Screening of virtual libraries for monoamine oxidase inhibitors

i

Screening of virtual libraries for monoamine

oxidase inhibitors

M Barkhuizen

21686866

B Pharm

Dissertation submitted in partial fulfillment of the requirements

for the degree Magister Scientiae in Pharmaceutical Chemistry at

the Potchefstroom Campus of the North-West University

Supervisor:

Prof JP Petzer

Co-supervisor:

Dr A Petzer

November 2013

ii

This work is based on the research supported in part by the Medical Research Council and National Research Foundation of South Africa [Grant specific reference numbers (UID) 85642 and 80647]. The Grantholders acknowledge that opinions, findings and conclusions or recommendations expressed in any publication generated by the NRF supported research are that of the authors and that the NRF accepts no liability whatsoever in this regard.

iii

Preface:

This dissertation is submitted in article format and contains an original research article. Those results not included in the article are presented in Chapters 3 and 4 of the dissertation. The article was submitted for publication to the academic journal Arzneimittelforschung/ Drug research. The author guidelines for the journal are included in annexure B. The research described in this dissertation was conducted by Ms. M. Barkhuizen at the North-West University, Potchefstroom campus.

iv

Abstract:

The traditional view of drug design is that a single drug should interact with a single molecular target. As science progressed, there was an understanding that most drugs interact with more than one target and that multiple targets may be responsible for either adverse effects or additional therapeutic effects. The idea of polypharmacology, which suggests that the focus of drug design should shift from a single drug that interacts with a single target to a single drug that can have interactions with multiple targets and multiple therapeutic effects, revolutionized the drug discovery process. Discovering new drugs is a long and costly process with years of research and development and clinical trials required before the drugs reach the market for much needed therapeutic applications. By repurposing drugs that are already on the market for a new therapeutic target, the discovery process is accelerated significantly.

One such a target disease, for which there is a great need for new effective therapies, is Parkinson’s disease (PD). PD is a progressive neurodegenerative disease that is caused by the death of dopaminergic neurons in the substantia nigra with the resulting loss of dopamine from the striatum. Degeneration in PD leads to varying degrees of motor difficulty and disability, along with other symptoms. Current therapies are focussed on symptomatic management and an improvement of the quality of life of patients, rather than on a cure. There are several therapeutic targets that are currently used in the treatment of PD. One of those targets is the monoamine oxidase (MAO) enzymes, in particular the MAO-B isoform. The MAO enzymes are responsible for the metabolism of amine neurotransmitters, such as dopamine, and inhibition of MAO-B has proven to be an effective strategy to increase the dopamine levels in the brain. Clinically, selective MAO-B inhibitors are administered concurrently with levodopa (a precursor of dopamine) to increase the levels of dopamine derived from levodopa. This approach prolongs the beneficial effects of levodopa.

Because MAO-A is responsible for the breakdown of noradrenalin, adrenalin, serotonin and tyramine, non-selective and selective MAO-A inhibitors have therapeutic applications in other neurological and psychiatric disorders such as depression. MAO-A inhibitors, particularly irreversible inhibitors, are also notable from a toxicological point of view. Irreversible MAO-A inhibitors may lead to potentially dangerous effects when combined with serotonergic drugs and certain foods containing tyramine, such as cheeses and processed meats. Selective MAO-B inhibitors and reversible MAO-A inhibitors appear to be free of these interactions.

v

Based on the considerations above, this study aimed to identify clinically used drugs which also inhibit the MAO enzymes as a secondary pharmacological property. Such drugs may, in theory, be repurposed as MAO inhibitors for therapeutic use in the treatment of PD and depression. The identification of potential MAO-A inhibitory properties among clinically used drugs are of further importance since the irreversible inhibition of MAO-A may lead to dangerous effects when combined with certain drugs and foods.

To screen clinically used drugs for potential MAO-A and MAO-B inhibitory activities, a pharmacophore approach was followed. A pharmacophore model is a virtual 3D representation of the common steric and electrostatic features of the interaction between an enzyme and a ligand. By identifying hydrogen bond acceptor, hydrogen bond donor and hydrophobic interactions between a reference ligand and an enzyme, a model is created that can search databases for other molecules that would have similar interactions with the enzyme and arguably also act as ligands. This enables the screening of a large amount of molecules in a short amount of time. To assist in the identification of MAO inhibitors, pharmacophore models of the MAO enzymes were constructed using the known crystallographic structures of MAO-A crystallized with harmine, and MAO-B co-crystallized with safinamide. The Discovery Studio® software package (Accelrys) was used for this purpose.

In this study, virtual libraries of United States Food and Drug Administration (FDA) approved drugs and the United States Environmental Protection Agency (EPA) maximum daily dose databases were screened with pharmacophore models of MAO-A and MAO-B. Among the hits, 26 drugs were selected on the basis of availability and cost, and were subjected to in

vitro bio-assays in order to determine their potencies (IC50 values) as inhibitors of recombinant human MAO-A and/or MAO-B. Among the drugs tested, 6 compounds exhibited inhibitory activity towards the MAO enzymes. Of the 6 compounds, pentamidine (IC50 = 0.61 µM for MAO-A and IC50 = 0.22 µM for MAO-B) and phenformin (IC50 = 41 µM for MAO-A) were selected for further analysis.

