the optimal precursor: TFA ratio was 2:1 which resulted in a maximum recovery of 96%. Another important factor is the selection of the solution “residence-time” needed to immerse the polysulfone membrane (PSM) into the imprinting sol-gel solution to generate a thin layer on the membrane surface. The optimal “pass-time” obtained was 5 times with a duration of 10 min as shown in Paper III. 3.2.2 Preparation of Sol-Gel MIP Coated Membrane (MIPM)

Figure 13 shows the preparation process of the modified PSM (MIPM) with help of an imprinting sol-gel solution. First, to remove any contamination, the PSM was washed with water and acetone. 500 μL of precursor (3PMTMOS), 100 μL of HA (0.1 mmol L-1) as template molecule and 400 μL of solvent (ACN) were mixed under sonication for 30 min to prepare the sol-gel MIP solution with the maximum binding probability. Then 400 μL TFA (100%) was slowly added in four steps (100 μL each step) and sonicated for another 10 min. Subsequently, drop by drop, 100 μL distilled water were added to start the hydrolysis process and to keep the solution in this state for 30 min. Finally, to form a sol-gel imprinted layer on both sides of the PSM, it was immersed in the sol-gel MIP solution for 10 min, and then placed in a desiccator for 10 min. The last two steps were repeated 5 times. Additional 24 h in the desiccator were needed for aging the MIPM. For a complete polycondensation, the MIPM was subjected to a temperature gradient from 50 to 150ºC for 5 hours. After that, the MIPM was washed with a mixture of methanol/acetic acid (80:20) for 2 hours to remove the trapped template and create a porous selective surface. A non-MIP membrane (NIPM) was prepared in the same way as the MIPM, but without the template. The modified polysulfone membrane was used to support a sol-gel molecularly imprinted sorbent in MEPS technique as represented in Figure 14.

31

Figure 14. Schematic picture of Sol-Gel MIP membrane (MIPM).

In Figure 15, scanning electron micrographs (SEM) images of the MIP membrane are shown. The modified thin layer of imprinted sol–gel on the membrane surface is clearly observed (A) as well as the surface (B) and cross-section (C). The surface porosity after the addition of the imprinted sol–gel and aging can also be noticed. As it can be perceived, a layer of imprinted sol–gel was created on the backbone PSM surface. 3.2.3 MIPM Procedure and Validation

The novel technique MIPM-MEPS was evaluated for extraction of hippuric acid from human plasma and urine samples by on-line connection with LC/MS/MS. The influence of different parameters on extraction efficiency such as solvent solutions, adsorption capacity, added salt, pH, extraction time were investigated. Conditioning, Washing and Elution Solutions

The selection of suitable solvent for washing and elution is a crucial issue. Thus, the effects of various conditioning, washing and eluting solutions on recovery were studied. The recovery was measured as the response from a spiked plasma or urine sample expressed as the target peak area. The results showed the best recovery was obtained when 100 μL ultra-pure water and ultra-pure water/FA (0.1%) (90:10) were used as the conditioning and washing solutions, respectively (Figure 16 A and B).

33

Figure 16. Test of conditioning (A), washing (B) and elution (C) solutions.

In addition, the effects of various elution solutions on extraction efficiency (EE) were also investigated. Solutions containing MeOH, ACN, water, acetic acid, formic acid and ammonium hydroxide were investigated as elution solutions. The elution efficiency was measured and compared with the pure analyte standard. The elution efficiency increased as the MeOH content in the eluent increased. A polar solvent such as MeOH can facilitate desorption process of analyte from sorbent. When 100 μL solution of ACN/MeOH (50:50) was used as the elution solution (Figure 16 C), the extraction efficiency was more than 91%. Effect of Added Salt, pH and Extraction Time

The influence of added salt on recovery of HA by MIPM extraction was examined. First, a range of NaCl concentrations (1–100 mmolL−1) were added to the solutions and extractions using MIPM were carried out. The results show that at a specific concentration of 20 mmolL−1, NaCl salt can help to increase the recovery of HA in plasma and urine samples and then the recovery decreased when the salt concentration increased. The results are shown in Figure 17 A. The addition of salt has commonly been used in microextraction techniques to improve the extraction efficiency. The addition of salt can increase the analyte partitioning into the

34

organic/solid phase by salting-out effect [Han and Row, 2012; Pawliszyn, 1997; Psillakis and Kalogerakis, 2003]. Moreover, the effect of salt addition may lead from improvement to deterioration of the extraction efficiency, depending on the nature of analytes and the sample matrix. In addition, the presence of high concentration of salt in sample solution may change the physical properties of the analyte or the sorbent decreasing the diffusion rate of the analyte into the organic/sorbent phase. The pH is a very important issue for obtaining stable and repeatable data. Because of that, the effect of pH on the rebinding efficiency of HA was studied by varying the solution pH from 2.0 to 12.0 (Figure 17 B). Several experiments were performed with 100 μL of solutions containing 100 nmol L−1 of HA within a wide pH range (0.1 mmolL−1 of acetic acid and ammonium hydroxide were used for preparing the exact desired pH). It was observed that HA underwent complete rebinding/elution from acidic to natural pH (i.e., 2–7). However, due to better selectivity and high recovery for the extraction of HA at pH 2.0–3.0 with MIPM compared to non-molecularly imprinted sol-gel membrane (NIPM) and PSM. HA has pKa value of 3.7 and at pH 2.0 the most of HA molecules will be in uncharged form and then the partitioning of the HA will be shifted toward the MIP sorbent giving higher recovery.

