Towards Mosquitocides for Prevention of Vector-Borne Infectious Diseases Discovery and Development of Acetylcholinesterase 1 Inhibitors Sofie Knutsson

Doctoral Thesis, Department of Chemistry Umeå University, 2016

Responsible publisher under Swedish law: the Dean of the Faculty of Science and Technology This work is protected by the Swedish Copyright Legislation (Act 1960:729) ISBN: 978-91-7601-492-9 Electronic version available at http://umu.diva-portal.org/ Tryck/Printed by: VMC-KBC Umeå Umeå, Sweden, 2016

“No one in the 21st century should die from the bite of a mosquito, asandfly, a blackfly or a tick.”

Dr. Margaret Chan

Director-General, World Health Organization

i

Table of Contents

Abstract iii List of Abbreviations v List of Publications ix Mot nya insekticider för bekämpning av sjukdomsbärande myggor - Identifiering och utveckling av acetylkolinesteras 1 inhibitorer xi Introduction 1 

Objectives of the thesis 2 Background 3 

How are vector-borne diseases transmitted? 3 Mosquitoes as vectors 4 Parasitic infections spread by mosquitoes 6 Arboviruses spread by mosquitoes 7 Vector control methods 9 Insecticide resistance 13 AChE as an insecticide target 14 AChEs in mosquitoes 17 

Methodology 21 The discovery and development process 21 Protein-ligand interactions and structure-activity relationships 21 HTS for hit identification 23 Chemical similarity 24 Biochemical evaluation of AChE inhibitors 26 Mosquito testing 29 

Chapter 1: Characterization of AChE1 from An. gambiae and Ae. aegypti (Paper I and Appendix I) 30 

Background 30 Sequence analysis of AChEs from different species 30 Characterization of the enzymes 32 Investigating the active site gorge with inhibitors 34 Cluster analysis 36 Summary of Chapter 1 37 

Chapter 2: Discovery of AChE1-inhibitors by HTS (Paper II and Appendix II) 38 

Background 38 Assay adaptation and screening for AChE1 inhibitors 38 Normalization of AChE1 HTS data 40 Assay quality evaluation 40 Identification and analysis of hits 41 Hit validation 42 Summary of Chapter 2 44 

ii

Chapter 3: Identification of selective AChE1-inhibitors by differential HTS analysis 45 

Selectivity of the AChE1 hits 45 Structure-selectivity relationship 46 Eight compound classes 48 Summary of Chapter 3 50 

Chapter 4: Homology modeling of AChE1 (Paper II and Paper III) 51 Background 51 Selection of a template for modeling of mosquito AChE1 51 Homology models of AgAChE1 and AaAChE1 53 The AgAChE1-G119S mutant 55 Summary of Chapter 4 57 

Chapter 5: Investigation of the structural basis for the selectivity of acetamide-based inhibitors (Paper II and Appendix III) 58 

Background 58 Binding modes in mAChE 58 Design and synthesis of analogues 60 Synthesis of compounds 61 Biochemical evaluation of analogues 64 Binding modes of propyl analogues 68 Structural basis for the observed SARs and SSRs 68 Summary of Chapter 5 72 

Chapter 6: Thiourea-based AChE1 inhibitors (Paper III) 73 Background 73 Design of analogues 74 Synthesis 75 SAR analysis of the thiourea-based AgAChE1- and AaAChE1 inhibitors 76 Structure-selectivity relationship over hAChE 80 The AgAChE1-G119S mutant 80 Study of bioactive conformations in mAChE 81 Study of possible binding modes in mosquito AChE1 using homology modeling 84 Mosquito testing 86 Summary of Chapter 6 88 

Concluding discussion 90 Is there a future for vector control? 90 Covalent or non-covalent inhibitors of mosquito AChE1? 91 Pursuing several compound classes 92 Design of mosquito selective insecticides 94 

Acknowledgements 97 References 100 Appendix I 1 Appendix II 8 Appendix III 12 

iii

Abstract Diseases such as malaria and dengue impose great economic burdens and are a serious threat to public health, with young children being among the worst affected. These diseases are transmitted by mosquitoes, also called disease vectors, which are able to transmit both parasitic and viral infections. One of the most important strategies in the battle against mosquito-borne diseases is vector control by insecticides and the goal is to prevent people from being bitten by mosquitoes. Today’s vector control methods are seriously threatened by the development and spread of insecticide-resistant mosquitos warranting the search for new insecticides. This thesis has investigated the possibilities of vector control using non-covalent inhibitors targeting acetylcholinesterase (AChE); an essential enzyme present in mosquitoes as well as in humans and other mammals. A key requirement for such compounds to be considered safe and suitable for development into new public health insecticides is selectivity towards mosquitos. The work presented here is focused on AChE1 from the disease-transmitting mosquitoes Anopheles gambiae (AgAChE1) and Aedes aegypti (AaAChE1), and their human (hAChE) and mouse (mAChE) counterparts. By taking a medicinal chemistry approach and utilizing high throughput screening (HTS), new chemical starting points have been identified. Analysis of the combined results of three different HTS campaigns targeting AgAChE1, AaAChE1, and hAChE allowed the identification of several mosquito-selective inhibitors and a number of compound classes were selected for further development. These compounds are non-covalent inhibitors of AChE1 and thereby work via a different mechanism compared to current anti-cholinergic insecticides, whose activity is the result of a covalent modification of the enzyme. The potency and selectivity of two compound classes have been explored in depth using a combination of different tools including design, organic synthesis, biochemical assays, protein X-ray crystallography and homology modeling. Several potent inhibitors with promising selectivity for the mosquito enzymes have been identified and the insecticidal activity of a few selected compounds have been confirmed by in vivo experiments on mosquitoes. The results presented here contribute to the field of public health insecticide discovery by demonstrating the potential of selectively targeting mosquito AChE1 using non-covalent inhibitors. Further, the presented compounds can be used as tools to study mechanisms important in insecticide development, such as exoskeleton penetration and other ADME processes in mosquitoes.

iv

v

List of Abbreviations Abbreviations

AaAChE1 Aedes aegypti acetylcholinesterase 1

ACh acetylcholine

AChE acetylcholinesterase

AChE1 acetylcholinesterase 1

AChR acetylcholine receptor

ACT Artemisinin-based combination therapies

ADME absorption, distribution, metabolism, excretion

Ae. aegypti Aedes aegypti

Ae. albopictus Aedes albopictus

AgAChE1 Anopheles gambiae acetylcholinesterase 1

An. gambiae Anopheles gambiae

ATCh acetylthiocholine

BTCh butyrylthiocholine

BuChE butyrylcholinesterase

CAS catalytic site

CHIKV chikungunya virus

Cys cysteine

DDT dichlorodiphenyltrichloroethane

DENV dengue virus

DmAChE Drosophila melanogaster acetylcholinesterase

DMF dimethylformamide

d.n.a. data not available

DNTB 5,5'-dithiobis(2-nitrobenzoic acid)

eq equivalents

Glu glutamic acid

Gly glycine

h hour/hours

hAChE Homo sapiens acetylcholinesterase

His histidine

HTS high throughput screening

IC50 half maximal inhibitory concentration

Ile isoleucine

IRS indoor residual spraying of insecticides

ITN insecticide-treated bed nets

vi

LLIN long lasting insecticidal nets

mAChE Mus musculus acetylcholinesterase

min minute/minutes

MWI microwave irradiation

mya million years ago

n.a. not applicable

n.d. not determined

n.t. not tested

nAChR nicotinic acetylcholine receptor OPLS-DA orthogonal partial least squares-discriminant

analysis

PAINS pan-assay interference compounds

PAS peripheral anionic site

PCA principal component analysis

pdb protein databank PEPPSI-iPr [1,3-bis(2,6-diisopropylphenyl)imidazol-2-

ylidene](3-chloropyridyl)palladium(II) dichloride

Phe phenylalanine

PTCh propionylthiocholine

rt room temperature

S.R. selectivity ratio

SAR structure-activity relationship

Ser serine

SS space spraying of insecticides

SSR structure-selectivity relationship

t.b.d. to be determined TBTU O-(benzotriazol-1-yl)-N,N,N′,N′-tetramethyluronium

tetrafluoroborate

T. californica Torpedo californica

TCDI 1,1′-thiocarbonyldiimidazole

TEA triethylamine

TFA trifluoroacetic acid

THF tetrahydrofuran

Trp tryptophan

Tyr tyrosine

vii

Notations

Enzyme kinetics

E enzyme

ES enzyme substrate complex

I inhibitor

P product

S substrate

kcat turnover number

kcat/KM catalytic efficiency

kf rate constant of forward reaction

ki inhibition constant

KM Michaelis constant

kobs pseudo first-order rate constant

kr rate constant of backward reaction

V rate of reaction

Vmax maximum rate of reaction

Multivariate data analysis

X data matrix

Y response matrix

T score vector matrix

Tp predictive score vector matrix

To orthogonal score vector matrix

t score values

P loading vector matrix

Pp predictive loading vector matrix

Po orthogonal loading vector matrix

p loading values

E residual

R2Y fraction of the Y variation modeled Q2 fraction of Y variation predicted by the model according to cross-

validation

viii

ix

List of Publications This thesis is based on the following papers, which are referred to in the text by Roman numerals in bold.

I. Engdahl, C.; Knutsson, S.; Fredriksson, S. A.; Linusson, A.; Bucht, G.; Ekström, F., Acetylcholinesterases from the Disease Vectors Aedes Aegypti and Anopheles Gambiae: Functional Characterization and Comparisons with Vertebrate Orthologues. PloS one 2015, 10 (10), e0138598.

II. Engdahl, C.;* Knutsson, S.;* Ekström, F.; Linusson, A., Discovery of Selective Inhibitors Targeting Acetylcholinesterase 1 from Disease-Transmitting Mosquitoes. Submitted.

III. Knutsson, S.; Kindahl, T.; Engdahl, C.; Forsgren, N.; Kitur, S.; Ekström, F.; Kamau, L.; Linusson, A., Structure-Activity and Structure-Selectivity Relationships of Thiourea-Based Inhibitors of Acetylcholinesterase 1. Manuscript.

*These authors contributed equally.

x

Papers by the author not included in the thesis.

Pemberton, N.; Pinkner, J. S.; Edvinsson, S.; Hultgren, S. J.; Almqvist, F., Synthesis and Evaluation of Dihydroimidazolo and Dihydrooxazolo Ring-Fused 2-Pyridones - Targeting Pilus Biogenesis in Uropathogenic Bacteria. Tetrahedron 2008, 64 (40), 9368-9376.

Chorell, E.; Edvinsson, S.; Almqvist, F., Improved Procedure for the Enantioselective Synthesis of Dihydrooxazolo and Dihydrothiazolo Ring-Fused 2-Pyridones. Tetrahedron Lett. 2010, 51 (18), 2461-2463.

Chorell, E.; Pinkner, J. S.; Phan, G.; Edvinsson, S.; Buelens, F.; Remaut, H.; Waksman, G.; Hultgren, S. J.; Almqvist, F., Design and Synthesis of C-2 Substituted Thiazolo and Dihydrothiazolo Ring-Fused 2-Pyridones: Pilicides with Increased Antivirulence Activity. J. Med. Chem. 2010, 53 (15), 5690-5695.

Chorell, E.; Pinkner, J. S.; Bengtsson, C.; Banchelin, T. S.-L.; Edvinsson, S.; Linusson, A.; Hultgren, S. J.; Almqvist, F., Mapping Pilicide Anti-Virulence Effect in Escherichia Coli, a Comprehensive Structure-Activity Study. Bioorg. Med. Chem. 2012, 20 (9), 3128-3142.

Chorell, E.; Pinkner, J. S.; Bengtsson, C.; Edvinsson, S.; Cusumano, C. K.; Rosenbaum, E.; Johansson, L. B. A.; Hultgren, S. J.; Almqvist, F., Design and Synthesis of Fluorescent Pilicides and Curlicides: Bioactive Tools to Study Bacterial Virulence Mechanisms. Chem. – Eur. J. 2012, 18 (15), 4522-4532.

Edvinsson, S.; Johansson, S.; Larsson, A., An Efficient Procedure for the Synthesis of Formylacetic Esters. Tetrahedron Lett. 2012, 53 (50), 6819-6821.

xi

Mot nya insekticider för bekämpning av sjukdomsbärande myggor - Identifiering och utveckling av acetylkolinesteras 1 inhibitorer Myggor sprider ett flertal parasit- och virussjukdomar t.ex. malaria, dengue, chikungunya och Zika. I dessa sammanhang kallas myggan för en vektor eftersom den fungerar som en länk mellan det sjukdomsframkallande smittämnet (t.ex. virus eller parasiter) och människan. Idag riskerar mer än hälften av jordens befolkning att drabbas av myggburna sjukdomar som ofta saknar vaccin eller specifik behandling. Myggan klassas därför ofta som världens farligaste djur. En viktig metod för att förebygga dessa sjukdomar är så kallad vektorkontroll med hjälp av insekticider (bekämpningsmedel mot insekter). Användandet av insekticider, t.ex. på impregnerade myggnät, leder till färre myggor, minskad risk för myggbett och därmed även minskad sjukdomsspridningen. Nu hotas effektiviteten av dessa metoder på grund av resistensutveckling bland myggor. Resistens innebär att myggorna inte är känsliga för bekämpningsmedlen längre, och detta är idag ett problem med alla insekticider som är godkända för vektorkontroll. Behovet av nya insekticider som är säkra att använda i närhet av människor är därför stort.

I denna avhandling presenteras forskning som syftar till att undersöka möjligheterna för vektorkontroll genom små organiska föreningar som hindrar funktionen av ett livsnödvändigt protein hos myggan som kallas acetylkolinesteras 1. Då acetylkolinesteras även finns hos människor och andra djur är det viktigt att nya acetylkolinesteras 1 hämmare är specifika för myggproteinet, d.v.s. att de endast påverkar funktionen hos myggproteinet men inte det mänskliga proteinet. Genom att experimentellt utvärdera flera tusen organiska substansers förmåga att blockera den biologiska funktion hos acetylkolinesteras 1 från två sjukdomsspridande myggor hittade vi ett antal substanser som var intressanta för vidare utveckling mot nya insekticider. De identifierade substanserna visade sig i ett tidigt skede vara selektiva för myggproteinet vilket möjliggjorde studier för att undersöka vilka faktorer som påverkar substansernas selektivitet. Ny substanser har därefter framställts genom organisk syntes och deras biologiska aktivitet har undersökts för att utröna hur substansernas uppbyggnad (kemiska struktur) relaterar till deras förmåga att selektivt hämma myggans acetylkolinesteras 1. För att understödja våra resultat har vi även tagit hjälp av metoder för att titta på den tredimensionella protein strukturen av acetylkolinesteras (både röntgenkristallografi och datorbaserade modelleringstekniker) där vi undersökt hur substanserna binder till proteinet och där igenom kan hämma dess funktion.

xii

De arbeten som presenteras i den här avhandlingen visar på möjligheten och potentialen i att utveckla selektiva insekticider mot acetylkolinesteras 1, något som inte finns på marknaden idag. De substanser som presenteras här skiljer sig från dagens insekticider, både vad det gäller kemisk struktur och verkningsmekanism, vilket kan vara fördelaktigt ur ett resistensutvecklingsperspektiv. Initiala försök på myggor har visat lovande resultat, men mycket forskning kvarstår ännu för att reda ut hur framtida säkra och effektiva insekticider ska utformas.

1

Introduction The latter half of the 1800s constituted an era of intense research and groundbreaking discoveries that laid the foundation for modern medicine. This included the revolutionary work of microbiologists Pasteur and Koch, whose work resulted in the discovery and experimental proof that microorganisms, such as bacteria and fungi, were able to cause disease. The realization that mosquitoes had the ability to transmit disease followed shortly after the establishment of the germ theory of disease. In 1880, the French army surgeon Alphonse Leveran was the first to identify protozoan parasites in the blood of a malaria patient and 18 years later Ronald Ross demonstrated that malaria was transmitted between birds by Anopheles mosquitoes.1-2 Soon thereafter, in 1900, following the work of Carlos Finley, the American army surgeon Walter Reed and his commission showed that yellow fever was transmitted via mosquitoes of the Aedes genus.3-5 Both Leveran (1907) and Ross (1902) were awarded the Nobel Prize for their discoveries.6-7

Today, organisms capable of transmitting infectious diseases between humans, or from animals to humans, are called vectors and the diseases they spread are known as vector-borne diseases. This group of diseases constitutes 17% of all infections worldwide8 and imposes immense health and economic burdens,9-10 especially in developing countries where young children are among the worst affected.11 The best known and studied vectors are the mosquitoes, which transmit both parasitic and viral diseases; however, diseases can also be transmitted by other animals such as flies, ticks, or snails. As this thesis is being written, the news is filled with alarming reports from South America, where outbreaks of Zika virus infections are being linked to a dramatic increase in children born with microcephaly.12-15 However, the Zika virus is just one example of the many pathogens which are transmitted to humans via the bite of an infected mosquito. There are a number of viral infections that are emerging or resurging. Dengue is one of the fastest spreading vector-borne diseases and it is estimated that 40% of the world’s population is at risk.16-17 It is encouraging to note that following a massive global effort over the last fifteen years the incidence of malaria has declined. Despite this, malaria is still the deadliest vector-borne disease and 1.2 billion people remain at high risk of infection.9, 18

With many mosquito-borne diseases lacking vaccines and/or specific treatments, the current strategy to reduce their incidence aims to prevent people from being bitten by mosquitoes. This strategy is termed vector control and relies on two main methods; indoor residual spraying of insecticides (IRS) and insecticide-treated bed nets (ITNs). IRS is used to reduce insect vectors in homes by spraying walls and furniture with insecticides. ITNs are mosquito nets that have been impregnated with a

2

pyrethroid insecticide; they were introduced in the 1980s. By sleeping under such a net, people are protected from mosquitoes which are night active and feed indoors. ITNs are cost-effective and can protect against malaria infection very efficiently.9, 18 In addition, insecticides are frequently used for outdoor space spraying (SS) and larviciding, i.e. killing mosquito larvae.19-20 Concerns such as safety for humans and the environment, durability, and costs have resulted in only four classes of insecticides being recommended for ITNs and IRS. This limited access to chemically diverse insecticides, in combination with heavy use and the fact that no new active ingredients have been brought to the market since lambda-cyhalothrin was registered in 1988, has allowed the development and spread of insecticide-resistant mosquito populations.21-25

Objectives of the thesis There is an urgent need for new insecticides for vector control in order to meet the demands posed by both insecticide resistance development and the spread of vectors to new areas and environments. In our work, we want to explore the possibilities of blocking the function of the essential enzyme acetylcholinesterase 1 (AChE1) by use of non-covalent inhibitors. AChE1 is a validated insecticide target. However, since AChE is an essential enzyme in mammals as well as in insects,26-27 a major challenge is to develop inhibitors that target the mosquito enzyme while being safe for humans and other non-target organisms.

The work presented in this thesis is focused on AChE1 from the disease-transmitting mosquitoes Anopheles gambiae (An. gambiae) and Aedes aegypti (Ae. aegypti), and their human (hAChE) and mouse (mAChE) counterparts. The thesis is intended to add to the understanding of what structural elements, both from a ligand- and an enzyme perspective, contribute to the potency and selectivity of non-covalent AChE1 inhibitors. The work outlined in Chapter 1 aimed to identify differences between AChE1 from An. gambiae (AgAChE1) and Ae. aegypti (AaAChE1), and vertebrate hAChE and mAChE that might be of importance for the development of selective insecticides. Thus, we studied and compared the protein sequences, the functional properties, and the ligand binding properties of AChE1 and vertebrate AChE. In Chapters 2 and 3 the results of three different high throughput screening (HTS) campaigns targeting AgAChE1, AaAChE1, and hAChE are analyzed. The aim of this study was to identify novel inhibitors that exhibit a high potential for being selective towards mosquito AChE1 over hAChE. In addition, we also wanted to investigate general differences between the enzymes based on the chemical structures of the identified hits. Chapters 4-6 describe our efforts to investigate and establish structure-activity relationships (SAR) and structure-selectivity relationships (SSR) of possible lead compounds.

3

Background

How are vector-borne diseases transmitted? The most common vectors are flies, ticks, snails, and mosquitoes, with the latter being the focus of this thesis. Transmission of vector-borne diseases can be described by different cycles. The epidemic (or urban) cycle describes how a disease is transmitted from human to human via a vector (A, Figure 1). In addition, some diseases also have an enzootic cycle (sometimes called a sylvatic or jungle cycle) which describes how the disease is transmitted between animals (B, Figure 1). As a result animals (often non-human primates) can act as reservoirs for a disease and humans can be infected via a spillover from the enzootic cycle (C, Figure 1).

Figure 1. Transmission of vector-borne diseases. In the epidemic cycle the pathogen

is carried from one human host to another via its vector (A). In the enzootic cycle the

disease is established in an animal population. For diseases affecting humans, the

animals are often non-human primates (B). Humans can become infected when

exposed to the enzootic cycle (C).

The recent surge in several vector-borne diseases has been linked to

environmental factors such as climate and weather changes, but also to changes in human behavior. A warmer climate in combination with increased travel and global trade have allowed vectors to populate new geographical areas.28-31 Rapid urbanization, high population density, inadequate housing and insufficient sanitation create environments favoring close contacts between pathogen, vector, and people. Even so, the complex transmission dynamics of vector-borne diseases, determined by

4

environmental, biological, and social factors, are not yet fully understood.28,

30, 32-33 Given the immense complexity of vector-host-pathogen relationships and transmission dynamics, a combination of several interventions, such as vaccines, specific treatments (antivirals and antimalarials) and vector control methods, are needed to achieve long term and large scale reductions in vector-borne diseases.9, 34

Mosquitoes as vectors Insects are the most diverse group of animals and throughout evolution they have lived in close relationships with both plants and mammals. One example of such a relationship is hematophagy – the strategy of feeding on blood - which occurs in many species across several different families of insects. The Culicidae family, or mosquitoes, are insects of the order Diptera and comprise just over 3,500 different species.35 Female mosquitoes depend on blood meals to obtain the necessary nutrients for egg production and reproduction. The oldest known mosquito fossils date as far back as to the Mesozoic Era (252-66 million years ago; mya)36-37 and a 46 million year old blood-engorged specimen shows that this behavior extends far back into the past.38

Hematophagy is a key requirement for the establishment of a second relationship: that between insects and disease agents such as viruses and parasites. Blood-feeding species can become contaminated with the disease-causing microorganisms during a blood meal and, if conditions are right, subsequently transfer the disease from one infected host to another previously uninfected one. To date, the earliest evidence of the parasite-host relationship between the malaria parasite (Plasmodium) and mosquitoes is of a fossil found in Dominican amber (uncertain dating, current estimates range from 20-15 mya to 45-30 mya).39 Table 1 lists some mosquito-borne diseases and their vectors,11, 33 of which Anopheles and Aedes mosquitoes are the focus of this thesis.

5

Table 1. Overview of different mosquito-borne diseases and their mosquito vectors.

Genus Disease Pathogen People at riska

Anopheles Malaria Parasite 3.2 billion

Lymphatic filariasis Parasite 1.2 billion

Aedes Dengue Flavivirus 2.5 billion

Chikungunya Alphavirus d.n.a.

Zika Flavivirus d.n.a

Yellow fever Flavivirus 900 million

Lymphatic filariasis Parasite 1.2 billion

Culex West Nile fever Flavivirus d.n.a

Japanese encephalitis Flavivirus 3 billion

Rift Valley fever Phlebovirus d.n.a

VEEb Alphavirus d.n.a

Lymphatic filariasis Parasite 1.2 billion

Haemagogus Yellow fever Flavivirus 900 million

Mayaro virus disease Alphavirus d.n.a

aData from WHO fact sheets40-44 d.n.a. = data not available bVenezuelan equine

encephalitis

The mosquito life cycle The bionomics of the mosquito, i.e. its habitat, behavior, and preferences, determine its vector competence. A mosquito passes through four different stages during its life cycle: egg, larvae, pupae, and adult. The time it takes from egg to adult varies depending on several factors such as species and temperature; for Ae. aegypti and An. gambiae typical timeframes are 8-10 days and 9-29 days, respectively.

Shortly after the adult female emerges she will mate and thereafter look for a blood meal. One mating is enough to allow the female to lay eggs throughout her lifetime, however a new blood meal is required for every oviposition. The biting behavior varies between different species of mosquitoes. Descriptions of both An. gambiae and Ae. aegypti indicate that they prefer to feed from humans (they are anthropophilic), and primarily bite and rest indoors (endophagic and endophilic). However, their biting cycles, i.e. the time of day when a mosquito mainly takes her blood meal, differ. An. gambiae is night active while Ae. aegypti is day active and bites more frequently. 45-46 However, it has been suggested that An. gambiae might be less specific and more opportunistic with respect to its host preferences and resting places.47-49

An. gambiae mosquitoes are known to lay their eggs in both natural and manmade water bodies. These can vary from large bodies such as at the edges of lakes, rice fields, or fish ponds, to small streams, wells, or temporary puddles, such as tire marks or hoof prints, which are exposed to sunlight.20,

6

50 Ae. aegypti have become increasingly domesticated and often lay their eggs in any manmade container that can hold water, such as flower pots, car tires, buckets, or rain gutters. Their eggs are very robust and can survive desiccation for at least six months.51 After the eggs hatch into larvae they go through four developmental stages, called instars, before transforming from larva to pupa. This final stage before the adult emerges is an inactive state and lasts 2-5 days for An. gambiae and 2-3 days for Ae. aegypti.20, 52 The differences in habitat and behavior between the species will not only affect the epidemiology of the diseases that they transmit, but are important factors to consider when developing and choosing vector control strategies to achieve the most effective disease prevention.

Parasitic infections spread by mosquitoes Parasites live in or on a host and depend on their host for nutrition. Protozoa (unicellular organisms), helminths (worms), and ectoparasites (arthropods such as lice and ticks) are the three main classes of parasites able to cause disease in humans. The helminths and protozoan parasites transmitted to humans via the bites of infected female Anopheles mosquitoes are discussed below.

Malaria In 2015 it was estimated that 214 million malaria infections occurred resulting in 438,000 deaths mostly among children under the age of five.9 Five different species of the Plasmodium parasite (protozoa) cause malaria in humans; P. falciparum, P. vivax, P. malariae, P. ovale, and P. knowlesi.40 P. falciparum is the cause of the most severe and deadly malaria infections (Figure 2). Today malaria is most often treated with Artemisinin-based combination therapies (ACT)53-54 and vaccines are being developed. 55-58

More than 80% of all malaria cases occur in Africa, where An. gambiae is considered the major vector.9 Other important vectors include Anopheles funestus (Africa), Anopheles stephensi (Asia) and Anopheles darlingi (Americas).50, 59-60 P. knowlesi differs from the other Plasmodium species by being transmitted via a enzootic cycle rather than an epidemic cycle.61

Even though the incident rate of malaria is still incredibly high, there has been a drastic global reduction in the number of infections and deaths since the year 2000.9, 18 This decrease of approximately 37% and 60% (when taking population growth into account) for incidence and mortality rates, respectively, is mainly the result of massive global initiatives to increase vector control coverage in malaria endemic countries.18

7

Figure 2. Plasmodium life cycle in man and mosquito. When a human is bitten by

an infected mosquito, the protozoan parasite is transmitted to its new host in the

form of sporozoites. The sporozoites travel to the liver where they mature into

schizonts. When the schizonts rupture, merozoites are released into the bloodstream

and infect the erythrocytes. This is where the parasites undergo asexual

multiplication (asexual blood cycle). Some merozoites will, however, differentiate

into a sexual stage and form male and female gametocytes. If the gametocytes are

ingested by a mosquito they develop into gametes which fertilize in the mosquito gut

to form zygotes. Zygotes develop into ookinetes which are able to penetrate the gut

wall before forming oocysts. After growth and division, sporozoites break out from

the oocyst and travel to the salivary glands from where they infect a new host upon

the next blood feed.62

Lymphatic filariasis Most cases of lymphatic filariasis are caused by the nematode parasite Wuchereria bancrofti.63 Culex, Anopheles, and Aedes mosquitoes are vectors of the parasite and it is estimated that at least 120 million people are currently suffering from lymphatic filariasis, mostly in South East Asia and Africa.11 Approximately one third of those infected are permanently disfigured by the infection because of damage to the lymphatic system. Elephantiasis, lymphedema, and scrotal swelling are the worst symptoms of the chronic disease, resulting in misery and suffering due to pain, social exclusion, financial losses, and poverty.

Arboviruses spread by mosquitoes Viruses that are dependent on blood-feeding arthropods such as mosquitoes, ticks, and flies, are called arboviruses (arthropod-borne). Today just over 500 known, probable or possible arboviruses are listed in the arbovirus catalogue and one third of these are able to cause disease in humans.64-65 The most common mosquito-borne viruses affecting human health are RNA viruses belonging to the Alphavirus and Flavivirus genera. The major vector

8

of these viruses is Ae. aegypti but other species such as Aedes albopictus (Ae. albopictus) can also be vectors.66 The viruses spread by Aedes mosquitoes cause febrile illnesses, with symptoms often including joint pain and skin rash, and it is often difficult to distinguish between the different infections based on their symptoms.

Yellow fever virus Yellow fever was a serious global health problem up until the mid-1900s.67 A vaccine against yellow fever was developed during the 1930s,68 but today yellow fever is still endemic in some parts of Africa and South America.11, 69 Ae. aegypti is often called the yellow fever mosquito due to its role in transmitting this disease. The virus has three different transmission cycles, one urban cycle between humans, one jungle/sylvatic cycle (enzootic cycle) between non-human primates and one intermediate/savannah cycle between human and non-human primates.42

Dengue viruses Dengue is considered the most important arboviral disease and the incidence of the disease has increased dramatically over the last 50 years.46 It has been estimated that 290-530 million cases of dengue occurred globally in 2010, with Asia carrying the heaviest burden (70%) followed by Africa (16%) and the Americas (14%).17 The great increase in both number of cases and endemic countries is attributed to the geographical spread of the major vector Ae. aegypti.66, 70-73

Dengue ranges from the relatively common asymptomatic or mildly symptomatic infections to the more severe forms known as dengue hemorrhagic fever and dengue shock syndrome.74 The infections are caused by one of four closely related dengue virus serotypes (DENV1-4).75 There are currently no antiviral treatments for dengue and vaccine developments have been challenged by the four virus serotypes.76-84 Still, one dengue vaccine has recently been registered in four countries and in early 2016, a public immunization program was launched in the Philippines.85

Chikungunya virus Chikungunya and Zika are both emerging viral infections that have recently seen a large increase in outbreaks. The chikungunya virus (CHIKV) manifests with fever, rash, muscle pain, and debilitating joint pains, which can persist for years; patients often require medical attention.86 There are no specific antiviral drugs or vaccines available. Chikungunya was first reported in Africa at the beginning of the 1950s and a few years later in Asia.87-88 Since 2004 a major increase in both the number and size of outbreaks has been recorded and chikungunya is now considered endemic and epidemic in Africa, Asia and the Indian Ocean.89 In addition to Ae. aegypti, Ae.

9

albopictus is an important vector and this has allowed chikungunya to spread to new geographic areas.90 Since 2007 and 2013, respectively, local transmission of chikungunya has been recorded in both Europe and the Americas.91-92

Zika virus Compared to the viruses discussed above, much less is known about the Zika virus. Zika infections are typically mild with symptoms such as fever, rashes, headache, conjunctivitis, joint and muscle pain. However, an increase in Guillain-Barré syndrome was noted during Zika outbreaks in French Polynesia (2013)93-96 and Brazil (2015).15 In addition, a 20-fold increase in children born with microcephaly in Brazil is of great concern.12-15 Still, more evidence is needed to establish conclusively the correlation between Zika virus infections and neurological disorders.

The first reported outbreak of Zika outside Africa or Asia occurred in 2007 on Yap in the Western Pacific Ocean and involved 49 confirmed cases.97 Larger outbreaks then occurred on other Pacific Islands from 2013 and South America from 2014. It has been estimated that 0.5-1.5 million people were infected with Zika virus in Brazil between late 2014 and early 2016. At the beginning of 2016, transmission of Zika virus had been documented in 61 countries.15 The virus is believed to be transmitted by Aedes mosquitoes,98-101 but further investigation is needed into possible vectors of the virus and vector statuses.102

Vector control methods Most vector control methods aim to reduce the disease burden by decreasing the vector density. The choice and effectiveness of different vector control methods depend on several factors including mosquito behavior and habitat, disease epidemiology, human activity, cost, socioeconomic conditions, acceptance, and cultural context.11 Different vector control interventions also target different stages of the mosquito life cycle.