An examination of the recoveries of the enzymatic activities after dilution and dialysis of the enzyme-inhibitor complexes showed that both pentamidine and phenformin interact reversibly with the MAO enzymes. A kinetic analysis suggests that pentamidine acts as a competitive inhibitor with estimated Ki values of 0.41 µM and 0.22 µM for the inhibition of MAO-A and MAO-B, respectively. An analysis of the available pharmacokinetic data and typical therapeutic doses of phenformin and pentamidine suggests that the MAO inhibitory potencies (and reversible mode of action) of phenformin are unlikely to be of pharmacological relevance in humans. Pentamidine, on the other hand, is expected to

vi

interact with both MAO-A and MAO-B at typical therapeutic doses. Because of its MAO-A inhibitory activity, pentamidine may thus, in theory, lead to a tyramine-associated hypertensive crisis when combined with tyramine-containing foods. However, pentamidine is unlikely to inhibit central MAO since it does not appear to penetrate the central nervous system to a large degree.

In an attempt to gain further insight into the mode of binding to MAO, pentamidine and phenformin were docked into models of the active sites of MAO-A and/or MAO-B. An analysis of the interactions between the enzyme models and the ligands were carried out and the results are discussed in the dissertation.

The results of this study show that the pharmacophore model approach may be useful in identifying existing drugs with potential MAO inhibitory effects. The search for new therapeutic MAO inhibitors, that can be used in the treatment of certain neurological disorders, including PD and depression, may be accelerated by employing a virtual screening approach. Such an approach may also be more cost effective than the de novo design of MAO inhibitors.

Keywords:

Monoamine oxidase, repurposing, Parkinson’s disease, virtual screening, toxicology, enzyme inhibition

vii

Uittreksel:

Die tradisionele beskouing van geneesmiddelontwerp was dat ʼn enkele geneesmiddel net met ʼn enkele molekulêre teiken interaksie ondergaan. Soos wat die wetenskap gevorder het, het die gedagterigting ontwikkel dat die meeste geneesmiddels met meer as een teiken interaksies het en dat meerdere teikens verantwoordelik mag wees vir beide die terapeutiese en newe-effekte. Die idee van polifarmakologie, wat voorstel dat die fokus van geneesmiddelontwerp behoort te verskuif vanaf ʼn enkele geneesmiddel wat net met ʼn enkele teiken ʼn interaksie het, na ʼn enkele geneesmiddel wat met verskeie teikens interaksies het om verskeie terapeutiese effekte te bewerkstellig, was verantwoordelik vir ʼn revolusie in die geneesmiddelontdekkingsproses. Om nuwe geneesmiddels te ontdek is ʼn lang en duur proses wat jare se navorsing, ontwikkeling en kliniese toetse vereis voordat geneesmiddels die mark bereik. Die geneesmiddelontdekkingsproses kan versnel word deur geneesmiddels wat alreeds op die mark is vir bestaande interaksies her aan te wend vir ʼn nuwe terapeutiese teiken.

Een so ʼn teikensiekte, waarvoor daar ʼn groot nood vir nuwe, effektiewe behandelings is, is Parkinson se siekte (PD). PD is ʼn progressiewe neurodegeneratiewe siekte wat veroorsaak word deur die afsterwe van dopaminergiese neurone in die substantia nigra en die gepaardgaande verlies aan dopamien. PD lei tot verskillende grade van motorgestremdheid en ander simptome. Huidige terapieë fokus op simptomatiese behandeling en ʼn verbetering in lewenskwaliteit, eerder as op genesing.

Daar bestaan verskeie terapeutiese teikens wat tans gebruik word vir die behandeling van PD. Een van dié teikens is die monoamienoksidase-ensieme (MAO), veral die MAO-B-isoform. Die MAO-ensieme is verantwoordelik vir die afbreek van amien-neuro-oordragstowwe, soos dopamien, en die inhibisie van MAO-B is ʼn effektiewe strategie om die dopamienvlakke in die brein te verhoog. Selektiewe MAO-B-inhibeerders word klinies gebruik saam met eksogeen toegediende levodopa (ʼn voorganger van dopamien) om die vlakke van dopamien wat uit die levodopa uit verkry word, te verhoog. Die benadering verleng die voordelige effekte van levodopaterapie.