Figure 17. Effect of salt (A), pH (B) and extraction time (C) on the extraction of HA.

35

In NIPM (non-specific interaction), HA molecular form (charged/uncharged) does not influence the recovery and therefore the pH has no impact on the recovery. A remarkable property of imprinted sol–gel materials is the capacity and high porosity. Thus, the maximum analyte binding by the MIPM depends on equilibrium between analyte and sorbent. To study adsorption capacity, a solution of HA was added to the MISM in from 1 to 6 min, and the maximum analyte absorption by the sorbent was reached after 4 min (Figure 17 C). Similar experiments were made for both NIPM and untreated PSM. After 1 min, both NIPM and PSM were saturated. The MISM needed more time for maximum adsorption, this is may be due to slow rate diffusion of the analyte into MIP sorbent. Method Validation

Concentrations from 1.0 to 1000 nmol L−1 were used to construct the calibration curve. For both urine and plasma, a close relationship was observed between concentration and peak area in the measured concentration range. For all analytes, the coefficient of determination was R2≥0.997 (n = 3). For n=15, the accuracy of QC samples for HA varied from 1.0 to 6.6% in plasma and from 2.2 to 7.1% in urine. The intra-day precisions (RSD) in plasma and urine for HA were 2.2–4.8% and 1.1–6.7%, respectively. Three washing solutions were applied between each run to avoid carry-over. Extraction recoveries were always above 91%.

Figure 18. MRM chromatograms of LOQ for HA in plasma (A) and urine (C) samples; and blank plasma (B) and urine (D) samples.

36

Table 5. Different methods for determination of HA including present study.

Extraction Technique

Detection method

LOD (ng mL-1)

Recovery (%)

Precision (RSD%)

Reference

μ-SPE GC/FID 16.5 98.0 1.6 [Ahmadi et al., 2009]

LLE GC/MS 4.0 92.7 3.6 [Ohashi et al., 2006]

SPE IC 1.0 93–98 1.4–2.3 [Zhao et al., 2011]

MIPM-MEPS LC/MS/MS 0.5 91–96 1.1–7.1 Paper III Typical chromatograms of LOQ for HA and blank chromatograms for plasma and urine samples are shown in Figure 18. The LOD values were 0.33 nmol L−1 and 0.3 nmol L−1 for plasma and urine samples, respectively. For human plasma and urine samples in three batches on different days, the LOQ value was 1.0 nmol L−1. The accuracies were 1-6.8% and 1-7% in human plasma and urine, respectively. The precision for plasma samples was obtained in the range from 2.2 to 7.3%, while values between 1.1 and 6.2% resulted for urine samples. Comparison with Other Extraction Techniques

Various techniques for the determination of the HA has been published including GC/FID [Ahmadi et al., 2009], GC/MS [Ohashi et al., 2006] and IC [Zhao et al., 2011]. However, sample preparation still remains as a bottleneck. It is still time consuming, it requires a lot of effort and there is an increased necessity for connecting it online. The sample preparation technique presented in this study, MIPM-MEPS, turns to be selective and robust, and it gives possibility to be connected online and performed fully automated. Moreover, it can handle small volumes of samples. As show in Table 5, MIPM-MEPS has a similar recovery and precision values as well as acceptable detection limit as compared to other published methods.

3.3 Development of a New Extraction Technique “MIPT” (Paper IV)

3.3.1 Preparation of MIP Sol-Gel Tablet (MIPT)

To increase the bioanalytical method selectivity and sensitivity, it is required to design and synthesis of more selective sorbents such as MIPs. In the present study sol-gel MIP was prepared in a new form as tablet. Polyethylene (PE) polymer with diameter and thickness of Ø6x1.2 mm was used as the backbone for the MIPT preparation (Paper IV).

37

Figure 19. Extraction tablets and applications.

Not only MIPs can be used in tablet form but also graphitic or silica sorbents can be used. In addition, the tablet can be prepared in different sizes from small to large depending on the sample volume. Figure 19 shows the extraction tablets and possible applications. The sol-gel MIP was prepared on the both the surfaces and the matrix of PE material. A PE material is carefully washed, first with methanol, and then with water and acetone to remove any contamination. For preparation of MIPT, the results described in Section 3.2.1 for preparation conditions of sol-gel MIP were used to select the optimal amounts of template/ precursor/catalyst to obtain microspheres on the PE material. Lastly, to replace the used template and to create a porous selective surface, the MIPT was washed with methanol for 2 h and with 0.2% formic acid in water for 30 min. The preparation of a non-molecularly imprinted polymer tablet (NIPT) was carried out in the same way but without template (methadone-d9) [Andersson et al., 1997; Andersson et al., 2004]. Figure 20 shows the SEM images of coated (sol-gel MIP) and un-coated (sol-gel NIP) PE surface. This investigation can verify layers of sol-gel imprinted materials covering the PE surface. In addition, the SEM image exhibits the uniformity in the structures and high surface area of the MIP particles.

38

A B

Figure 20. Scanning electron micrographs of a PE surface

NIP-sol-gel tablet (A) and MIP-sol-gel tablet (B) MIPT micro-solid phase extraction technique is a novel method for sample preparation including the following steps (Figure 21). First, the loading by inserting the MIPT directly in the biological sample solution followed by adsorption of the bioanalytes specifically onto the MIPT by shaking the samples with the help of a vortex during max 10 min.

Figure 21. Steps involved in the new extraction technique using MIPT.

43

Table 6. Accuracy and precision of methadone (MT) in human plasma.