Insecticides are one of the key pillars in vector control both for immature control (larviciding) and adult control (ITNs, IRS, and SS).11, 19, 46 During the ten year period 2000-2009, over 6000 tons of insecticides were used annually for vector control purposes (Figure 3).19 Only a few chemical classes of insecticides are recommended for this103-106 and most of these target the nervous system of the mosquito. Organochlorines and pyrethroids primarily target sodium gated ion channels.107 Organophosphates and carbamates inhibit the essential enzyme AChE, and spinosyns target the nicotinic acetylcholine receptor (nAChR).108 Both AChE and nAChR are vital for cholinergic neurotransmission. The last category of public health insecticides are insect growth regulators, such as juvenile hormone mimics

10

(pyriproxyfen) and chitin biosynthesis inhibitors (benzoylureas),107 which are only recommended as larvicides.

Figure 3. Pie charts showing the proportions of insecticides used annually for vector

control purposes during 2000-2009 (in tons of active ingredients). The yearly

average use was 6248 tons of insecticides.19 A) The proportions of different chemical

classes: organochlorines (OC), organophosphates (OP), carbamates (C), and

pyrethroids (P). It is noteworthy that pyrethroids constitute ~80% of the insecticide

use in terms of spray coverage (only ~1/60 of the active ingredient (weight)) is

needed of pyrethroids to achieve the same coverage as organophosphates and

carbamates. B) The proportions used for different vector control methods: larviciding

(L), indoor residual spraying (IRS), space spraying (SS), and insecticide-treated bed

nets (ITN).

Immature control Immature vector control refers to methods targeting the aquatic stages of the mosquito. Reducing mosquito density by interfering with breeding habitats is especially important for controlling Aedes mosquitoes that breed in close proximity to humans,46 but Anopheles populations can also be reduced by larval source management.20 Prior to the advent and large scale use of synthetic insecticides in the 1950s,109 environmental management to reduce larval habitats was the primary method for preventing malaria, yellow fever, and dengue. In fact, from around 450 BCE and through the Middle ages Romans and other early civilizations performed sanitation work by diverting streams and draining marshes and this continued to modern times.110

Today, effective control is often the result of a combination of strategies including habitat modification and manipulation, biological control, and larviciding. Examples include flushing of streams, proper waste

11

management, removing and/or preventing access to water storage containers, introducing larvivorous fish or other larvivorous predators, and planting aquatic plants that hinder larvae from breathing at the water surface.20 Larviciding is performed by treating water with oils or surface films,111 biocides, or insecticides. Two bacterial larvicides and eight larvicidal compounds are recommended for vector control (Table 2).106 Treating drinking water with insecticide is typically considered a last resort and should be implemented with extreme care to achieve a balance between the human intake of the insecticide from drinking-water and the control of larvae.

Table 2. Insecticides recommended for larviciding.106

Insecticidea Classb Targetc Methodd

Chlorpyrifos (1965) OP AChE OWB

Temephos (1965) OP AChE OWB/CB/DW

Pirimiphos-methyl (1967) OP AChE OWB/CB

Fenitrothion (1975) OP AChE OWB

Diflubenzuron (1975) B Chitin biosynthesis OWB/CB/DW

Novaluron (1996) B Chitin biosynthesis OWB/CB/DW

Pyriproxyfen (1995) JHM unknown OWB/CB/DW

Spinosad (1997) S nAChR OWB/CB/DW

aYear of introduction or first report given in parentheses; bOP = organophosphate, B

= benzoylureas, JHM = juvenile hormone mimics, S = spinosyns; cAChE =

acetylcholinesterase, nAChR = nicotinic acetylcholine receptor; dOWB = open water

bodies, CB = container breeding, DW = drinking water.

12

Adult control Spraying the insides of houses with residual insecticides (IRS) to reduce malaria transmission was first implemented in the early 1930s.112 Between 1940 and 1960 residual spraying with DDT greatly reduced and even eliminated malaria in many countries. In 1955 the WHO launched the Global Eradication Program, but in 1969 this program was discontinued and thus malaria resurged in many areas.113 As a result of, inter alia, environmental awareness and resistance development in the late 1980s and 1990s, DDT was replaced by pyrethroids for IRS,112 however resistance to the pyrethroids led to the re-introduction of DDT during the 2000s in many African countries. Today, IRS is the most widely used method to control adult mosquitoes, but only four classes of insecticides are recommended for IRS (Table 3).104 To obtain maximum protection at least 80% of the houses in an area need to be treated and the insecticide needs to be applied 1-4 times per year.11

Spraying insecticides in and around houses is known as perifocal or peridomestic space spraying. This method is often used to control Ae. aegypti during dengue outbreaks, however its effectiveness has been questioned.114-115 During space spraying, an insecticide is launched into the air as an aerosol, with the aim of killing adult mosquitoes; it can be applied in two forms: cold or thermal fog. Insecticides usually need to be reapplied every few days, however the effectiveness depends on both weather conditions and environmental factors such as street and building layouts. The five insecticides recommended for space spraying are listed in Table 3.103

ITNs were first developed during the 1980s116 and today, together with long lasting insecticidal nets (~3 years durability; LLINs), are the most important method for malaria prevention.9, 18 Because of low human toxicity, rapid knockdown effect, low cost, and longevity, only pyrethroids are currently recommended for the treatment of bed nets (Table 3). However, proper distribution, user compliance and net maintenance are important factors for the effectiveness of the LLINs.117-120

13

Table 3. Insecticides recommended for adult vector control showing the chemical

structure of one insecticide from each class.103-105

Insecticidea Classb Targetc Methodd

DDT (1939) OC SGIC IRS

Malathion (1952) OP AChE IRS/SS

Pirimiphos-methyl (1967) OP AChE IRS

Fenitrothion (1975) OP AChE IRS

Propoxur (1959) C AChE IRS

Bendiocarb (1971) C AChE IRS

Permethrin (1973) P SGIC ITN/SS

Deltamethrin (1974) P SGIC ITN/IRS/SS

Etofenprox (1981) P SGIC ITN/IRS

alpha-Cypermethrin (1985) P SGIC ITN/IRS

Bifenthrin (1985) P SGIC IRS

trans-Cyphenothrin (1986) P SGIC SS

Cyfluthrin (1987) P SGIC ITN/IRS

lambda-Cyhalothrin (1988) P SGIC ITN/IRS/SS

aThe year of introduction or EPA registration given in parenthesis; bOC =

organochlorine, OP = organophosphate, C = carbamate, P = pyrethroids; cSGIC =

sodium gated ion channels, AChE = acetylcholinesterase; dIRS = indoor residual

spraying, SS = space spraying, ITN = insecticide-treated bed net, in this table ITN

also includes long lasting insecticidal net (LLIN).

Insecticide resistance One of the major threats to sustainable vector control is the rapid development of insecticide resistance in mosquitoes.24, 108, 121 Resistance has been defined by the Insecticide Resistance Action Committee as “the selection of a heritable characteristic in an insect population that results in the repeated failure of an insecticide product to provide the intended level of control when used as recommended”.108 Four types of resistance mechanisms are commonly considered relevant to current vector control strategies: target-, metabolic-, behavioral-, and cuticular resistance. Behavioral resistance,122-125 e.g. changes in when and where the mosquito takes its blood meal, and cuticular resistance, 126-128 i.e. a thickening of the

14

mosquito exoskeleton leading to reduced penetration of the insecticide, are not yet fully understood and will not be further discussed in this thesis.

Target site resistance Target site resistance involves structural modifications of the insecticide’s molecular target, which result in reduced affinity for the insecticide and ultimately allows the mosquito to evade the effect of the insecticide and survive.108 Mutations in the insecticide binding site, i.e. the exchange of one amino acid for another, can lead to insecticide insensitivity.21 However such resistance mutations can also impair the original function of the target protein. The negative effects of the target site resistance on the life history traits of the mosquito are referred to as the fitness cost. Target site resistance is widely spread across different mosquito species and has been recorded for both voltage gated sodium channels (pyrethroids and DDT)129-135 and AChE (organophosphates and carbamates).136-140

Metabolic resistance The most common form of insecticide resistance is metabolic resistance, where the mosquito adapts and becomes more efficient at detoxifying insecticides. Altered activity, due to gene up-regulation or mutations in the amino acid sequence, of three main metabolic enzyme systems have been implicated in insecticide resistance.21, 24 Cytochrome P450-dependent monooxygenases, which are capable of oxidizing a wide variety of substrates, are very important for metabolism of pyrethroids, DDT, carbamates and organophosphates.141 Monooxygenases belonging to the CYP6 and CYP9 families, among others, are known to be overexpressed in both An. gambiae142-145 and Ae. aegypti.146-148 Detoxification of pyrethroids and organophosphates can also be achieved by mosquito esterases.146-147, 149 Esterases hydrolyze esters to alcohols and carboxylic acids, but can also hinder organophosphates from reaching their target (AChE) by sequestering the insecticide.150-151 The last enzyme system to be addressed here are the glutathione S-transferases.152-153 These enzymes facilitate excretion of lipophilic compounds by conjugation of glutathione, thereby making them more hydrophilic. Glutathione S-transferases are mainly important for metabolism of DDT and carbamates.154-157

AChE as an insecticide target The work presented in this thesis is focused on the validated insecticide target AChE; the insecticidal activity of both organophosphates and carbamates is the result of AChE inhibition.158-159

15

Function and structure of AChE AChE (E.C. 3.1.1.7) is an essential enzyme belonging to the serine hydrolase superfamily. In animals and humans, AChE terminates cholinergic nerve signaling by hydrolyzing the neurotransmitter acetylcholine (ACh) to acetic acid and choline (Figure 4).26-27 The hydrolysis of ACh by AChE is remarkably specific, rapid, and efficient.160-161 Inhibition of AChE leads to continuous nerve signaling due to accumulation of ACh in the synaptic cleft, and eventually to paralysis and death of the organism (Figure 4).

AChE has been thoroughly studied and the structures of the enzyme from Pacific electric ray,162 mouse,163 fruit fly,164 and human165 have been determined by X-ray crystallography. AChE is an α/β protein consisting of a 12-stranded central mixed β-sheet and 14 α-helices (Figure 5a). The enzymatic hydrolysis of acetylcholine by AChE takes place at the bottom of a 20 Å deep gorge, where the catalytic triad consisting of residues Ser203, Glu334, and His447 (hAChE numbering is used throughout this thesis unless otherwise is stated) are located (Figure 5b).162 Many aromatic residues are found both lining the narrow gorge and by its entrance (the peripheral anionic site, PAS), and are believed to be important for transient binding of cationic ligands prior to their diffusion down to the catalytic site (CAS).166 During the hydrolysis, the catalytic serine residue performs a nucleophilic attack on the carbonyl carbon of acetylcholine (Figure 4b, 1). The negative carbonyl oxygen thus formed is stabilized by the backbone NH of Gly121, Gly122, and Ala204 (oxyanion hole) while the acyl is placed in the acyl binding pocket made up of the aromatic side chains of Trp236, Phe295, Phe297, and Phe338 (Figure 5b).167-168 The acyl pocket has been shown to be important for substrate specificity.169 The positively charged amine of acetylcholine forms activated CH···arene interactions mainly with the indole of Trp86 in the choline binding site (Figure 5b). Following cleavage of the bond between acetate and choline (Figure 4b, 2), a water molecule completes the hydrolysis by deacylating Ser203 (Figure 4b, 3-4).

Reported inhibitors of vertebrate AChEs exhibit different mode of action, great structural diversity and span a large physicochemical property space.170-171 Non-covalent AChE inhibitors such as propidium iodide172 have been shown to bind in PAS, while huperzine173 and galantamine174 bind in CAS. There are also examples of ligands, such as bis-tacrine175-176 and donepezil,177 that span the length of the gorge and thereby bind simultaneously to both PAS and CAS.

16

Figure 4. (a) Illustration of cholinergic neurotransmission. When an action

potential reaches the synapse, acetylcholine is released from the presynaptic nerve

into the synaptic cleft. The impulse is transmitted to the postsynaptic nerve upon

binding of acetylcholine to the acetylcholine receptors (AChR). AChE terminates the

neurotransmission by hydrolyzing acetylcholine to acetate and choline (i). Inhibition

of AChE results in accumulation of acetylcholine in the synaptic cleft and continuous

nerve signaling which may be fatal to the organism (ii). (b) Hydrolysis of

acetylcholine by AChE. 1) Nucleophilic attack of Ser203 on carbonyl carbon of

acetylcholine. 2) Cleavage of choline. 3) Nucleophilic attack of H2O on acylated

Ser203. 4) Deacylation and release of acetic acid.

Figure 5. The structure of hAChE (pdb code: 4EY4). (a) The secondary structure

shown as a ribbon with the deep and narrow active site gorge shown by a molecular

surface. (b) Close up of the active site gorge. Amino acids (hAChE numbering) are

colored according to the different subsites: the catalytic triad is shown in green, the

choline binding site in yellow, the acyl binding pocket in red, and PAS in blue.

17

AChEs in mosquitoes Mosquitoes and many other insects have two genes encoding AChE enzymes; ace-1 and ace-2.178-180 The two genes are believed to be the result of an old duplication taking place before the radiation of arthropods.179 Notably, true flies such as Drosophila melanogaster and Musca domestica only have one gene (ace-2), probably due to an evolutionary loss.180 In mosquitoes, the enzyme encoded by ace-1 (AChE1) has the greater catalytic activity.181 In An. gambiae, AChE1 and AChE2 have 53% amino acid sequence identity,179 however the function of AChE2 is still not fully understood.182-183 The ace-1 and ace-2 gene sequences from both An. gambiae183-184 and Ae. aegypti185-186 have been determined and recombinant AgAChE1 has been expressed and characterized.187-188

Organophosphates and carbamates as inhibitors of AChE1 The organophosphates used for vector control are phosphorothioates which undergo oxidative desulfurization (P=S to P=O) by CYP enzymes in the mosquito in order to reach their full insecticidal potential; the resulting oxon analogues inhibit AChE more strongly than the parent compound.141, 189-190 Organophosphates and carbamates exert their insecticidal action by covalent modification of the catalytic serine in a similar fashion to the acylation described in Figure 4b. In the case of carbamates, the process is reversible and enzyme activity can be restored over time.158 The phosphonylated serine, on the other hand, can undergo a process known as aging, where a second leaving group is cleaved from the phosphorus resulting in an irreversibly inhibited enzyme.191

The conserved function of AChE in insects and mammals has given rise to concerns about the safety of organophosphate- and carbamate-based insecticides.192-193 Several studies indicate that organophosphate exposure has negative effects on prenatal development.194-198 In addition, organophosphate pesticide self-poisoning is a serious problem worldwide, especially in developing countries.199-202

Target site resistance in mosquito AChE1 In mosquitoes, AChE1 is the enzyme isoform responsible for AChE-mediated insecticide resistance.136-138 The most frequent mutation is a glycine to serine substitution in the oxyanion hole known as G119S (for this mutation T. californica AChE numbering is used throughout this thesis to facilitate comparison with the literature; corresponding to G121S using hAChE numbering). This mutation has been independently detected several times in both Anopheles203-206 and Culex mosquitoes,136, 207 but not Aedes. This is probably due to there being a different codon for the glycine in Aedes.138 The G119S mutation confers resistance to both organophosphate- and

18

carbamate-based insecticides. It has been shown that the G119S mutation leads to a decrease in the enzymatic activity of AChE1137, 178, 187, 208 and the mutation has been associated with fitness costs for several of the mosquito’s traits.209-213 Duplication of the ace gene can compensate for the decreased activity and ease the fitness cost.214-216 Other mutations detected in the mosquito AChE1 active site that have been linked to insecticide resistance include F290V and F331W (T. californica AChE numbering, corresponding to F297V and F338W in hAChE, respectively). 139-140

Ongoing research on AChE1 as a target for new insecticides Public health insecticide development is challenged by the need to address issues related to both human and environmental safety, and insecticide resistance. Are AChEs from different species sufficiently diverse to allow the development of mosquito selective compounds that leave the cholinergic transmission in off-target species unaffected? As will be discussed further in Chapter 1, sequence alignments of AChEs from different species have shown that, even when the overall amino acid sequence identity is 30-40%, the active site gorge and overall fold of different AChEs are highly conserved.217-220 Despite this, some specific differences between insect AChE1 and vertebrate AChE have been noted. In AChE1, Phe295 in the acyl pocket is exchanged for a cysteine217-218 and Tyr72 located in PAS is exchanged for Ile.220 Both Phe295 and Tyr72 are known to interact with ligands of vertebrate AChE.176, 221-223 There are currently no crystal structures available for any insect AChE1, 3D structures of AgAChE1 have, however, been developed by computational methods (homology modeling).137, 218, 220 The models show that one of the loops in PAS (loop 1, Figure 5), could be of particular interest. This loop is shorter compared to vertebrate AChE and affects the structure of the gorge’s rim. Homology modeling of mosquito AChE1 is described in Chapter 4.

In comparison to the numerous studies on vertebrate AChEs, the scope of mosquito AChE1 inhibitors is much less explored. The majority of the reports focus on covalent inhibitors, which have been investigated for the development of insecticides with improved selectivity.220, 224-228 One strategy being explored is to target the free cysteine present in insect AChE1 (corresponding to Phe295 in hAChE) with covalent inhibitors (10 and 11, Figure 6).217-218, 225, 229 Likewise, studies aiming to improve potential human safety by re-designing aryl methyl carbamates that covalently modify the catalytic serine residue have shown that substantial selectivity ratios can be attained for mosquito over vertebrate enzymes by increasing the steric bulk of the inhibitors (12, Figure 6).226-227 To overcome the carbamate insensitivity conferred by the G119S mutation in AgAChE1, small core carbamates (13, Figure 6) have been developed.187, 230-231 In addition to carbamates, difluoromethyl ketones (14, Figure 6) can also inhibit both wild

19

type AgAChE1 and the G119S mutant by reacting with the catalytic serine residue.232 However, the resistance-breaking carbamate-based inhibitors discussed above do not exhibit useful selectivity for AgAChE1 over hAChE. At high concentrations, several of these novel carbamates and difluoromethyl ketones exhibit in vivo toxicity towards An. gambiae, but issues, probably attributed to penetration, transport, and/or metabolism have been noted.187, 226, 228, 230-232

N

O

O

N (CH2)20

O

S

(CH2)18 SSO2MeN

NH

O

NN

ONH

ON

NHF2C

O

N

N O

O

O

O

N

CN

O

O

N

NH2

ON

O

HO

N

S

N

10 11

12 13 14

15 16

17 18 19 Figure 6. Examples of mosquito AChE1 inhibitors reported in the recent literature.

Covalent inhibitors: 10-11 covalent inhibitors targeting the free cysteine,225, 229 12

aryl methyl carbamate with selectivity for AgAChE1 over hAChE,226 13 small core

carbamate able to inhibit AgAChE1 and AgAChE1-G119S,187 14 difluoromethyl ketone

able to inhibit AgAChE1 and AgAChE1-G119S.232 Non-covalent inhibitors: 15

Alkylidene barbiturate able to inhibit AgAChE1 and preferentially AgAChE1-

G119S,233 16 inhibitor of AaAChE1,234 17 tacrine and 18 galanthamine non-selective

inhibitors of AgAChE1,220 and 19 ethopropazine showing 40 times higher inhibition

of AgAChE1 than hAChE.220

Examples of non-covalent mosquito AChE1 inhibitors are sparse. Alkylidene barbiturates (15, Figure 6) were identified in a screening against AgAChE1-G119S233 and girgensohnine analogs (16, Figure 6) have been designed and evaluated for AChE1 inhibition in Ae. aegypti (AaAChE1).234 It is noteworthy, however, that selectivity over hAChE was not investigated in these two studies. In addition, the known hAChE inhibitors tacrine (17,

20

Figure 6) and galanthamine (18, Figure 6) are also inhibitors of AgAChE1.220 Interestingly, the butyrylcholinesterase (BuChE, EC 3.1.1.8) inhibitor ethopropazine (19, Figure 6) has been shown to inhibit AgAChE1 approximately 40 times more strongly than hAChE.220

21

Methodology

The discovery and development process Today’s insecticide research and development activities can be compared to the drug discovery process (Figure 7). HTS, structure-based design, combinatorial chemistry, and fragment-based design, are methodologies perhaps most associated with drug discovery programs, but they are also applicable to insecticide discovery.235-240 However, there are factors that need special consideration when developing insecticides, such as environmental fate, ecotoxicology, residual activity (half-life), and costs of the active compound.240

Figure 7. A comparative overview of the drug discovery and insecticide discovery

processes. The early development of new insecticides can be very similar to that of

new drugs.241-242 In insecticide discovery, the compounds of interest are generally

tested on the target organism much sooner compared to drug candidates. Human

and environmental safety are paramount and are evaluated from an early stage in the

process.243 The field trials (corresponding to clinical trials for drugs) include

laboratory testing (phase I), small scale field trials using experimental huts (phase

II), and large scale field trials on a community level (phase III).244

Protein-ligand interactions and structure-activity relationships Modern insecticides (i.e. post 1980s) do not differ that much from modern drugs, in that they generally consist of a small organic molecule (ligand) able to bind to and modulate the activity of one enzyme or receptor (target). The ligand often binds in a pocket formed on the surface of the target, such as the active site of an enzyme or areas for protein-protein interactions of receptors or signaling molecules. The protein-ligand binding is dependent upon a number of factors, such as shape complementarity, electrostatic forces, and

22

hydrophobic interactions, which influence how strong the interaction is. Examples of specific types of interactions that are important for the recognition process are classical and non-classical hydrogen bonds, arene···arene interactions, and halogen bonds (Figure 8). 223, 245-246

Figure 8. Examples of ligand-protein interactions: classical hydrogen bond (A),

cation···arene interaction (non-classical hydrogen bond, B), activated C-H···arene

interaction (non-classical hydrogen bond, C), C-H···arene interaction (non-classical

hydrogen bond, D), T-shaped arene···arene interaction (E), and halogen bond (F).

When a biologically active compound has been identified, the next steps in the development process most often involve the identification of the structural elements necessary for activity or selectivity, i.e. the SAR or SSR.247-248 SAR and SSR are based on the theory that closely related chemical analogues should also display related biological activity. When exploring the SAR and/or SSR, new compound analogues are carefully designed, synthesized, and biochemically evaluated to obtain information of how changes in the chemical structure affect the biological activity. Chapters 5 and 6 explore the SARs and SSRs of two different classes of Ag/AaAChE1 inhibitors. A 3D-structural model of the target protein is of great help during this process, as it can both guide the design and facilitate the interpretation of the data.249-253 Protein structures can be determined by NMR or by X-ray crystallography.254-257 In protein X-ray crystallography, protein crystals (with or without bound ligands) are subjected to intense X-rays and the resulting scattering gives rise to a diffraction pattern. The collected data is processed to allow calculation of an electron density map into which the 3D-structure of the protein or protein-ligand complex is modeled.256-257 In the research described herein, protein X-ray crystallography has been applied to determine the binding mode of inhibitors in mAChE (Chapters 5 and 6). In cases where the 3D-structure of the target protein has not been determined, the structure of the target can be predicted using computational methods and known structures of similar proteins (templates).258-259 The process, known as homology modeling, involves the following steps: template selection, sequence alignment of target and template, model building, model refinement, and model evaluation.260 As described in Chapter 4, homology modeling has here been used to predict the structures of AgAChE1 and

23

AaAChE1. In the following sections, some of the methods employed during the work presented in this thesis are discussed in more detail.

HTS for hit identification HTS is the general term for an automated system in which large numbers of compounds are tested for their ability to modulate a pharmacologically interesting target. Compounds exhibiting the desired biological activity are identified as hits and often constitute the starting points for a medicinal chemistry project. Since the middle of the 1980s rapid development has resulted in HTS becoming an integral part of both the pharmaceutical and agrochemical discovery process, where ~1-3 million compounds may be screened at a rate of between 10,000 and 100,000 compounds per day.235-236,

261-264 More recently, HTS has also become available in academic settings.265-

267 The aim of HTS is to screen compounds rapidly and cost effectively while

still generating good quality data. To achieve this, the instrumentation is automated (to varying levels depending on the facility) and the chosen assay is adapted to a microtiter plate format, thereby significantly reducing the assay volume (Table 4). Adapting and optimizing the assay for screening are often the most time consuming steps in any HTS campaign. A drawback of small assay volumes is that the readout, often detected as light, might be weak and subject to variation. Several statistical methods have been used to evaluate HTS assays, such as signal to noise, signal to background, and Z’-factor,268 of which the latter was used in the work described in Chapter 2.

Table 4. Different sizes of microtiter plates used for screening.

Bench top Microtiter plate format

Number of wells n.a.a 96 384 1536

Assay volume (µl) >1000 100-200 ~50 ~5

an.a. not applicable

HTS generates vast amounts of data which need to be processed and analyzed. Several methods have been developed for HTS data normalization.269-271 Control based normalization methods rely on the availability of good negative controls and can result in a poor estimate of the mean due to the limited number of control wells per plate and their placement on the plate (spatial effects due to placement along the edge). These issues can be addressed by the use of sample based control methods; since in most screenings the majority of the tested compounds will be inactive, the wells containing screening compounds can be considered to represent negative controls. This yields a larger sample set, spaced over the entire plate, from which the mean can be estimated. Both the mean and the median can be used for percent of control and percent of sample methods.

24

The median is considered the more robust option since the mean is sensitive to outliers. When the data normalization is complete, hits can be identified in a number of ways such as selection of a discrete cut-off value, mean/median ± n standard deviations cut-off, or quartile-based cut-off.269-

270

Chemical similarity As mentioned above, the ability to compare and assess different molecules in relation to each other or to their biological activity is central to both drug and insecticide development. Thus, many different methods for describing the structural, physical, and chemical properties of a molecule have been developed. This section describes the methods used in this thesis to characterize chemical structures and properties, and assess chemical similarity.

Fingerprints and distance coefficients The chemical structure of a molecule can be described by so called fingerprints.272 Fingerprints describe the chemical structure in terms of binary vectors denoting either the presence (1) or absence (0) of pre-defined structural fragments or elements (Figure 9). Molecule Binary fingerprint vector

a 0 1 1 0 1 0 1 o o 1 0 1 1 0 1 0 1 1

b 0 0 0 0 0 1 1 0 1 1 1 0 1 0 0 1 1 0

presence of –NO2 absence of -CH2CH2CH2-

Figure 9. An example of two molecules (a and b) and their 18 bit fingerprint vectors.

In addition to methods for describing chemical structures, several coefficients have been developed to compute similarity or dissimilarity (distance) between molecules based on both binary and continuous descriptors.273 In the work described in Chapter 3, the Soergel distance (Equation 1) was used to compute the distance between different hit molecules, whose chemical structures had been described by fingerprints. Dab = (a+b-2c)/(a+b-c) (Equation 1)

Where a is the number of bits set to “1” in the fingerprint vector of molecule a, b is the number of bits set to “1” in the fingerprint vector of hit b, and c is the number of bits set to “1” in the fingerprint vectors of both a and b. Thus,

25

according to the Soergel distance, only fragments that are present can support similarity (matched absence is not considered). The Soergel distance ranges between 0 and 1 and is the complement to the widely used Tanimoto (Jaccard) similarity coefficient.273-274

Molecular descriptors and multivariate data analysis methods Molecular descriptors, based either on the 2D or 3D representations of a molecule’s chemical structure, can be used to describe, for example, electronic, hydrophobic, steric, and size related properties. To enable analysis and interpretation, molecular descriptors have been used here in conjunction with multivariate data analysis methods. Various projection methods have been developed to analyze multivariate data, since mining information from large and complex data sets can be difficult. These methods are used to extract and visualize systematic variation in the data set, thereby reducing the complexity of the data and facilitating its interpretation.

Principal component analysis Principal component analysis (PCA) is an unsupervised multivariate projection method.275-276 PCA is used to compute a number of orthogonal principal components (PCs) from a data matrix (X), consisting of observations and variables (e.g. molecules and descriptors, respectively). In the resulting model, the first PC describes the maximum variance in the data, the second PC the second largest variance and so on. The generated PCs are referred to as score vectors (T), and each object obtains its new score values (t1, t2,…tn) by projection of its position in the original variable space on to the new PCs. The original variables’ influence on the PCs are described by the loadings (P). Thus PCA models the original data matrix X as:

X = TP’+E (Equation 2)

Where P’ is the transposed loadings and E the remaining unexplained variation (residuals).

In this work, PCA was applied to compound sets whose physicochemical properties had been described by calculated molecular descriptors. PCA was then used to obtain an overview of the compounds and to discern how compounds varied in relation to each other (Chapter 2).

Orthogonal partial least squares-discriminant analysis Orthogonal partial least squares-discriminant analysis (OPLS-DA) is a supervised multivariate regression method, which can used for the interpretation of differences between pre-defined groups of observations and for classification of observations.277 The group identity of each observation is

26

described by the response matrix Y, which is used in OPLS-DA to decompose the X matrix into Tp, To and E according to:

X = TpPp’+ ToPo’ + E (Equation 3)

Where Tp and Pp describe the predictive scores and loadings which are related to the response, respectively. To and Po describe the scores and loadings which are orthogonal to the modeled response, respectively, and E is the residual variation.278 In this fashion, OPLS-DA allows the largest variance between the classes (predictive component) to be interpreted separately from the within-class variance (orthogonal components). OPLS-DA was here used to investigate the correlation between the physicochemical properties of a set of compounds and their AChE1- or hAChE selectivity (Chapter 3).

Biochemical evaluation of AChE inhibitors Accurate methods to determine and compare the (inhibitory) activity of compounds on the studied biochemical or biological function are central to every medicinal chemistry project. The kinetic- and inhibition constants discussed in this thesis were determined using the Ellman assay279 and are further described below.

The Ellman assay for determination of AChE activity The Ellman assay279 is a colorimetric method that is widely used to evaluate the activity of serine hydrolases such as AChE. In this assay, the natural substrate acetylcholine is substituted with acetylthiocholine (ATCh; 20, Figure 10) and, upon hydrolysis by AChE, thiocholine is formed (22, Figure 10). The formation of thiocholine is monitored using the reagent 5,5'-dithiobis(2-nitrobenzoic acid) (DNTB; 23 Figure 10). The reaction between DNTB and thiocholine yields the mixed disulfide 24 and 5-thio-2-nitrobenzoic acid 25, which gives a yellow color. Using a spectrophotometer, the color change can be monitored by measuring the change in absorbance at 412 nM. The assay is commonly performed in a time-dependent manner where the linear increase in absorbance over time (i.e. the slope) is monitored.

27

Figure 10. The Ellman assay used for determination of AChE enzyme activity.

Enzyme kinetics The Michaelis-Menten kinetics280 model is used to describe the activity of an enzyme (E) that transforms the substrate (S) to a product (P) via a transition complex (ES) according to:

Where kf is the rate constant for the formation of the ES complex, kr is the rate constant for the dissociation of the ES complex and kcat is the turnover number, i.e. the number of substrates converted to product per enzyme and unit time. When studying AChE, the substrate would typically be ATCh (20) and the products acetic acid (21) and thiocholine (22). For many enzymatic reactions the rate of catalysis (V) is dependent on the concentration of the substrate [S] up to the concentration at which the enzyme is fully saturated and the maximal rate (under the specified conditions) is reached (Vmax). If we make the following assumptions that (i) the reactants behave ideally and (ii) the system is under steady-state, i.e. that the concentration of ES ([ES]) is constant, an enzymatic reaction can be described as follows. At a very early stage (close to time t0 when the impact of accumulated P on the reaction is negligible) E and S form the complex ES at the rate kf[E][S], and ES dissociates at rate kr[ES] or forms P at the rate kcat[ES]. The rate of catalysis is then given by the Michaelis-Menten equation (Equation 4).

Vo = Vmax ([S]/([S]+KM)) (Equation 4)

Where the Michaelis constant (KM) is defined as:

KM = (kr + kcat)/kf (Equation 5)

28

The Michaelis-Menten equation suggests that at very low substrate concentrations ([S]<<KM) the rate of catalysis is directly proportional to the substrate concentration and that when [S] = KM, the rate of catalysis is half of Vmax. The catalytic efficiency (Equation 6) can be used to compare the efficiency of enzymes under conditions similar to physiological conditions ([S]<KM). Catalytic efficiency = kcat/KM (Equation 6)

Determination of inhibition constants The half maximal inhibitory concentration (IC50) is frequently used to determine how efficiently a non-covalent ligand inhibits the target of interest. When studying enzymes such as AChE, a dose-response plot is generated by measuring the enzymatic activity in the presence of an inhibitor at different concentrations and then plotting the activity versus the concentration of inhibitor (log[I]) (Figure 11a). From this the IC50 value can be determined as the inhibitor concentration resulting in 50% of the maximum activity of the enzyme (i.e. the lower the IC50, the more potent the inhibitor). It should be noted that the IC50 value depends on experimental conditions such as the concentration of the substrate, ionic strength, and pH. Thus, caution is advised when comparing IC50 values determined under different conditions. The IC50 values presented in this thesis have been determined under comparable conditions (e.g. substrate concentration, temperature, and final DMSO concentration). In this thesis, a compound’s selectivity for mosquito AChE1 over hAChE is described by its selectivity ratio. The IC50-based selectivity ratios are computed by taking a compound’s IC50 value for hAChE and dividing it by the higher of the compound’s IC50 values for AgAChE1 and AaAChE1.