MAO-A is verantwoordelik vir die metabolisme van noradrenalien, adrenalien, serotonien en tiramien. Beide nie-selektiewe en selektiewe MAO-A-inhibeerders word gebruik in die behandeling van neurologiese en sielkundige afwykings soos depressie. MAO-A-inhibeerders, veral onomkeerbare MAO-A-inhibeerders, is ook belangrik vanuit ʼn toksikologiese oogpunt, want onomkeerbare MAO-A-inhibeerders kan gevaarlike interaksies hê indien dit met serotonergiese geneesmiddels en kossoorte wat tiramien bevat, soos kase en verwerkte

viii

vleis, gekombineer word. Selektiewe MAO-B-inhibeerders en omkeerbare MAO-A inhibeerders toon nie dié interaksies nie.

Bogenoemde stellings in ag geneem, is met hierdie studie gepoog om klinies bruikbare geneesmiddels wat reeds in gebruik is, te identifiseer wat, benewens hulle primêre werking, ook die MAO-ensieme mag inhibeer as ʼn sekondêre farmakologiese eienskap. Sulke middels kan teoreties heraangewend word as MAO-inhibeerders vir gebruik in die terapie van PD en depressie. Die identifisering van potensiële MAO-A-inhiberende eienskappe van middels wat reeds gebruik word, is ook belangrik omdat die onomkeerbare inhibisie van MAO-A tot gevaarlike interaksies met sekere geneesmiddels en voedsel mag lei.

ʼn Farmakofoorbenadering is gevolg om deur geneesmiddels wat klinies gebruik word, te sif vir middels wat moontlik MAO-A en MAO-B mag inhibeer. ʼn Farmakofoormodel is ʼn virtuele 3D-voorstelling van die algemene ruimtelike en elektrostatiese eienskappe van die interaksie tussen ʼn ensiem en ʼn ligand. Deur waterstofbinding ontvanger, waterstofbinding skenker en hidrofobiese interaksies tussen ʼn verwysingsligand en ʼn ensiem te identifiseer, word ʼn model geskep wat gebruik kan word om deur ander databasisse te soek vir ander molekules wat soortgelyke interaksies met die ensiem sal hê en moontlik ook as ligande kan optree. Dit maak die sifting van groot hoeveelhede molekules in ʼn beperkte tyd moontlik. Om te help met die identifisering van MAO inhibeerders, is farmakofoormodelle van die MAO-ensieme geskep met behulp van bekende kristallografiese strukture van MAO-A wat mede-gekristalliseer is met harmien en van MAO-B wat mede-mede-gekristalliseer is met safienamied. Die Discovery Studio® sagtewarepakket van Accelrys is vir die doeleinde gebruik.

Vir die studie is virtuele biblioteke van die Verenigde State Voedsel en Geneesmiddel Administrasie se aanvaarde geneesmiddels en die Verenigde State Omgewingsbeskermingsagentskap se maksimum daaglikse dosis databasis gesif met die farmakofoormodelle van MAO-A en MAO-B. Van al die molekules wat deur die modelle geïdentifiseer is, is 26 geneesmiddels gekies op grond van beskikbaarheid en koste, en dié geneesmiddels is onderwerp aan in vitro biotoetse om hulle sterktes (IC50 waardes) as inhibeerders van rekombinante menslike MAO-A en MAO-B te bepaal. Van al die geneesmiddels wat getoets is, het 6 verbindings MAO-inhiberende aktiwiteit getoon. Van die 6 verbindings is pentamidien (IC50 = 0.61 µM vir MAO-A en IC50 = 0.22 µM vir MAO-B) en fenformien (IC50 = 41 µM vir MAO-A) gekies vir verdere analise.

ʼn Ondersoek na die herstel van die ensimatiese aktiwiteite, na die verdunning en dialise van die ensiem-inhibeerder-komplekse, toon dat beide pentamidien en fenformien omkeerbare interaksies met die MAO-ensieme ondergaan. ʼn Kinetiese analise dui aan dat pentamidien as ʼn kompeterende inhibeerder optree met geskatte Ki-waardes van 0.41 µM en 0.22 µM vir

ix

die inhibisie van MAO-A en MAO-B onderskeidelik. ʼn Analise van die beskikbare farmakokinetiese data en die tipiese terapeutiese doserings van pentamidien en fenformien dui daarop dat die MAO-inhiberende eienskappe (en omkeerbare binding) van fenformien waarskynlik nie van farmakologiese belang in mense is nie. Daar word egter verwag dat pentamidien interaksies met beide MAO-A en MAO-B sal hê teen normale terapeutiese dosisse. Omdat pentamidien MAO-A inhibeer, kan dit teoreties lei tot ʼn tiramien-geassosieerde hipertensiewe krisis as dit gekombineer word met tiramien-bevattende voedsel. Dit is egter onwaarskynlik dat pentamidien MAO sentraal sal inhibeer omdat die mate waartoe pentamidien die senustelsel binnedring beperk is.

In ʼn poging om verdere insig in die manier waarop binding aan MAO plaasvind te verkry, is pentamidien en fenformien vasgemeer in modelle van die aktiewe setels van MAO-A en/of MAO-B. ʼn Analise van die interaksies tussen die ensiem-modelle en die ligande is uitgevoer en die resultate word in die verhandeling bespreek.