QC samples Mean accuracy Intra-assay Inter-assay

(ng mL-1) % Mean RSD Mean RSD (n=12) (n=6) (%) (n=12) (%)

15 9.7 16.10 8.0 16.46 7.0 2000 3.6 2046 4.0 2073 8.0 4000 5.9 4389 6.0 4238 5.0

Method Validation

The bioanalytical method for the detection of MT in human plasma utilizing MIPT was validated according to FDA and European Medicines Agency guidelines [US FDA, 2014; EU EMA, 2011]. The validation parameters, including the calibration and extraction recovery, accuracy, precision, carry-over, and LOD and LLOQ, selectivity and matrix effects were evaluated using standard human plasma samples. Calibration and extraction recovery. The plasma standard calibration curves were in the range of 5–5000 ng mL-1 and described by a quadratic regression (see section 3.1.4). Each calibration curve consists of eight calibration points. The coefficient of determination was r2 > 0.999 in all runs (n=3). The lower limit of quantification (LLOQ) was achieved as the lowest point of standard calibration curves (5 ng mL-1) and the limit of detection (LOD) was 1 ng mL-1. The method extraction recovery was 80% using concentration levels at 10 ng mL-1 and 2500 ng mL-1 (n=3, RSD = 5-7%). Accuracy and precision. The method accuracy was determined by the percentage of the equation: [(found/theoretical concentration) – 1] using six replicates of QC samples at three concentration levels: 15 ng mL-1, 2000 ng mL-1 and 4000 ng mL-1. The LLOQ was 5.0 ng mL-1. The accuracy was in the range of 3.6 – 9.7%. The method precision was evaluated by the relative standard deviation (RSD) of six runs of each QC concentration level, in one day (intra-day assay) and in three days (inter-day assay). Intra-assay precision was ranged from 4 to 8% while the inter-assay precision was 5 – 8%. The inter-day precision of LLOQ was 11 % (n=6). The accuracy and precision results are summarized in Table 6. Method selectivity and matrix effect. The method selectivity was established by comparing blank plasma sample and spiked plasma sample processed by the MIPT extraction. Blank plasma samples did not show any significant peaks of interferences Figure 28.

44

Figure 28. MRM transitions obtained from the analysis of MT at LLOQ with internal standard (A) and blank plasma sample (B).

45

The matrix effect was evaluated using post-extraction addition method at two concentration levels (QCH and QCL). No significant matrix effects were observed and the variations were less than 10%. Carry-over effect. The carry-over effect was investigated by injecting the pure mobile phase after the injection of highest concentration sample (5000 ng mL-1). Each MIPT was reused up to twenty extractions without any changing in the extraction efficiency. To decrease the carry-over, the MIPT was washed with methanol and with water after each extraction. No carry-over was detected when the tablet reused. Comparison with Other Extraction Techniques

Table 7 demonstrates a comparison between the results of the MIPT (Paper IV), MEPS (Paper II) and published data obtained from other extraction techniques (SPME and SBSE). Both MEPS and MIPT techniques considerably reduced the extraction time compared to SPME (decreased by 3-15 fold) and SBSE (decreased by 9-45 fold). In addition, the sample volume was reduced by 5 times and 25 times using SPME and SBSE, respectively. This work demonstrates that quantitation can be performed using MIPT with lower extraction time and higher selectivity. The limits of detection and quantification of the present method are improved in comparison to the SPME. Table 7. Comparison of LOD, LLOQ, extraction time and precision between

different extraction techniques.

Methadone SPME SBSE MEPS MIPT

Matrix Plasma Urine Urine Plasma Sample Volume (mL) 1.0 5.0 1.0 0.2 Detection Method GC/MS GC/MS GC/MS LC/MS/MS Linear Range (ng mL-1)

50-2000 No data 2.3-3100 5-5000

LOD (ng mL-1)

9.0 No data 0.4 1.0

LLOQ (ng mL-1)

30 No data 2.3 5.0

Extraction time (min)

30 90 2 10

Precision 5.0 No data 11 4-8

Reference [Bermejo et al., 2000]

[Tienpont et al., 2003]

Paper II Paper IV

46

4 Conclusions and Future Aspects

New microextraction techniques for the assay of neurotransmitters and biomarkers in biological fluids were developed and validated. In Paper I and Paper II, the extraction of neurotransmitters such as dopamine, serotonin and methadone from human urine was performed by microextraction in packed sorbent (MEPS) and connected online with LC/MS/MS or GC/MS. In Paper III, the surface of a porous polysulfone membrane (PSM) was modified by sol-gel based molecularly imprinting methodology. The molecularly imprinted polymer membrane (MIPM) was subsequently used as sorbent for online extraction of a lung cancer biomarker, hippuric acid, from human plasma and urine samples using LC/MS/MS. In Paper IV, a new sol-gel MIP technique was introduced in a tablet form (MIPT). MIPT was used for micro-solid-phase extraction in small sample volumes to extract and enrich neurotransmitters such as amphetamine and methadone from plasma using LC/MS/MS. The acceptance criteria for the validation of the investigated techniques were well in accordance with the international guidelines [EU EMA, 2011; US FDA, 2014; Shah et al., 2000]. During calibration, regression correlation coefficients were over 0.99 in all experiments. MEPS showed values of precision and accuracy that are comparable with other techniques. Compared to previously used methods, such as LLE-LC/MS/MS [Kushnir et al., 2002], SPME-GC/NPD [Myung et al., 1999], LLE-LC/UV [He et al., 2005], SPE(C18)-LC/UV [Quintela et al., 2006] and HPLC [Xiao et al., 1998], both MEPS-LC/MS/MS and MEPS-GC/MS improved the limits of detection (LOD) and reduced the extraction time significantly. MEPS, as a sample preparation method, proved to be suitable for the fully automated determination of biologically active analytes in complex matrices requiring only small sample and solvent volumes thus reducing the experimental costs. In addition, MEPS technique resulted to be relatively selective due to the use of different sorbents with variety of selectivity. It is easy to use and fast. Only small amounts (1 mg) of a solid packing material were needed, and any adsorption material such as silica-based (C2, C8, C18) or molecularly imprinted polymers (MIP) could be used. Even though, there were some serious issues related to the loss, disorganization or formation of imperfections in the solid sorbent matrix, thus affecting the repeat lifetimes of the MEPS common sorbents. After few consecutive extractions of biological samples, the sorbent should be replaced. Consequently, the selection and preparation of stable, non-deteriorating, reusable and long-lasting sorbents is an important challenge during sample preparation process. MIPM as extraction sorbent showed higher selectivity towards analytes as they were designed for more than the common sorbents used in MEPS technique. Moreover,