Figure 11. (a) Example of a dose-response curve used for determination of the IC50

value (indicated by dotted lines). (b) Plot of experimentally determined kobs vs. the

concentration of inhibitor ([I]). The ki corresponds to the slope of the line.

29

For covalent inhibitors of AChE, IC50 values are inappropriate as they do not contain a time dimension, which is necessary to describe the chemical reactivity. Instead, the inhibition constant (ki) can be used to describe inhibition potency. For carbamates (as studied in the work described in Chapter 1) the mechanism of inhibition can be seen as a progressive inactivation, where ki is the overall rate constant of the inhibition. The ki constant is experimentally determined by measuring enzyme activity at different concentrations of inhibitor and after different incubation times. For each inhibitor concentration the pseudo first-order rate constant kobs is determined as the slope of the straight line formed when plotting ln[enzyme activity] versus time. The ki rate constant is obtained by plotting the calculated kobs constant vs the inhibitor concentration [I] (Figure 11b) according to Equation 7. More potent inhibitors have larger ki values compared to weaker inhibitors.

kobs = ki[I] (Equation 7)

Mosquito testing Two mosquito strains from Kenya have been used in the work presented in this thesis (Chapter 6) to evaluate the intrinsic insecticidal activity of the compounds; the Anopheles gambiae s.s. Kisumu strain and the Aedes aegypti Mombasa strain. The larvae and adult mosquitoes used for testing were reared in an insectary and from these, five day old female mosquitoes (non-blood fed) and third instar larvae were selected for testing of adulticides and larvicides, respectively.

For the evaluation of adulticides the mosquitoes are rapidly anesthetized at -20 °C before a compound at different concentrations in acetone was applied to the pronotum (back plate). The mosquitos are then kept under normal conditions in the insectary. Mortality rates were recorded after 24 and 48 hours and compared with negative controls (application of acetone). To test the insecticidal activity on larvae a compound was dissolved at different concentrations directly in the water used for rearing. The larvae was then kept in the compound treated water under normal conditions in the insectary. Mortality rates were recorded after 24 and 48 hours and compared with negative controls (untreated water).

30

Chapter 1: Characterization of AChE1 from An. gambiae and Ae. aegypti (Paper I and Appendix I)

Background In the work described in Paper I and Appendix I we focused on the structural and functional properties of mosquito AChE1 compared to vertebrate orthologues from human and mouse. By studying the basal kinetics of the enzymes and their ligand binding properties we aimed to gain a better understanding of similarities and differences between the mosquito and vertebrate enzymes, which could be of value for future development of selective insecticides. Specifically, we wanted to investigate the effects of a set of non-covalent inhibitors on the different enzymes. The known resistance conferring mutant in An. gambiae, AgAChE1-G119S, was also included in order to investigate how this glycine to serine mutation affects the ligand binding properties of the enzyme.

Sequence analysis of AChEs from different species An amino acid sequence analysis was performed on a selection of AChEs from 16 species representing mammals, birds, fish, nematodes, and insects (Table 5). The AgAChE1 and AaAChE1 enzymes are very similar; the sequence identity is 92% (Appendix I). On the other hand, the sequence identity between AgAChE1 and AChE2 from the honey bee was only 42%. This is of interest from an insecticide point of view since the honey bee is an important pollinator and relies on AChE2 for its main catalytic activity.181

The sequence similarities in relation to hAChE (Table 5) were very high for other mammals (89-94%), followed by fish (~60%), insect AChE1 (43-47%), and insect AChE2 (34-38%). These trends are visualized in the phylogenetic tree in Figure 12, showing a clear clustering of the mammal AChEs while insect AChE1 and AChE2 form two distinct clusters (Appendix I).

31

Table 5. Sequence identity compared to hAChE for the different species included in

the analysis. If a species was known to express different isoforms of AChE the isoform

included is stated. Sequence identity was calculated as described in Appendix I.

Sequence identity (%)

Species Enzyme All Gorge CASa PASb

An. gambiae (mosquito) AChE1 46 68 91 33

Ae. aegypti (mosquito) AChE1 47 68 91 33

Cx. pipiens (mosquito) AChE1 45 68 91 33

B. germanica (German cockroach) AChE1 43 65 86 33

A. mellifera (honey bee) AChE2 38 54 82 13

M. domestica (house fly) AChE 35 54 82 13

D. melanogaster (fruit fly) AChE 34 54 82 13

T. californica (Pacific electric ray) AChE 57 84 91 73

D. rerio (zebra fish) AChE 58 84 95 67

M. musculus (mouse) AChE 89 97 100 93

R. norvegicus (brown rat) AChE 89 97 100 93

B. taurus (European cattle) AChE 94 97 100 93

O. cuniculus (European rabbit) AChE 93 95 100 87

G. gallus (chicken) AChE 48 76 91 53

C. elegans (round worm) AChE1 40 57 82 20

aBased on 22 residues found in the lower part of the gorge (including CAS); bBased on

15 residues by the entrance of the gorge (including PAS)

Figure 12. Phylogenetic tree showing AChEs from 16 different species and the

resistance conferring mutant AgAChE1-G119S (Appendix I). AChE1 enzymes are

indicated by the light grey shading.

32

A focused analysis of the active site gorge showed that the residues adjacent to the gorge were generally highly conserved (Table 5). In particular, residues in the lower half of the gorge (including the choline binding site, the catalytic triad, the oxyanion hole and the acyl pocket) exhibited high sequence identity across all species (>80%). The entrance of the gorge (including PAS) exhibited more diversity between the species. Both insect AChE1 and AChE2 exhibited only 10-30% sequence identity compared to hAChE for these residues. These findings are in agreement with previous studies of AChEs from different species.217-218, 220 Based on this, one could hypothesize that inhibitors binding in the upper region (PAS) could have greater chances of delivering species selectivity. However, further studies are needed to understand how these differences affect the shape and ligand binding properties of the entire active site gorge.

Characterization of the enzymes Recombinant AgAChE1, AaAChE1, and AgAChE1-G119S were expressed by Sf9 insect cells using a baculovirus expression system and their functional properties were evaluated using Michaelis-Menten kinetics (Paper I). It was found that, analogously to hAChE and mAChE, the wild-type mosquito enzymes are subject to substrate inhibition at high ATCh (20) concentrations and the preferred substrate was the smaller ATCh (20, Figure 13) over propionylthiocholine (26, PTCh) and butyrylthiocholine (27, BTCh).

Figure 13. To investigate the substrate preference of the enzymes, the KM constant,

Vmax, and in some cases kcat were determined for three different substrates of

increasing size; acetylthiocholine (20, ATCh), propionylthiocholine (26, PTCh), and

butyrylthiocholine (27, BTCh). Thio-analogues of the natural substrates are used in

agreement with the Ellman assay.

We noted that AgAChE1 and AaAChE1 exhibited a higher affinity for

ATCh compared to hAChE and mAChE; KM values were 27, 25, 146 and 84 µM, respectively (Table 6). However, both kcat and the catalytic efficiency (kcat/KM) was lower compared to the vertebrate AChEs. The kcat values presented here are also lower than those previously reported for AgAChE1.187-188 However, the catalytic efficiency of AgAChE1 is in agreement with that previously determined by Wong et al.187

33

Table 6. Kinetic parameters of AaAChE1, AgAChE1, AgAChE1-G119S, mAChE, and

hAChE (Paper I). Values obtained from the literature are given in parentheses.

Substrate AaAChE1 AgAChE1 AgAChE1-G119S mAChE hAChE

KM

a

ATCh 0.025 0.027 0.058 0.084 0.15

PTCh 0.020 0.025 0.30 0.059 0.15

BTCh 0.045 0.036 0.43 (0.049-0.093)f (0.30)h

k ca

tb

ATCh 0.084 0.074 (0.053)d (1.4-1.7)g (3.7-4.2)i

PTCh 0.055 0.057 n.d.e n.d.e (1.6)j

BTCh 0.0078 0.0084 n.d.e (0.0085-0.011)f (0.075)h

k ca

t/K

Mc ATCh 3.4 2.7 (0.41)d (30-44)g (26-29)i

PTCh 2.8 2.3 n.d.e n.d.e (6.4)j

BTCh 0.17 0.23 n.d.e (0.12-0.17)f (0.25)i

a(mM); b(*10-5*min-1); c(*10-8*M-1*min-1); dPreviously reported187; en.d. = not

determined; fPreviously reported281-282; gPreviously reported281-283; hPreviously

reported169, 284; iPreviously reported169, 284-289; jPreviously reported.169

Table 7. Comparison of substrate specificity based on kinetic constant ratios (Paper

I). Values obtained from the literature are given in parentheses.

Ratioa AaAChE1 AgAChE1 mAChEb hAChEb

Vm

ax P/A 0.66 0.77 0.51 0.52

B/A 0.094 0.11 n.d. n.d.

k ca

t P/A 0.65 0.77 n.d. (0.43)d

B/A 0.093 0.11 (0.0065-0.0079)c (0.02)e

k ca

t/K

M

P/A 0.82 0.85 n.d. (0.24)d

B/A 0.05 0.082 (0.0039-0.004)c (0.0096-0.011)e

aP/A: PTCh/ATCh, B/A: BTCh/ATCh; bn.d. = not determined; cPreviously

reported281-282; dPreviously reported169; ePreviously reported.169, 284

Interestingly, our data indicate that the wild type mosquito enzymes are

less sensitive to substrate size than either mAChE or hAChE (Table 7). The relative decrease in both kcat and the catalytic efficiency is lower for the

34

mosquito enzymes than the vertebrate enzymes, especially when going from ATCh to PTCh, but also to some extent BTCh (Table 7). The substrate specificity of insect AChEs has been described as being intermediate between AChE and BuChE.27, 290 In addition, it is known that the acyl pocket is important for substrate specificity. For example, mutants of mAChE and hAChE, where Phe295 (Cys295 in mosquitoes) is exchanged for a smaller residue such as alanine or leucine, are better at hydrolyzing larger substrates compared to the wild type enzymes.169, 281-282, 284-285

Compared to the wild-type AgAChE1, the AgAChE1-G119S mutant exhibited a more pronounced sensitivity to substrate size; for PTCh KM was 25 µM and 315 µM for wild-type and mutant AgAChE1, respectively and the Vmax ratio between PTCh and ATCh was 0.77 and 0.33 for wild-type and mutant AgAChE1, respectively (see Table 2, Paper I).

Investigating the active site gorge with inhibitors A set of seven known cholinesterase inhibitors was selected to study the ligand binding properties of the mosquito and vertebrate enzymes (Table 8). Two of these were covalent inhibitors that contain a carbamate functionality able to react with the catalytic serine residue, propoxur (28)291 and the natural product eserine (29).292-293 The remaining five included non-covalent inhibitors ethopropazine (19),281, 294 the Alzheimer’s drug donepezil (30),177,

295-296 HTS hit 31297 and enantiomers 32R and 32S.221, 297 Comparison of the wild-type mosquito and vertebrate enzymes showed

that the two covalent inhibitors were 10-15 times more potent inhibitors of wild-type AChE1 than mAChE or hAChE. The BuChE inhibitor ethopropazine (19) was the only one of the studied non-covalent ligands that inhibited AgAChE1 and AaAChE1 (IC50 values of 8.3 µM and 4.6 µM, respectively) but not mAChE and hAChE. BuChE is a cholinesterase present in vertebrates, which is able to hydrolyze larger substrates, such as butyrylcholine, more efficiently than AChE.26, 298 This stronger inhibition of AChE1 than hAChE by 19 has been reported before,220 and, being a BuChE inhibitor, is in agreement with the trends discussed above regarding substrate specificity.

35

Table 8. Ligands used to study the binding properties of mosquito AChE1 and

vertebrate AChE.

AaAChE1 AgAChE1

AgAChE1

-G119S mAChE hAChE

k i (

µM-1

s-1

)a

28 1.2 (15) 1.11 (14) n.d. 0.092 (1) 0.081 (1)

29 24.8 (9) 27.26 (10) 1.18 (0.4) 1.36 (0.5) 2.66 (1)

IC 5

0 (

µM)b

19 4.6 (217) 8.3 (120) 1000 (1) 1000 (1) 1000 (1)

30 0.28 (0.03) 0.24 (0.03) 0.31 (0.03) 0.007 (1) 0.008 (1)

31 0.36 (1) 0.44 (0.8) 1.3 (0.3) 0.2c (2) 0.36 (1)

32R >30 (0.04) >30 (0.04) >30 (0.04) 0.7d (2) 1.3d (1)

32S >30 (0.05) >30 (0.05) >30 (0.05) 0.7d (2) 1.4d (1)

aSelectivity ratio compared to hAChE given in parentheses, computed as

ki(xyAChE(1))/ki(hAChE). n.d. = not determined; bSelectivity ratio compared to

hAChE given in parentheses, computed as IC50(hAChE)/IC50(xyAChE(1)); cPreviously reported297; dPreviously reported.221

Structural analysis The binding modes of 30 and 31 in vertebrate AChEs have been studied using X-ray crystallography.173, 177, 223 Both ligands span the entire active site gorge, with the indanone of 30 and dichlorophenyl of 31 binding in PAS, and the benzyl (30) and piperidine (31) interacting with Trp86 at the bottom of

36

the gorge. Both of these ligands are potent inhibitors of hAChE, with IC50 values in the nanomolar range, and both ligands also exhibit strong inhibition of the mosquito AChE1, including the G119S mutant. It should be noted however, that 30 inhibits hAChE approximately 30 times more efficiently than AChE1. Finally, the two enantiomers 32R and 32S, exhibited stronger inhibition of the vertebrate enzymes. For both enantiomers, the aromatic nitro group forms a hydrogen bond to the back bone NH of Phe295 in mAChE.221 Since Phe295 is part of loop 1 (Figure 5), which is three residues shorter in mosquito AChE1, it is possible that this interaction is lost or altered in AgAChE1 and AaAChE1 due to a different conformation of the loop. In addition, as mentioned earlier, Phe295 is one of the residues in the active site which is not conserved between vertebrate AChE and insect AChE1 (cysteine).217-218, 220

The insensitivity of the AgAChE1-G119S mutant towards covalent inhibition by propoxur (28) was clear and only a marginal decrease in enzyme activity could be detected. In addition, 29 proved to be approximately 20 times less potent an inhibitor of the mutant compared to the wild-type enzyme. Similar observations have been made previously for AChE1 from the sandfly.299 The mutant AgAChE1-G119S exhibited similar or slightly higher IC50 values compared to the wild-type enzyme for all non-covalent ligands except 19, for which only weak inhibition could be observed.

Cluster analysis The trends of the kinetic and inhibitory parameters were investigated by cluster analyses visualized in the form of functional trees (Figure 14). The data were first divided into two groups including (i) constants related to affinity and/or reactivity of molecules that react with the catalytic serine or (ii) the inhibition efficiency of the five non-covalent ligands (19, 30-32R and S). The most notable difference between the two subsets of data was the shift in properties induced by the G119S substitution. For substrates and covalent inhibitors, the ligand binding properties of the mutant enzyme were more similar to the properties of the vertebrate AChEs. For non-covalent inhibitors, on the other hand, a shift towards AgAChE1 and AaAChE1 indicates that the ligand binding properties were more similar to the wild type AChE1 enzymes. Thus, the small set of non-covalent ligands studied in this work indicated that (i) there are differences in the ligand binding properties of mosquito and vertebrate AChEs and (ii) it might be possible to target the resistance conferring mutant with non-covalent inhibitors.

37

Figure 14. Cluster analysis results presented as unrooted trees. The end of each

branch represents one enzyme and the distance corresponds to the similarity

between two enzymes based on the constants included in the analysis. (A) Cluster

analysis based on constants related to affinity and/or reactivity of molecules that

react with the catalytic serine (kcat, KM, Vmax and ki). (B) Cluster analysis based on the

inhibition capacity (IC50 values) of the five non-covalent ligands (19, 30-32R and S).

Summary of Chapter 1 The mosquito AChE1s show 44-45% overall sequence identity to

hAChE The active site gorge is highly conserved between species, especially

close to the catalytic triad The mosquito AChE1 enzymes share both functional properties and

structural features with AChE from mouse and human The ligand binding properties of the mosquito and the vertebrate

enzymes differ The G119S substitution induced significant changes to the properties

of AgAChE1, especially for substrates and ligands that react with the catalytic serine

Further studies of non-covalent inhibitors of AChE1 are warranted

38

Chapter 2: Discovery of AChE1-inhibitors by HTS (Paper II and Appendix II)

Background To build on the findings described in Chapter 1, we considered HTS to be a suitable first step to identify new chemical starting points for investigating the scope for developing non-covalent inhibitors of mosquito AChE1.

HTS enables the evaluation of large numbers of compounds resulting in the opportunity to investigate a broad physicochemical property space. In addition, our in-house library has previously been successfully screened against hAChE resulting in the identification of several interesting hAChE inhibitors.221-223, 297 An overview of the work discussed in Chapters 2 and 3 (Paper II and Appendix II) is given in Figure 15 and aimed to (i) identify non-covalent inhibitors of AgAChE1 and AaAChE1 (Chapter 2) and (ii) investigate the potential for utilizing the corresponding hAChE screening data297 to identify and analyze compounds with potential selectivity for mosquito AChE1 over hAChE, i.e. differential HTS (Chapter 3).

Assay adaptation and screening for AChE1 inhibitors To identify non-covalent inhibitors of AgAChE1, AaAChE1 or both (from here on AChE1 will refer to either or both of the enzymes unless otherwise is stated) we decided to screen an in-house library for compounds able to inhibit the activity of AChE1. For this purpose the Ellman assay279 was miniaturized for screening in 96-well plates and liquid handling was facilitated by the use of an automated workstation. The assay conditions were adjusted until adequate assay sensitivity and reproducibility was achieved on the basis of repeat measurements of a reference plate containing eight compounds (eight replicates of each) with varying inhibition capacities (IC50 values ranged between 0.05 µM to >1000 µM). Finally, 17 500 compounds were screened against each protein at a concentration of 50 µM in singlicate. The reference plate was run and analyzed continuously during the compound screening to monitor assay stability and robustness.

39

Figure 15. Work-flow of the differential HTS process discussed in Chapters 2 and

3. The screening against hAChE has previously been reported297 and its results are

used here in combination with the results obtained from the screenings against

AgAChE1 and AaAChE1.

40

Normalization of AChE1 HTS data Data normalization was performed per plate and several methods for normalization269-271 were investigated in relation to hit identification (Appendix II). Hits were identified as compounds that reduced the activity of the enzyme by at least three times the standard deviation of the mean for the 17 500 compounds. Comparison of both control based methods (percent of control, robust percent of control, and normalized percent inhibition) and sample based methods (percent of sample and robust percent of sample) showed that the investigated methods gave very consistent results. No clear differences were observed either when comparing control- versus sample based methods or mean versus median based methods. In total, the five methods identified 352 hits of AChE1. Of these 87% were scored as a hit by all five methods. Finally, percent of sample was chosen as the normalization method, in accordance with the normalization performed in the HTS against hAChE,297 resulting in the identification of 338 AChE1-hits.

Assay quality evaluation The quality of the HTS assay and data was evaluated in several steps. First, trends and shifts in the raw data were visualized in plate order effect plots of the raw data and plate median (Appendix II). Second, any systematic spatial effects were identified by visual inspection of plate heat maps. No clear trends or effects could be detected and hence, no statistical adjustments were made to compensate for the systematic variability in the screens.

The Z’-factor is a coefficient developed to determine the quality of an assay, which corresponds to the separation band between active and inactive compounds.268 Z’ ranges from 0 to 1, where 1 is considered an ideal assay, 1 >Z’≥0.5 corresponds to an excellent assay, and 0.5 >Z’>0 to a case where double assays are required. In the screenings against AgAChE1 and AaAChE1, we used the data obtained from the reference plates to determine the Z’-factors. Compounds 29 and 33 were selected as the negative control and positive control, respectively (Figure 16). The resulting average Z’-factors were 0.65 and 0.73 for AgAChE1 and AaAChE1 showing that both screenings were able to distinguish between active and inactive compounds. 

41

AgAChE1 AaAChE1 AgAChE1 AaAChE1 IC50 (µM) 0.005 0.007 >1000 >1000 Ref. plate (%) 95±8 94±3 2±7 1±5

Figure 16. Reference plate compounds 29 and 33 selected for determination of the

Z’-factor. The reference plate %-inhibition values at 50 µM are shown as mean ± the

standard deviation of the mean for the 27 and 12 runs for AgAChE1 and AaAChE1,

respectively.

In addition to the Z’-factors, the reference plate data also allowed the

estimation of the proportions of false positives and false negatives in the screenings. This analysis estimated 6% and 11% false positives for AgAChE1 and AaAChE1, respectively, while the proportion of false negatives was 0.5% in both cases (Paper II).

Identification and analysis of hits The overall hit rates for AgAChE1 and AaAChE1 were 1.3% and 1.6%, respectively, resulting in 338 unique compounds being scored as AChE1-hits. The AChE1 hits comprised a diverse set of compounds with respect to their chemical structures. Many of the compounds contained a basic amine, that can form cationarene or activated CHY hydrogen bonds (Y = O or arene) typical for AChE,223 but neutral or negatively charged compounds were also identified. Analysis of the physicochemical properties of the hits also revealed diversity with regards to molecular size, lipophilicity, and flexibility: the molecular weights were between 199 and 629; the logP values were between 0.6 and 8.7; and the numbers of rotatable bonds were between 0 and 10.

Pan-assay interference compounds (PAINS) are a known problem in HTS.300 PAINS contain structures whose activity does not depend on specific ligand-protein interactions, instead these false positives can be the result of aggregation, protein reactivity, or assay interference. The 338 hits were passed through a PAINS filter 301 and 16 hits (5%) were flagged as PAINS. Most of these hits showed 30-70% inhibition in the screenings and contained tertiary anilines or phenolic Mannich bases, but problematic barbiturates, hydrazones, catechols, and acridines were also identified (Appendix II). Due to the low number of PAINS, the identified compounds were not excluded from our continued analyses, but flagged as potentially problematic.

42

Hit validation A subset of hits was selected for follow-up evaluation to verify their inhibitory activity by determination of their IC50 values. The selection of diverse and representative compounds was facilitated by first dividing the compounds according to their activity in the screenings, and then selecting compounds from each activity range (set A: 70-100% inhibition; set B: 30-70% inhibition; set C: < 30% inhibition; sets D1-D2: ≥ 30% difference in measured inhibition between the two enzymes). For each set, the main variability in physicochemical properties of the compounds was extracted by construction of PCA models based on 2D molecular descriptors. The results were visualized in the form of score plots and from each set, representative compounds were manually selected (Figure 17). In total 76 compounds were selected for dose-response experiments and IC50 determinations.

Figure 17. Chemical space of AChE1 hits showing ≥ 70% inhibition. Score plots from

PCA of the physicochemical properties of the 55 hits showing ≥ 70% inhibition, the

24 hits manually selected for IC50 determination colored in black (set A). The first

and second components describe the size and lipophilicity of the hits (a) and the third

and fourth components show diversity relating to flexibility and charge (b).

The full dose-response analyses of the 76 selected compounds generally

agreed well with the screening data (Table 9, for more details see Table 2 in Paper II). Of the 47 hits in sets A and B, 34 were confirmed with IC50 values ranging from 0.2 to 128 µM (examples are shown in Table 10). Compounds with IC50 values above 200 µM, or which were insufficiently soluble to provide reliable experimental data, were considered to be inactive. Most of

43

the confirmed inhibitors were found in set A and, in comparison, set B appeared to have a higher false positive rate than set A.

Table 9. Overview of the follow-up evaluation of the screenings targeting AChE1.

HTS

inhibition No. compounds AChE1 inhibitors n.d.d

Set Aa 70-100% 24 21 0

Set Ba 30-70% 23 13 4

Set Ca < 30% 14 4 2

Set D1a,b 37 – 63% 9 0 0

Set D2a,c 47- 80% 6 2 2 aThe compounds were assigned to the group corresponding to their highest level of

inhibition in the AgAChE1 or AaAChE1 screens; bHits with ≥ 30% higher inhibition of

AgAChE1 than AaAChE1. The tabulated HTS inhibition corresponds to AgAChE1

inhibition; cHits with ≥ 30% higher inhibition of AaAChE1 than AgAChE1. The

tabulated HTS inhibition corresponds to AaAChE1 inhibition; dCould not be

determined due to poor solubility.

The most potent inhibitors in the follow up assay were found in set A (IC50

values of 0.21 – 86 µM for AgAChE1 and 0.22 – 66 µM for AaAChE1). In addition, several inhibitors were confirmed, with IC50 values between 2.4 and 22 µM, among the hits showing lower levels of inhibition in the HTS (set B). Compound 40, and three additional compounds from set C, proved to be false negatives. These four compounds had high structural similarity to some hits from sets A and B, but were in general less potent than their analogues (IC50 values ranging from 11-128 µM compared to 0.21-22 µM for compounds from sets A and B).

Comparison of the AgAChE1 and AaAChE1 screening data suggested that there was no distinct difference between the two mosquito enzymes. Indeed, only 35 of the 338 hits displayed a clear difference in inhibition (≥30%) between the two enzymes. Analysis of the IC50 data for set D showed that most compounds were false positives and the only two active compounds were equally potent against AgAChE1 and AaAChE1 (cf. compound 35).

44

Table 10. Examples of compounds studied in the follow-up evaluation.

ON

O

N

N

F

NO

O

N

NS

N

O

S

N N

N SO

NH

S

O

O

O NHF

N

O

N

O

N

N

F

ON

O

N

N

N N

N S

O

N

NO

N

O

34 35

36

3738

39

40

41

42

Compound Set

HTS (% inhibition) IC50 (µM)

AgAChE1 AaAChE1 AgAChE1 AaAChE1

34 A 92 76 0.21 0.22

35 A 86 90 >100 >100

36 A 94 93 8 9

37 A 76 82 >1000 >1000

38 B 23 38 >500 >500

39 B 32 33 17 12

40 C 24 18 21 19

41 D1 63 1 >1000 >1000

42 D2 41 80 3 3

Summary of Chapter 2 17 500 compounds were screened for their ability to inhibit the

activity of AgAChE1 and AaAChE1 According to various statistical analyses, the experimental

conditions generated reliable and high quality screening data HTS proved a successful approach for identifying non-covalent

inhibitors of mosquito AChE1 No significant differences related to binding of non-covalent

inhibitors could be detected between the two mosquito enzymes Several hits were confirmed to inhibit the activity of AChE1 in a

dose-responsive manner The identified hits displayed chemical diversity both in terms of

structure and physicochemical properties

45

Chapter 3: Identification of selective AChE1-inhibitors by differential HTS analysis

Selectivity of the AChE1 hits To investigate the selectivity of the identified AChE1 hits, we also included the data from a HTS targeting hAChE (i.e. differential HTS) that was performed previously.297 This approach was possible since the assay conditions were comparable (e.g. same plate format, buffer conditions, and concentration of substrate and compound) and since the data normalization and hit identification was performed independently for each screening based on its own statistics. Comparison of the AChE1 and hAChE HTS data showed that the majority of the identified AChE1 hits exhibited low inhibition in the screening against hAChE. Indeed, only 10% of the AChE1 hits were also scored as hits against hAChE. To assess this potential mosquito over human selectivity, the IC50 values of the compounds from sets A-D were also determined against hAChE (examples are given in Table 17). Table 17. Assessment of mosquito over human selectivity.

Compound

IC50 (µM) S.Ra

AgAChE1 AaAChE1 hAChE IC50b HTSc

34 0.21 0.22 31 141 6.3

35 >100 >100 >200 n.a. 2.6

36 8 9 5 0.6 1.7

37 >1000 >1000 >1000 n.a. 76

38 >500 >500 >500 n.a. 23

39 17 12 >200 >12 11

40 21 19 43 2 n.a.

41 >1000 >1000 >1000 n.a. n.a.

42 3 3 >200 >67 2.7

aS.R: = selectivity ratios, n.a. = not applicable; bBased on the IC50 values. The

selectivity ratios were computed by taking the compound’s IC50 value for hAChE and

dividing by the higher of the its IC50 values for AgAChE1 and AaAChE1; cBased on the

HTS % inhibition (50 µM). The selectivity ratios were computed by taking the lower

of the compound’s inhibition (%) values for AgAChE1 and AaAChE1 and dividing by

its inhibition (%) value for hAChE. If a compound exhibited ≤ 0% inhibition in a

HTS, the inhibition was set to 1% prior to calculation.

Ultimately, this effort resulted in the identification of a number of

selective compounds warranting further investigation and development. We

46

confirmed the selectivity of 26 compounds that exhibited IC50 based selectivity ratios (S.RIC50) of at least ten (i.e. exhibited at least ten times lower IC50 values for AChE1 compared to hAChE), indicating that there are differences in the ligand binding properties of the mosquito and human enzymes that can be employed by non-covalent inhibitors to achieve selectivity.

One such compound is 34, a potent inhibitor of AChE1 (210 nM and 220 nM for AgAChE1 and AaAChE1, respectively), which exhibited a 140-fold loss of inhibitory activity against hAChE. No linear correlation between the % inhibition at 50 µM in the screenings and the corresponding IC50 values could be seen for the studied compounds. Nevertheless, this analysis demonstrated that it was possible to identify potentially AChE1-selective hits from their HTS inhibition profiles; hits that inhibited AChE1 at least five times more strongly than hAChE in the HTS were found to exhibit S.R.IC50 values of ten or more.

Structure-selectivity relationship Combining all the AChE1- and hAChE-hits resulted in a list of 425 unique compounds. Classification of their selectivity on the basis of the HTS inhibition profiles (i.e. the single point 50 µM % inhibition for AgAChE1, AaAChE1, and hAChE) yielded 163 AChE1-selective hits, 74 hAChE-selective hits, and 37 non-selective hits. The remaining hits were classified as ‘miscellaneous’ (see Paper II for more details). As described below, we next set out to explore whether we could detect any trends or patterns relating the chemical structures of the hit compounds to their selectivity on the basis of the HTS screening data.

Physicochemical properties and selectivity The physicochemical properties of the AChE1- and hAChE selective hits were described by calculated 2D molecular descriptors and an OPLS-DA model was used to correlate the properties of the compounds (the X matrix, i.e. 58 molecular descriptors) to the selectivity data (Y, i.e. AChE1- or hAChE selective). Evaluation of the resulting model showed that the predictive component described 52% of the total variation in Y (R2Y = 0.52) and had an internal prediction ability of 38% (Q2 = 0.38). Interpretation of the scores and loading plots (Figure 17) indicated that there were chemical differences between the two groups of compounds. The results suggested that the AChE1-selective hits tended to be smaller and more flexible than the hAChE-selective hits.

47

Figure 17. Score- and loading plot from the OPLS-DA model. a) Plot showing the

score vector of the predictive component (t[1]). Hits exhibiting potential selectivity

for AChE1 and hAChE are shown as black dots and grey squares, respectively. b) Plot

showing the loading vector of the predictive component (p[1]). The predictive

component mainly revealed separation due to size and flexibility.

Fragments and selectivity The investigation was continued by looking at fragments and substructures found among the hit compounds. The chemical structures of the AChE1- and hAChE hits were described by structural fingerprints, and cluster analysis yielded a chemical structural tree that revealed the similarities between the different hits (Figure 18). Three major groups were visible in the tree (I-III). The first group was generally characterized by more flexible molecules, the second group had a higher contribution of heteroatoms (N, O, S), and the last group contained hits with more aromatic and/or rigid substructures.

The relationship between different chemotypes and selectivity was studied by adding the selectivity profile of each hit compound to the tree. The resulting tree revealed smaller clusters of branches (i.e. compounds with similar structures) in which compounds were either potentially selective for AChE1, potentially selective for hAChE, or had mixed selectivity profiles (Figure 18c). For example, compounds 34, 39 and 42, which all show a higher inhibition of AChE1 (S.R.IC50: 141, >12, and >67, respectively), were located in clusters where most of the compounds were potentially AChE1 selective. On the other hand, compounds 31 and 36, with S.R.IC50: <1, were located in clusters in which the hits were potentially non-selective or exhibited mixed selectivity. By looking more closely at the chemical structures giving rise to these different types of clusters we were able to

48

recognize fragments and patterns, which appear to be important for selectivity.