Die resultate van die studie toon dat die farmakofoormodel-benadering bruikbaar kan wees om bestaande geneesmiddels met potensiële MAO-inhiberende effekte te identifiseer. Die soektog na nuwe terapeutiese MAO-inhibeerders wat vir die behandeling van sekere neurologiese toestande, soos PD en depressie, gebruik kan word, kan versnel word deur ʼn virtuele siftingsbenadering te volg. So ʼn benadering mag ook meer koste-effektief wees as die de novo ontwerp van MAO-inhibeerders.

Sleutelwoorde:

Monoamienoksidase, heraanwending, Parkinson se siekte, virtuele sifting, toksikologie, ensiem-inhibisie

x

Table of contents:

Abstract p. iv

List of abbreviations p. xiii

List of figures p. xvii

List of tables p. xxiv

Chapter 1: Introduction

1.1 Title

1.1.1 Short title p. 1

1.1.2 Expanded title p. 1

1.2 Introduction and overview

1.2.1 Parkinson’s disease p. 1 1.2.2 Monoamine oxidase p. 2 1.2.3 Pharmacophore modelling p. 3 1.2.4 Polypharmacology p. 4 1.3 Hypothesis p. 5 1.4 Rationale p. 5 1.5 Objectives p. 6

Chapter 2: Literature Study

2.1 Parkinson’s disease p. 8

2.1.1 Aetiology p. 8

2.1.2 Relevant normal brain anatomy and physiology p. 9

2.1.3 Pathophysiology in Parkinson’s disease p. 11

2.1.3.1 Sites of neurodegeneration p. 11

2.1.3.2 Mechanisms of neurodegeneration p. 13

2.2 Epidemiology and genetics p. 18

2.3 Chemical compounds of importance in PD p. 20

2.3.1 Dopamine p. 20

2.3.2 Levodopa p. 22

2.3.2.1 Rationale of levodopa therapy p.22

2.3.2.2 Absorption and metabolism p. 22

2.3.2.3 Adverse effects of levodopa therapy p. 25

2.4 Monoamine oxidase p. 26

2.4.1 Physiology of MAO p. 26

2.4.2 Composition and structure of MAO p. 26

2.4.3 The MAO catalytic cycle p. 33

2.4.3.1 The ternary complex mechanism p. 33

2.4.3.2 Mechanisms for the cleavage of CH bonds p. 34

2.4.3.3 Heterolytic proton abstraction mechanisms p. 36

2.4.4 MAO inhibitors p. 38

2.4.4.1 Introduction p. 38

2.4.4.2 The mechanisms of inhibition of MAO by selected inhibitors p. 39 2.4.4.3 Interactions and adverse effects of MAO inhibition p. 47 2.4.4.4 The clinical significance of MAO inhibition p. 48

2.5 Summary p. 53

Chapter 3: Pharmacophore modelling studies

xi

3.2 Experimental methods p. 61

3.2.1 Construction and screening of the pharmacophore models p. 61

3.2.2 Molecular docking p. 63

3.3 Results p. 65

3.3.1 Structure-based pharmacophore model of MAO-A p. 65 3.3.2 Structure-based pharmacophore model of MAO-B p. 80

3.4 Summary p. 89 Chapter 4: Enzymology 4.1 Introduction p. 90 4.1.1 Enzyme kinetics p. 90 4.1.2 Overview p. 95 4.1.3 Enzymology p. 96

4.2 Chemicals and instrumentation. p. 97

4.3 Determining the IC50 values. p. 97

4.3.1 Method p. 99

4.4 Recovery of enzyme activity after dilution studies. p. 101

4.5 Lineweaver-Burk Plots. p. 103

4.5.1 Method p. 103

4.6 Recovery of enzyme activity after dialysis studies. p. 105

4.7 Results of the MAO inhibition studies with drugs that mapped to the structure-based pharmacophore models of MAO-A and MAO-B, but proved not to be inhibitors in vitro, or were not potent enough inhibitors to be of clinical significance.

p. 107 4.7.1 2-Ethoxybenzamide p. 107 4.7.2 Isoxsuprine p. 108 4.7.3 Phenytoin p. 109 4.7.4 2-Benzyl-2-imidazoline p. 110 4.7.5 Mebeverine p. 111 4.7.6 Amodiaquine p. 112 4.7.7 Amlodipine p. 113 4.7.8 Zafirlukast p. 114 4.7.9 Dicumarol p. 115 4.7.10 Sulpiride p. 116 4.7.11 Cefotaxime p. 117 4.7.12 Cefuroxime p. 118 4.7.13 Sumatriptan p. 119 4.7.14 Valpromide p. 120 4.7.15 Papaverine p. 121 4.7.16 Ranolazine p. 122 4.7.17 Clofibrate p. 122 4.7.18 Fursultiamine p. 123

xii 4.7.19 Griseofulvin p. 124 4.7.20 Bisoprolol p. 125 4.7.21 Pentamidine p. 127 4.7.22 Phenformin p. 137 4.7.23 Metoprolol p. 143 4.7.24 Fluoxetine p. 144 4.7.25 Terfenadine p. 145 4.7.26 Lansoprazole p. 146