47

the sol-gel imprinting technique showed the capacity to use strong bonding to bind to surfaces through hydrogen bonding. This increased the robustness and lifetime of the sorbent. Furthermore, the cross-linking capacity and high rigidity of silica precursor components allowed to keep the size and shape of the created cavities in the MIPM matrix, which was definitive for using it in several tens of extraction runs. The main disadvantage observed was related to the sensibility of the MIP cavities to the sample matrix, being that the impurities present in plasma and urine samples could destroy or obstruct the membrane in the syringe, and therefore resulting in a decreased extraction recovery after 50 extractions. With MIPT, various effective parameters, including adsorption capacity, extraction and desorption time, addition of salt and impact of pH were investigated and optimized. The validated method showed good selectivity for the extraction of the selected biologically active analytes from plasma samples and demonstrated that the method is accurate, precise and sensitive. The MIPT was used for about twenty extractions without any noticeable change in the results. The MIPT procedures are very simple, fast and robust. The MIPT can be prepared in different size to be suitable for large volume samples from environmental and food analysis. Due to the advantages over other sample preparation techniques, the microextraction techniques presented in this thesis will undoubtedly allow an increased application in the analysis of numerous different classes of substances in a variety of matrices. The development of sorbents with high selectivity, easy-to-use sorbents with simpler processing steps, as well as a reduction of error risks are the main goals of companies involved in new extraction sorbent materials and methods. Considering that there will always be a high demand for rapid, selective, reliable and sensitive techniques, it is convenient to propose the following issues for future development studies:

� Development of the synthesis of new membranes for MEPS for industrial production.

� Development of tablet manufacture for a number of well-known neurotransmitters and biomarkers for industrial production.

� Produce tablets for all kinds of molecules.

� Utilization of molecular dynamics simulations as a theoretical tool for development of molecularly imprinted systems in order to find suitable monomers and cross-linkers.

48

Acknowledgements

A great number of people and institutions have colored my professional life in different ways through collaboration and discussion. Hence, I am pleased to acknowledge all of them for making this work possible. The following are openly listed and others are thanked but not forgotten.

I would like to express my sincere gratitude to Professor Cynthia de Wit and Professor Anders Colmsjö for the opportunity that allowed me to perform the research work at the Department of Environmental Science and Analytical Chemistry (ACES), Stockholm University, Sweden.

To my supervisor, Professor Mohamed Abdel Rehim, I highly praise his excellent guidance and supervision at AstraZeneca and at Stockholm University. I am overwhelmed by his endless generosity offering his time to listen to my questions, putting his advice towards the right target, guiding scattered ideas to a converging point, sharing his precious expertise in the field of sample preparation, encouraging and supporting in various ways this endeavor. I am deeply grateful for everything he did.

I thank Professor Lars Blomberg at Karlstad University for his invaluable support, as well as Professor Lars Andersson for kindly proofreading the manuscript and for the scientific discussions.

I express my appreciation to PhD Osvaldo Pino Garcia for his enthusiasm and willingness to helping during the preparation of my dissertation.

I am indebted to all project collaborators and co-authors for the successful cooperation we had. Without you, this work would be hard to be accomplished. My thanks go to Mohammad Mahdi Moein, Anders Kussak, Mehran Javanbakht, Mohammad Karimi, Behrouz Akbari-adergani, Hana Vlčková, Lucie Nováková, Petr Solich, Abbi Abdel-Rehim, Amin Jeppsson-Dadoun, Rana Said and Mohamed Kamel.

I would like to say thank you to Professor Ulrika Nilsson, and to the entire Department of Environmental Science and Analytical Chemistry (ACES), Stockholm University, for constant support and for creating such a wonderful and stimulating working environment.

My gratitude to administrative and technical staff Jonas Rutberg, and Lena Elfver for their utmost help and presence whenever there was a need for that.

Special thanks go to PhD students in our Department: Francesco Iadaresta for the scientific ideas we shared and nice collaboration we had in the lab and to Aljona Saleh for sharing her knowledge with me in mass spectrometry. Trifa Mohammad Ahmed and Liying Jiang for nice conversations; Giovanna Luongo bringing Italian coffee and nice conversation; Ioannis Sadiktsis for professional computer assistance when needed; Nadia Zguna (Kiselova) and Pedro Sousa for being nice officemates; Farshid Mashayekhy Rad for discussions about life and science; Ahmed Ramzy and Hatem Elmongy for sharing knowledge, happiness and frustration in the lab. Hwanmi Lim, Javier Zurita Perez,

49

Alessandro Quaranta, Jenny Aasa, Rozanna Avagyan and Mazaher Ahmadi I have no doubts that your future is scientifically very bright!