Figure 18. a) Chemical structural tree of the 425 AChE1 and hAChE hits structurally

characterized by MACCS fingerprints. Compound similarity was described in terms of

Soergel distances and the tree was generated based on cluster analysis. The end of

each branch represents one hit molecule and the lengths of the branches are

proportional to the distance, i.e. similarity, between the hits. Three major groups (I-

III) of hits are shown, each against a grey background. Based on their HTS inhibition

data, hits exhibiting potential selectivity for AChE1 are shown as blue dots, non-

selective hits as purple squares, and hits exhibiting potential selectivity for hAChE as

yellow triangles. Compounds 31, 34, 36, 39, and 42 are labeled and their symbols

outlined in black. b) Chemical structures of 31, 34, 36, 39, and 42. c) Closer view of

part of the tree illustrating the presence of smaller clusters with different selectivity

profiles.

Eight compound classes Based on their overall chemical structures and scaffolds, the 425 hits were manually divided into approximately 70 different classes. Considering the results from the analyses described above, the AChE1 inhibition capacity (HTS data),

49

Figure 19. A representation of the eight different classes prioritized for further

investigation. The size of the circle is representative of the number of compounds in

each class. The coloring of the circle is representative of the selectivity profiles

exhibited by the compounds in the class, where AChE1 selectivity is represented by

blue, non-selectivity by purple, and hAChE selectivity by yellow. The 50 µM

inhibition capacity of the compounds in each class is shown for both AgAChE1 and

AaAChE1 (%). The diversity among the classes is exemplified by structural features,

median molecular weight and calculated logP.

and the selectivity profiles, eight of these classes were prioritized for further investigation. An illustration of the eight selected compound classes is shown in Figure 19. The number of compounds in each class varied from 2-21 and the classes were diverse both in terms of structural features and physicochemical properties. Some classes contained smaller, fragment-like302 compounds (classes 1, 4 and 8) while the others contained larger and more lipophilic compounds, although still obeying the ‘Rule of 5’.303

The activity and selectivity were confirmed for several compounds in each class, based on either newly synthesized (classes 1, 3, 4, 6, and 8) or purchased material (classes 2, 5, and 7). During the biochemical evaluation, class 2 was deemed as problematic since the compounds had solubility and

50

stability issues. In addition, class 7 consisted of false positives and was therefore disregarded from further investigation. The compounds in four of the remaining classes showed potential selectivity towards the mosquito enzyme (classes 1, 3, 6 and 8), while the remainder (classes 4 and 5) showed mixed selectivity profiles. The binding mode in mAChE has been studied using X-ray crystallography for representative compounds of classes 1, 3, 4, 6, and 8. The seven protein-ligand complexes obtained show that the different compound classes have distinct binding modes and are able to bind in different locations of the active site gorge (examples are shown in Papers II and III, and Appendix III).

Summary of Chapter 3 The data obtained from the three screenings against AgAChE1,

AaAChE1 and hAChE were combined and analyzed (using differential HTS)

Several potent AChE1 inhibitors with promising selectivity for AChE1 over hAChE were identified

Trends in physicochemical properties and chemical structure were related to selectivity

Six compound classes were identified and prioritized for further investigation

51

Chapter 4: Homology modeling of AChE1 (Paper II and Paper III)

Background Since the 3D structures of AgAChE1 and AaAChE1 have still not been determined experimentally we decided to use homology modeling to predict the structures of AgAChE1, AaAChE1, and AgAChE1-G119S (Paper II and III). Our aim was that structural information provided by the homology models would facilitate the explanation of experimental data (e.g. ligand potency and mosquito AChE1 versus hAChE selectivity), but also to investigate specific features of AgAChE1 and AaAChE1 potentially useful for the design of selective inhibitors.

In homology modeling, a known 3D structure of a similar protein is used as a template to model the sequence of interest (target). Thus, the resulting model is highly dependent on both the alignment of the template and target sequences and the template 3D structure. As a consequence, it is often difficult to obtain accurate predictions of the fold of loops, as well as of conformational changes induced by ligand binding. Homology models of AgAChE1 have previously been constructed.208, 218, 220 However, due to a number of unconserved loops, it is challenging to obtain high resolution models of AChE1. In addition, it is known that some residues (e.g. Tyr337) in the active site of AChE are flexible and that their conformations are ligand dependent.163-164, 221, 304 Therefore, we chose to include a ligand for induced fit during the modeling procedure to obtain homology models more relevant for the study of protein-ligand interactions.

This chapter describes the generation of the homology models, from the important template selection to the final models, but also includes structural comparisons to mAChE. The sequence and structure similarity between mAChE and hAChE is high, and mAChE was used here for comparison since this was the enzyme used for crystallization and soaking with AChE1 ligands throughout the work described in this thesis.

Selection of a template for modeling of mosquito AChE1 Both the general fold as well as the amino acids lining the active site are highly conserved among AChEs from different species (Figure 20). However, the two loops at the entrance of the active site (loop 1 and loop 2, Figure 20) differ between AChE from Drosophila melanogaster (DmAChE) and AChE from H. sapiens, M. musculus, and T. californica. In DmAChE, loop 1 is two residues shorter compared to vertebrate AChE and loop 2 contains one additional residue. It should be noted that DmAChE is encoded by the ace-2 gene.

52

Figure 20. Superposition

of AChEs shown as a ribbon

from H. sapiens (4EY4,173

orange), M. musculus

(1J06,163 cyan), T.

californica (1EA5,304 blue),

and D. melanogaster

(1QO9,164 magenta). Amino

acids defined as loop 1, loop

2, catalytic triad (shown as

ball and stick in black), and

the three loops where

mAChE was used as the

template are indicated by

arrows.

Being located at the rim of the gorge, changes to one or both of loops 1 and

2 could affect the ligand binding properties of the enzyme. The first loop is adjacent to Trp286, a key residue found at the entrance of the gorge, and extends from Leu289 to Phe297. According to multiple sequence alignment, loop 1 in AgAChE1 and AaAChE1 is three residues shorter compared to vertebrate AChE (i.e. one residue shorter than DmAChE; Table 11). Another important difference is the presence of the free cysteine in loop 1 of AgAChE1 and AaAChE1 (replacing Phe295). This cysteine is conserved among many insect AChE1 and has been targeted by covalent inhibitors.217-218, 224, 229, 305

Table 11. Amino acid sequence of loops 1 and 2 after multiple alignment.

Species Loop 1 Loop 2

An. gambiae LG - - -ICEF YLTELLRK

Aa. aegypti LG - - -ICEF YLTELLRK

H. sapiens LPQESVFRF - GAPGFSK

M. musculus LPQESIFRF - GVPGFSK

T. californica LPFDSIFRF - GAPGFSK

D. melanogaster YS - - GILSF DFIDYFDK

The identities of the considered templates, mAChE, hAChE, and DmAChE and the sequences to be modeled were 48%, 48%, and 39%, respectively, for AgAChE1 and 48%, 48%, and 38%, respectively, for AaAChE1. In analogy to DmAChE, the second loop (Gly342 to Lys348) in AgAChE1 and AaAChE1 contains one additional amino acid compared to vertebrate AChE (Table 11). Being a conformationally restricted amino acid, proline is often found in loops and turns and can influence their structures. In vertebrate AChE, loops 1 and 2 each contain one proline residue, however these prolines have been

53

replaced in loop 1 of Ag/AaAChE1 and DmAChE by glycine and serine, respectively, and in loop 2 of Ag/AaAChE1 and DmAChE glutamic acid and aspartic acid, respectively. Based on the observations described above, DmAChE was selected as the main template for modeling AgAChE1 and AaAChE1. This template was used to model all residues with the exception of three loops: Trp102-Thr112, Gly487-Gln499, and Ser512-Val520, for which mAChE was used as a template (Figure 20). Compound 31 was selected as the most suitable ligand to include in the modeling; 31 is a potent inhibitor of the mosquito, human, and mouse enzymes and its binding mode in mAChE (pdb code: 5FOQ223) has been studied by X-ray crystallography.

Homology models of AgAChE1 and AaAChE1 The generated homology models of AgAChE1 and AaAChE1 (Paper II) have very similar secondary structures (Figure 21a) and no structural differences were found in the active site gorge (Figure 21b).

The most obvious difference between AChE1 and mAChE is the different conformation of loop 1 (Figure 22). In the homology models, loop 1 makes a tighter turn compared to mAChE, and the shorter length of the loop alters the structure of the gorge’s rim (see below). The phenylalanine to cysteine replacement at position 295 also affects the structure of the gorge, particularly that of the acyl pocket.

Furthermore, the homology models suggest that the differences between mAChE and AChE1 in the sequence of loop 2 could alter the position of the α-helix consisting of residues Tyr337, Phe338, and Tyr341, among others. These aromatic residues are important for the shape of the gorge and their movement makes the binding pocket of AChE1 slightly larger than that of mAChE. The potentially larger CAS of AChE1 compared to that in mAChE is consistent with our previous finding that the former enzymes can hydrolyze larger substrates more effectively (Chapter 1) and that bulky substituents can improve the AgAChE1 over hAChE selectivity of aryl carbamates.226-227

54

Figure 21. 3D structures of AgAChE1 (cyan) and AaAChE1 (light pink) generated by

homology modeling. a) Superposition of AgAChE1 and AaAChE1 illustrating the

conserved fold and secondary structures. The RMSD between the heavy atoms of the

two models was 1.2 Å. b) Close up view of the active site view showing the side chains

of residues lining the gorge.

Figure 22. The homology model of AaAChE1 (light pink) and crystal structure of

mAChE (grey, pdb code: 5FOQ223). a) Superposition of AaAChE1 and mAChE

illustrating their similar secondary structures. One key difference is the structure of

loop 1 at the rim of the gorge. The RMSD between the backbone atoms of the two

structures was 1.7 Å. B) Close up view of the active site showing the side chains of

residues lining the gorge. Many of the amino acids in the binding site are identical.

55

A closer look at the entrance of the gorge As illustrated in Figure 23, the shorter loop 1 of AChE1 has a different conformation than the corresponding loop in mAChE. Loop 1 contains Cys295 and Phe295, for AChE1 and mAChE respectively. It has been reported that several inhibitors of mAChE and hAChE form a hydrogen bond to the backbone NH of Phe295.221, 223 However, in our models, the backbone NH of Cys295 is turned away from the active site suggesting that this interaction might not be as easily accessible in AChE1. The shorter length of loop 1 also makes room for a larger residue at the top of the adjacent α-helix, as seen for Tyr342 in AChE1 and the corresponding Gly342 in mAChE (indicated by an arrow in Figure 23). On the opposite side of the gorge, close to the important Trp286, Tyr72 in mAChE is exchanged for an isoleucine in AChE1. The smaller isoleucine side chain makes this part of the opening somewhat wider and has previously been implicated as an important factor for the selectivity reported for bis-tacrine ligands of vertebrate AChE.220 It is also possible that the Tyr/Ile substitution has an effect on both the position and the dynamics of Trp286 in AChE1.

The AgAChE1-G119S mutant The resistance conferring G119S mutation in AgAChE1 is located in the oxyanion hole of the active site gorge.137 To get an idea of how this mutation would affect the binding site, we performed an in silico point mutation of the generated AgAChE1 homology model (Paper III). In addition to decreasing the size of the oxyanion hole, our models also indicated a shift of the aromatic residue Tyr124 (Figure 24). The presence of the larger serine side chain, forced the aromatic ring to move in the opposite direction to the serine residue (the phenolic oxygen moved 2.1 Å away from its original position in the wild type model). This shift resulted in a narrower waist in the active site gorge compared to the wild type enzymes.

56

Figure 23. A close up of the entrance of the gorge as seen from above of (a)

AaAChE1 and (b) mAChE (pdb code 5FOQ223). Amino acids differing in size between

AaAChE1 and mAChE (positions 72 and 342) are indicated by arrows.

Figure 24. Molecular surface of the active site gorge. a) Homology model of wild

type AgAChE1 (cyan). The two glycines (Gly118 and Gly119) in the oxyanion hole are

marked in magenta. b) Homology model of mutant AgAChE1-G119S (orange). Gly118

and the mutated Ser119 in the oxyanion hole are marked in magenta. The differences

seen in the shape and size of the oxyanion hole and the acyl pocket resulting from the

larger serine residue and the altered shape of the waist due to a conformational

change of Tyr124 are indicated by arrows.

57

Summary of Chapter 4 Homology models of AgAChE1 and AaAChE1 were generated using

DmAChE as the main template The overall secondary structure of AChE is very similar across

different species The major differences between AChE1 and vertebrate AChE related

to ligand binding are seen in two loops located by the rim of the active site gorge

According to the homology models, the altered structure of one of these loops of AChE1 could modify the shape further down the gorge, making the active site slightly larger compared to mAChE

The G119S mutation affects the size and shape of the oxyanion hole and also, potentially, the waist located midway down the gorge

58

Chapter 5: Investigation of the structural basis for the selectivity of acetamide-based inhibitors (Paper II and Appendix III)

Background During the differential HTS study, compound 34 belonging to class 3 (Figure 19) emerged as a potent and selective AChE1 inhibitor (Figure 23). Interestingly, compound 31, which has a number of structural elements in common with 34, was very potent against all three enzymes. We therefore decided to use these two compounds as tools to investigate the selectivity properties of Ag/AaAChE1 and hAChE further (Paper II). As a second step, we wanted to investigate the SAR and SSR of compound 34 by design, synthesis, and biochemical evaluation of a number of analogues (Appendix III).

Figure 23. Chemical structures and IC50 values of 34 and 31.

Binding modes in mAChE The structural basis for the difference in selectivity of 34 and 31 was studied by comparing their binding modes in mAChE using X-ray crystallography. The compounds’ potency against mAChE revealed similar trends as those seen for hAChE (Figure 23); the IC50 values of 34 and 31 for mAChE were 18 µM and 0.026 µM, respectively (31 µM and 0.030 µM against hAChE, respectively).

The crystal structure of 34 in complex with mAChE was determined to a resolution of 2.8 Å (34mAChE, pdb code: 5FUM). This complex showed that 34 spanned the entire active site gorge, however, the electron density map allowed modeling of 34 in two different conformations (Figure 24a-b). One in which the ethylpiperazine was buried deep in the active site gorge and interacting with Trp86 (pose A; Figure 24a), and one where the biphenyl fragment was located at the bottom of the gorge close to Trp86 (pose B; Figure 24b). Both of these poses enabled a water mediated hydrogen bond

59

between the amide carbonyl and the backbone NH of Phe295. In pose A, the “proximal” phenyl ring (i.e. the phenyl closest to the amide) appears to form an aromatic stacking interaction with Trp286 in PAS, while the positioning of the piperidine-piperazine moiety likely allows activated CH···arene interactions with Tyr337 and Trp86.223 The piperidine linker is positioned in the narrow waist of the active site gorge formed by Tyr124 and Try337. It is possible that this results in unfavorable steric contacts with the enzyme. In pose B, the piperazine extends towards the opening of the gorge and is exposed to the solvent. However, the close proximity to Trp286 suggest favorable interactions to this residue. The “distant” phenyl ring appears to form an aromatic stacking with Trp86. The bulky piperidine linker is, according to pose B, positioned in a wider part of the gorge closer to the entrance, and thereby does not appear experience any steric hindrance.

Figure 24. Binding poses of inhibitors in the active site of mAChE. Where a) shows

pose A of the 34mAChE complex (pdb code: 5FUM), b) shows pose B of the

34mAChE complex (pdb code: 5FUM), and c) shows the 31mAChE complex (pdb

code: 5FOQ223). The side chains of mAChE are shown in grey (backbone N of Phe295

included due to hydrogen bond involvement), water is shown as red spheres, and

ligands 34 and 31 in cyan and yellow, respectively.

The bioactive conformation of 31 in mAChE (31mAChE, pdb code:

5FOQ,223 Figure 24c) can be compared to pose A of 34; both compounds form aromatic stacking interactions with Trp286 and the positively charged amines are located at the base of the gorge enabling activated CH···arene hydrogen bond interactions known to be important in AChE-ligand complexes.163, 173, 222-223 Still, 31 is almost 700 times more potent on mAChE than 34. Could this difference in potency be the due to 34 actually binding according to pose B, and thereby be lacking the important activated

60

CH···arene hydrogen bond interactions? Or is the lower potency of 34 the result of the less optimal interactions to Trp86 and/or steric effects due to the bulky piperidine linker (benzyl in 31)? To seek answers to these questions we set out to design analogues of 34, aiming to elucidate their SAR, SSR, and binding modes in mAChE and mosquito AChE1.

Design and synthesis of analogues

To investigate the possible binding modes of 34 in mAChE further, but also to study the SAR and SSR of 34, a small library of analogues was designed based on the chemical structure of 34. The design focused around three structural elements of 34; the importance of the ethylpiperazine (A, Figure 25), the influence of the piperidine linker (B, Figure 25), and finally the effect of the biphenyl moiety (C, Figure 25).

Figure 25. Three elements (A-C) of the parent molecule were varied in the design of

analogues.

Assuming that the bioactive conformation of 34 in hAChE corresponds to the one in mAChE, the design was based on the and following hypotheses: According to pose A,

the tertiary amine (element A) is important for the potency of the compounds

the size of the tertiary amine can affect the selectivity of the compounds

the piperidine linker (element B) is important for selectivity due to steric hinder

changes to the biphenyl fragment (element C) will not drastically effect the binding mode in mAChE/hAChE

According to pose B, the tertiary amine (element A) only has a moderate effect on potency the biphenyl fragment (element C) is important for the potency in

hAChE selectivity might be due to significantly different binding modes in

hAChE/mAChE and AChE1 To challenge these hypotheses, compounds were designed by varying the tertiary amines, incorporating a more flexible propyl linker, and by adding a 4’ substituent to the biphenyl. The designed compounds were synthesized as described below (Appendix III).

61

Synthesis of compounds The HTS hit 34 was synthesized according to Scheme 1. In the first step, compound 44a was prepared by means of a Williamson ether synthesis using 4-phenylphenol (43) and chloroacetic acid. The resulting carboxylic acid was coupled with 1-ethyl-4-pipiridin-4-yl piperazine using TBTU to yield the amide 34. Scheme 1.a

aReagents and conditions: (a) chloroacetic acid (5 eq), NaOH (s) (7 eq), MeOH,

reflux, 24h, yield of 44a 29%, yield of 44b 75%; (b) 44a, 1-ethyl-4-piperidin-4-yl-

piperazine (1.2 eq), TBTU (1.2 eq), TEA (3 eq), DMF, rt, 5 days; (c) Half sat. HCl in

CH2Cl2, 0 °C to rt, 20h, 47% yield over two steps.

The synthesis of analogue 47 was accomplished via the route described in

Scheme 2. Amide coupling of 4-morpholinopiperidine 45 and chloroacetic acid yielded amide 46, which was reacted with 4-phenylphenol (43) under basic conditions and using KI as a catalyst to give the target compound 47 in moderate yield. Scheme 2.a

aReagents and conditions: (a) chloroacetic acid (1.3 eq), TEA (1.3 eq), CH2Cl2, rt 1.5 h,

52% yield (b) 4-phenylphenol (43) (1.5 eq), K2CO3 (2 eq), KI (0.05 eq), DMF, rt, 3

days, 48% yield (c) half sat. HCl in CH2Cl2, 0 °C to rt, 4 h, 91% yield. Good yields of the truncated piperidine analogue 48a, and compound

48b, which is lacking the piperidine linker, were obtained from 44b after conversion to the acid chloride followed by addition of the corresponding amine (Scheme 3). The ethyl-piperazine analogue 48c was obtained from 44b by conversion to the acid chloride and then performing the amide coupling using a two-phase system formed by the use of CH2Cl2 and water as solvents.

62

Scheme 3.a

aReagents and conditions: (a) i) oxalyl chloride (5 eq), DMF (cat.), THF, rt, 1h ii)

piperidine or morpholine (2.2 eq), THF, rt, overnight, 68-72% yield; (b) i) oxalyl

chloride (2 eq), DMF (cat.) CH2Cl2, rt, 1h, ii) NaHCO3 (1 eq), N-ethylpiperazine (1 eq),

CH2Cl2:H2O 4:1, rt overnight, 69% yield; (c) 1M HCl in Et2O, CH2Cl2, rt, 10 min, 90%

yield.

Compounds 49a-c were prepared in low to good yields following the same

procedure as described for 48a-b (Scheme 4). The lower yield achieved for 49b (20%) was probably due to difficulties during the purification procedure. The TFA salt of 3-(4-ethylpiperazin-1-yl)propylamine proved more difficult to couple and the attention was again turned to the CH2Cl2:H2O two phase reaction used to synthesize 48c. Even though this method resulted in a low yield of 49d (Scheme 4), enough material was obtained for biochemical evaluation. Work is currently ongoing to understand the reasons for the significantly lower yield achieved for 49d compared to 48c.

Scheme 4.a

aReagents and conditions: (a) i) oxalyl chloride (5 eq), DMF (cat.), THF, rt, 1h ii)

corresponding amine (2.2 eq), THF, rt, overnight, 20-68% yield; (b) half-sat. HCl in

CH2Cl2, rt 10 min; (c) i) oxalyl chloride (2 eq), DMF (cat.) CH2Cl2, rt, 1h, ii) NaHCO3

(4.5 eq), 3-(4-ethylpiperazin-1-yl)propylamine 3*TFA (1.1 eq), CH2Cl2:H2O 4:1, rt

overnight, iii) 1M HCl in Et2O, CH2Cl2, rt 10 min, 11% yield

63

Finally, it was envisioned that using the Suzuki coupling reaction would give access to different para substituted aryl ethers. For this purpose, 4-iodophenoxy acetic acid (50) was converted to the acid chloride and coupled with various amines to produce the halogenated intermediates 51a-c, achieving a yield of 9-85% yield (Scheme 5). The Suzuki coupling with 4-methoxyphenylboronic acid using PEPPSI-iPr as the palladium source and microwave irradiation proceeded smoothly for the target compounds 52a-b, which were isolated in good yields. As experienced previously, good yields of the corresponding 3-(4-ethylpiperazin-1-yl)propylamine analogue proved elusive and, due to lack of material, 51c was not subjected to the Suzuki coupling.

Scheme 5.a

aReagents and conditions: (a) i) oxalyl chloride (5 eq), DMF (cat.), THF, rt, 1h ii)

corresponding amine (2.2 eq), THF, rt, overnight, 76-85% yield; (b) Half sat. HCl in

CH2Cl2, CH2Cl2 or MeOH, rt, 10 min, 83-100% yield; (c) i) oxalyl chloride (2 eq),

DMF (cat.) CH2Cl2, rt, 1h, ii) NaHCO3 (4.5 eq), 3-(4-ethylpiperazin-1-yl)propylamine

3*TFA (1.1 eq), CH2Cl2:H2O 4:1, rt overnight, iii) 1M HCl in Et2O, CH2Cl2, rt 10 min,

9% yield; (d) 2M Na2CO3 (3 eq), 4-metoxyphenylboronic acid (2 eq), PEPPSI-iPr

(0.05 eq), DMF, MWI 110 °C, 10 min, 86-98% yield; (e) 1M HCl in Et2O, CH2Cl2, rt,

10 min, 57-85% yield.

To obtain the 4-metoxyphenyl substituted version of 34, we first followed

the procedure described in Scheme 1 using 4-iodophenoxy acetic acid (50). Unfortunately, due in part to the poor solubility of 1-ethyl-4-piperidin-4-yl-piperazine, this approach failed to yield the desired halogenated amide. Instead, the methyl ester of 4-iodophenoxy acetic acid (53) was prepared and subsequently coupled with 4-methoxyphenylboronic acid by adapting a previously developed protocol (Scheme 6).306-307 Hydrolysis of the methyl ester yielded acid 54, which could be employed in the amide coupling with 1-ethyl-4-piperidin-4-yl-piperazine using TBTU to give the target compound 55 in 13% yield. Poor solubility of both the acid (54) and the triamine could be the reason for the low yield.

64

The route described for 55 was also used for the preparation of compound 56, the only difference being the conditions for the amide coupling reaction. For the coupling with the TFA salt of 3-(4-ethylpiperazin-1-yl)propylamine, the conditions used for 49d and 51c were slightly modified by increasing the amount of amine and changing the base to Na2CO3. Under these conditions 56 was obtained in 27% yield, which was a small improvement compared to the previous 11% and 9% yields of 49d and 51c, respectively.

Scheme 6.a

O

O

Y

O

O

O

I

OO

O

OH

O

a, b) c or d)

55: Y = N N N

NH

N

N

56: Y =

53 54

2*HCl

2*HCl aReagents and conditions: (a) 4-metoxyphenylboronic acid (2 eq), PEPPSI-iPr (0.01

eq), KF (2 eq), MeOH, MWI 110 °C, 10 min, 90% yield; (b) 0.1 M LiOH (1.3 eq)

overnight, 99% yield; (c) i) 1-ethyl-4-piperidin-4-yl-piperazine (1 eq), TBTU (1.3 eq),

TEA (3 eq), DMF, rt, 3 days, ii) 1M HCl in Et2O, CH2Cl2, rt, 20 min, 13% yield; (d) i)

oxalyl chloride (2 eq), DMF (cat.) THF, rt, 1h, ii) Na2CO3 (7.5 eq), 3-(4-

ethylpiperazin-1-yl)propylamine 3*TFA (2 eq), CH2Cl2:H2O 4:1, rt overnight, iii) 1M

HCl in Et2O, CH2Cl2, rt, 20 min, 27% yield.

Biochemical evaluation of analogues The inhibitory activity of the synthesized analogues was evaluated by determination of their IC50 values for AgAChE1, AaAChE1, and hAChE (Table 13). The following sections describe the SAR for AgAChE1 and AaAChE1 and the SSR over hAChE.

SAR for AChE1 of the synthesized analogues The biochemical evaluation showed that replacing the ethylpiperazine in 34 with a morpholine (47) was not tolerated by the AChE1 enzymes and resulted in a complete loss of activity. Compounds 48a-b further demonstrated the importance of the ethylpiperazine fragment; the truncated analogues showed no inhibition of AChE1. In addition, the activity could not be regained by replacing the piperidine in 48a with ethylpiperazine (48c), indicating that the placement of the piperazine fragment is also important.

65

In general, exchanging the piperidine linker for a less bulky and more flexible propyl chain resulted in a significant drop in activity against AChE1 (49a-d). For this set of compounds we also varied the size of the tertiary amine, going from a small dimethylamine (49a), via piperidine (49b), to the larger methyl- and ethylpiperazines (49c-d). Although these compounds exhibited reduced potency compared to 34, compounds 49b, 49c, and 49d still inhibited the activity of the AChE1 enzymes. The ethylpiperazine analogue 49d was marginally more potent than the other two; the IC50 values of 49b, 49c, and 49d were, respectively, 10 µM, 17 µM, and 3.7 µM for AgAChE1 and 10 µM, 14 µM, and 3.2 µM for AaAChE1.

The iodo intermediates 51a-c were also included in the evaluation. Of these, the piperidine (51a) and ethylpiperazine (51c) compounds proved to be equally potent (IC50 values around 40 µM), while 51b only showed weak inhibition of AChE1. Both 51a and 51c were less potent than their biphenyl analogues 49b and 49d (IC50 values were 3-10 times higher). The introduction of a methoxy group in the para position on the biphenyl element of 49b-d gave compounds 52a-b and 56. This modification had only minor effects on the potency the propyl linker analogues. Also the methoxy substituted analogue of 34, compound 55, exhibited very similar IC50 values as the parent compound (0.77 µM and 0.31 µM for AgAChE1 and AaAChE1, respectively), and was thus the most potent of all the designed compounds.

To summarize, from the SAR of this set of compounds it is clear that the ethylpiperazine (element A, Figure 25) is vital for the activity of 34 (cf. 47 and 48a). It could also be seen that the piperidine linker (element B, Figure 25) is very important for the potency; exchanging this fragment for a propyl chain is accompanied by a loss of activity (~15 times greater IC50 values). The trend for the propyl chain analogues (49a-d, 51a-c, 52a-b, 56) is that the ethylpiperazine or piperidine amines are more beneficial for activity compared to methylpiperazine, and especially the dimethylamine. Changes to the biphenyl moiety also effects the potency of the compounds (element C). The largest difference was seen for the iodo intermediates (cf. 51a-c versus 49b-d), which lost activity, while only minor effects was seen for the methoxy substituted analogues (52a-b, 55, and 56).

66

Table 13. Biochemical evaluation of designed and synthesized analogues. Parent

compound 34 is included for comparison (marked in grey).

IC50 (µM)a

ID Structure AgAChE1 AaAChE1 hAChE S.Rb

34 N

O

O

N

N2*HCl

0.21 (0.12-0.37)

0.22 (0.12-0.38)

31 (29-34)

141

47

>500 >500 >500 n.a.

48a

>1000 >1000 >1000 n.a.

48b

>1000 >1000 >1000 n.a.

48c

>1000 >1000 >1000 n.a.

49a

119 (89-159)

149 (118-189)

207 (161-267)

1.4

49b

10 (7.9-14)

10 (7.6-14)

56 (53-60)

6

49c

17 (16-19)

14 (13-15)

195 (163-235)

12

49d

3.7 (3.2-4.2)

3.2 (2.7-3.9)

94 (62-141)

25

51a

55 (28-107)

36 (21-61)

15 (11-19)

0.3

51b

226 (146-351)

317 (187-537)

297 (136-652)

1.0

51c

37 (33-43)

43 (33-56)

66 (33-131)

1.5

52a

5.2 (3.6-7.7)

4.2 (2.1-8.6)

12 (8.7-16)

2.3

67

Table 13. Continued.

52b

31 (24-42)

28 (16-51)

>400 >13

56

6.4 (4.9-8.4)

4.8 (3.1-7.5)

108 (74-158)

17

55

0.77 (0.60-1.0)

0.31 (0.18-0.53)

16 (8.9-29) 21

a95% confidence interval given in parentheses; bS.R. = selectivity ratio, computed by

taking the compound’s IC50 value for hAChE and dividing by the higher of its IC50

values for AgAChE1 and AaAChE1. n.a. = not applicable

Selectivity for AChE1 over hAChE The parent compound 34 exhibited a 141 times greater potency for AChE1 than hAChE. Compounds 47 and 48a-c, which were inactive on AChE1, also showed no inhibition of hAChE. The propyl chain analogues exhibited decreased selectivity for AChE1 compared to 34; 49a-d had 6-25 times greater IC50 values for hAChE than AChE1 (IC50 values in the range of 50-200 µM), with 49d being the most selective for AChE1. These compounds exhibited similar SAR trends for hAChE as for AChE1 with 49a and 49c being the least active. Interestingly, iodo substituted 51a proved to be more active against hAChE than AChE1; the IC50 value was 15 µM for hAChE compared to 36-55 µM for AChE1. The other to iodo containing compounds (51b-c) were equally potent on hAChE and AChE1.

Finally, the introduction of a para methoxy substituent resulted in an increase in potency of almost five times on hAChE for 52a compared to 49b, resulting in 52a being non-selective for AChE1. The methyl- and ethylpiperazine analogues 52b and 56 on the other hand exhibited more than 13 times higher IC50 values for hAChE than AChE1 (S.R. >13 and 17 for 52b and 56, respectively). The methoxy analogue of 34, compound 55, was slightly more potent on hAChE compared to 34. The IC50 value of 55 for hAChE was 16 µM, resulting in a decrease in selectivity ratio from 141 to 25.

The general SSR for AChE1 over hAChE seen for the compounds in this study suggests that the amine (element A, Figure 25) is very important for selectivity. The ethylpiperazine contributes most to selectivity, while the piperidine seems to be good for hAChE activity and thereby reduces the selectivity of the compounds. The limited number of analogues in this study prevents definite conclusions regarding the effect of the piperidine linker (element B). Comparison of 34 and 49d suggests that this fragment is very important for selectivity, but on the other hand 55 and 56 exhibit very

68

similar selectivity ratios. Finally, the aromatic moiety (element C) can have pronounced effects on the selectivity; exchanging the biphenyl for a para iodophenyl (51a-c), resulted in a decreased, or even reversed, selectivity. A much smaller effect on selectivity was observed with respect to the introduction of a methoxy substituent (cf. 49b-d versus 52a-b and 56).

Binding modes of propyl analogues The bioactive conformations of 49b and 49c when binding to mAChE was studied using X-ray crystallography. The crystal structures of 49bmAChE and 49cmAChE were determined at resolutions of 2.7 Å and 2.5 Å, respectively (Appendix III). The electron density maps showed that both compounds bound in elongated conformations spanning the entire active site gorge (Figure 26). The binding modes of 49b and 49c were very similar. The biphenyls were extending towards the rim of the gorge indicating aromatic interactions between their proximal phenyls and Trp286. The positions of amide carbonyl groups of 49b and 49c suggested a hydrogen bond with the backbone NH of Phe295, while the piperidine (49b) or the piperazine (49c) were buried deep at the base the gorge presumably participating in activated CH···arene interactions with Tyr337 and/or Trp86 (Figure 26).

Structural basis for the observed SARs and SSRs The information obtained from the crystal structures of 31, 34, 49b, and 49c in mAChE was used in combination with homology modeling to investigate the structural basis for the observed trends in activity and selectivity of the compounds in this series. Differences in binding modes to mAChE and possible binding modes in mosquito AChE1 are discussed below.