4.8 Results of the MAO inhibition studies with known MAO inhibitors

4.8.1 Toloxatone p. 148 4.8.2 Lazabemide p. 148 4.9 Summary p. 149 Chapter 5: Article p. 150 Chapter 6: Summary p. 171 Bibliography p. 175 Acknowledgements p. 180 Annexures p. 181

xiii LIST OF ABBREVIATIONS 2-BFI 5-Isothiocyanato-2-benzofuranyl-imidazoline 3D Three dimensional 4-HQ 4-Hydroxyquinoline [ ] Concentration of A A Actives Aβ β-Amyloid Acc Accuracy AdoHcy S-Adenosyl-homocysteine AdoMet S-Adenosyl-methionine Ala Alanine

APP Amyloid precursor protein

Arg Arginine

Asn Asparagine

B

Bcl2 B-Cell lymphoma 2

BDNF Brain-derived neurotrophic factor

C COMT Catechol-O-methyltransferase Cys Cysteine D D Dopamine receptor DA Dopamine DMF N,N-Dimethylformamide Dopa 3,4-Dihydroxyphenylalanine E E Enzyme

xiv

EPA United States Environmental Protection Agency

F

FAD Flavin adenine dinucleotide

FDA United States Food and Drug Administration. F-dopa Fluorinated levodopa

G

GABA Gamma-aminobutyric acid

GAPDH Glyceraldehyde-3-phosphate dehydrogenase GDNF Glial cell line-derived neurotrophic factor

Gln Glutamine

Gly Glycine

GP Globus pallidus

GPe Globus pallidus externa

GPi Globus pallidus interna

GSH Glutathione

H

HLA-DR Human leukocyte antigen-DR

Hcy Homocysteine

HRMS High resolution mass spectrometry

Hsp Heat shock protein

HTS High-throughput screening

I

I Inhibitor

I2 Imidazoline type 2 receptor

IC50 Inhibitor concentration that produces 50% inhibition of an enzyme

IL-1 Interleukin-1

xv K Ki Inhibitor constant KI Equilibrium constant KM Michaelis constant L

lbal Balanced labelling performance

Leu Leucine

LRRK2 Leucine-repeat rich kinase 2

Lys Lysine

M

MAO Monoamine oxidase

Met Methionine

MPP+ 1-Methyl-4-phenylpyridinium

MPTP 1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine mRNA Messenger ribonucleic acid

N

n Number of selected hits

N Total number of hits

NGF Neuronal growth factor NMDA N-Methyl-D-aspartate NOS Nitric oxide synthase

NTF Neurotrophic factor

P

PD Parkinson’s disease

PET Positron emission tomography

Phe Phenylalanine

PKC Protein kinase C

xvi ROS Reactive oxygen species

ROC curve Receiver operating characteristics curve

S S Substrate SD Standard deviation Se Sensitivity Ser Serine SN Substantia nigra

SNpc Substantia nigra pars compacta SNpr Substantia nigra pars reticularis

Sp Specificity

T

tHcy Total homocysteine

Thr Threonine

TNF-α Tumor necrosis factor α

Tyr Tyrosine

Trp Tryptophan

U

UCHL1 Ubiquitin carboxyhydrolase L1

V

vi Initial velocity

V Reaction velocity

Vmax Maximal velocity

Val Valine

Y

xvii

LIST OF FIGURES

Chapter 2

Figure 2.1.1 Schematic activity of the basal ganglia-thalamocortical motor circuit.

p. 10

Figure 2.1.2 Neuropathology of PD. p. 12

Figure 2.1.3 Mechanisms of neurodegeneration in PD. p. 14 Figure 2.1.4 The mechanism of neurotoxicity of hydrogen peroxide induced

by the Fenton reaction.

p. 16

Figure 2.3.1 The synthesis of dopamine. p. 20

Figure 2.3.2 The main pathways of dopamine metabolism in the brain. p. 21 Figure 2.3.3 Schematic illustration of the metabolism of levodopa. p. 24 Figure 2.4.1 The overall structure of MAO-A drawn in ribbon mode. p. 27 Figure 2.4.2 The structure of 1,4-diphenylbutene bound to the active site of

MAO-B.

p. 30

Figure 2.4.3 Ribbon diagrams of the structures of human MAO-A and human MAO-B.

p. 30

Figure 2.4.4 A comparison of the active sites of human MAO-A, human MAO-B and rat MAO-A in complex with selected inhibitors.

p. 31

Figure 2.4.5 A comparison of the active site cavities of human MAO-A and MAO-B.