I would like to express my thanks to Carina Norsten-Höög and Eva Klasson-Wehler for kindly giving me the opportunity to use the excellent practical facilities during part of the research work carried out at the Department of DMPK & BAC at AstraZeneca R&D, Södertälje. To all my colleagues and friends at AstraZeneca for the motivation I got from their cheerful atmosphere, unfailing encouragement and advice respectively.

To PhD Atia Daoud, PhD Ahlam Maqour, Mehdi Gharbi, Roger Bydler, Margareta Grandinson, Sandra Löfgren, Karin Barman and Mats Hansson; I would like to express my deepest thanks for their continuous encouragement.

I directed my thanks to all my former teachers at University Sidi Mohamed Ben Abdellah (USMBA) in Fès, Morocco, for the solid education received from them and for being a source of inspiration that helped me to continue developing professionally. Merci beaucoup!

My appreciation goes to all teachers at Södertörn University, Stockholm, during my Master studies in Biomolecular Chemistry and in Education.

Thanks to all my friends, I can always count on and for unconditional support.

Finally, to my dear ones; to the whole big family in Sweden and Morocco. My gratefulness goes to my husband my lovely daughter and my son for being always next to me when I needed them; thank you for being my sunshine! For my 50th birthday present, my grandchild, and forthcoming grandchildren; I am and will be happy to welcome you all to this exciting world with a new title! To my parents for what they did for me to be raised as a nice human being and for always believing in me; and to my brothers & sisters for sharing everything as a large family. I care about them all in different ways but at the same level.

Finally yet importantly, this chance in life could not happen at this scale if I did not have this opportunity to live in this great country, which is Sweden. I profoundly like the Swedish educational system and environment and acknowledge the Swedish universities for giving the chance evenly to everyone who really likes to make his way through towards the realm of knowledge!

50

References

1 Abdel-Rehim M., AstraZeneca Application: Syringe for solid phase microextraction. Current Patents Gazette, week 0310, WO 03019149, p.77 (2003).

2 Abdel-Rehim M., New trends in sample preparation: Online microextraction in packed syringe for liquid and gas chromatography applications. I. Determination of local anaesthetics in human plasma samples using gas chromatography – mass spectrometry. J. Chromatogr. B (2004) 801, 317-321.

3 Abdel-Rehim M., Recent advances in microextraction by packed sorbent for bioanalysis. J. Chromatogr. A (2010) 1217, 2569–2580.

4 Abdel-Rehim M., Microextraction by packed sorbent (MEPS): A tutorial. Anal. Chim. Acta (2011) 701, 119–128.

5 Agilent Technologies, Basics of LC/MS. Primer (2001) http://ccc.chem.pitt.edu/wipf/Agilent%20LC-MS%20primer.pdf

6 Ahmadi F., Asgharloo H., Sadeghi S., Gharehbagh-Aghababa V. and Adibi H., Post-derivatization procedure for determination of hippuric acid after extraction by an automated micro solid phase extraction system and monitoring by gas chromatography. J. Chromatogr. B (2009) 877, 2945–2951.

7 Alexander C., Andersson H.S., Andersson L.I., Ansell R.J., Kirsch N., Nicholls I.A., O’Mahony J. and Whitcombe M.J., Molecular imprinting science and technology: a surveyof the literature for the years up to and including 2003. J. Mol. Recognit. (2006) 19, 106–180.

8 Ali W.H., Derrien D., Alix F., Pérollier C., Lépine O., Bayoudh S., Hugon F.C. and Pichon V., J. Chromatogr. A (2010) 1217, 6668.

9 An Z., Chen Y., Zhang R., Song Y., Sun J., He J., Bai J., Dong L., Zhan Q. and Abliz Z., Integrated ionization approach for RRLC–MS/MS-based metabonomics: finding potential biomarkers for lung cancer. J. Proteome Res. (2010) 9, 4071–4081.

10 Andersson L.I., Application of molecular imprinting to the development of aqueous buffer and organic solvent based radioligand binding assays for (S)-propranolol. Anal. Chem. (1996) 68, 111-117.

11 Andersson L.I., Hardenborg E., Sandberg-Ställ M., Möller K., Henriksson J., Bramsby-Sjöström I., Olsson L.I. and Abdel-Rehim M., Development of a molecularly imprinted polymer based solid-phase extraction of local anaesthetics from human plasma. Anal. Chim. Acta (2004) 526, 147-154.

12 Andersson L.I, Paprica A. and Arvidsson T., A highly selective solid phase extraction sorbent for pre- concentration of sameridine sade by molecular imprinting. Chromatographia (1997) 46, 57-62.

13 Apffel A., Fischer S, Goldberg G., Goodley P.C. and Kuhlmann F.E., Enhanced sensitivity for peptide mapping with electrospray liquid chromatography-mass spectrometry in the presence of signal suppression due to trifluoroacetic acid-containing mobile phases. J. Chromatogr. A (1995) 712, 177.

14 Arthur C.L. and Pawliszyn J., Solid phase microextraction with thermal desorption using fused silica optical fibers. Anal. Chem. (1990) 62, 2145-2148.

15 Ashri N.Y. and Abdel-Rehim M., Sample treatment based on extraction techniques in biological matrices. Bioanalysis (2011) 3, 2003-2018.