Rationalization of the SAR for hAChE To explore the observed SAR for hAChE, which is important for the understanding of selectivity, the different binding modes of 31, 34, 49b and 49c in mAChE were compared. The increased potency of 55 compared to 34 for hAChE strongly suggested that 34 is binding in the active site of hAChE, and most likely also mAChE, according to pose A (Figure 24a). Pose B is improbable, since the introduction of a para methoxy substituent would result in steric clashes between the substituent and the residues at the base of the gorge and, thereby, a reduced potency for 55 compared to 34.

A comparison of the binding modes of 34 and 31 showed that the piperidine in 31 was located approximately 2.5 Å deeper down in the gorge compared to the piperazine in 34 (Figure 27a). It is possible that this shift in position leads to weaker interactions with Trp86, and thereby a lower activity of 34 compared to 31. In addition, the piperidine linker of 34 is

69

located in the narrow waist of the gorge indicating that conformational strain and/or unfavorable steric contacts could negatively affect the potency of 34.

Figure 26. Binding poses of inhibitors in the active site of mAChE. Where a) shows

the 49bmAChE complex, and b) shows the 49cmAChE complex. The side chains of

mAChE are shown in grey (backbone N of Phe295 included due hydrogen bond

involvement) and ligands 49b and 49c in green and magenta, respectively.

Figure 27. Superposition of the binding poses seen in the active site of mAChE.

Where a) shows 34 and 31, b) shows 49b and 49c, c) shows 49b and 31, d) shows

49b and 34, and e) shows 49c and 34. The ligands 31, 34, 49b, and 49c are shown

in yellow, cyan, magenta, and cyan, respectively.

70

Compounds 49a-d were designed to investigate the role of the bulky piperidine linker in 34, with the motivation that a propyl linker could alleviate possible strain or steric hindrance induced by the piperidine. However, the biochemical evaluation showed that exchanging the piperidine for a propyl chain was accompanied by a loss of activity for hAChE. The observed binding modes of 34, 49b, and 49c are similar, which is consistent with the observed moderate reduction of potency (2-6 times higher IC50 values). With respect to the important interactions with Trp86, the propyl linker did not alter the position of the piperidine (49b) or methylpiperazine (49c) compared to the ethylpiperazine in 34 (Figure 27d-e). As illustrated in Figure 27, the position of amide carbonyls of 49b and 49c differs slightly from that seen in 34 and 31 (Figure 27b-e). This alteration allows 49b and 49c to form a hydrogen bond directly with the backbone NH of Phe295; in the complexes of 34 and 31 this hydrogen bond is mediated by a water molecule. Thus, it can be said that changing the linker from a piperidine to a propyl chain does not significantly alter the binding mode of the compounds. Instead it appears that either the piperidine is able to lock the ligand in a favorable conformation and/or that favorable interactions between the piperidine and the residues lining the gorge are lost by the introduction of the more flexible propyl linker.

Possible binding modes in AChE1 The ligand complexes in mAChE contributed to the understanding of the SAR observed for hAChE. We therefore wanted to explore the possibility of using homology modeling to, on a structural level, rationalize the SAR for AChE1. However, generation of biologically relevant poses in AChE can be challenging due to the properties of the binding site; the gorge is lined by aromatic residues and their conformations are ligand dependent.297 One explanation for the modeling difficulties could be the inability of current force fields to estimate correctly the importance and strength of the activated CH···arene interactions that are key in AChE-ligand complexes.223 Since many of the residues in the active sites of AChE1 and mAChE are conserved, it is reasonable to assume that the ligands studied here would bind to AChE1 in a similar manner to that seen in mAChE. We therefore manually docked compound 49c in AaAChE1 based on it binding mode in mAChE. The generated complex was then subjected to energy minimization, during which the ligand was allowed to move freely (Appendix III).

The resulting model suggested that 49c may be able to bind to AChE1 in a slightly different pose compared to mAChE. It appears that the smaller residue in position 72 (Ile in AChE1 and Tyr in hAChE) could allow 49c to adopt a different binding pose in AChE1 compared to mAChE (Figure 28a). The change in the position of the biphenyl group in PAS could potentially lead to a T-shaped stacking between the distant phenyl of the ligand and

71

Trp286, as well as a CH···arene interaction between said phenyl and Ile72. These interactions could contribute to the higher inhibition capacity seen against AChE1 compared to hAChE. In addition, such an interaction pattern would be lost by exchanging the distant phenyl group for an iodine and could potentially explain the loss of activity on AChE1 seen for 51c compared to 49c. In the 49c•mAChE complex it is the proximal phenyl ring that is responsible for the interaction with Trp286 (Figure 26b). Substituting the distant phenyl for an iodine would therefore, in agreement with the IC50 data, not have the same dramatic effect on hAChE activity.

Figure 28. Possible binding poses of 49c and 34 in the homology model of

AaAChE1. a) Shows the suggested pose of ligand 49c (magenta), where the biphenyl

group was interacting with Trp286 and Ile72 and the piperazine fragment was

located deeper in the gorge compared to the binding mode in mAChE, and b) shows

the ligand 34 (cyan) adopting a similar conformation to that modeled for 49c. The

side chains of AaAChE1 are shown in light pink and the backbone N of Cys295 is

included for comparison. In the AaAChE1 model, the piperazine fragment of 49c is located

approximately 2.5 Å further down in the gorge compared to its position in 49c•mAChE. This altered position of the tertiary amine could lead to stronger interactions with Trp286 and potentially also Tyr337. In addition, this pose also suggests the possibility of a hydrogen bond to Glu202. Altered interactions at the base of the gorge, could thereby potentially contribute to the selectivity for AChE1 over hAChE. The binding mode and interactions presented in Figure 28a are also applicable for the piperidine and ethyl-piperazine analogues 49b and 49d, as well as for 34 and 56 (Figure 28b). It is therefore plausible that altered interactions both by the rim and the base

72

of the gorge are contributing to the AChE1 selectivity observed for this series of compounds. Still, the suggested models do not explain all of the SAR and SSR trends seen in Table 13 and should only be considered to represent one possible binding mode. Additional analogues should be designed and evaluated to explore the suggested hypothesis and more advance modeling could yield a more comprehensive picture.

Summary of Chapter 5 Compound 34 was identified in the differential HTS study as a

potent and selective inhibitor of mosquito AChE1 X-ray crystallography suggested two possible binding modes of 34 in

mAChE A number of analogues were designed and synthesized to explore the

SAR and SSR of 34 Biochemical evaluation showed that the piperidine linker in 34 is

pivotal for potency, but that the amine and the biphenyl fragments are also influential with respect to both the activity and selectivity of the studied compounds

Modeling of analogue 49c in mosquito AChE1 suggested possible differences in the interaction pattern of the selective compounds in mosquito AChE1 compared to mAChE

73

Chapter 6: Thiourea-based AChE1 inhibitors (Paper III)

Background One of the interesting classes identified in the differential HTS study was based on a thiourea scaffold (class 6, Figure 19 and Table 14). In the work described in Paper III the SAR and SSR of this class of compounds was explored further by design, synthesis, and biological evaluation of a small library of compounds.

Table 14. Hits belonging to class 6 identified in the differential HTS study (Chapter

3).

AgAChE1 AaAChE1 hAChE S.Ra,b

57 HTS (%) 86 90 -1 86

IC50 (µM) 0.96 0.63 12 13

58 HTS (%) 65 67 0 65

IC50 (µM) 7.8 5.5 90 12

59 HTS (%) 46 51 22 2.1

IC50 (µM) 7.1 6.0 120 17

60 HTS (%) 48 49 -4 48

IC50 (µM) 11 11 >1000 >91

61 HTS (%) 44 50 1 44

IC50 (µM)c >1000 >1000 >1000 n.a.

aHTS-based selectivity ratios were determined by taking the lower of the compound’s

inhibition (%) values for AgAChE1 and AaAChE1 and dividing by its inhibition (%)

value for hAChE. If a compound exhibited ≤ 0% inhibition in a HTS the inhibition

was set to 1% prior to calculation; bIC50-based selectivity ratios were computed by

taking the compound’s IC50 value against hAChE and dividing by the higher of its IC50

values against AgAChE1 and AaAChE1. n.a. = not applicable; cSolubility issues

experienced during IC50 determinations.

74

As described in Chapter 3, five thiourea-based compounds with potential selectivity towards mosquito AChE1 over hAChE were identified in the differential HTS study (Figure 19 and Table 14). These hits consisted of asymmetric thioureas carrying one aryl substituent, and one tertiary amine connected to the thiourea via an ethyl linker. Four of the hits (57-60) exhibited IC50 values for AChE1 between 0.63 and 11 µM and selectivity over hAChE (S.RIC50 ≥12). Compound 61, which exhibited poor solubility in the follow-up assays, did not inhibit any of the enzymes in a dose-dependent manner and was considered to be false positive.

The confirmed inhibitors all carried different aromatic substituents and different tertiary amines (Table 14). We therefore decided to design a balanced set of compounds to investigate how the aromatic substitution pattern (i.e. the position and the properties of the substituents) and the size of the amine effect the activity and selectivity of the compounds. In this study we also wanted to evaluate the compounds’ ability to inhibit the insecticide-resistant form of AChE1 found in An. gambiae, AgAChE1-G119S. The naturally occurring G119S mutation confers resistance towards current anti-cholinergic insecticides (organophosphates and carbamates) and is therefore highly relevant also for non-covalent inhibitors of AChE1. An important step towards our long term goal to develop safe insecticides for vector control purposes is to explore and verify the in vivo insecticidal activity of our compounds. Thus, a few of the designed compounds were selected for in vivo experiments on larvae and adult An. gambiae and Ae. aegypti mosquitoes.

Design of analogues Three of the hits (57-59) were selected to form the basis of a design intended to explore SAR differences between AgAChE1, AaAChE1, and hAChE, but also the insecticide insensitive AgAChE1-G119S mutant. The design strategy was to vary the aromatic moiety and the tertiary amine, while keeping the ethyl-linker fixed. The final sets of compounds were designed around the aromatic moiety of each of the hits (sets A-C, Figure 29). The tertiary amines were commonly selected across all three sets and were varied so that each building block returned several times, but in combination with different aromatic moieties.

75

Figure 29. Representative chemical structures of the three design sets A-C. X and Y

show the substitution positions on the aromatic moiety for the different sets. The

tertiary amines were varied cross all three sets (R1 and R2).

Synthesis Depending on commercial availability of the starting materials, ease of purification, and safety, thioureas 57-60 and 62-77 were synthesized using one of three methods (Scheme 7). The thioureas could be readily accessed from commercially available phenylisothiocyanates and primary amines in yields of 70-98% by microwave assisted heating for 10-15 minutes (method A).

In other cases, the thioureas were synthesized in a one-pot two-step reaction, where the isothiocyanate was formed in situ using thiocarbonyldiimidazole (TCDI, method B) or thiophosgene (method C). In method B, a substituted aniline was allowed to react with TCDI, after which a primary amine was added to yield the thiourea after heating (17-77% yield). Unfortunately, in some cases the use of TCDI and the required heating resulted in poor yields and the formation of byproducts, which greatly hampered the purification of the product. Therefore, two of the thioureas were synthesized using thiophosgene, allowing the reaction to proceed at -10 °C to rt (method C). Thiophosgene is a toxic and volatile liquid, but here produced fewer byproducts than TCDI.

Scheme 7.a

NCSR H2N

N RNH

NH

N

S

Method A

Methods B and C

NH2

RX X

S

B: X = 1-imidazolylC: X = Cl

NCSR R

NH

NH

N

S

+

+

a)

b or c) d or e)

R2

R1

R2

R1

R2

R1

aReagents and conditions: (a) CH2Cl2, MWI 70 °C, 10-50 min, 70-98% yield; (b)

MeCN, MWI 100-120 °C, 10-20 min; (c) MeCN, -10 °C to rt, 5-10 min; (d)

corresponding amine, MWI 100-120 °C, 10-20 min, 17-77% yield; -10 °C to rt, 2 h, 52-

56% yield.

76

SAR analysis of the thiourea-based AgAChE1- and AaAChE1 inhibitors The inhibition capacity of the synthesized thioureas was investigated by determining their IC50 values against AgAChE1, AaAChE1, hAChE, and AgAChE1-G119S. In this section, the SARs for sets A-C are described (Tables 15-17).

Set A The di-substituted parent compound 57 was the most potent of the thiourea-based HTS hits; it had IC50 values of 0.96 µM and 0.63 µM for AgAChE1 and AaAChE1, respectively. The importance of both substituents became instantly clear; removal of either the chloro- or methoxy substituent resulted in a complete loss of activity (62 and 63, Table 15). Interestingly, as seen for compound 64, and to some extent 65, some of the activity could be regained for the mono-substituted analogues by altering the tertiary amine from a methylpiperidine in 57, to a slightly smaller piperidine. The effect of the amine was further demonstrated by compounds 60, 66-71. The morpholine analogues (60, 66, and 67), proved to be even more potent than both their methylpiperidine and piperidine counterparts. On the other hand, the introduction of the smaller dimethylamine (69-71) was accompanied by a decrease in potency.

The most potent compound in this set was the morpholine analogue (66) of the parent compound. The IC50 values of 66 for AgAChE1 and AaAChE1 were 0.12 µM and 0.090 µM, respectively. As seen for 57, removal of one of the aromatic substituents from 66 resulted in a substantial loss of activity with compounds 60 and 67 exhibiting approximately 100 and 600 times higher IC50 values, respectively. However, the loss was less pronounced compared to that seen for the methylpiperidine compounds 57, 62, and 63. The importance of the aromatic substitution pattern was further demonstrated by compound 68; swapping the position of the 3-chloro-4-methoxy phenyl substituents to 3-methoxy-4-chloro was not tolerated by the AChE1 enzymes.

Based on these results, the SAR of set A showed that the potency of the compounds can be increased by changing the tertiary amine from methylpiperidine to morpholine. It was also seen that, independent of the tertiary amine, the substituents on the phenyl moiety have a great effect on the activity of the compounds with the di-substituted 3-methoxy-4-chloro phenyl being the most beneficial for potency.

77

Table 15. Structures and IC50 values of compounds in design set A, the parent

compound 57 is included for comparison (marked in grey).

No Structure AgAChE1

(M) AaAChE1

(M) hAChE

(M) S.Ra

57 NH

NH

S

N

Cl

O

0.96 (0.67-1.4)

0.63 (0.27-1.5)

12 (8.5-17) 13

62

inactiveb inactiveb inactiveb n.a.c

63 NH

NH

S

N

Cl

inactiveb inactiveb inactiveb n.a.c

64

29 (19-45)

28 (18-44)

120 (82-160) 4

65

140 (110-180)

120 (86-170)

190 (121-302) 1

66 NH

NH

S

N

O

Cl

O

0.12 (0.11-0.14)

0.090 (0.083-0.098)

14 (13-15)

117

60

11 (8.7-13)

11 (9.2-14) >1000 >91

67 NH

NH

S

N

O

Cl

68 (41-110)

53 (45-62) >1000* >15

68

99 (81-120)

72 (51-100) >1000 >10

69

18 (17-20)

16 (14-19)

46 (39-53) 3

70 NH

NH

S

N

O

>1000 >1000 >500 n.a.c

71

96 (85-110)

83 (70-99)

140 (120-170) 1

aS.R. = selectivity ratio computed as hAChE/AgAChE1; binactive up to 1 mM; cn.a. =

not applicable.

78

Set B In the design of set B we aimed to investigate the effect of the para substituent on the phenyl ring further (Table 16). More specifically we wanted to explore how the size, polarity, and electronic properties, of the para substituent contributed to potency and selectivity.

Table 16. Structures and IC50 values of compounds in design set B, the parent

compound 58 is included for comparison (marked in grey).

No Structure AgAChE1

(M) AaAChE1

(M) hAChE

(M) S.Ra

58

7.8 (6.9-8.7)

5.5 (4.8-6.2)

90 (72-110)

12

72 NH

NH

S

N

F3C

>200 >200 >200 n.a.c

73

~200 ~200 >1000 n.a.c

74

inactiveb inactiveb inactiveb n.a.c

aS.R. = selectivity ratio computed as hAChE/AgAChE1; binactive up to 1 mM; cn.a. = not applicable. Parent compound 58, containing a sulfonamide in the para position and a piperidine as the tertiary amine had IC50 values of 7.8 M and 5.5 M for AgAChE1 and AaAChE1, respectively. Compound 58 with its electron withdrawing para substituent was thus more potent than its analogue 64, carrying the electron donating methoxy-substituent. Exchanging the sulfonamide for a trifluoromethyl, another electron withdrawing but smaller and more hydrophobic substituent, led to a significant decrease in potency as seen for compound 72. With respect to the amine, changing the piperidine (58) to a morpholine (73) also led to a significant loss of activity. In addition, compound 74 carrying a para chloro substituent and morpholine exhibited no enzyme inhibition when tested up to a concentration of 1 mM.

Although the number of para substituted compounds are limited, the results indicate that the activity of 58 is not merely an electronic effect of the aromatic substituent. The sulfonamide can act both as a hydrogen bond acceptor and donor, and it is possible that these abilities are important for the activity of 58. Additional compounds, specifically designed to investigate the importance of the hydrogen bond ability, could help further the

79

understanding of the SAR of this set. With respect to the tertiary amine, the results indicate that a piperidine is more beneficial for activity than morpholine.

Set C The design of the final set was based on compound 59, which had IC50 values of 7.1 and 6.0 M for AgAChE1 and AaAChE1, respectively. Compound 59 differed from both 57 and 58 in that it had both an ortho substituent on the phenyl ring (methoxy) and a smaller amine (dimethyl). The results of the biochemical evaluation of set C are presented in Table 17.

Table 17. Structures and IC50 values of compounds in design set C. The parent

compound 59 is included for comparison (marked in grey).

No Structure AgAChE1

(M) AaAChE1

(M) hAChE

(M) S.Ra

59

7.1 (6.5-7.6)

6.0 (5.6-6.5)

120 (100-130) 17

75

17 (14-22)

14 (13-16)

>200 >12

76

29 (26-33)

22 (19-24)

32 (29-35) 1

77

>1000 >1000 >1000 n.a.c

aS.R. = selectivity ratio computed as hAChE/AgAChE1; binactive up to 1 mM; cn.a. =

not applicable.

It was observed that changing the methoxy group for a chlorine (75)

resulted in a small decrease in activity. The same trend was also seen when the dimethylamine in 59 was exchanged for a piperidine (76). The effect of the amine was even more pronounced for the morpholine analogue, as seen for compound 77 which failed to inhibit the enzymes up to a concentration of 1 mM. The limited data for set C indicate that for ortho substituted thioureas and smaller amines are beneficial for activity.

The general trends seen over all three sets were that both the properties of the aromatic substituents and their positions have dramatic effects on the inhibitory capacity of the studied compounds. Similar observations have previously been noted for mAChE.222 We also conclude that with respect to the tertiary amine the three sets had different SARs (compare for example 57 and 66 with 58 and 73 or 59, 76, and 77 with 70, 64, and 60).

80

Structure-selectivity relationship over hAChE The parent compound in set A (57) exhibited a 13 times greater inhibition capacity of AChE1 than hAChE. For AChE1, an increase in activity was seen when exchanging the methylpiperidine in this compound for a morpholine (66). The same trend was not seen for hAChE, instead 66 and 57 had very similar IC50 values for hAChE (14 µM and 12 µM, respectively) resulting in an increased selectivity of 66 for AChE1 (S.R = 117) compared to 57 (Table 15). For hAChE it was also found that exchanging the piperidine (64 and 65) for a morpholine (60 and 67) was, in contrast to AChE1, accompanied by a loss of activity. In general, the piperidine and dimethylamine containing analogues exhibited similar potencies towards hAChE and AChE1 (S.R. = 1-4). In set B, compound 58 was less potent on hAChE compared to AChE1 (S.R. = 12). The remaining compounds in the set (72-74) exhibited very weak to no inhibition of hAChE, in agreement with the data for AChE1 (Table 16). The meta methoxy substituted 59 and meta chloro substituted 75 (set C) exhibited similar selectivity over hAChE (S.R. = 17 and >12, respectively; Table 17). However, going from the dimethylamine in 59 to the piperidine led to an increase in potency against hAChE, and thus a non-selective compound (76). Exchanging the piperidine for a morpholine resulted in a significant loss of activity also on hAChE (77).

The SSR for set A shows that, independent of the aromatic substituents, large amines are needed to obtain selectivity of the compounds. In particular, the morpholine is beneficial for selectivity. The compounds in set B exhibited very similar SARs for AChE1 and hAChE, but the potency of 58 was lower for hAChE compared to AChE1. Finally, for the ortho substituted compounds (set C), the small dimethylamine was crucial for selectivity. It appears that with regards to the tertiary amine, the different sets display different SSRs, as well as different SARs.

The AgAChE1-G119S mutant The determination of the compounds’ potencies on mutated enzyme was affected by the enzyme’s the lower catalytic activity, and resulted in larger uncertainties in the IC50 values for AgAChE1-G119S (see Tables 1-3 in Paper III). However, it could generally be seen that all of the studied compounds exhibited a decrease in potency for AgAChE1-G119S compared to the wild type. The data indicate that compounds 57, 66, and 71 (set A) were equally active on this enzyme, with IC50 values of approximately 20 µM (see Table 1 in Paper III). These values were more similar to those seen for hAChE and correspond to 57 and 66 having ~25 and ~200 times greater IC50 values for the mutant compared to the wild type, respectively. For the unselective 69, with the smaller dimethylamine, the IC50 for AgAChE1-G119S was in the same range as those for AgAChE1, AaAChE1 and hAChE (Table 15; and Table 1 in Paper III). For compound 58 (set B) the IC50 value (~30 µM)

81

suggested an inhibition capacity on the mutant that was intermediate between AgAChE1 and hAChE (see Table 2 in Paper III). The same trend was seen for 59 (set C), which had an IC50 value of ~50 µM for AgAChE1-G119S (see Table 3 in Paper III).

Overall, these data indicate that the SARs of sets A-C for AgAChE1-G119S were more similar to the SAR:s seen for hAChE than for AChE1. For example, it appears that loss of activity for the mutant compared to the wild type is more pronounced for the larger amines in set A. These findings are in agreement with what has previously been reported for aryl methylcarbamates, where it was found that selectivity for AgAChE1-G119S over hAChE can be challenging.230-232 Still, compounds 58 and 59 show weaker inhibition of hAChE compared to both AgAChE1-G119S and AgAChE1, indicating a possibility to develop non-covalent inhibitors which are targeting both the wild type and the mutant, but which are safe for humans.

Study of bioactive conformations in mAChE We envisioned that analysis of the binding modes in mAChE could provide a structural explanation for the trends observed in the hAChE IC50 data. In addition, the analysis would give information regarding possible binding modes in mosquito AChE1, which in combination with homology modeling could help to explain the structural basis for the selectivity of the compounds. The binding modes of 57 and 59 in mAChE were studied using X-ray crystallography and the complexes were determined to a resolution of 2.3 Å and 2.1 Å for 57•mAChE and 59•mAChE, respectively. The IC50 values of 57 and 59 for mAChE were 26 µM and 217 µM, respectively.

Binding mode of compound 57 from set A In the 57•mAChE complex it could be seen that the ligand spanned the entire active site gorge (Figure 30a). The thiocarbonyl was most likely forming a hydrogen bond to the backbone NH of Phe295, while the position of the di-substituted phenyl ring suggested a π-π stacking with the indole of Trp286 at the entrance (Figure 30b). It is possible that the meta chloro substituent also has contacts with both Trp286 and the phenol of Tyr72, however the exact positioning of the substituents on the phenyl ring could not be determined from the obtained electron density map. In addition, the electron density corresponding to the methylpiperidine (specifically C3-C5 and the methyl substituent) was mostly undefined suggesting that this fragment is mobile. Still, the location of the piperidine ring indicated that an activated CH···arene interaction between the piperidine and Tyr337 is possible.

It is evident from the IC50 data that the aromatic substituents of the compounds has great influence on the activity. Assuming that all compounds

82

in set A exhibit similar bioactive conformations in mAChE, it is not clear from the 57•mAChE complex why removal of either the methoxy or chloro substituent has such a dramatic impact on the activity as seen for 62 and 63. The chloro substituent in 57 is able to interact with both Trp286 and Tyr72, but the general trend in set A is that this substituent is less important than the methoxy group. It is possible that the electronic properties of the aromatic rings affect their aromatic π-stacking interactions.246 The para methoxy substituent renders both the phenyl ring and the thiocarbonyl in 60 and 64 slightly more electron rich compared to 65 and 67. This could potentially be beneficial for both the interaction with Trp286 and the hydrogen bond to Phe295.

Binding mode of compound 59 from set C In the crystal structure of mAChE in complex with 59 it could be seen that 59 adopted a very different binding pose compared to 57 (Figure 30c). In the 59•mAChE complex the ligand was binding at the bottom of the gorge with its phenyl ring forming favorable contacts with Trp86 (Figure 30d). The dimethylamine was extending towards the opening of the gorge, and its proximity to Trp341 indicated that activated CH···arene interactions are possible to this residue (Figure 30d). The conformation of 59 was stabilized by an intramolecular hydrogen bond between the positively charged amine and the thiocarbonyl. Several water molecules were found in the vicinity of the ligand, one of which formed a hydrogen bond to the NH of the thiourea functionality (Figure 30d).

Compound 59 exhibited an approximately eight times higher IC50 value for mAChE than 57. The lower activity of 59 could be due to the lack of hydrogen bonds and the complete absence of interactions with Trp286 at the entrance of the gorge. The ortho substituent on the phenyl ring forces the thiourea to adopt a different conformation compared to meta and para substituted thioureas; the plane of the phenyl ring is perpendicular to the plane of the thiourea functionality (59) as compared to in the same plane (57). This restricted rotation of the phenyl ring in 59 could possibly hinder the ligand from binding in PAS and interacting with Trp286, as seen for 57. In addition, the reduced potency of 59 may be due to the positioning of the dimethylamine. This fragment is located in the waist part of the gorge and steric clashes with Tyr124 may contribute to the low activity in hAChE and mAChE.

83

Figure 30. The binding modes of 57 and 59 in mAChE. a) Showing the extended

binding mode of 57. Due to the disordered electron density map corresponding to the

aromatic moiety, 57 was modeled in two conformations. mAChE is shown as a grey

ribbon with residues Trp86 and Trp286 included for spatial recognition; b) A close

up of the active site gorge with 57 bound within it; c) Showing the binding mode of

59. mAChE is shown as grey ribbon with residues Trp86 and Trp286 included for

spatial recognition; d) A close up of the active site gorge with 59 bound within it. The

ligands 57 and 59 are shown in orange and pink respectively, the side chains of

mAChE are shown in grey (backbone N of Phe295 is included due to hydrogen bond

involvement), and water molecules are represented by red spheres.

84

Study of possible binding modes in mosquito AChE1 using homology modeling To further our understanding of the structural basis for the observed SSRs we used homology models of AgAChE1 and AgAChE1-G119S to explore possible structural differences between mAChE and mosquito AChE1. The analysis was based on our homology models and the assumption that 57 and 59 would exhibit very similar binding modes in mosquito AChE1 as those seen in mAChE.

Possible binding modes of 57 in AgAChE1 As illustrated in Figure 31a, most of the amino acids in close proximity to 57 are conserved between mAChE and AgAChE1 and no new potential interactions that could contribute to the higher potency in AgAChE1 could be identified. One difference between mAChE and AgAChE1, which might be related to the difference in potency, is found at position 295. In the 57•mAChE, the thiocarbonyl appears to form a hydrogen bond to the backbone of Phe295. In our AChE1 models, neither the side chain nor the backbone NH of Cys295 are accessible for ligand interactions. However, the exact position of this residue is unknown; for instance the work by Pang et al. have strongly indicated that the Cys295 thiol could be targeted by covalent inhibitors.218, 224-225, 308 It can therefore not be ruled out that the thiourea functionality of the inhibitors may interact with either the backbone or the side chain of Cys295. That such an interaction alone could be the source for the selectivity is, however, unlikely since it is known that thiocarbonyl groups are, in general, poorer hydrogen bond acceptors than carbonyls.309 It is possible that the differences in PAS between AgAChE1 and mAChE, such as the differences in the conformation of loop 1 and the Tyr/Ile change at position 72, affects the interactions of both the thiocarbonyl and the 3-chloro-4-methoxyphenyl groups of 57. However, more advanced modeling is required to explore such hypotheses.

Possible binding modes of 59 in AgAChE1 Based on the homology modeling and the assumption that 59 would bind to AgAChE1 according to the pose in mAChE (Figure 31b), it is difficult to understand the almost 30 fold increase in potency of 59 seen for AgAChE1 compared to mAChE; the residues at the bottom of the gorge are completely conserved between mAChE and AgAChE1. As discussed in Chapters 1, 4, and 5, and other work concerning aryl methylcarbamates,226-227 it is possible that more room at the bottom of the gorge (CAS) in AChE1 could be the source of the selectivity observed for 59. More room could possibly relieve potential internal strain of 59 and potentially allow the phenyl ring to form a more optimal T-shaped stacking with Trp86 (Figure 31b). However, the large difference in potency between mosquito and vertebrate enzymes together

85

with the different SARs for AChE1 and hAChE observed for 59, could be an indication that 59 has a completely different binding mode in AgAChE1.

Figure 31. Possible binding modes of 57 (orange) and 59 (pink) in the homology

models of AgAChE1 and AgAChE1-G119S. a) Compound 57 modeled in the active site

of AgAChE1. b) Compound 59 modeled in the active site of AgAChE1. c) Compound

57 modeled in the active site of AgAChE1-G119S. d) Compound 59 modeled in the

active site of AgAChE1-G119S. The side chains of AgAChE1 and AgAChE1-G119S are

shown in cyan and dark green, respectively. The backbones of Gly120, Gly121 and

Ser121 (shown in magenta) are included for clarity and the backbone N of Cys295 is

included for comparison.

86

Possible effects of the G119S mutation in AgAChE1 With respect to the insecticide-resistant enzyme AgAChE1-G119S, our modeling revealed no obvious steric clashes for 57 (Figure 31c). This compound was able inhibit AgAChE1-G119S, although the inhibition was weaker compared to the wild type; IC50 values for AgAChE1-G119S and AgAChE1 are ~25 µM and 0.96 µM, respectively. The reduced potency on AgAChE1-G119S may be due to steric effects induced by the larger serine residue, which would also explain the more pronounced effects on the larger amines in set A compared to the smaller ones (cf. 57 and 66 vs 69, Table 1 in Paper III).

From the binding pose seen in mAChE, one would assume that 59 would be seriously affected by the glycine to serine mutation; the thiocarbonyl points directly towards the oxyanion hole. However, the effect of the mutation is smaller for 59 than 57; 59 exhibits ~7 times greater IC50 value for the mutated enzyme compared to the wild type compared to ~25 times difference for 59. The IC50 data and the modeling of 59 in AgAChE1-G119S indicate that the thiocarbonyl could still fit and possibly even form a hydrogen bond to the alcohol of the serine (Figure 31d). It is possible that the loss of activity is instead the result of the narrower waist (caused by a shift of Tyr124), resulting in steric clashes with the dimethylamine group. However, this could also be an additional indication that 59 has an entirely different binding mode in AgAChE1 compared to that observed in mAChE.

Mosquito testing The insecticidal activity of compounds 59, 66, and 70 was evaluated on both larvae and adults of An. gambiae and Ae. aegypti (Tables 18 and 19). In the larvae experiments (Table 18), compound 66 resulted in 100% and 97% mortality of An. gambiae and Ae. aegypti, respectively, after 24 hours when tested at 500 µM. However, at a lower concentration (50 µM) no significant mortality was observed for either of the species. Compounds 59 and 70 (IC50 values ~7 and >1000 µM for AChE1, respectively) did not have any effect on larval mortality at the highest concentration tested (500 µM).

The adult mosquitoes were exposed to the compounds via topical application (Table 19). Here, clear differences were seen between the two species. For An. gambiae compound 66 produced a dose-dependent mortality when tested at concentrations 0.1 mM, 2 mM, and 10 mM, with 70% mortality achieved at the highest concentration. For Ae. aegypti, application of 66 at a concentration of 10 mM did not result in any mortality.

87

Table 18. The insecticidal activity of compounds 59, 66, and 70 on third instar An.

gambiae and Ae. aegypti larvae after 24h. Compounda

Species Concentration 59 66 70

An

. ga

mbi

ae

(% m

orta

lity

) 50 µM 5 1 4

300 µM n.t. t.b.d. n.t.

500 µM 3 100 3 A

e. a

egyp

ti

(% m

orta

lity

) 50 µM 7 8 4

300 µM n.t. t.b.d. n.t.

500 µM 5 97 1

an.t. = not tested by choice; t.b.d. = to be determined.