p. 32

Figure 2.4.6 The catalytic pathway of the MAO-enzymes. p. 33 Figure 2.4.7 Reaction pathway for MAO catalysis. p. 33 Figure 2.4.8 General mechanism of C-H bond cleavage in the hydride

mechanism and the homolytic hydrogen atom transfer mechanism.

p. 35

Figure 2.4.9 Single electron transfer mechanism of MAO catalysis. p. 36 Figure 2.4.10 The polar nucleophilic mechanism for MAO catalysis. p. 37 Figure 2.4.11 The active site cavity structures of human MAO-B in

combination with various inhibitors.

p. 40

Figure 2.4.12 Hypothetical mechanism for the inactivation of MAO by tranylcypromine.

p. 41

Figure 2.4.13 Hypothetical mechanism for the inactivation of MAO by (R)-deprenyl.

p. 42

Figure 2.4.14 The structures of lazabemide and N-(2-aminoethyl)-p-chlorobenzamide.

p. 42

Figure 2.4.15 The structure of the flavin adduct with the MAO-B catalyzed oxidation product of N-(2-aminoethyl)-p-chlorobenzamide.

p. 43

Figure 2.4.16 The structure of safinamide. p. 43

Figure 2.4.17 The structure of coumarin derivatives. p. 44 Figure 2.4.18 MAO-B in complex with safinamide and

7-(3-chlorobenzyloxy)-4-carboxaldehyde-coumarin.

p. 44

Figure 2.4.19 The structure of 8-(3-chlorostyryl)caffeine. p. 45 Figure 2.4.20 The structure of trans,trans-farnesol. p. 45 Figure 2.4.21 The structure of 1,4-diphenyl-2-butene. p. 45 Figure 2.4.22 The complex of MAO-B with 1,4-diphenyl-2-butene. p. 46

xviii

Figure 2.4.23 The structure of isatin. p. 46

Figure 2.4.24 The structure of harmine. p. 47

Figure 2.4.25 The metabolism of (R)-deprenyl and rasagiline. p. 51

Chapter 3

Figure 3.1.1 A structure-based pharmacophore model. p. 55 Figure 3.1.2 Selection of n molecules from a database containing N entries. p. 56 Figure 3.1.3 Theoretical distributions for active molecules and decoys

according to their score.

p. 58

Figure 3.1.4 The ROC curves for ideal and overlapping distributions of actives and decoys.

p. 59

Figure 3.1.5 The 3D pharmacophore-based screening workflow. p. 60 Figure 3.2.1 Workflow for the construction of a structure-based

pharmacophore model and screening of a virtual library.

p. 62

Figure 3.2.2 Workflow for docking ligands into the active site of the MAO enzymes.

p. 64

Figure 3.3.1 Graphical representation of the structure-based

pharmacophore model of MAO-A, which was constructed using the structure of the co-crystallized ligand, harmine.

p. 65

Figure 3.3.2 Graphical representation of the MAO-A pharmacophore model derived from the structure of harmine using the structure-based approach. In this representation, only the exclusion constraint features are illustrated.

p. 66

Figure 3.3.3 Graphical representation of the MAO-A pharmacophore model derived from the structure of harmine using the structure-based approach. In this representation, the exclusion constraint features as well as the hydrogen bond acceptor, hydrogen bond donor and hydrophobic features are shown.

p. 66

Figure 3.3.4 A two-dimensional representation of the binding of harmine in the MAO-A active site.

p. 67

Figure 3.3.5 A three-dimensional representation of the interaction between harmine and the selected residues in the active site of MAO-A

p. 69

Figure 3.3.6 A three-dimensional representation of acceptor features and their corresponding residues.

p. 70

Figure 3.3.7 A three-dimensional representation of donor features and their corresponding residues.

p. 71

Figure 3.3.8 A ROC curve for the MAO-A pharmacophore model. p. 79 Figure 3.3.9 Graphical representation of the structure-based

pharmacophore model of MAO-B, which was constructed using the structure of the co-crystallized ligand, safinamide.

p. 80

Figure 3.3.10 Graphical representation of the MAO-B pharmacophore model derived from the structure of safinamide using the

structure-based approach. In this representation, only the exclusion

constraint features are illustrated..

p. 80

Figure 3.3.11 Graphical representation of the MAO-B pharmacophore model derived from the structure of safinamide using the

structure-based approach. In this representation, the exclusion

xix

constraint features as well as the hydrogen bond acceptor, hydrogen bond donor and hydrophobic features are shown. Figure 3.3.12 A two-dimensional representation of the binding of safinamide

in the MAO-B active site.

p. 82

Figure 3.3.13 A three-dimensional representation of the interactions between safinamide and selected residues in the active site of MAO-B.

p. 83

Figure 3.3.14 A three-dimensional representation of acceptor features and their corresponding interacting residues.

p. 83

Figure 3.3.15 A three-dimensional representation of donor features and their corresponding interacting residues.