16 Ashri N.Y., Daryanavard M. and Abdel-Rehim M., Biomed. Chromatogr. (2013) 27, 396.

17 Bahrami A., Jonidi-Jafari A., Folladi B., Mahjub H., Sadri Q. and Zadeh M.M., Comparison of urinary o-cresol and hippuric acid in drivers, gasoline station workers and painters exposed to toluene in West of Iran. Pakistan J. Biol. Sci. (2005) 8, 1001-1005.

18 Bell J., Pharmacological maintenance treatments of opiate addiction. Br. J. Clin. Pharmacol. (2012) 77, 253-263.

19 Bermejo A.M., Seara R., dos Santos Lucas A.C., Tabernero M.J., Fernandez P. and Marsili R., J. Anal. Toxicol. (2000) 24, 66-69.

51

20 Berridge C.W. and Devilbiss D.M., Psychostimulants as cognitive enhancers: the prefrontal cortex, catecholamines, and attention-deficit/hyperactivity disorder. Biol. Psychiatry (2011) 69, e101–111.

21 Brinker C.J. and Scherer G.W., Sol–Gel Science: The Physics and Chemistry of Sol–Gel Processing. Academic Press, San Diego (1990).

22 Cech N.B. and Enke C.G., Practical implications of some recent studies in electrospray ionization fundamentals. Mass Spectrom. Rev. (2001) 20, 362-387.

23 Chen F.F., Xie X.Y. and Shi Y.P., J. Chromatogr. A (2013) 1300, 112.

24 Chen L., Xu S. and Li J., Recent advances in molecular imprinting technology: current status, challenges and highlighted applications. Chem. Soc. Rev. (2011) 40, 2922-2942.

25 Cherkaoui S., Rudaz S., Varesio E. and Veuthey J.L., Electrophoresis (2001) 22, 3308-3315.

26 Cody R.B., Laramée J.A. and Durst H.D., Versatile new ion source for analysis of materials in open air under ambient conditions. Anal. Chem. (2005) 77, 2297-2302.

27 Cole R.B. (Ed.) Electrospray Ionisation Mass Spectrometry: Fundamentals, Instrumentation and Applications. Wiley, Chichester (1997).

28 Corriu J.P., Ceramics and nanostructures from molecular precursors. Angew. Chem. Int. Ed Engl, (2000) 39, 1376–1380.

29 Daryanavard S.M., Jeppsson-Dadoun A., Andersson L.I., Hashemi M., Colmsjö A. and Abdel-Rehim M., Molecularly imprinted polymer in microextraction by packed sorbent for the simultaneous determination of local anesthetics: lidocaine, ropivacaine, mepivacaine and bupivacaine in plasma and urine samples. Biomed. Chromatogr. (2013) 27, 1481–1488.

30 De Hoffman E. and Stroobant V., Mass Spectrometry: Principles and Applications, 3rd Ed., Wiley (2007).

31 De Jong W.H.W.M., de Vries E.G. and Kema I.P., Automated mass spectrometric analysis of urinary and plasma serotonin. Anal. Bioanal. Chem. (2010) 396, 2609-2616.

32 De la Mora J.F., Van Berkel G.J., Enke C.G. et al., Electrochemical processes in electrospray ionization mass spectrometry. J. Mass Spectrom. (2000) 35, 939-952.

33 De Zeeuw R.A., Drug screening in biological fluids. The need for a systematic approach. J. Chromatogr. B (1997) 689, 71-79.

34 EU EMA, European Committee for Medicinal Products for Human Use (CHMP), Guideline on Bioanalytical Method Validation, London, UK (2011).

35 Feldman J.M. and Lee E.M., Serotonin content of foods: effect on urinary excretion of 5-hydroxyindoleacetic acid. Am. J. Clin. Nutrition (1985) 42, 639–643.

36 Fritz J. In: Analytical Solid Phase Extraction, Wiley-VCH, New York (1999).

37 Han D. and Row K.H., Trends in liquid-phase microextraction, and its applicationto environmental and biological samples. Mirochim. Acta (2012) 176, 1–22.

38 Hansen S.H. and Pedersen-Bjergaard S. (Eds.) Bioanalysis of Pharmaceuticals. Sample Preparation, Separation Techniques and Mass Spectrometry. Wiley, Chichester (2015).

39 He H., Sun C., Wang X.R., Pham-Huy C., Chikhi-Chorfi N., Galons H., Thevenin M., Claude J.R. and Warnet J.M., J. Chromatogr. B (2005) 814, 385–391.

40 Hench L. And West J., The sol-gel process. Chem. Rev. (1990) 90, 33-72.

41 Hsieh Y. and Korfmacher W.A., Increasing speed and throughput when using HPLC-MS/MS systems for drug metabolism and pharmacokinetic screening. Curr. Drug Metabolism (2006) 7, 479-489.

42 Hu X., Cai Q., Fan Y., Ye T., Cao Y. and Guo C., Molecularly imprinted polymercoated solid-phase microextraction fibers for determination of Sudan I-IV dyesin hot chili powder and poultry feed samples. J. Chromatogr. A (2012) 1219, 39-46.

43 Huang B.Y., Chen Y.C., Wang G.R. and Liu C.Y., J. Chromatogr. A (2011) 1218, 849-855.

52

44 Ingle P.K., L-DOPA bearing plants. Natural Product Radiance (2003) 2, 126–133.

45 Inoue K., Kitade M., Hino T. and Oka H., Screening assay of angiotensin-converting enzyme inhibitory activity from complex natural colourants and foods using high-throughput LC–MS/MS. Food Chem. (2011) 126, 1909–1915.