Compounds 59 and 70 yielded similar results when tested on An. gambiae. Both compounds exhibited lower in vivo activity compared to 66, and produced 55-57% mortality when tested at 100 mM. Neither of the compounds had a significant effect on the mortality of Ae. aegypti.

The concentrations required here to generate in vivo activity on larvae and adult mosquitoes are ~4000 and 8000-14000 times higher than the determined IC50 values, respectively. Similar trends have been observed for both non-covalent233 and covalent AChE1 inhibitors.187, 226, 230, 232 A possible explanation is that exoskeleton penetration and/or other absorption, distribution, metabolism, excretion (ADME) issues affect the compounds’ ability to reach the target. Differences in the exoskeleton between adult An. gambiae and Ae. aegypti could also be a reason for the lack of effect on Ae. aegypti. The in vivo effect on adult An. gambiae seen for 70 could potentially be a result of the high concentrations (70 exhibits ~30% inhibition of AgAChE1 at 1 mM) and/or improved penetration or ADME of this compound compared to 66 and 59. It is also possible that the observed effect is the result of a different mode of action.

88

Table 19. The insecticidal activity of compounds 59, 66, and 70 on adult An.

gambiae and Ae. aegypti mosquitoes after 24h.

Compounda

Species Concentrationb 59 66 70

An

. ga

mbi

ae

(% m

orta

lity

) 0.1 mM n.t. 0 n.t.

2 mM 16 22 0

10 mM 0 70 32

100 mM 57 n.d. 55 A

e. a

egyp

ti

(% m

orta

lity

) 0.1 mM n.t. n.t. n.t.

2 mM 8 5 1

10 mM 10 0 0

100 mM 10 n.d. 12

an.t. = not tested by choice; n.d. = not determined due to solubility issues. b0.1 µl of

the compound solution with the indicated concentration was applied to each

mosquito.

Further studies are needed to investigate how the physicochemical

properties of the compounds and their formulations are related to exoskeleton penetration and other ADME properties in mosquito. Still, the results of the in vivo experiments presented here are encouraging and motivate further investigations of non-covalent AChE1 inhibitors and their potential for insecticide development.

Summary of Chapter 6 A set of thiourea-based hits was identified in the differential HTS

study and the three most promising compounds were selected for further study

A design, consisting of sets A-C, was developed to study the SAR and SSR of the compound class

The designed analogues were obtained via one of three synthetic pathways

Biochemical evaluation showed that the SAR and SSR of the compounds were greatly influenced by the substitution pattern on the phenyl ring

The different sets displayed different SAR:s Most of the compounds showed lower inhibition of the insecticide-

resistant mutant AgAChE1-G119S compared to the wild type enzyme

89

X-ray crystallography showed that compounds 57 (set A) and 59 (set C) have different binding modes in mAChE, which could be an explanation for the different SAR:s

Modeling of the compounds in AgAChE1 did not give a full explanation for the observed selectivity over hAChE

In vivo insecticidal activity was observed for the most potent compound on both larvae and adult An. gambiae and larvae Ae. aegypti

90

Concluding discussion This thesis focuses on the vectors An. gambiae and Ae. aegypti, mosquitoes responsible for transmission of malaria, lymphatic filariasis, dengue, yellow fever, chikungunya, and the Zika virus. Today more than half of the world’s population are at risk of vector-borne diseases, especially people living in tropical or sub-tropical areas. Not only do vector-borne diseases take approximately 1 000 000 lives per year, they also impose heavy economic burdens and cause enormous pain and suffering.11 Worst affected are the poorest of the poor, and other already vulnerable parts of the population such as young children, pregnant women, and immunocompromised people.11 We are living in a rapidly developing and ever changing world, where the fight against vector-borne diseases is constantly challenged by numerous factors including climate change, urbanization, deforestation, human behavior, global travel and trade, conflicts and war, and natural disasters. Recent decades have seen a rapid spread of arboviral infections in particular and only now are we beginning to discern the possible lasting impact that these diseases will have on human health and society.

Is there a future for vector control? In light of the challenges discussed above, along with increasing insecticide resistance, and advances in the development of vaccines and more effective drugs, one may ask whether vector control has had its day? Although effective vaccines are highly desirable and would be a valuable tool in the fight against vector-borne diseases, it would probably take a long period of time before everyone living at risk are protected. In addition, a vaccine only protects against one disease. In many areas several vector-borne diseases are present at the same time, making several different vaccines necessary. Effective vector control methods on the other hand, offer the possibility of protection against several diseases. This is particularly useful when one vector is responsible for transmitting several different pathogens, as in the case with Ae. aegypti. But it has, for instance, been shown that the mosquito nets (LLINs) used against malaria can also protect against leishmaniasis (transmitted by sandflies).310-314

It is also important to note that one treatment or prevention method does not eliminate the need for others. It has been shown that combining different vector control methods with rapid and accurate treatment as well as information and education can effectively reduce disease burden.315-320 The vector-borne disease burden is a multifactorial problem including the complexity of disease transmission, the different environments and settings where they act, and the varying disease symptoms and care requirements. Thus, the consensus is that integrated approaches, where different

91

interventions are combined, are needed to achieve sustainable and large scale reductions in vector-borne diseases.9, 11, 34, 46, 321

Covalent or non-covalent inhibitors of mosquito AChE1? Covalent inhibition of AChE by organophosphates or carbamates is a proven insecticidal strategy. However, these compounds are limited in their value (only recommended for IRS) and their effectiveness is seriously compromised by insecticide resistance. Could there be benefits of using non-covalent AChE1 inhibitors instead of or as a complement to the covalent inhibitors? This section aims to discuss some of the different aspects of covalent versus non-covalent inhibition and their potential implications for vector control.

Are non-covalent inhibitors sufficiently potent? One advantage of covalent inhibition is that, since it is not subject to equilibrium kinetics, lower and/or less frequent doses are typically needed to attain biological activity.322-324 Can sufficient potency also be achieved by non-covalent inhibition of AChE1, and how potent do the compounds need to be? Today, no non-covalent, AChE1 targeting insecticides are in use and reference compounds are lacking in the literature. The insecticides propoxur and bendiocarb are covalent reversible inhibitors and their IC50 values for mosquito AChE1 have been determined to be 350-500 nM and 130-220 nM, respectively (after 10 min incubation).228 These data provide a rough estimate of the required potency, however these inhibitors have a different mode of action compared to the non-covalent inhibitors discussed in this thesis and direct comparisons, therefore, cannot be made. The most potent compounds presented in this thesis display IC50 values in the range 100-200 nM on AChE1 (Chapters 3, 5, and 6). One of these compounds exhibited an insecticidal effect in vivo on both larvae and adult An. gambiae mosquitoes; however, as will be discussed in more detail below, pharmacokinetic factors may be equally or even more influential on in vivo activity than the target potency.

Safety and selectivity over non-target organisms As discussed in Chapters 1 and 4, the current understanding is that the active site close to the catalytic triad is highly conserved between AChEs from different species. However, the work by Carlier et al. has shown that useful selectivity for AgAChE1 over hAChE can be achieved by introducing larger and more branched aryl substituents on the aryl methylcarbamates targeting the catalytic serine residue (S.R.: 200-600).226-227 Targeting the insect AChE1 specific Cys295 by covalent modification is another approach to achieving selectivity over hAChE.224-225 One issue of concern regarding covalent inhibitors is the potential risk of toxicity due to the reactivity of the

92

compounds. Care must be taken when selecting the reactive functionality in order to obtain an inhibitor that is sufficiently electrophilic to react with the target, but not so reactive that they react with cellular proteins at random.323

As illustrated in this thesis, reversible inhibitors do not depend on a covalent bond to the catalytic serine residue or a free cysteine for their potency, but instead achieve target affinity through a pattern of non-covalent interactions. They can therefore bind in different parts of the active site gorge, thus allowing, at least in theory, for more freedom to exploit sites that can provide selectivity and/or potency. Our differential HTS study identified several compound classes with diverse chemical structures and physicochemical properties (Chapter 3). Importantly, they also exhibit different binding modes in mAChE, indicating that selectivity can be obtained in different locations of the active site gorge (Chapters 3, 5, and 6).

Resistance development The fact that covalent inhibitors rely heavily on one covalent link to the enzyme also renders these inhibitors sensitive to mutations in the active site. One example is the cross-resistance obtained for many organophosphate- and carbamate insecticides via the G119S mutation in AgAChE1.136-137, 140, 179,

208 Recent research has shown that small core carbamates can overcome this resistance, however the development of carbamates that are selective for AgAChE1 over hAChE and also active on the resistance conferring mutant AgAChE1-G119S seems challenging.187, 230-232 As discussed above, non-covalent inhibitors rely on a pattern of protein-ligand interactions and can bind in different parts of the active site gorge (and potentially also outside the active site gorge). This makes it very difficult to predict how target site mutations such as G119S, F290V, or F331W will affect the potency of non-covalent inhibitors. As seen in Chapter 1, although both 31223 and donepezil (30)177 span the entire active site gorge from PAS to CAS, both of these inhibitors maintain most of their activity on AgAChE1-G119S compared to the wild type enzyme. On the other hand, the studied thiourea-based compounds (Chapter 6) were more sensitive to the G119S mutation and generally exhibited lower potencies on the mutant enzyme compared to the wild-type.

Pursuing several compound classes The Innovative Vector Control Consortium 325-326 was founded in 2005 and aims to develop three novel public health insecticides with different modes of action. The reasoning is that using different types of insecticide either in rotation or in mixtures will reduce the development of insecticide resistance. The same reasoning has been applied in other fields, i.e. combination therapy protocols are well established in medicine as a way of achieving the

93

desired therapeutic effect and circumventing resistance issues. Examples include treatment of malaria (ACT), 327 and other complex diseases, such as cancer, asthma, type 2 diabetes mellitus, bacterial and viral infections.328-330 Although the compound classes presented in this thesis have the same molecular target (AChE1), there are several potential benefits of carrying forward work on more than one compound class.

Insecticide resistance As mentioned above, different compound classes could be affected differently by target site mutations depending on their binding mode. In addition to target site resistance, metabolic resistance is perhaps an even bigger threat to the effectiveness of vector control insecticides. Would the risk of metabolic and/or cross-resistance be lower if several insecticides with diverse chemical structures and physicochemical properties were available? Possibly, and at least in theory it would also be beneficial if new insecticides have different chemical structures compared to the insecticides in use today. In addition, a lot is known from drug discovery on how to modify non-covalent compounds to make them more metabolically stable.331-332 Since mosquitoes and humans rely on the same detoxification systems (P450 enzymes, esterases, and glutathione S-transferases), the methods developed to enhance metabolic stability of drugs could also be applicable in insecticide research and should be considered early in the development process.

Insecticide entry and reaching the target In 1997 Lipinski reported on the ‘Rule of 5’,303 stating that orally available drugs at that time generally did not have more than five hydrogen bond donors, ten hydrogen bond acceptors, a molecular weight more than 500, and an octanol-water coefficient (logP) no greater than five. A corresponding study on insecticides reached similar conclusions with the exception that insecticides generally have fewer hydrogen bond donors (≤2).333 It is encouraging to note that the chemical classes identified in Chapter 3 fulfill the criteria outlined by Tice.333 However, some caution is warranted since it is not clear whether these trends are a reflection of suitable ADME properties of insecticides or the result of the insecticide discovery process and industry.

Our initial in vivo experiments indicate that penetration across the exoskeleton of the mosquito, and possibly other ADME problems as well, might be a major hurdle when developing new compounds. Similar observations have been made concerning the in vivo activity of both non-covalent AChE1 inhibitors233 and novel carbamates on An. gambiae.187, 226,

230, 232 In a study by Conley et al. of dopamine receptor agonists, female Ae. aegypti were subjected to the most promising compounds by microinjection.334 Again, working simultaneously with several different

94

compound classes could increase the likelihood of finding the “right” chemical properties to reduce penetration problems and other pharmacokinetic issues.

Different vector control products Due to behavioral differences, different mosquitoes require different vector control strategies and new tools are needed, especially to control the day active Aedes mosquitoes.34 Furthermore, the environmental setting and community acceptance will affect both the choice and the effectiveness of a vector control product. In addition to the currently used methods (IRS and ITNs), insecticide-treated materials such as curtains, wall hangings, and screens are under development, as are lethal ovitraps (i.e. oviposition traps equipped with insecticides) and toxic sugar baits.34, 46 It is likely that as the number of insecticide-based vector control products grows, so will the need for different compounds that meet their requirements with respect to formulation, durability, solubility and attractiveness.

Design of mosquito selective insecticides Obtaining selectivity between structurally similar proteins is a major challenge not only in insecticide development but also in drug discovery.335 The work in this thesis aimed to investigate the possibilities of selectively inhibiting mosquito AChE1 with non-covalent inhibitors to improve human and environmental safety of future insecticides. In this section the major findings regarding selectivity are summarized and discussed.

The sequence comparison in Chapter 1 showed that, irrespective of species, the residues neighboring the catalytic triad at the base of the active site gorge are conserved, whereas the amino acids at the entrance of the gorge exhibit greater variation. Differing amino acids by the entrance of the active site gorge, such as Phe/Cys295224-225 and Tyr/Ile72,220 have previously been implicated as contributors to mosquito over human selectivity. The work presented in this thesis indicates that these residues could also be of importance for the selectivity of the compounds studied here (Chapter 5 and 6).

It is also possible that the differences seen in the PAS region have more far-reaching effects on the size and shape of the active site gorge. As discussed in Chapters 4, 5, and 6, the differences in the two loops at the rim of the gorge (loops 1 and 2) likely also change the structure, and potentially the dynamics, of residues surrounding the catalytic triad at bottom of the gorge. Indeed, there are a number of indications that the mosquito AChE1 binding site is larger and/or has a different shape than that in hAChE. In particular, the CAS region, including the acyl pocket and choline binding site, appears to differ between the mosquito and vertebrate enzymes, as indicated by the more effective hydrolysis of larger substrates

95

(Chapter 1), the increased potency of ethopropazine (19) compared to hAChE (Chapter 1 and Carlier et al.220), the increased selectivity observed for larger amines such as ethylpiperazines (Chapter 5), and the AgAChE1 selectivity observed for bulky aryl methylcarbamates,226-227. Assuming that the thiourea-based compound 59 binds to AChE1 in a similar fashion to that seen in mAChE, AChE1 selectivity could be achieved by ligands binding exclusively in CAS (Chapter 6). On the other hand, if the AChE1 active site gorge is larger than in hAChE, it would seem contradictory that the OPLS-DA model of the HTS hits indicated that AChE1 selective hits are smaller and more flexible than the hAChE selective hits (Chapter 3). It can generally be assumed that smaller compounds will generally have fewer possible interaction points compared to larger compounds and so, to achieve high affinity, each interaction needs to be strong. One explanation for the trends described in Chapter 3 could therefore be that the selective compounds are small enough to bind deep in the gorge of AChE1 and there form strong and specific interactions with the enzyme. The lower potency in hAChE may be explained by steric exclusion by the slightly smaller catalytic site, forcing the ligands to instead form weaker interactions to the residues surrounding the entrance.

The work presented in this thesis shows that the relationships between the chemical structure of non-covalent AChE1 inhibitors and their selectivity over hAChE are not trivial, but most likely the result of intricate interaction patterns that are highly dependent on the chemical structure of the ligands. One example of this is the case of the morpholine. In the thiourea study, the introduction of a morpholine proved beneficial for both the potency and the selectivity of the compounds in design set A (Chapter 6). That morpholines can have a negative effect on the activity of hAChE inhibitors has been reported previously,336-338 supporting the idea that this could be a design strategy for selective AChE1 inhibitors. However, this is not a general trend. As described in Chapter 5, exchanging the ethylpiperazine fragment in compound 34 for a morpholine resulted in a complete loss of activity for both the mosquito and human enzymes.

While the work described here shows that it is possible to selectively target mosquito AChE1 with non-covalent inhibitors, it is also clear that more studies are needed to guide the design and optimization of selective inhibitors. Such efforts could include other compound classes, more advanced modeling, and additional assays, for instance to study entropic and enthalpic contributions to ligand binding. In addition, experimentally determined 3D structures of mosquito AChE1 would, needless to say, be of great help to determine the structural basis for selectivity. Furthermore, studies using designed mutants of the enzymes could also contribute valuable information regarding structural differences between the mosquito and vertebrate enzymes.

96

Studies are also needed to investigate exoskeleton penetration and other ADME processes in the mosquito. The compounds presented here could be used tools to study such processes and thereby contribute to the understanding of how potential insecticides’ physicochemical and pharmacokinetic properties are related.

97

Acknowledgements This project has been a challenge as much for the heart as for the brain. I would therefore like to express my appreciation and sincere gratitude to every single one of you who have contributed and made this work possible. Thank you.

Det här projektet har varit en utmaning lika mycket för hjärtat som för hjärnan. Jag vill därför uttrycka min mest ödmjuka och uppriktiga tacksamhet till de personer som på ett eller annat sätt varit med och gjort det här möjligt. Tack. Jag vill speciellt nämna, Min handledare Anna Linusson – för att du alltid ser och bryr sig om helheten, framför allt när det kommer till forskningen men också allt annat som hör livet till. Jag upphör aldrig att förvånas över din förmåga att veta precis när du ska stötta, pusha, utmana eller hjälpa. Tack vare dig har jag lyckats med sådant som jag aldrig vågat drömma om att ens försöka mig på. Jag finner inga ord för att beskriva hur tacksam jag är för att du tog mig under dina vingar. Fredrik Ekström – för att ingen kan ge mitt självförtroende en skjuts så som du. För att du introducerat mig till kristallografins underbara värld, ställt upp i vått och torrt och med kort varsel, varit excellent resesällskap och fått mig att se saker från helt nya infallsvinklar. Min biträdande handledare Fredrik Almqvist – för att du funnits med i en eller annan skepnad ända sedan studietiden. Din genuina glädje och entusiasm för forskning är både inspirerande och smittande. Mina fantastiska och kompetenta gruppmedlemmar som ställt upp bortom all rim och reson för att detta skulle nå i mål. Cecilia Engdahl – för att du ska (kommer) rädda världen. Jag är så glad och tacksam för att du välkomnade mig in i projektet med öppna armar och gjorde mig till din kompanjon. Bättre samarbetspartner än dig hade jag aldrig kunnat få. Lotta Berg – för att du är min bästa alltiallo - vän, syster, kollega, bollplank, klagomur, kritiker, hejaklack och allt däremellan. Ja, helt enkelt bäst! David Andersson – för att du är en fantastisk kontorskompis som alltid tar dig tid att hjälpa, diskutera och svara på frågor, stöta och blöta statistik om och om igen, och inte minst för att du drog med mig på yoga. Cecilia Lindgren – för att du är så väldigt bra att resonera med och för att du utan att tveka hoppade in på labb som en räddande ängel. Tomas Kindahl – för att du

98

sedan första början varit en suverän kontors- och labb kompis som jag alltid kunnat rådfråga och visa både snygga och hemska NMR för. Och ja Tomas, jag har börjat med avhandlingen nu! Hela gänget på FOI, framför allt Nina Forsgren – för suveräna IC50 bestämningar, Lund resor och hjälp med kristall strukturer. Luna Kamau and her group at KEMRI - for their amazing work with the mosquitoes and for realizing my dream of having one of my compounds tested in vivo! Diploma workers and students, especially Darius Nikjoo, Peter Demirel, Svetlana Ivanova, Malin Johansson, Tommy Orre, and Hans Stenlund, for their contributions to this work. Lennart Johansson, den evige sekreteraren, och Dan Johnels, den evige ordföranden – för att ni ställt upp vid alla mina uppföljningsmöten. Erik Chorell och Nils Pemberton - för att ni för snart tio år sedan gav mig chansen och släppte in mig till ert labb, lät mig göra spännande kemi och tog er tid att lära mig i princip allt jag kan om syntes. To everyone in the corridor (both old and new) for creating such a good working environment. Speciellt tack till Lina Mören (ibland måste man bara få ”ringa och gnälla”), Christoffer Bengtsson (namnreaktionernas mästare), Caroline Zetterström (mjölkklubben är viktig) och Hanna Uvell (som gjorde mig till cellkvinna). Ett stort tack också till Liz, Maria, Barbro, Tina, Carina, L-G, Sara, Rosita, Pelle, Mattias, Tobias och Kemiförrådet – för att ni hållit koll på allt från kurser, ekonomi, datorer, anställningar, instrumentering och kemikalier. Min fina och omtänksamma vän Hanna - för trevliga luncher och överraskningar när man som mest behöver dom. The Thevathasans - for making me part of your family and for all the love, support, and encouragement I have been given. Kerstin, Björn och Thomas – inte minst för fantastiska söndagsmiddagar och besök i Ramsta och Hemavan. Kerstin, du kommer aldrig förstå hur många dagar din matlagning inte bara har räddat utan förgyllt.

99

Mormor och morfar, farmor och farfar – för att ni generöst delar med er av likväl hemmaodlad mat som egna historier, erfarenheter och tankar som ger välbehövligt perspektiv på tillvaron. Mamma och Pappa – för att ni förbjöd mig att gå vårdprogrammet på gymnasiet men i övrigt alltid låtit mig ta mina egna beslut och sedan stöttat dom. Robin och Malin, Emil och Elin, Lotta och Patrik – för att om det så är Idol/Mello/Grillning, vinterstudion eller Middagstips så gör ni livet både så mycket lättare och roligare! Underbara Annie och Alfred - för att oavsett hur mycket vi busar, åker ’låtsas rutschkana’ eller leker ’Häng med’ så har jag alltid mer energi och ork efter att ha hängt er. Michael - för att du är min enda, min lycka, och min sista pusselbit. Och mitt lilla Sprattel, som hållit mig sällskap såväl sena kvällar som helger sedan första ordet i den här avhandlingen, och som redan kommit med så mycket hopp och glädje.

100

References 1. Ross, R., The Role of the Mosquito in the Evolution of the Malaria Parasite. Lancet 1898, 488-489. 2. Manson, P., Surgeon-Major Ronald Ross's Recent Investigations on the Mosquito-Malaria Theory. Br. Med. J. 1898, 1, 1575-7. 3. Reed, W.; Carroll, J.; Agramonte, A.; Lazear, J. W., The Etiology of Yellow Fever-a Preliminary Note. Public health papers and reports 1900, 26, 37-53. 4. Reed, W.; Carroll, J.; Agramonte, A., The Etiology of Yellow Fever. An Additional Note. JAMA, J. Am. Med. Assoc. 1901, Feb 16. 5. Reed, W.; Carroll, J., The Prevention of Yellow Fever. Public health papers and reports 1901, 27, 113-29. 6. Alphonse Laveran - Nobel Lecture: Protozoa as Causes of Diseases. http://www.nobelprize.org/nobel_prizes/medicine/laureates/1907/laveran-lecture.html (accessed: 19 Feb. 2016). 7. Ronald Ross - Nobel Lecture. http://www.nobelprize.org/nobel_prizes/medicine/laureates/1902/ross-lecture.html (accessed: 19 Feb. 2016). 8. WHO, WHO Fact Sheet on Vector-Borne Diseases. http://www.who.int/mediacentre/factsheets/fs387/en/ (accessed: 18 Feb. 2016). 9. WHO, World Malaria Report 2015. Geneva: World Health Organization 2015. 10. Suaya, J. A.; Shepard, D. S.; Siqueira, J. B.; Martelli, C. T.; Lum, L. C. S.; Tan, L. H.; Kongsin, S.; Jiamton, S.; Garrido, F.; Montoya, R.; Armien, B.; Huy, R.; Castillo, L.; Caram, M.; Sah, B. K.; Sughayyar, R.; Tyo, K. R.; Halstead, S. B., Cost of Dengue Cases in Eight Countries in the Americas and Asia: A Prospective Study. Am. J. Trop. Med. Hyg. 2009, 80, 846-855. 11. WHO, A Global Brief on Vector-Borne Diseases. Geneva: World Health Organization 2014, (WHO/DCO/WHD/2014.1). 12. WHO, Zika Situation Report. Available at: http://www.who.int/emergencies/zika-virus/situation-report/19-february-2016/en/ (accessed: 29 Feb. 2016). 13. Brasil, P.; Pereira, J., Jose P.; Raja Gabaglia, C.; Damasceno, L.; Wakimoto, M.; Ribeiro Nogueira, R. M.; Carvalho de Sequeira, P.; Machado Siqueira, A.; Abreu de Carvalho, L. M.; Cotrim da Cunha, D.; Calvet, G. A.; Neves, E. S.; Moreira, M. E.; Rodrigues Baião, A. E.; Nassar de Carvalho, P. R.; Janzen, C.; Valderramos, S. G.; Cherry, J. D.; Bispo de Filippis, A. M.; Nielsen-Saines, K., Zika Virus Infection in Pregnant Women in Rio De Janeiro — Preliminary Report. N. Engl. J. Med., in press doi:10.1056/NEJMoa1602412. 14. Ventura, C. V.; Maia, M.; Bravo-Filho, V.; Gois, A. L.; Belfort, R., Jr., Zika Virus in Brazil and Macular Atrophy in a Child with Microcephaly. Lancet 2016, 387, 228-228. 15. WHO, Zika Virus Microcephaly and Guillain-Barré Syndrome Situation Report 24 March 2016. 2016, Available at http://www.who.int/emergencies/zika-virus/situation-report/24-march-2016/en/ (accessed: 25 Mar. 2016).

101

16. WHO, Dengue: Guidelines for Diagnosis, Treatment, Prevention and Control. New Edition. Geneva: World Health Organization 2009, (WHO/HTM/NTD/DEN/2009.1). 17. Bhatt, S.; Gething, P. W.; Brady, O. J.; Messina, J. P.; Farlow, A. W.; Moyes, C. L.; Drake, J. M.; Brownstein, J. S.; Hoen, A. G.; Sankoh, O.; Myers, M. F.; George, D. B.; Jaenisch, T.; Wint, G. R. W.; Simmons, C. P.; Scott, T. W.; Farrar, J. J.; Hay, S. I., The Global Distribution and Burden of Dengue. Nature 2013, 496, 504-507. 18. Bhatt, S.; Weiss, D. J.; Cameron, E.; Bisanzio, D.; Mappin, B.; Dalrymple, U.; Battle, K. E.; Moyes, C. L.; Henry, A.; Eckhoff, P. A.; Wenger, E. A.; Briet, O.; Penny, M. A.; Smith, T. A.; Bennett, A.; Yukich, J.; Eisele, T. P.; Griffin, J. T.; Fergus, C. A.; Lynch, M.; Lindgren, F.; Cohen, J. M.; Murray, C. L. J.; Smith, D. L.; Hay, S. I.; Cibulskis, R. E.; Gething, P. W., The Effect of Malaria Control on Plasmodium Falciparum in Africa between 2000 and 2015. Nature 2015, 526, 207-211. 19. WHO, Global Insecticide Use for Vector-Borne Disease Control. A 10-Year Assessment (2000-2009) 5th Edition Geneva: World Health Organization 2011, (WHO/HTM/NTD/VEM/WHOPES/2011.6). 20. WHO, Larval Source Management: A Supplementary Measure for Malaria Vector Control: An Operational Manual. Geneva: World Health Organization 2013. 21. Hemingway, J.; Hawkes, N. J.; McCarroll, L.; Ranson, H., The Molecular Basis of Insecticide Resistance in Mosquitoes. Insect Biochem. Mol. Biol. 2004, 34, 653-665. 22. Rivero, A.; Vezilier, J.; Weill, M.; Read, A. F.; Gandon, S., Insecticide Control of Vector-Borne Diseases: When Is Insecticide Resistance a Problem? PLoS Pathog. 2010, 6. 23. Ranson, H.; N'Guessan, R.; Lines, J.; Moiroux, N.; Nkuni, Z.; Corbel, V., Pyrethroid Resistance in African Anopheline Mosquitoes: What Are the Implications for Malaria Control? Trends Parasitol. 2011, 27, 91-98. 24. WHO, Global Plan for Insecticide Resistance Management in Malaria Vectors. Geneva: World Health Organization 2012, (WHO/HTM/GMP/2012.5 Available at www.who.int/malaria/publications/atoz/gpirm/en/ (accessed: 28 Feb. 2016)). 25. Ffrench-Constant, R. H., The Molecular Genetics of Insecticide Resistance. Genetics 2013, 194, 807-815. 26. Massoulie, J.; Pezzementi, L.; Bon, S.; Krejci, E.; Vallette, F. M., Molecular and Cellular Biology of Cholinesterases. Prog. Neurobiol. 1993, 41, 31-91. 27. Toutant, J. P., Insect Acetylcholinesterase - Catalytic Properties, Tissue Distribution and Molecular-Forms. Prog. Neurobiol. 1989, 32, 423-446. 28. Weaver, S. C.; Barrett, A. D. T., Transmission Cycles, Host Range, Evolution and Emergence of Arboviral Disease. Nat. Rev. Microbiol. 2004, 2, 789-801. 29. Benedict, M. Q.; Levine, R. S.; Hawley, W. A.; Lounibos, L. P., Spread of the Tiger: Global Risk of Invasion by the Mosquito Aedes Albopictus. Vector Borne Zoonotic Dis. 2007, 7, 76-85.

102

30. Stoddard, S. T.; Morrison, A. C.; Vazquez-Prokopec, G. M.; Soldan, V. P.; Kochel, T. J.; Kitron, U.; Elder, J. P.; Scott, T. W., The Role of Human Movement in the Transmission of Vector-Borne Pathogens. PLoS Neglected Trop. Dis. 2009, 3. 31. Fischer, D.; Thomas, S. M.; Suk, J. E.; Sudre, B.; Hess, A.; Tjaden, N. B.; Beierkuhnlein, C.; Semenza, J. C., Climate Change Effects on Chikungunya Transmission in Europe: Geospatial Analysis of Vector's Climatic Suitability and Virus' Temperature Requirements. Int. J. Health Geogr. 2013, 12. 32. Minakawa, N.; Sonye, G.; Mogi, M.; Githeko, A.; Yan, G. Y., The Effects of Climatic Factors on the Distribution and Abundance of Malaria Vectors in Kenya. J. Med. Entomol. 2002, 39, 833-841. 33. Weaver, S. C.; Reisen, W. K., Present and Future Arboviral Threats. Antiviral Res. 2010, 85, 328-345. 34. Achee, N. L.; Gould, F.; Perkins, T. A.; Reiner, R. C., Jr.; Morrison, A. C.; Ritchie, S. A.; Gubler, D. J.; Teyssou, R.; Scott, T. W., A Critical Assessment of Vector Control for Dengue Prevention. PLoS Neglected Trop. Dis. 2015, 9. 35. Valid Species List. Mosquito Taxonomic Inventory. http://mosquito-taxonomic-inventory.info/ (accessed: 18 Feb. 2016). 36. Borkent, A.; Grimaldi, D. A., The Earliest Fossil Mosquito (Diptera : Culicidae), in Mid-Cretaceous Burmese Amber. Ann. Entomol. Soc. Am. 2004, 97, 882-888. 37. Poinar, G. O.; Zavortink, T. J.; Pike, T.; Johnston, P. A., Paleoculicis Minutus (Diptera: Culicidae) N. Gen., N. Sp., from Cretaceous Canadian Amber, with a Summary of Described Fossil Mosquitoes Acta Geol. Hisp. 2000, 35, 119-128. 38. Greenwalt, D. E.; Goreva, Y. S.; Siljestroem, S. M.; Rose, T.; Harbach, R. E., Hemoglobin-Derived Porphyrins Preserved in a Middle Eocene Blood-Engorged Mosquito. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 18496-18500. 39. Poinar, G., Plasmodium Dominicana N. Sp (Plasmodiidae : Haemospororida) from Tertiary Dominican Amber. Syst. Parasitol. 2005, 61, 47-52. 40. WHO, WHO Fact Sheet on Malaria. http://www.who.int/mediacentre/factsheets/fs094/en/ (accessed: 4 Apr. 2016). 41. WHO, WHO Fact Sheet on Dengue and Severe Dengue. http://www.who.int/mediacentre/factsheets/fs117/en/ (accessed: 4 Apr. 2016). 42. WHO, WHO Fact Sheet on Yellow Fever. http://www.who.int/mediacentre/factsheets/fs100/en/ (accessed: 29 Feb. 2016). 43. WHO, WHO Fact Sheet on Lymphatic Filariasis. http://www.who.int/mediacentre/factsheets/fs102/en/ (accessed: 4 Apr. 2016). 44. WHO, WHO Fact Sheet on Japanese Encephalitis. http://www.who.int/mediacentre/factsheets/fs386/en/ (accessed: 4 Apr. 2016).