p. 84

Figure 3.3.16 A ROC curve for the MAO-B pharmacophore model. p. 88

Chapter 4

Figure 4.1.1 A graph showing the relationship between Vmax and Km. p. 91 Figure 4.1.2 An example of a Lineweaver-Burk plot. p. 92 Figure 4.1.3 Lineweaver-Burk plots illustrating competitive inhibition. p. 93 Figure 4.1.4 Graphical representation of the IC50 value. p. 94 Figure 4.1.5 The oxidative deamination of kynuramine by MAO-A or MAO-B

to yield 4-hydroxyquinoline.

p. 97

Figure 4.3.1 Diagrammatic representation of the method for determining IC50 values for the inhibition of MAO-A and MAO-B.

p. 98

Figure 4.3.2 An example of the calibration curves routinely obtained in this study.

p. 99

Figure 4.4.1 Diagrammatic representation of the method followed for the dilution studies.

p. 102

Figure 4.5.1 Diagrammatic presentation of the method used for the construction of Lineweaver-Burk plots

p. 104

Figure 4.6.1 Diagrammatic overview of the protocol followed for the dialysis studies.

p. 106

Figure 4.7.1 The structure of 2-ethoxybenzamide. p. 107 Figure 4.7.2 The recombinant human MAO-A and MAO-B catalyzed

oxidation of kynuramine in the presence of various concentrations of 2-ethoxybenzamide.

p. 107

Figure 4.7.3 The structure of isoxsuprine. p. 108

Figure 4.7.4 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of isoxsuprine.

p. 108

Figure 4.7.5 The structure of phenytoin. p. 109

Figure 4.7.6 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of phenytoin.

p. 109

Figure 4.7.7 The structure of 2-benzyl-2-imidazoline. p. 110 Figure 4.7.8 The recombinant human MAO-A and MAO-B catalyzed

oxidation of kynuramine in the presence of various concentrations of 2-benzyl-2-imidazoline.

p. 110

Figure 4.7.9 The structure of mebeverine. p. 111

xx

oxidation of kynuramine in the presence of various concentrations of mebeverine.

Figure 4.7.11 The structure of amodiaquine. p. 112

Figure 4.7.12 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of amodiaquine.

p. 112

Figure 4.7.13 The structure of amlodipine. p. 113

Figure 4.7.14 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of amlodipine.

p. 113

Figure 4.7.15 The structure of zafirlukast. p. 114

Figure 4.7.16 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of zafirlukast.

p. 114

Figure 4.7.17 The structure of dicumarol. p. 115

Figure 4.7.18 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of dicumarol.

p. 115

Figure 4.7.19 The structure of sulpiride. p. 116

Figure 4.7.20 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of sulpiride.

p. 116

Figure 4.7.21 The structure of cefotaxime. p. 117

Figure 4.7.22 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of cefotaxime.

p. 117

Figure 4.7.23 The structure of cefuroxime. p. 118

Figure 4.7.24 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of cefuroxime.

p. 118

Figure 4.7.25 The structure of sumatriptan. p. 119

Figure 4.7.26 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of sumatriptan.

p. 119

Figure 4.7.27 The structure of valpromide. p. 120

Figure 4.7.28 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of valpromide.

p. 120

Figure 4.7.29 The structure of papaverine. p. 121

Figure 4.7.30 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of papaverine.

p. 121

Figure 4.7.31 The structure of ranolazine. p. 122

Figure 4.7.32 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of ranolazine.

xxi

Figure 4.7.33 The structure of clofibrate. p. 122

Figure 4.7.34 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of clofibrate.

p. 123

Figure 4.7.35 The structure of fursultiamine. p. 123

Figure 4.7.36 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of fursultiamine.

p. 124

Figure 4.7.37 The structure of griseofulvin. p. 124

Figure 4.7.38 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of griseofulvin.

p. 125

Figure 4.7.39 The structure of bisoprolol. p. 125

Figure 4.7.40 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of bisoprolol.

p. 126

Figure 4.7.41 The structure of pentamidine. p. 127

Figure 4.7.42 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of pentamidine.

p. 128

Figure 4.7.43 Reversibility of inhibition of MAO-A and MAO-B by pentamidine (dilution method).

p. 129

Figure 4.7.44 Reversibility of inhibition of MAO-A and MAO-B by pentamidine (dialysis method).

p. 130

Figure 4.7.45 Lineweaver-Burk plots of human MAO-A and MAO-B activities in the absence and presence of various concentrations of pentamidine.

p. 131

Figure 4.7.46 The docked orientation of pentamidine within the MAO-A active site. The two-dimensional representation of the binding mode in MAO-A is also given.

p. 132

Figure 4.7.47 The docked orientation of pentamidine within the MAO-B active site. The two-dimensional representation of the binding mode in MAO-B is also given.

p. 133

Figure 4.7.48 Pentamidine mapped to the pharmacophore model of MAO-A. p. 136 Figure 4.7.49 Pentamidine mapped to the pharmacophore model of MAO-B. p. 136

Figure 4.7.50 The structure of isethionic acid. p. 137

Figure 4.7.51 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of isethionic acid.

p. 137

Figure 4.7.52 The structure of phenformin. p. 137

Figure 4.7.53 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of phenformin.

p. 138

Figure 4.7.54 Reversibility of inhibition of MAO-A by phenformin (dilution method).

p. 139

xxii method).