46 Isaza C., Henao J., Velez J., Rodriguez M.A., Sierra J., Beltran L. and Sepulveda A., Evaluation of the methadone maintenance program of the Risaralda Mental Hospital. Rev. Colomb. Psiquiat. (2014) 43, 96-105.

47 Javanbakht M., Moein M.M. and Akbari-adergani B., On-line clean-up and determination of tramadol in human plasma and urine samples using molecularly imprinted monolithic column coupling with HPLC. J. Chromatogr. B (2012) 911, 49–54.

48 Karlsson J.G, Andersson L.I and Nicholls I.A, Probing the molecular basis for ligand-selective recognition in molecularly imprinted polymers selective for the local anaesthetic bupivacaine. Analytica Chimica Acta (2001) 435, 57–64.

49 Kebarle P. and Tang L., From ions in solution to ions in the gas phase: The mechanism of electrospray mass spectrometry. Anal. Chem. (1993) 65, 972A.

50 Khorrami A.R. and Rashidpur A., Development of a fiber coating based on molecular sol–gel imprinting technology for selective solid-phase micro extraction of caffeine from human serum and determination by gas chromatography/mass spectrometry. Anal. Chim. Acta (2012) 727, 20–25.

51 Kintz P., Eser H.P., Tracqui A., Moeller M., Cirimele V. and Mangin B. J. Forensic Sci. (1997) 42, 291-295.

52 Koeber R., Fleischer C., Lanza F., Boos K.S., Sellergren B. and Barceló D., Evaluation of a multidimensional solid-phase extraction platform for highly selective on-line cleanup and high-throughput LC/MS analysis of triazines in river water samples using molecularly imprinted polymers. Anal. Chem. (2001) 73, 2437–2444.

53 Kole P.L., Venkatesh G., Kotecha J. and Sheshala R., Recent advances in sample preparation techniques for effective bioanalytical methods. Biomed. Chromatogr. (2011) 25, 199–217.

54 Krasia-Christoforou T., Hybrid and hierarchical composite materials, in: Kim C.S., Randow C. and Sano T. (Eds.) Organic–Inorganic Polymer Hybrids: Synthetic Strategies and Applications, Springer International Publishing, Switzerland (2015) 11–63

55 Kulma A. and Szopa J., Catecholamines are active compounds in plants. Plant Science (2007) 172, 433–440.

56 Kushnir M.M., Urry F.M., Frank E.L., Roberts W.L. and Shushan B., Analysis of catecholamines in urine by positive-ion electrospray tandem mass spectrometry. Clin. Chem. (2002) 48, 323-331.

57 Lin Z., Wang J., Tan X., Sun L., Yu R., Yang H. and Chen G., J. Chromatogr. A (2013) 1319, 141.

58 Lodish H., Berk A. and Zipursky S.L., Neurotransmitters, synapses, and impulse transmission. Molecular Cell Biology, 4th ed., W. H. Freeman and Co., New York (2000).

59 Moein M.M., El-Beqqali A., Abdel-Rehim A., Jeppsson-Dadoun A. and Abdel-Rehim M., Preparation of monolithic molecularly imprinted polymer sol-gel packed tips for high-throughput bioanalysis: Extraction and quantification of L-tyrosine in human plasma and urine samples utilizing liquid chromatography and tandem mass spectrometry. J. Chromatogr. B (2014) 967, 168-173.

60 Moein M.M., Javanbakht M. and Akbari-Adergani B., Molecularly imprinted polymer cartridges coupled on-line with high performance liquid chromatographyfor simple and rapid analysis of dextromethorphan in human plasma samples. J. Chromatogr. B (2011) 879, 777–782.

61 Moein M.M., Javanbakht M. and Akbari-adergani B., Molecularly imprinted polymer cartridges coupled on-line with high performance liquid chromatography for simple and rapid analysis of human insulin in plasma and pharmaceutical formulations. Talanta (2014) 121, 30–36.

62 Monaghan P.J., Brown H.A., Houghton L.A. and Keevil B.G., Measurement of serotonin in platelet depleted plasma by liquid chromatography tandem mass spectrometry. J. Chromatogr. B (2009) 877, 2163-2167.

53

63 Montesano C., Sergi M., Odoardi S., Simeoni M.C., Compagnone D. and Curini R., A μ-SPE procedure for the determination of cannabinoids and their metabolites in urine by LC-MS/MS. J. Pharm. Biomed. Anal. (2014) 91, 169-175.

64 Moriarty M., Lee A., O’Connell B., Kelleher A., Keeley H. and Furey A., Development of an LC-MS/MS method for the analysis of serotonin and related compounds in urine and the identification of a potential biomarker for attention deficit hyperactivity/hyperkinetic disorder. Anal. Bioanal. Chem. (2011) 401, 2481-2493.

65 Myung S.W., Kim S., Park J.H., Kim M., Lee J.C. and Kim T.J., Solid-phase microextraction for the determination of pethidine and methadone in human urine using gas chromatography with nitrogen-phosphorus detection. Analyst (1999) 124, 1283–1286.

66 Ohashi Y., Mamiya T., Mitani K., Wang B., Takigawa T., Kira S. and Kataok H., Simultaneous determination of urinary hippuric acid, o-, m- and p-methylhippuricacids, mandelic acid and phenylglyoxylic acid for biomonitoring of volatileorganic compounds by gas chromatography–mass spectrometry. Anal. Chim.Acta (2006) 566, 167–171.