103

45. Harrington, L. C.; Edman, J. D.; Scott, T. W., Why Do Female Aedes Aegypti (Diptera : Culicidae) Feed Preferentially and Frequently on Human Blood? J. Med. Entomol. 2001, 38, 411-422. 46. WHO, Global Strategy for Dengue Prevention and Control, 2012-2020. Geneva: World Health Organization 2012, (WHO/HTM/NTD/VEM/2012.5). 47. Diatta, M.; Spiegel, A.; Lochouarn, L.; Fontenille, D., Similar Feeding Preferences of Anopheles Gambiae and a-Arabiensis in Senegal. Trans. R. Soc. Trop. Med. Hyg. 1998, 92, 270-272. 48. Duchemin, J. B.; Tsy, J.; Rabarison, P.; Roux, J.; Coluzzi, M.; Costantini, C., Zoophily of Anopheles Arabiensis and An. Gambiae in Madagascar Demonstrated by Odour-Baited Entry Traps. Med. Vet. Entomol. 2001, 15, 50-57. 49. Bogh, C.; Clarke, S. E.; Pinder, M.; Sanyang, F.; Lindsay, S. W., Effect of Passive Zooprophylaxis on Malaria Transmission in the Gambia. J. Med. Entomol. 2001, 38, 822-828. 50. Sinka, M. E.; Bangs, M. J.; Manguin, S.; Coetzee, M.; Mbogo, C. M.; Hemingway, J.; Patil, A. P.; Temperley, W. H.; Gething, P. W.; Kabaria, C. W.; Okara, R. M.; Van Boeckel, T.; Godfray, H. C. J.; Harbach, R. E.; Hay, S. I., The Dominant Anopheles Vectors of Human Malaria in Africa, Europe and the Middle East: Occurrence Data, Distribution Maps and Bionomic Precis. Parasites Vectors 2010, 3. 51. CDC, Factsheet: Dengue and the Aedes Aegypti Mosquito. Available at http://www.cdc.gov/dengue/entomologyecology/index.html (accessed: 16 Mar. 2016). 52. CDC, Mosquito Life Cycle : Aedes Aegypti. Available at http://stacks.cdc.gov/view/cdc/24391 (accessed: 16 Mar. 2016). 53. Anthony, M. P.; Burrows, J. N.; Duparc, S.; Jmoehrle, J.; Wells, T. N. C., The Global Pipeline of New Medicines for the Control and Elimination of Malaria. Malar. J. 2012, 11. 54. WHO, Guidelines for the Treatment of Malaria - 3rd Edition. Geneva: World Health Organization 2015. 55. Schwartz, L.; Brown, G. V.; Genton, B.; Moorthy, V. S., A Review of Malaria Vaccine Clinical Projects Based on the Who Rainbow Table. Malar. J. 2012, 11. 56. Riley, E. M.; Stewart, V. A., Immune Mechanisms in Malaria: New Insights in Vaccine Development. Nat. Med. 2013, 19, 168-178. 57. Tinto, H.; D'Alessandro, U.; Sorgho, H.; Valea, I.; Tahita, M. C.; Kabore, W.; Kiemde, F.; Lompo, P.; Ouedraogo, S.; Derra, K. et al., Efficacy and Safety of RTS,S/As01 Malaria Vaccine with or without a Booster Dose in Infants and Children in Africa: Final Results of a Phase 3, Individually Randomised, Controlled Trial. Lancet 2015, 386, 31-45. 58. WHO, Malaria Vaccine: WHO Position Paper – January 2016. Weekly Epidemiological Record 2016, 91, 33-52. 59. Sinka, M. E.; Rubio-Palis, Y.; Manguin, S.; Patil, A. P.; Temperley, W. H.; Gething, P. W.; Van Boeckel, T.; Kabaria, C. W.; Harbach, R. E.; Hay, S. I., The Dominant Anopheles Vectors of Human Malaria in the Americas: Occurrence Data, Distribution Maps and Bionomic Precis. Parasites Vectors 2010, 3.

104

60. Sinka, M. E.; Bangs, M. J.; Manguin, S.; Chareonviriyaphap, T.; Patil, A. P.; Temperley, W. H.; Gething, P. W.; Elyazar, I. R. F.; Kabaria, C. W.; Harbach, R. E.; Hay, S. I., The Dominant Anopheles Vectors of Human Malaria in the Asia-Pacific Region: Occurrence Data, Distribution Maps and Bionomic Precis. Parasites Vectors 2011, 4. 61. Cox-Singh, J.; Singh, B., Knowlesi Malaria: Newly Emergent and of Public Health Importance? Trends Parasitol. 2008, 24, 406-410. 62. Josling, G. A.; Llinas, M., Sexual Development in Plasmodium Parasites: Knowing When It's Time to Commit. Nat. Rev. Microbiol. 2015, 13, 573-587. 63. Taylor, M. J.; Hoerauf, A.; Bockarie, M., Lymphatic Filariasis and Onchocerciasis. Lancet 2010, 376, 1175-1185. 64. Gubler, D. J., Human Arbovirus Infections Worldwide. West Nile Virus: Detection, Surveillance, and Control 2001, 951, 13-24. 65. Centers for Disease Control and Prevention, Arbovirus Catalog. Available at https://wwwn.cdc.gov/arbocat/ (accessed: 15 Feb. 2016). 66. Lambrechts, L.; Scott, T. W.; Gubler, D. J., Consequences of the Expanding Global Distribution of Aedes Albopictus for Dengue Virus Transmission. PLoS Neglected Trop. Dis. 2010, 4. 67. Taylor, M. J., Viruses and Man: A History of Interactions. Springer International Publishing Cham, 2014. 68. Staples, J. E.; Monath, T. P., Yellow Fever: 100 Years of Discovery. JAMA, J. Am. Med. Assoc. 2008, 300, 960-962. 69. Jentes, E. S.; Poumerol, G.; Gershman, M. D.; Hill, D. R.; Lemarchand, J.; Lewis, R. F.; Staples, J. E.; Tomori, O.; Wilder-Smith, A.; Monath, T. P.; Informal, W. H. O. W. G. G., The Revised Global Yellow Fever Risk Map and Recommendations for Vaccination, 2010: Consensus of the Informal WHO Working Group on Geographic Risk for Yellow Fever. Lancet Infect. Dis. 2011, 11, 622-632. 70. Almeida, A. P. G.; Goncalves, Y. M.; Novo, M. T.; Sousa, C. A.; Melim, M.; Gracio, A. J. S., Vector Monitoring of Aedes Aegypti in the Autonomous Region of Madeira, Portugal. Eurosurveillance 2007, 12, E071115.6-E071115.6. 71. Medlock, J. M.; Hansford, K. M.; Schaffner, F.; Versteirt, V.; Hendrickx, G.; Zeller, H.; Van Bortel, W., A Review of the Invasive Mosquitoes in Europe: Ecology, Public Health Risks, and Control Options. Vector Borne Zoonotic Dis. 2012, 12, 435-447. 72. Schaffner, F.; Medlock, J. M.; Van Bortel, W., Public Health Significance of Invasive Mosquitoes in Europe. Clin. Microbiol. Infect. 2013, 19, 685-692. 73. Sousa, C. A.; Clairouin, M.; Seixas, G.; Viveiros, B.; Novo, M. T.; Silva, A. C.; Escoval, M. T.; Economopoulou, A., Ongoing Outbreak of Dengue Type 1 in the Autonomous Region of Madeira, Portugal: Preliminary Report. Eurosurveillance 2012, 17, 15-18. 74. Simmons, C. P.; Farrar, J. J.; Nguyen Van Vinh, C.; Wills, B., Current Concepts Dengue. N. Engl. J. Med. 2012, 366, 1423-1432. 75. Henchal, E. A.; Putnak, J. R., The Dengue Viruses. Clin. Microbiol. Rev. 1990, 3, 376-396.

105

76. Dejnirattisai, W.; Jumnainsong, A.; Onsirisakul, N.; Fitton, P.; Vasanawathana, S.; Limpitikul, W.; Puttikhunt, C.; Edwards, C.; Duangchinda, T.; Supasa, S.; Chawansuntati, K.; Malasit, P.; Mongkolsapaya, J.; Screaton, G., Cross-Reacting Antibodies Enhance Dengue Virus Infection in Humans. Science 2010, 328, 745-748. 77. Whitehead, S. S.; Blaney, J. E.; Durbin, A. P.; Murphy, B. R., Prospects for a Dengue Virus Vaccine. Nat. Rev. Microbiol. 2007, 5, 518-528. 78. Capeding, M. R.; Ngoc Huu, T.; Hadinegoro, S. R. S.; Ismail, H. I. H. J. M.; Chotpitayasunondh, T.; Chua, M. N.; Chan Quang, L.; Rusmil, K.; Wirawan, D. N.; Nallusamy, R.; Pitisuttithum, P.; Thisyakorn, U.; Yoon, I.-K.; van der Vliet, D.; Langevin, E.; Laot, T.; Hutagalung, Y.; Frago, C.; Boaz, M.; Wartel, T. A.; Tornieporth, N. G.; Saville, M.; Bouckenooghe, A.; Grp, C. Y. D. S., Clinical Efficacy and Safety of a Novel Tetravalent Dengue Vaccine in Healthy Children in Asia: A Phase 3, Randomised, Observer-Masked, Placebo-Controlled Trial. Lancet 2014, 384, 1358-1365. 79. Gunther, V. J.; Putnak, R.; Eckels, K. H.; Mammen, M. P.; Scherer, J. M.; Lyons, A.; Sztein, M. B.; Sun, W., A Human Challenge Model for Dengue Infection Reveals a Possible Protective Role for Sustained Interferon Gamma Levels During the Acute Phase of Illness. Vaccine 2011, 29, 3895-3904. 80. Guy, B.; Briand, O.; Lang, J.; Saville, M.; Jackson, N., Development of the Sanofi Pasteur Tetravalent Dengue Vaccine: One More Step Forward. Vaccine 2015, 33, 7100-7111. 81. Poo, J.; Galan, F.; Forrat, R.; Zambrano, B.; Lang, J.; Dayan, G. H., Live-Attenuated Tetravalent Dengue Vaccine in Dengue-Naive Children, Adolescents, and Adults in Mexico City Randomized Controlled Phase 1 Trial of Safety and Immunogenicity. Pediatr. Infect. Dis. J. 2011, 30, E9-E17. 82. Sabchareon, A.; Wallace, D.; Sirivichayakul, C.; Limkittikul, K.; Chanthavanich, P.; Suvannadabba, S.; Jiwariyavej, V.; Dulyachai, W.; Pengsaa, K.; Wartel, T. A.; Moureau, A.; Saville, M.; Bouckenooghe, A.; Viviani, S.; Tornieporth, N. G.; Lang, J., Protective Efficacy of the Recombinant, Live-Attenuated, Cyd Tetravalent Dengue Vaccine in Thai Schoolchildren: A Randomised, Controlled Phase 2b Trial. Lancet 2012, 380, 1559-1567. 83. Sun, W.; Eckels, K. H.; Putnak, J. R.; Lyons, A. G.; Thomas, S. J.; Vaughn, D. W.; Gibbons, R. V.; Fernandez, S.; Gunther, V. J.; Mammen, M. P., Jr.; Statler, J. D.; Innis, B. L., Experimental Dengue Virus Challenge of Human Subjects Previously Vaccinated with Live Attenuated Tetravalent Dengue Vaccines. J. Infect. Dis. 2013, 207, 700-708. 84. Wilder-Smith, A.; Gubler, D. J., Dengue Vaccines at a Crossroad. Science 2015, 350, 625-626. 85. Press Release: World’s First Public Dengue Immunization Program Starts in the Philippines. Avalaibe at http://www.dengue.info/#overlay=content/world%25E2%2580%2599s-first-public-dengue-immunization-program-starts-philippines-0 (accessed: 15 Apr. 2016). 86. Pialoux, G.; Gauzere, B.-A.; Jaureguiberry, S.; Strobel, M., Chikungunya, an Epidemic Arbovirosis. Lancet Infect. Dis. 2007, 7, 319-327.

106

87. Robinson, M. C., An Epidemic of Virus Disease in Southern Province, Tanganyika Territory, in 1952-53. I. Clinical Features. Trans. R. Soc. Trop. Med. Hyg. 1955, 49, 28-32. 88. Lumsden, W. H., An Epidemic of Virus Disease in Southern Province, Tanganyika Territory, in 1952-53. Ii. General Description and Epidemiology. Trans. R. Soc. Trop. Med. Hyg. 1955, 49, 33-57. 89. Powers, A. M.; Logue, C. H., Changing Patterns of Chikungunya Virus: Re-Emergence of a Zoonotic Arbovirus. J. Gen. Virol. 2007, 88, 2363-2377. 90. de Lamballerie, X.; Leroy, E.; Charrel, R. N.; Ttsetsarkin, K.; Higgs, S.; Gould, E. A., Chikungunya Virus Adapts to Tiger Mosquito Via Evolutionary Convergence: A Sign of Things to Come? Virol. J. 2008, 5. 91. Rezza, G.; Nicoletti, L.; Angelini, R.; Romi, R.; Finarelli, A. C.; Panning, M.; Cordioli, P.; Fortuna, C.; Boros, S.; Magurano, F.; Silvi, G.; Angelini, P.; Dottori, M.; Ciufolini, M. G.; Majori, G. C.; Cassone, A.; Grp, C. S., Infection with Chikungunya Virus in Italy: An Outbreak in a Temperate Region. Lancet 2007, 370, 1840-1846. 92. Cassadou, S.; Boucau, S.; Petit-Sinturel, M.; Huc, P.; Leparc-Goffart, I.; Ledrans, M., Emergence of Chikungunya Fever on the French Side of Saint Martin Island, October to December 2013. Eurosurveillance 2014, 19, 6-9. 93. Besnard, M.; Lastere, S.; Teissier, A.; Cao-Lormeau, V. M.; Musso, D., Evidence of Perinatal Transmission of Zika Virus, French Polynesia, December 2013 and February 2014. Eurosurveillance 2014, 19, 13-16. 94. Cao-Lormeau, V.-M.; Roche, C.; Teissier, A.; Robin, E.; Berry, A.-L.; Mallet, H.-P.; Sall, A. A.; Musso, D., Zika Virus, French Polynesia, South Pacific, 2013. Emerging Infect. Dis. 2014, 20, 1085-1086. 95. Oehler, E.; Watrin, L.; Larre, P.; Leparc-Goffart, I.; Lastere, S.; Valour, F.; Baudouin, L.; Mallet, H. P.; Musso, D.; Ghawche, F., Zika Virus Infection Complicated by Guillain-Barre Syndrome - Case Report, French Polynesia, December 2013. Eurosurveillance 2014, 19, 4-6. 96. Cao-Lormeau, V.-M.; Blake, A.; Mons, S.; Lastère, S.; Roche, C.; Vanhomwegen, J.; Dub, T.; Baudouin, L.; Teissier, A.; Larre, P.; Vial, A.-L.; Decam, C.; Choumet, V.; Halstead, S. K.; Willison, H. J.; Musset, L.; Manuguerra, J.-C.; Despres, P.; Fournier, E.; Mallet, H.-P.; Musso, D.; Fontanet, A.; Neil, J.; Ghawché, F., Guillain-Barré Syndrome Outbreak Associated with Zika Virus Infection in French Polynesia: A Case-Control Study. Lancet 2016, 387, 1531-1539. 97. Duffy, M. R.; Chen, T.-H.; Hancock, W. T.; Powers, A. M.; Kool, J. L.; Lanciotti, R. S.; Pretrick, M.; Marfel, M.; Holzbauer, S.; Dubray, C.; Guillaumot, L.; Griggs, A.; Bel, M.; Lambert, A. J.; Laven, J.; Kosoy, O.; Panella, A.; Biggerstaff, B. J.; Fischer, M.; Hayes, E. B., Zika Virus Outbreak on Yap Island, Federated States of Micronesia. N. Engl. J. Med. 2009, 360, 2536-2543. 98. Dick, G. W. A.; Kitchen, S. F.; Haddow, A. J., Zika Virus .1. Isolations and Serological Specificity. Trans. R. Soc. Trop. Med. Hyg. 1952, 46, 509-520. 99. Haddow, A. J.; Williams, M. C.; Woodall, J. P.; Simpson, D. I. H.; Goma, L. K. H., 12 Isolations of Zika Virus from Aedes ( Stegomyia )

107

Africanus ( Theobald ) Taken in and above a Uganda Forest. Bull. W. H. O. 1964, 31, 57-&. 100. Marchett, N. J.; Garcia, R.; Rudnick, A., Isolation of Zika Virus from Aedes Aegypti Mosquitoes in Malaysia. Am. J. Trop. Med. Hyg. 1969, 18, 411-&. 101. WHO, WHO Fact Sheet on Zika Virus. Available at http://www.who.int/mediacentre/factsheets/zika/en/ (accessed: 21 Mar. 2016). 102. Ayres, C. F. J., Identification of Zika Virus Vectors and Implications for Control. Lancet Infect. Dis. 2016, 16, 278-279. 103. WHO, WHO Recommended Insecticides for Space Spraying against Mosquitoes. Available at http://www.who.int/whopes/en/ (accessed: 16 Feb. 2016). 104. WHO, WHO Recommended Insecticides for Indoor Residual Spraying against Malaria Vectors. Available at http://www.who.int/whopes/en/ (accessed: 16 Feb. 2016). 105. WHO, WHO Recommended Insecticide Products for Treatment of Mosquito Nets for Malaria Vector Control. Available at http://www.who.int/whopes/en/ (accessed: 16 Feb. 2016) 106. WHO, WHOPES-Recommended Compounds and Formulations for Control of Mosquito Larvae. Available at http://www.who.int/whopes/en/ (accessed: 16 Feb. 2016). 107. Insecticide Resistance Action Committee (IRAC), IRAC Mode of Action Classification Scheme Version 8.0. 2015. 108. Insecticide Resistance Action Committee (IRAC), Prevention and Management of Insecticide Resistance in Vectors of Public Health Importance. 2nd Edition. 2011. 109. Casida, J. E.; Quistad, G. B., Golden Age of Insecticide Research: Past, Present, or Future? Annu. Rev. Entomol. 1998, 43, 1-16. 110. Najera, J. A., Malaria Control : Achievements, Problems and Strategies. Geneva: World Health Organization 1999, (WHO/MAL/99.1087 Available at http://www.who.int/iris/handle/10665/66640 (accessed: 18 Feb. 2016)). 111. Nayar, J. K.; Ali, A., A Review of Monomolecular Surface Films as Larvicides and Pupicides of Mosquitoes. J. Vector Ecol. 2003, 28, 190-199. 112. Mabaso, M. L. H.; Sharp, B.; Lengeler, C., Historical Review of Malarial Control in Southern African with Emphasis on the Use of Indoor Residual House-Spraying. Trop. Med. Int. Health 2004, 9, 846-856. 113. Najera, J. A.; Gonzalez-Silva, M.; Alonso, P. L., Some Lessons for the Future from the Global Malaria Eradication Programme (1955-1969). PLoS Med. 2011, 8. 114. Reiter, P.; Nathan, M., Guidelines for Assessing the Efficacy of Insecticidal Space Sprays for Control of the Dengue Vector Aedes Aegypti. Geneva: World Health Organization 2001, (WHO/CDS/CPE/PVC/2001.1 Available at http://www.who.int/iris/handle/10665/67047 (accessed: 21 Feb. 2016)). 115. Esu, E.; Lenhart, A.; Smith, L.; Horstick, O., Effectiveness of Peridomestic Space Spraying with Insecticide on Dengue Transmission; Systematic Review. Trop. Med. Int. Health 2010, 15, 619-631.

108

116. Snow, R. W.; Lindsay, S. W.; Hayes, R. J.; Greenwood, B. M., Permethrin-Treated Bed Nets (Mosquito Nets) Prevent Malaria in Gambian Children. Trans. R. Soc. Trop. Med. Hyg. 1988, 82, 838-842. 117. Willey, B. A.; Paintain, L. S.; Mangham, L.; Car, J.; Schellenberg, J. A., Strategies for Delivering Insecticide-Treated Nets at Scale for Malaria Control: A Systematic Review. Bull. W. H. O. 2012, 90, 672-684. 118. Singh, M.; Brown, G.; Rogerson, S. J., Ownership and Use of Insecticide-Treated Nets During Pregnancy in Sub-Saharan Africa: A Review. Malar. J. 2013, 12. 119. Obembe, A.; Anyaele, O. O.; Oduola, A. O., Lessons from the Implementation of Llin Distribution Campaign in Ilorin Kwara State, Nigeria. BMC Public Health 2014, 14. 120. Strachan, C. E.; Nuwa, A.; Muhangi, D.; Okui, A. P.; Helinski, M. E. H.; Tibenderana, J. K., What Drives the Consistent Use of Long-Lasting Insecticidal Nets over Time? A Multi-Method Qualitative Study in Mid-Western Uganda. Malar. J. 2016, 15. 121. Hemingway, J., The Role of Vector Control in Stopping the Transmission of Malaria: Threats and Opportunities. Philos. Trans. R. Soc., B 2014, 369. 122. Russell, T. L.; Govella, N. J.; Azizi, S.; Drakeley, C. J.; Kachur, S. P.; Killeen, G. F., Increased Proportions of Outdoor Feeding among Residual Malaria Vector Populations Following Increased Use of Insecticide-Treated Nets in Rural Tanzania. Malar. J. 2011, 10. 123. Reddy, M. R.; Overgaard, H. J.; Abaga, S.; Reddy, V. P.; Caccone, A.; Kiszewski, A. E.; Slotman, M. A., Outdoor Host Seeking Behaviour of Anopheles Gambiae Mosquitoes Following Initiation of Malaria Vector Control on Bioko Island, Equatorial Guinea. Malar. J. 2011, 10. 124. Moiroux, N.; Gomez, M. B.; Pennetier, C.; Elanga, E.; Djenontin, A.; Chandre, F.; Djegbe, I.; Guis, H.; Corbel, V., Changes in Anopheles Funestus Biting Behavior Following Universal Coverage of Long-Lasting Insecticidal Nets in Benin. J. Infect. Dis. 2012, 206, 1622-1629. 125. Gatton, M. L.; Chitnis, N.; Churcher, T.; Donnelly, M. J.; Ghani, A. C.; Godfray, H. C. J.; Gould, F.; Hastings, I.; Marshall, J.; Ranson, H.; Rowland, M.; Shaman, J.; Lindsay, S. W., The Importance of Mosquito Behavioural Adaptations to Malaria Control in Africa. Evolution 2013, 67, 1218-1230. 126. Wood, O. R.; Hanrahan, S.; Coetzee, M.; Koekemoer, L. L.; Brooke, B. D., Cuticle Thickening Associated with Pyrethroid Resistance in the Major Malaria Vector Anopheles Funestus. Parasites Vectors 2010, 3. 127. Fang, F.; Wang, W.; Zhang, D.; Lv, Y.; Zhou, D.; Ma, L.; Shen, B.; Sun, Y.; Zhu, C., The Cuticle Proteins: A Putative Role for Deltamethrin Resistance in Culex Pipiens Pallens. Parasitol. Res. 2015, 114, 4421-4429. 128. Vannini, L.; Reed, T. W.; Willis, J. H., Temporal and Spatial Expression of Cuticular Proteins of Anopheles Gambiae Implicated in Insecticide Resistance or Differentiation of M/S Incipient Species. Parasites Vectors 2014, 7. 129. Martinez-Torres, D.; Chandre, F.; Williamson, M. S.; Darriet, F.; Berge, J. B.; Devonshire, A. L.; Guillet, P.; Pasteur, N.; Pauron, D., Molecular Characterization of Pyrethroid Knockdown Resistance (Kdr) in the Major Malaria Vector Anopheles Gambiae S.S. Insect Mol. Biol. 1998, 7, 179-184.

109

130. Martinez-Torres, D.; Chevillon, C.; Brun-Barale, A.; Berge, J. B.; Pasteur, N.; Pauron, D., Voltage-Dependent Na+ Channels in Pyrethroid-Resistant Culex Pipiens L Mosquitoes. Pestic. Sci. 1999, 55, 1012-1020. 131. Chang, C.; Shen, W.-K.; Wang, T.-T.; Lin, Y.-H.; Hsu, E.-L.; Dai, S.-M., A Novel Amino Acid Substitution in a Voltage-Gated Sodium Channel Is Associated with Knockdown Resistance to Permethrin in Aedes Aegypti. Insect Biochem. Mol. Biol. 2009, 39, 272-278. 132. Singh, O. P.; Dykes, C. L.; Das, M. K.; Pradhan, S.; Bhatt, R. M.; Agrawal, O. P.; Adak, T., Presence of Two Alternative Kdr-Like Mutations, L1014f and L1014s, and a Novel Mutation, V1010l, in the Voltage Gated Na+ Channel of Anopheles Culicifacies from Orissa, India. Malar. J. 2010, 9. 133. Srisawat, R.; Komalamisra, N.; Eshita, Y.; Zheng, M.; Ono, K.; Itoh, T. Q.; Matsumoto, A.; Petmitr, S.; Rongsriyam, Y., Point Mutations in Domain Ii of the Voltage-Gated Sodium Channel Gene in Deltamethrin-Resistant Aedes Aegypti (Diptera: Culicidae). Appl. Entomol. Zool. 2010, 45, 275-282. 134. Jones, C. M.; Liyanapathirana, M.; Agossa, F. R.; Weetman, D.; Ranson, H.; Donnelly, M. J.; Wilding, C. S., Footprints of Positive Selection Associated with a Mutation (N1575y) in the Voltage-Gated Sodium Channel of Anopheles Gambiae. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 6614-6619. 135. Stenhouse, S. A.; Plernsub, S.; Yanola, J.; Lumjuan, N.; Dantrakool, A.; Choochote, W.; Somboon, P., Detection of the V1016g Mutation in the Voltage-Gated Sodium Channel Gene of Aedes Aegypti (Diptera: Culicidae) by Allele-Specific Pcr Assay, and Its Distribution and Effect on Deltamethrin Resistance in Thailand. Parasites Vectors 2013, 6. 136. Weill, M.; Lutfalla, G.; Mogensen, K.; Chandre, F.; Berthomieu, A.; Berticat, C.; Pasteur, N.; Philips, A.; Fort, P.; Raymond, M., Insecticide Resistance in Mosquito Vectors. Nature 2003, 423, 136-137. 137. Weill, M.; Malcolm, C.; Chandre, F.; Mogensen, K.; Berthomieu, A.; Marquine, M.; Raymond, M., The Unique Mutation in Ace-1 Giving High Insecticide Resistance Is Easily Detectable in Mosquito Vectors. Insect Mol. Biol. 2004, 13, 1-7. 138. Weill, M.; Berthomieu, A.; Berticat, C.; Lutfalla, G.; Negre, V.; Pasteur, N.; Philips, A.; Leonetti, J. P.; Fort, P.; Raymond, M., Insecticide Resistance: A Silent Base Prediction. Curr. Biol. 2004, 14, R552-R553. 139. Nabeshima, T.; Mori, A.; Kozaki, T.; Iwata, Y.; Hidoh, O.; Harada, S.; Kasai, S.; Severson, D. W.; Kono, Y.; Tomita, T., An Amino Acid Substitution Attributable to Insecticide-Insensitivity of Acetylcholinesterase in a Japanese Encephalitis Vector Mosquito, Culex Tritaeniorhynchus. Biochem. Biophys. Res. Commun. 2004, 313, 794-801. 140. Alout, H.; Berthomieu, A.; Hadjivassilis, A.; Weill, M., A New Amino-Acid Substitution in Acetylcholinesterase 1 Confers Insecticide Resistance to Culex Pipiens Mosquitoes from Cyprus. Insect Biochem. Mol. Biol. 2007, 37, 41-47. 141. Feyereisen, R., Insect P450 Enzymes. Annu. Rev. Entomol. 1999, 44, 507-533. 142. Djouaka, R. F.; Bakare, A. A.; Coulibaly, O. N.; Akogbeto, M. C.; Ranson, H.; Hemingway, J.; Strode, C., Expression of the Cytochrome P450s, Cyp6p3 and Cyp6m2 Are Significantly Elevated in Multiple

110

Pyrethroid Resistant Populations of Anopheles Gambiae S.S. From Southern Benin and Nigeria. BMC Genomics 2008, 9. 143. Müller, P.; Warr, E.; Stevenson, B. J.; Pignatelli, P. M.; Morgan, J. C.; Steven, A.; Yawson, A. E.; Mitchell, S. N.; Ranson, H.; Hemingway, J.; Paine, M. J. I.; Donnelly, M. J., Field-Caught Permethrin-Resistant Anopheles Gambiae Overexpress Cyp6p3, a P450 That Metabolises Pyrethroids. PLoS Genet. 2008, 4. 144. Mitchell, S. N.; Stevenson, B. J.; Mueller, P.; Wilding, C. S.; Egyir-Yawson, A.; Field, S. G.; Hemingway, J.; Paine, M. J. I.; Ranson, H.; Donnelly, M. J., Identification and Validation of a Gene Causing Cross-Resistance between Insecticide Classes in Anopheles Gambiae from Ghana. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 6147-6152. 145. Stevenson, B. J.; Bibby, J.; Pignatelli, P.; Muangnoicharoen, S.; O'Neill, P. M.; Lian, L.-Y.; Mueller, P.; Nikou, D.; Steven, A.; Hemingway, J.; Sutcliffe, M. J.; Paine, M. J. I., Cytochrome P450 6m2 from the Malaria Vector Anopheles Gambiae Metabolizes Pyrethroids: Sequential Metabolism of Deltamethrin Revealed. Insect Biochem. Mol. Biol. 2011, 41, 492-502. 146. Marcombe, S.; Poupardin, R.; Darriet, F.; Reynaud, S.; Bonnet, J.; Strode, C.; Brengues, C.; Yebakima, A.; Ranson, H.; Corbel, V.; David, J.-P., Exploring the Molecular Basis of Insecticide Resistance in the Dengue Vector Aedes Aegypti: A Case Study in Martinique Island (French West Indies). BMC Genomics 2009, 10. 147. Marcombe, S.; Mathieu, R. B.; Pocquet, N.; Riaz, M.-A.; Poupardin, R.; Selior, S.; Darriet, F.; Reynaud, S.; Yebakima, A.; Corbel, V.; David, J.-P.; Chandre, F., Insecticide Resistance in the Dengue Vector Aedes Aegypti from Martinique: Distribution, Mechanisms and Relations with Environmental Factors. PLoS One 2012, 7. 148. David, J.-P.; Ismail, H. M.; Chandor-Proust, A.; Paine, M. J. I., Role of Cytochrome P450s in Insecticide Resistance: Impact on the Control of Mosquito-Borne Diseases and Use of Insecticides on Earth. Philos. Trans. R. Soc., B 2013, 368. 149. Poupardin, R.; Srisukontarat, W.; Yunta, C.; Ranson, H., Identification of Carboxylesterase Genes Implicated in Temephos Resistance in the Dengue Vector Aedes Aegypti. PLoS Neglected Trop. Dis. 2014, 8. 150. Ketterman, A. J.; Jayawardena, K. G. I.; Hemingway, J., Purification and Characterization of a Carboxylesterase Involved in Insecticide Resistance from the Mosquito Culex-Quinquefasciatus. Biochem. J. 1992, 287, 355-360. 151. Karunaratne, S.; Jayawardena, K. G. I.; Hemingway, J.; Ketterman, A. J., Characterization of a B-Type Esterase Involved in Insecticide Resistance from the Mosquito Culex-Quinquefasciatus. Biochem. J. 1993, 294, 575-579. 152. Enayati, A. A.; Ranson, H.; Hemingway, J., Insect Glutathione Transferases and Insecticide Resistance. Insect Mol. Biol. 2005, 14, 3-8. 153. Lumjuan, N.; Stevenson, B. J.; Prapanthadara, L.-a.; Somboon, P.; Brophy, P. M.; Loftus, B. J.; Severson, D. W.; Ranson, H., The Aedes Aegypti Glutathione Transferase Family. Insect Biochem. Mol. Biol. 2007, 37, 1026-1035. 154. Ranson, H.; Rossiter, L.; Ortelli, F.; Jensen, B.; Wang, X. L.; Roth, C. W.; Collins, F. H.; Hemingway, J., Identification of a Novel Class of Insect