Figure 4.7.56 The docked orientation of phenformin within the MAO-A active site. The two-dimensional representation of the binding mode in MAO-A is also given.

p. 141

Figure 4.7.57 Phenformin mapped to the pharmacophore model of MAO-A. p. 143

Figure 4.7.58 The structure of metoprolol. p. 143

Figure 4.7.59 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of metoprolol.

p. 144

Figure 4.7.60 The structure of fluoxetine. p. 144

Figure 4.7.61 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of fluoxetine.

p. 145

Figure 4.7.62 The structure of terfenadine. p. 145

Figure 4.7.63 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of terfenadine.

p. 146

Figure 4.7.64 The structure of lansoprazole. p. 146

Figure 4.7.65 The recombinant human MAO-A and MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of lansoprazole.

p. 147

Figure 4.8.1 The recombinant human MAO-A catalyzed oxidation of kynuramine in the presence of various concentrations of toloxatone.

p. 148

Figure 4.8.2 The recombinant human MAO-B catalyzed oxidation of kynuramine in the presence of various concentrations of lazabemide.

p. 148

Chapter 5

Fig. 1 The structures of phenformin and pentamidine. p. 153 Fig. 2 The sigmoidal concentration-inhibition curve for the inhibition of

human MAO-A by phenformin.

p. 157

Fig. 3 The sigmoidal concentration-inhibition curves for the inhibition of human MAO-A and MAO-B by pentamidine.

p. 158

Fig. 4 Reversibility of inhibition of MAO-A by phenformin and pentamidine (dilution method).

p. 159

Fig. 5 Reversibility of inhibition of MAO-B by pentamidine (dilution method).

p. 160

Fig. 6 Reversibility of inhibition of MAO-A by phenformin (dialysis method).

p. 161

Fig. 7 Reversibility of inhibition of MAO-A and MAO-B by pentamidine (dialysis method).

p. 162

Fig. 8 Lineweaver-Burk plots of human MAO-A and MAO-B activities in the absence and presence of various concentrations of pentamidine.

p. 163

Fig. 9 Lineweaver-Burk plots of human MAO-A activity in the absence and presence of various concentrations of phenformin

xxiii

xxiv

LIST OF TABLES

Chapter 2

Table 2.4.1 MAO inhibitors. p. 39

Chapter 3

Table 3.3.1 The interaction energies between harmine and the active site residues and waters of MAO-A.

p. 68

Table 3.3.2 A list of the compounds in the DrugBank which mapped to the pharmacophore model derived from the structure of MAO-A (with harmine co-crystallized) and from the structure of MAO-B (with safinamide co-crystallized) using the structure-based approach. Also given are the fit-values of the respective compounds. These compounds represent drugs which are used systemically by humans. It is important to note that these are only the hits that were selected for further in vitro evaluation as MAO inhibitors.

p. 72

Table 3.3.3 Compounds that were four feature hits in the screening of a virtual library of 18 test compounds with the pharmacophore model of MAO-A.

p. 75

Table 3.3.4 A virtual library of 18 test compounds (9 inhibitors and 9 non-inhibitors of MAO-A) was screened with the MAO-A pharmacophore model. Below is given a list of compounds that were found not to be hits. The pharmacophore model used was derived from the structure of harmine. The compounds not shaded are known not to inhibit MAO-A.

p. 77

Table 3.3.5 The interaction energies between safinamide and the active site residues and waters of MAO-B.

p. 82

Table 3.3.6 A virtual library of 17 test compounds (10 known inhibitors and 7 non-inhibitors of MAO-B) was screened with the MAO-B pharmacophore model. Below is given a list of compounds that were found to be hits. The pharmacophore model used was derived from the structure of safinamide. The compounds not shaded are known not to inhibit MAO-B.

p. 85

Table 3.3.7 A virtual library of 17 test compounds (10 known inhibitors and 17 non-inhibitors of MAO-B) was screened with the MAO-B pharmacophore model. Below is given a list of compounds that were found not to be hits. The pharmacophore model used was derived from the structure of safinamide. The compounds not shaded are known not to inhibit MAO-B.

p. 86

Chapter 4

Table 4.7.1 The principal interaction energies of pentamidine with the active site residues and waters of MAO-A.

p. 134

Table 4.7.2 The principal interaction energies of pentamidine with the active site residues and waters of MAO-B.

p. 135

Table 4.7.3 The principal interaction energies of phenformin with the active site residues and waters of MAO-A.

xxv

Chapter 6

Table 6.1.1 The structures of compounds that inhibited MAO-A and/or MAO-B in vitro.