67 Owens G.J., Singh R.K., Foroutan F., Alqaysi M., Han C.M., Mahapatra C., Kim H.W. and Knowles J.C., Sol-gel based materials for biomedical applications. Progress Mater. Sci. (2016) 77, 1-19.

68 Pan L. and Pawliszyn J., Anal. Chem. (1997) 69, 190.

69 Pawliszyn J., Solid Phase Microextraction Theory and Practice, Wiley-VCH, New York (1997).

70 Peterson Z.D., Lee M.L. and Graves S.W., Determination of serotonin and its precursors in human plasma by capillary electrophoresis-electrospray ionization-time-of-flight mass spectrometry. J. Chromatography B (2004) 810, 101-110.

71 Poole C.F., New trends in solid-phase extraction. Trends in Anal. Chem. (2003) 22, 362-373.

72 Prasad B.B., Srivastava A. and Tiwari M.P., Highly selective and sensitive analysis of dopamine by molecularly imprinted stir bar sorptive extraction technique coupled with complementary molecularly imprinted polymer sensor. J. Coll. Interf. Sci. (2013a) 396, 234–241.

73 Prasad B.B., Srivastava A. and Tiwari M.P., Highly sensitive and selective hyphenated technique (molecularly imprinted polymer solid-phase microextraction–molecularly imprinted polymer sensor) for ultra trace analysis of aspartic acid enantiomers. J. Chromatogr. A (2013b) 1283, 9–19.

74 Prasad B.B., Tiwari K., Singh M., Sharma P.S., Patel A.K. and Srivastava S., Ultra-trace analysis of uracil and 5-fluorouracil by molecularly imprinted polymer brushes grafted to silylated solid-phase microextraction fiber in combination with complementary molecularly imprinted polymer-based sensor. Biomed. Chromatogr. (2009) 23, 499–509.

75 Psillakis E. and Kalogerakis N., Developments in liquid-phase microextraction, Trends Anal. Chem. (2003) 22, 565–574.

76 Qiu L., Liu W., Huang M. and Zhang L., J. Chromatogr. A (2010) 1217, 7461.

77 Quintela O., Lopez P., Bermejo A.M. and Lopez-Rivadulla M., J. Chromatogr. B (2006) 834, 188–194.

78 Rohner T.C., Lion N. and Girault H.H., Electrochemical and theoretical aspects of electrospray ionization. Phys. Chem. Chem. Phys. (2004) 6, 3056-3068.

79 Rudaz S., Calleri E., Geiser L., Cherkaoui S., Prat J. and Veuthey, J.L., Electrophoresis (2003) 24, 2633-2641.

80 Schweitz L., Andersson L.I. and Nilsson S., Capillary electrochromatography with predetermined selectivity obtained through molecular imprinting. Anal. Chem. (1997) 69, 1179-1183.

81 Shea K.J. and Loy D.A., Bridged polysilsesquioxanes. Molecular-engineered hybrid organic−inorganic materials. Chem. Mater, (2001) 13, 3306–3319.

82 Smeraglia J., Baldrey S.F. and Watson D., Matrix effect and selectivity issues in LC-MS/MS. Chromatographia (2002) 55, S95-S99.

83 Sporkert F. and Pragst F., J. Chromatogr. B (2000) 746, 255-264.

54

84 Takats Z., Wiseman J.M., Gologan B. and Cooks R.G., Mass spectrometry sampling under ambient conditions with desorption electrospray ionization. Science (2004) 5695, 471-473.

85 Temesi D. and Law B., The effect of LC eluent composition on MS response using electrospray ionization. LC-GC (1999) 17, 626.

86 Tienpont B., David F., Benijts T. and Sandra P., J. Pharm. Biomed. Anal. (2003) 32, 569-579.

87 US FDA, US Department of Health and Human Services, CDER, Guidance for Industry and FDA Staff, Qualification Process for Drug Development Tools, Silver Spring, MD, USA (2014) http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM230597.pdf

88 Vaghela A., Patel A., Patel A., Vyas A. and Patel N., Sample preparation in bioanalysis: A review. Int. J. Scient. & Technol. Research (2016) 5, 6-10.

89 Vuorensola K.S.H. and Karjalainen U., Determination of dopamine and methoxycatecholamines in patient urine by liquid chromatography with electrochemical detection and by capillary electrophoresis coupled with spectrophotometry and mass spectrometry. J. Chromatography B (2003) 788, 277-289.

90 Widstrand C., Björk H. and Yilmaz E., Analysis of analytes – the use of MIPs in solid-phase extraction increases efficiency and improves detection limits. Laboratory News (2006) 14-15.

91 Wilkins D.G., Nagasawa P.R., Gygi S.P., Foltz R.L. and Rollins D.E., J. Anal. Toxicol. (1996) 20, 355-361.

92 Wood P.B., Role of central dopamine in pain and analgesia. Expert Review of Neurotherapeutics (2008) 8, 781–797.

93 Young S.N., How to increase serotonin in the human brain without drugs. J. Psychiatry Neurosci. (2007) 32, 394–399.

94 Zhao F., Wang Z., Wang H. and Ding M., Determination of hippuric acid in humanurine by ion chromatography with conductivity detection. J. Chromatogr. B (2011) 879, 296–298.

95 Zhu X., Cai J., Yang J., Su Q., Gao Y., J. Chromatogr. A (2006) 1131, 37.

55

Original Papers (I-IV)