111

Glutathione S-Transferases Involved in Resistance to DDT in the Malaria Vector Anopheles Gambiae. Biochem. J. 2001, 359, 295-304. 155. Lumjuan, N.; McCarroll, L.; Prapanthadara, L. A.; Hemingway, J.; Ranson, H., Elevated Activity of an Epsilon Class Glutathione Transferase Confers DDT Resistance in the Dengue Vector, Aedes Aegypti. Insect Biochem. Mol. Biol. 2005, 35, 861-871. 156. Lumjuan, N.; Rajatileka, S.; Changsom, D.; Wicheer, J.; Leelapat, P.; Prapanthadara, L.-a.; Somboon, P.; Lycett, G.; Ranson, H., The Role of the Aedes Aegypti Epsilon Glutathione Transferases in Conferring Resistance to DDT and Pyrethroid Insecticides. Insect Biochem. Mol. Biol. 2011, 41, 203-209. 157. Riveron, J. M.; Yunta, C.; Ibrahim, S. S.; Djouaka, R.; Irving, H.; Menze, B. D.; Ismail, H. M.; Hemingway, J.; Ranson, H.; Albert, A.; Wondji, C. S., A Single Mutation in the Gste2 Gene Allows Tracking of Metabolically Based Insecticide Resistance in a Major Malaria Vector. Genome Biol. 2014, 15. 158. Casida, J. E., Mode of Action of Carbamates. Annu. Rev. Entomol. 1963, 8, 39-58. 159. Casida, J. E., Pest Toxicology: The Primary Mechanisms of Pesticide Action. Chem. Res. Toxicol. 2009, 22, 609-619. 160. Wilson, I. B.; Harrison, M. A., Turnover Number of Acetylcholinesterase. J. Biol. Chem. 1961, 236, 2292-&. 161. Quinn, D. M., Acetylcholinesterase - Enzyme Structure, Reaction Dynamics, and Virtual Transition-States. Chem. Rev. 1987, 87, 955-979. 162. Sussman, J. L.; Harel, M.; Frolow, F.; Oefner, C.; Goldman, A.; Toker, L.; Silman, I., Atomic-Structure of Acetylcholinesterase from Torpedo-Californica - a Prototypic Acetylcholine-Binding Protein. Science 1991, 253, 872-879. 163. Bourne, Y.; Taylor, P.; Radic, Z.; Marchot, P., Structural Insights into Ligand Interactions at the Acetylcholinesterase Peripheral Anionic Site. EMBO J. 2003, 22, 1-12. 164. Harel, M.; Kryger, G.; Rosenberry, T. L.; Mallender, W. D.; Lewis, T.; Fletcher, R. J.; Guss, J. M.; Silman, I.; Sussman, J. L., Three-Dimensional Structures of Drosophila Melanogaster Acetylcholinesterase and of Its Complexes with Two Potent Inhibitors. Protein Sci. 2000, 9, 1063-1072. 165. Kryger, G.; Harel, M.; Giles, K.; Toker, L.; Velan, B.; Lazar, A.; Kronman, C.; Barak, D.; Ariel, N.; Shafferman, A.; Silman, I.; Sussman, J. L., Structures of Recombinant Native and E202q Mutant Human Acetylcholinesterase Complexed with the Snake-Venom Toxin Fasciculin-Ii. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2000, 56, 1385-1394. 166. Dvir, H.; Silman, I.; Harel, M.; Rosenberry, T. L.; Sussman, J. L., Acetylcholinesterase: From 3D Structure to Function. Chem.-Biol. Interact. 2010, 187, 10-22. 167. Harel, M.; Quinn, D. M.; Nair, H. K.; Silman, I.; Sussman, J. L., The X-Ray Structure of a Transition State Analog Complex Reveals the Molecular Origins of the Catalytic Power and Substrate Specificity of Acetylcholinesterase. J. Am. Chem. Soc. 1996, 118, 2340-2346. 168. Bourne, Y.; Radic, Z.; Sulzenbacher, G.; Kim, E.; Taylor, P.; Marchot, P., Substrate and Product Trafficking through the Active Center Gorge of

112

Acetylcholinesterase Analyzed by Crystallography and Equilibrium Binding. J. Biol. Chem. 2006, 281, 29256-29267. 169. Ordentlich, A.; Barak, D.; Kronman, C.; Flashner, Y.; Leitner, M.; Segall, Y.; Ariel, N.; Cohen, S.; Velan, B.; Shafferman, A., Dissection of the Human Acetylcholinesterase Active-Center Determinants of Substrate-Specificity - Identification of Residues Constituting the Anionic Site, the Hydrophobic Site, and the Acyl Pocket. J. Biol. Chem. 1993, 268, 17083-17095. 170. Houghton, P. J.; Ren, Y. H.; Howes, M. J., Acetylcholinesterase Inhibitors from Plants and Fungi. Nat. Prod. Rep. 2006, 23, 181-199. 171. Tumiattit, V.; Bolognesi, M. L.; Minarini, A.; Rosini, M.; Milelli, A.; Matera, R.; Melchiorre, C., Progress in Acetylcholinesterase Inhibitors for Alzheimer's Disease: An Update. Expert Opin. Ther. Pat. 2008, 18, 387-401. 172. Radic, Z.; Taylor, P., Interaction Kinetics of Reversible Inhibitors and Substrates with Acetylcholinesterase and Its Fasciculin 2 Complex. J. Biol. Chem. 2001, 276, 4622-4633. 173. Cheung, J.; Rudolph, M. J.; Burshteyn, F.; Cassidy, M. S.; Gary, E. N.; Love, J.; Franklin, M. C.; Height, J. J., Structures of Human Acetylcholinesterase in Complex with Pharmacologically Important Ligands. J. Med. Chem. 2012, 55, 10282-10286. 174. Greenblatt, H. M.; Kryger, G.; Lewis, T.; Silman, I.; Sussman, J. L., Structure of Acetylcholinesterase Complexed with (-)-Galanthamine at 2.3 Angstrom Resolution. FEBS Lett. 1999, 463, 321-326. 175. Carlier, P. R.; Han, Y. F.; Chow, E. S. H.; Li, C. P. L.; Wang, H.; Lieu, T. X.; Wong, H. S.; Pang, Y. P., Evaluation of Short-Tether Bis-Tha AChE Inhibitors. A Further Test of the Dual Binding Site Hypothesis. Bioorg. Med. Chem. 1999, 7, 351-357. 176. Rydberg, E. H.; Brumshtein, B.; Greenblatt, H. M.; Wong, D. M.; Shaya, D.; Williams, L. D.; Carlier, P. R.; Pang, Y.-P.; Silman, I.; Sussman, J. L., Complexes of Alkylene-Linked Tacrine Dimers with Torpedo Californica Acetylcholinesterase: Binding of Bis(5)-Tacrine Produces a Dramatic Rearrangement in the Active-Site Gorge. J. Med. Chem. 2006, 49, 5491-5500. 177. Kryger, G.; Silman, I.; Sussman, J. L., Structure of Acetylcholinesterase Complexed with E2020 (Aricept (R)): Implications for the Design of New Anti-Alzheimer Drugs. Structure with Folding Des. 1999, 7, 297-307. 178. Bourguet, D.; Raymond, M.; Fournier, D.; Malcolm, C. A.; Toutant, J. P.; Arpagaus, M., Existence of Two Acetylcholinesterases in the Mosquito Culex Pipiens (Diptera: Culicidae). J. Neurochem. 1996, 67, 2115-2123. 179. Weill, M.; Fort, P.; Berthomieu, A.; Dubois, M. P.; Pasteur, N.; Raymond, M., A Novel Acetylcholinesterase Gene in Mosquitoes Codes for the Insecticide Target and Is Non-Homologous to the Ace Gene in Drosophila. Proc. R. Soc. B 2002, 269, 2007-2016. 180. Huchard, E.; Martinez, M.; Alout, H.; Douzery, E. J. P.; Lutfalla, G.; Berthomieu, A.; Berticat, C.; Raymond, M.; Weill, M., Acetylcholinesterase Genes within the Diptera: Takeover and Loss in True Flies. Proc. R. Soc. B 2006, 273, 2595-2604.

113

181. Kim, Y. H.; Lee, S. H., Which Acetylcholinesterase Functions as the Main Catalytic Enzyme in the Class Insecta? Insect Biochem. Mol. Biol. 2013, 43, 47-53. 182. Malcolm, C. A.; Bourguet, D.; Ascolillo, A.; Rooker, S. J.; Garvey, C. F.; Hall, L. M. C.; Pasteur, N.; Raymond, M., A Sex-Linked Ace Gene, Not Linked to Insensitive Acetylcholinesterase-Mediated Insecticide Resistance in Culex Pipiens. Insect Mol. Biol. 1998, 7, 107-120. 183. Zhao, P.; Wang, Y.; Jiang, H., Biochemical Properties, Expression Profiles, and Tissue Localization of Orthologous Acetylcholinesterase-2 in the Mosquito, Anopheles Gambiae. Insect Biochem. Mol. Biol. 2013, 43, 260-271. 184. Mongin, E.; Louis, C.; Holt, R. A.; Birney, E.; Collins, F. H., The Anopheles Gambiae Genome: An Update. Trends Parasitol. 2004, 20, 49-52. 185. Mori, A.; Lobo, N. F.; deBruyn, B.; Severson, D. W., Molecular Cloning and Characterization of the Complete Acetylcholine Sterase Gene (Ace1) from the Mosquito Aedes Aegypti with Implications for Comparative Genome Analysis. Insect Biochem. Mol. Biol. 2007, 37, 667-674. 186. Anthony, N.; Rocheleau, T.; Mocelin, G.; Lee, H. J.; Ffrenchconstant, R., Cloning, Sequencing and Functional Expression of an Acetylcholinesterase Gene from the Yellow-Fever Mosquito Aedes-Aegypti. FEBS Lett. 1995, 368, 461-465. 187. Wong, D. M.; Li, J.; Chen, Q.-H.; Han, Q.; Mutunga, J. M.; Wysinski, A.; Anderson, T. D.; Ding, H.; Carpenetti, T. L.; Verma, A.; Islam, R.; Paulson, S. L.; Lam, P. C. H.; Totrov, M.; Bloomquist, J. R.; Carlier, P. R., Select Small Core Structure Carbamates Exhibit High Contact Toxicity to "Carbamate-Resistant" Strain Malaria Mosquitoes, Anopheles Gambiae (Akron). PLoS One 2012, 7. 188. Jiang, H.; Liu, S.; Zhao, P.; Pope, C., Recombinant Expression and Biochemical Characterization of the Catalytic Domain of Acetylcholinesterase-1 from the African Malaria Mosquito, Anopheles Gambiae. Insect Biochem. Mol. Biol. 2009, 39, 646-653. 189. Feyereisen, R., Insect Molecular Biology and Biochemistry. First edition ed.; Elsevier/Academic Press Inc: London, 2012; p 574. 190. Casida, J. E.; Quistad, G. B., Why Insecticides Are More Toxic to Insects Than People: The Unique Toxicology of Insects. J. Pestic. Sci. 2004, 29, 81-86. 191. Fukuto, T. R., Mechanism of Action of Organophosphorus and Carbamate Insecticides. Environ. Health Perspect. 1990, 87, 245-254. 192. Casida, J. E.; Quistad, G. B., Organophosphate Toxicology: Safety Aspects of Nonacetylcholinesterase Secondary Targets. Chem. Res. Toxicol. 2004, 17, 983-998. 193. Risher, J. F.; Mink, F. L.; Stara, J. F., The Toxicologic Effects of the Carbamate Insecticide Aldicarb in Mammals - a Review. Environ. Health Perspect. 1987, 72, 267-281. 194. Eskenazi, B.; Bradman, A.; Castorina, R., Exposures of Children to Organophosphate Pesticides and Their Potential Adverse Health Effects. Environ. Health Perspect. 1999, 107, 409-419.

114

195. Eskenazi, B.; Harley, K.; Bradman, A.; Weltzien, E.; Jewell, N. A.; Barr, D. B.; Furlong, C. E.; Holland, N. T., Association of in Utero Organophosphate Pesticide Exposure and Fetal Growth and Length of Gestation in an Agricultural Population. Environ. Health Perspect. 2004, 112, 1116-1124. 196. Eskenazi, B.; Marks, A. R.; Bradman, A.; Harley, K.; Barr, D. B.; Johnson, C.; Morga, N.; Jewell, N. A., Organophosphate Pesticide Exposure and Neurodevelopment in Young Mexican-American Children. Environ. Health Perspect. 2007, 115, 792-798. 197. Bouchard, M. F.; Chevrier, J.; Harley, K. G.; Kogut, K.; Vedar, M.; Calderon, N.; Trujillo, C.; Johnson, C.; Bradman, A.; Barr, D. B.; Eskenazi, B., Prenatal Exposure to Organophosphate Pesticides and Iq in 7-Year-Old Children. Environ. Health Perspect. 2011, 119, 1189-1195. 198. Rauh, V. A.; Perera, F. P.; Horton, M. K.; Whyatt, R. M.; Bansal, R.; Hao, X.; Liu, J.; Barr, D. B.; Slotkin, T. A.; Peterson, B. S., Brain Anomalies in Children Exposed Prenatally to a Common Organophosphate Pesticide. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 7871-7876. 199. Van der Hoek, W.; Konradsen, F.; Athukorala, K.; Wanigadewa, T., Pesticide Poisoning: A Major Health Problem in Sri Lanka. Soc. Sci. Med. 1998, 46, 495-504. 200. Eddleston, M., Patterns and Problems of Deliberate Self-Poisoning in the Developing World. QJM 2000, 93, 715-731. 201. Eddleston, M.; Phillips, M. R., Self Poisoning with Pesticides. Br. Med. J. 2004, 328, 42-44. 202. Gunnell, D.; Eddleston, M.; Phillips, M. R.; Konradsen, F., The Global Distribution of Fatal Pesticide Self-Poisoning: Systematic Review. BMC Public Health 2007, 7, 15. 203. Djogbenou, L.; Weill, M.; Hougard, J. M.; Raymond, H. M.; Akogbeto, M.; Chandre, F., Characterization of Insensitive Acetyleholinesterase (Ace-1(R)) in Anopheles Gambiae (Diptera : Culicidae): Resistance Levels and Dominance. J. Med. Entomol. 2007, 44, 805-810. 204. Djogbenou, L.; Dabire, R.; Diabate, A.; Kengne, P.; Akogbeto, M.; Hougard, J. M.; Chandre, F., Identification and Geographic Distribution of the Ace-1(R) Mutation in the Malaria Vector Anopheles Gambiae in South-Western Burkina Faso, West Africa. Am. J. Trop. Med. Hyg. 2008, 78, 298-302. 205. Dabire, K. R.; Diabate, A.; Namontougou, M.; Djogbenou, L.; Kengne, P.; Simard, F.; Bass, C.; Baldet, T., Distribution of Insensitive Acetylcholinesterase (Ace-1(R)) in Anopheles Gambiae S.L. Populations from Burkina Faso (West Africa). Trop. Med. Int. Health 2009, 14, 396-403. 206. Ahoua Alou, L. P.; Koffi, A. A.; Adja, M. A.; Tia, E.; Kouassi, P. K.; Kone, M.; Chandre, F., Distribution of Ace-1r and Resistance to Carbamates and Organophosphates in Anopheles Gambiae S.S. Populations from Cote D'ivoire. Malar. J. 2010, 9, 167-167. 207. Cui, F.; Raymond, M.; Berthomieu, A.; Alout, H.; Weill, M.; Qiao, C.-L., Recent Emergence of Insensitive Acetylcholinesterase in Chinese Populations of the Mosquito Culex Pipiens (Diptera : Culicidae). J. Med. Entomol. 2006, 43, 878-883.

115

208. Alout, H.; Djogbenou, L.; Berticat, C.; Chandre, F.; Weill, M., Comparison of Anopheles Gambiae and Culex Pipiens Acetycholinesterase 1 Biochemical Properties. Comp. Biochem. Physiol., Part B: Biochem. Mol. Biol. 2008, 150, 271-277. 209. Lenormand, T.; Raymond, M., Resistance Management: The Stable Zone Strategy. Proc. R. Soc. B 1998, 265, 1985-1990. 210. Berticat, C.; Boquien, G.; Raymond, M.; Chevillon, C., Insecticide Resistance Genes Induce a Mating Competition Cost in Culex Pipiens Mosquitoes. Genet. Res. 2002, 79, 41-47. 211. Bourguet, D.; Guillemaud, T.; Chevillon, C.; Raymond, M., Fitness Costs of Insecticide Resistance in Natural Breeding Sites of the Mosquito Culex Pipiens. Evolution 2004, 58, 128-135. 212. Djogbenou, L.; Noel, V.; Agnew, P., Costs of Insensitive Acetylcholinesterase Insecticide Resistance for the Malaria Vector Anopheles Gambiae Homozygous for the G119s Mutation. Malar. J. 2010, 9. 213. Brown, Z. S.; Dickinson, K. L.; Kramer, R. A., Insecticide Resistance and Malaria Vector Control: The Importance of Fitness Cost Mechanisms in Determining Economically Optimal Control Trajectories. J. Econ. Entomol. 2013, 106, 366-374. 214. Labbe, P.; Berthomieu, A.; Berticat, C.; Alout, H.; Raymond, M.; Lenormand, T.; Weill, M., Independent Duplications of the Acetylcholinesterase Gene Conferring Insecticide Resistance in the Mosquito Culex Pipiens. Mol. Biol. Evol. 2007, 24, 1056-1067. 215. Djogbenou, L.; Labbe, P.; Chandre, F.; Pasteur, N.; Weill, M., Ace-I Duplication in Anopheles Gambiae: A Challenge for Malaria Control. Malar. J. 2009, 8. 216. Assogba, B. S.; Djogbenou, L. S.; Milesi, P.; Berthomieu, A.; Perez, J.; Ayala, D.; Chandre, F.; Makoutode, M.; Labbe, P.; Weill, M., An Ace-1 Gene Duplication Resorbs the Fitness Cost Associated with Resistance in Anopheles Gambiae, the Main Malaria Mosquito. Sci. Rep. 2015, 5. 217. Pezzementi, L.; Rowland, M.; Wolfe, M.; Tsigelny, I., Inactivation of an Invertebrate Acetylcholinesterase by Sulfhydryl Reagents: The Roles of Two Cysteines in the Catalytic Gorge of the Enzyme. Invertebr. Neurosci. 2006, 6, 47-55. 218. Pang, Y.-P., Novel Acetylcholinesterase Target Site for Malaria Mosquito Control. PLoS One 2006, 1. 219. Wiesner, J.; Kriz, Z.; Kuca, K.; Jun, D.; Koca, J., Acetylcholinesterases - the Structural Similarities and Differences. J. Enzyme Inhib. Med. Chem. 2007, 22, 417-424. 220. Carlier, P. R.; Anderson, T. D.; Wong, D. M.; Hsu, D. C.; Hartsel, J.; Ma, M.; Wong, E. A.; Choudhury, R.; Lam, P. C. H.; Totrov, M. M.; Bloomquist, J. R., Towards a Species-Selective Acetylcholinesterase Inhibitor to Control the Mosquito Vector of Malaria, Anopheles Gambiae. Chem.-Biol. Interact. 2008, 175, 368-375. 221. Berg, L.; Niemiec, M. S.; Qian, W.; Andersson, C. D.; Wittung-Stafshede, P.; Ekström, F.; Linusson, A., Similar but Different: Thermodynamic and Structural Characterization of a Pair of Enantiomers Binding to Acetylcholinesterase. Angew. Chem., Int. Ed. 2012, 51, 12716-12720.

116

222. Andersson, C. D.; Forsgren, N.; Akfur, C.; Allgardsson, A.; Berg, L.; Engdahl, C.; Qian, W.; Ekström, F.; Linusson, A., Divergent Structure-Activity Relationships of Structurally Similar Acetylcholinesterase Inhibitors. J. Med. Chem. 2013, 56, 7615-7624. 223. Berg, L.; Mishra, B. K.; Andersson, C. D.; Ekström, F.; Linusson, A., The Nature of Activated Non-Classical Hydrogen Bonds: A Case Study on Acetylcholinesterase-Ligand Complexes. Chemistry 2016, 22, 2672-2681. 224. Pang, Y.-P.; Ekström, F.; Polsinelli, G. A.; Gao, Y.; Rana, S.; Hua, D. H.; Andersson, B.; Andersson, P. O.; Peng, L.; Singh, S. K.; Mishra, R. K.; Zhu, K. Y.; Fallon, A. M.; Ragsdale, D. W.; Brimijoin, S., Selective and Irreversible Inhibitors of Mosquito Acetylcholinesterases for Controlling Malaria and Other Mosquito-Borne Diseases. PLoS One 2009, 4. 225. Dou, D.; Park, J. G.; Rana, S.; Madden, B. J.; Jiang, H.; Pang, Y.-P., Novel Selective and Irreversible Mosquito Acetylcholinesterase Inhibitors for Controlling Malaria and Other Mosquito-Borne Diseases. Sci. Rep. 2013, 3. 226. Hartsel, J. A.; Wong, D. M.; Mutunga, J. M.; Ma, M.; Anderson, T. D.; Wysinski, A.; Islam, R.; Wong, E. A.; Paulson, S. L.; Li, J.; Lam, P. C. H.; Totrov, M. M.; Bloomquist, J. R.; Carlier, P. R., Re-Engineering Aryl Methylcarbamates to Confer High Selectivity for Inhibition of Anopheles Gambiae Versus Human Acetylcholinesterase. Bioorg. Med. Chem. Lett. 2012, 22, 4593-4598. 227. Jiang, Y.; Swale, D.; Carlier, P. R.; Hartsel, J. A.; Ma, M.; Ekström, F.; Bloomquist, J. R., Evaluation of Novel Carbamate Insecticides for Neurotoxicity to Non-Target Species. Pestic. Biochem. Physiol. 2013, 106, 156-161. 228. Swale, D. R.; Carlier, P. R.; Hartsel, J. A.; Ma, M.; Bloomquist, J. R., Mosquitocidal Carbamates with Low Toxicity to Agricultural Pests: An Advantageous Property for Insecticide Resistance Management. Pest Manage. Sci. 2015, 71, 1158-1164. 229. Pang, Y.-P., Species Marker for Developing Novel and Safe Pesticides. Bioorg. Med. Chem. Lett. 2007, 17, 197-199. 230. Wong, D. M.; Li, J.; Lam, P. C. H.; Hartsel, J. A.; Mutunga, J. M.; Totrov, M.; Bloomquist, J. R.; Carlier, P. R., Aryl Methylcarbamates: Potency and Selectivity Towards Wild-Type and Carbamate-Insensitive (G119s) Anopheles Gambiae Acetylcholinesterase, and Toxicity to G3 Strain An. Gambiae. Chem.-Biol. Interact. 2013, 203, 314-318. 231. Verma, A.; Wong, D. M.; Islam, R.; Tong, F.; Ghavami, M.; Mutunga, J. M.; Slebodnick, C.; Li, J.; Viayna, E.; Lam, P. C. H.; Totrov, M. M.; Bloomquist, J. R.; Carlier, P. R., 3-Oxoisoxazole-2(3h)-Carboxamides and Isoxazol-3-Yl Carbamates: Resistance-Breaking Acetylcholinesterase Inhibitors Targeting the Malaria Mosquito, Anopheles Gambiae. Bioorg. Med. Chem. 2015, 23, 1321-1340. 232. Camerino, E.; Wong, D. M.; Tong, F.; Koerber, F.; Gross, A. D.; Islam, R.; Viayna, E.; Mutunga, J. M.; Li, J.; Totrov, M. M.; Bloomquist, J. R.; Carlier, P. R., Difluoromethyl Ketones: Potent Inhibitors of Wild Type and Carbamate-Insensitive G119s Mutant Anopheles Gambiae Acetylcholinesterase. Bioorg. Med. Chem. Lett. 2015, 25, 4405-4411.

117

233. Alout, H.; Labbe, P.; Berthomieu, A.; Djogbenou, L.; Leonetti, J.-P.; Fort, P.; Weill, M., Novel AChE Inhibitors for Sustainable Insecticide Resistance Management. PLoS One 2012, 7. 234. Carreno Otero, A. L.; Vargas Mendez, L. Y.; Duque L, J. E.; Kouznetsov, V. V., Design, Synthesis, Acetylcholinesterase Inhibition and Larvicidal Activity of Girgensohnine Analogs on Aedes Aegypti, Vector of Dengue Fever. Eur. J. Med. Chem. 2014, 78, 392-400. 235. Ridley, S. M.; Elliott, A. C.; Yeung, M.; Youle, D., High-Throughput Screening as a Tool for Agrochemical Discovery: Automated Synthesis, Compound Input, Assay Design and Process Management. Pestic. Sci. 1998, 54, 327-337. 236. Harrison, W., Agrochemical Research and Development in the 21(St) Century: Bloom or Bust. J. Biomol. Screening 1999, 4, 61-65. 237. Stetter, J.; Lieb, F., Innovation in Crop Protection: Trends in Research. Angew. Chem., Int. Ed. 2000, 39, 1724-1744. 238. Delaney, J.; Clarke, E.; Hughes, D.; Rice, M., Modern Agrochemical Research: A Missed Opportunity for Drug Discovery? Drug Discovery Today 2006, 11, 839-845. 239. Lindell, S. D.; Pattenden, L. C.; Shannon, J., Combinatorial Chemistry in the Agrosciences. Bioorg. Med. Chem. 2009, 17, 4035-4046. 240. Lamberth, C.; Jeanmart, S.; Luksch, T.; Plant, A., Current Challenges and Trends in the Discovery of Agrochemicals. Science 2013, 341, 742-746. 241. Lombardino, J. G.; Lowe, J. A., The Role of the Medicinal Chemist in Drug Discovery - Then and Now. Nat. Rev. Drug Discovery 2004, 3, 853-862. 242. Blass, B. E., Chapter 1 - Drug Discovery and Development: An Overview of Modern Methods and Principles. In Basic Principles of Drug Discovery and Development, Academic press/Elsevier: 2015; pp 1-34. 243. CropLife, Innovation in Support of Sustainable Agriculture, Brussels, Available at: https://croplife.org/crop-protection/stewardship/research-development/ (accessed: 23 Mar. 2016). 2007. 244. WHO, Guidelines for Testing Mosquito Adulticides for Indoor Residual Spraying and Treatment of Mosquito Nets. Geneva: World Health Organization 2006, (WHO/CDS/NTD/WHOPES/GCDPP/2006.3). 245. Gohlke, H.; Klebe, G., Approaches to the Description and Prediction of the Binding Affinity of Small-Molecule Ligands to Macromolecular Receptors. Angew. Chem., Int. Ed. 2002, 41, 2644-76. 246. Bissantz, C.; Kuhn, B.; Stahl, M., A Medicinal Chemist's Guide to Molecular Interactions. J. Med. Chem. 2010, 53, 5061-5084. 247. Kubinyi, H., Similarity and Dissimilarity: A Medicinal Chemist's View. Perspect. Drug Discovery Des. 1998, 9-11, 225-252. 248. Hughes, J. P.; Rees, S.; Kalindjian, S. B.; Philpott, K. L., Principles of Early Drug Discovery. Br. J. Pharmacol. 2011, 162, 1239-1249. 249. Blundell, T. L.; Jhoti, H.; Abell, C., High-Throughput Crystallography for Lead Discovery in Drug Design. Nat. Rev. Drug Discovery 2002, 1, 45-54. 250. Walter, M. W., Structure-Based Design of Agrochemicals. Nat. Prod. Rep. 2002, 19, 278-291.

118

251. Davis, A. M.; Teague, S. J.; Kleywegt, G. J., Application and Limitations of X-Ray Crystallographic Data in Structure-Based Ligand and Drug Design. Angew. Chem., Int. Ed. 2003, 42, 2718-2736. 252. Congreve, M.; Murray, C. W.; Blundell, T. L., Keynote Review: Structural Biology and Drug Discovery. Drug Discovery Today 2005, 10, 895-907. 253. Zheng, H.; Handing, K. B.; Zimmerman, M. D.; Shabalin, I. G.; Almo, S. C.; Minor, W., X-Ray Crystallography over the Past Decade for Novel Drug Discovery - Where Are We Heading Next? Expert Opin. Drug Discovery 2015, 10, 975-989. 254. Wüthrich, K., Protein-Structure Determination in Solution by Nmr-Spectroscopy. J. Biol. Chem. 1990, 265, 22059-22062. 255. Cavalli, A.; Salvatella, X.; Dobson, C. M.; Vendruscolo, M., Protein Structure Determination from Nmr Chemical Shifts. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 9615-9620. 256. Chruszcz, M.; Wlodawer, A.; Minor, W., Determination of Protein Structures - a Series of Fortunate Events. Biophys. J. 2008, 95, 1-9. 257. Wlodawer, A.; Minor, W.; Dauter, Z.; Jaskolski, M., Protein Crystallography for Aspiring Crystallographers or How to Avoid Pitfalls and Traps in Macromolecular Structure Determination. FEBS J. 2013, 280, 5705-5736. 258. Hillisch, A.; Pineda, L. F.; Hilgenfeld, R., Utility of Homology Models in the Drug Discovery Process. Drug Discovery Today 2004, 9, 659-669. 259. Cavasotto, C. N.; Phatak, S. S., Homology Modeling in Drug Discovery: Current Trends and Applications. Drug Discovery Today 2009, 14, 676-683. 260. Marti-Renom, M. A.; Stuart, A. C.; Fiser, A.; Sanchez, R.; Melo, F.; Sali, A., Comparative Protein Structure Modeling of Genes and Genomes. Annu. Rev. Biophys. Biomol. Struct. 2000, 29, 291-325. 261. Pereira, D. A.; Williams, J. A., Origin and Evolution of High Throughput Screening. Br. J. Pharmacol. 2007, 152, 53-61. 262. Bleicher, K. H.; Bohm, H. J.; Muller, K.; Alanine, A. I., Hit and Lead Generation: Beyond High-Throughput Screening. Nat. Rev. Drug Discovery 2003, 2, 369-378. 263. Mayr, L. M.; Fuerst, P., The Future of High-Throughput Screening. J. Biomol. Screening 2008, 13, 443-448. 264. Macarron, R.; Banks, M. N.; Bojanic, D.; Burns, D. J.; Cirovic, D. A.; Garyantes, T.; Green, D. V. S.; Hertzberg, R. P.; Janzen, W. P.; Paslay, J. W.; Schopfer, U.; Sittampalam, G. S., Impact of High-Throughput Screening in Biomedical Research. Nat. Rev. Drug Discovery 2011, 10, 188-195. 265. Stein, R. L., High-Throughput Screening in Academia: The Harvard Experience. J. Biomol. Screening 2003, 8, 615-619. 266. Nelson, S. L., Academic Hts: Diverse Portraits. Drug Discovery Today: Technol. 2008, 5, e1-e34. 267. Frearson, J. A.; Collie, I. T., Hts and Hit Finding in Academia - from Chemical Genomics to Drug Discovery. Drug Discovery Today 2009, 14, 1150-1158.

119

268. Zhang, J. H.; Chung, T. D. Y.; Oldenburg, K. R., A Simple Statistical Parameter for Use in Evaluation and Validation of High Throughput Screening Assays. J. Biomol. Screening 1999, 4, 67-73. 269. Malo, N.; Hanley, J. A.; Cerquozzi, S.; Pelletier, J.; Nadon, R., Statistical Practice in High-Throughput Screening Data Analysis. Nat. Biotechnol. 2006, 24, 167-175. 270. Birmingham, A.; Selfors, L. M.; Forster, T.; Wrobel, D.; Kennedy, C. J.; Shanks, E.; Santoyo-Lopez, J.; Dunican, D. J.; Long, A.; Kelleher, D.; Smith, Q.; Beijersbergen, R. L.; Ghazal, P.; Shamu, C. E., Statistical Methods for Analysis of High-Throughput Rna Interference Screens. Nat. Methods 2009, 6, 569-575. 271. Goktug, A. N.; Chai, S. C.; Chen, T., Data Analysis Approaches in High Throughput Screening. In Drug Discovery, Ed. El-Shemy, H., published by InTech: 2013; 201-226. 272. Bajorath, J., Molecular Crime Scene Investigation - Dusting for Fingerprints. Drug Discovery Today: Technol. 2013, 10, e491-8. 273. Willett, P.; Barnard, J. M.; Downs, G. M., Chemical Similarity Searching. J. Chem. Inf. Comput. Sci. 1998, 38, 983-996. 274. Hankinson, S. E.; Willett, W. C.; Colditz, G. A.; Hunter, D. J.; Michaud, D. S.; Deroo, B.; Rosner, B.; Speizer, F. E.; Pollak, M., Circulating Concentrations of Insulin-Like Growth Factor-I and Risk of Breast Cancer. Lancet 1998, 351, 1393-1396. 275. Pearson, K., On Lines and Planes of Closest Fit to Systems of Points in Space. Philos. Mag. Series 6 1901, 2, 559-572. 276. Wold, S.; Esbensen, K.; Geladi, P., Principal Component Analysis. Chemom. Intell. Lab. Syst. 1987, 2, 37-52. 277. Bylesjö, M.; Rantalainen, M.; Cloarec, O.; Nicholson, J. K.; Holmes, E.; Trygg, J., OPLS Discriminant Analysis: Combining the Strengths of PLS-DA and Simca Classification. J. Chemom. 2006, 20, 341-351. 278. Trygg, J.; Wold, S., Orthogonal Projections to Latent Structures (O-Pls). J. Chemom. 2002, 16, 119-128. 279. Ellman, G. L.; Courtney, K. D.; Andres, V.; Featherstone, R. M., A New and Rapid Colorimetric Determination of Acetylcholinesterase Activity. Biochem. Pharmacol. 1961, 7, 88-95. 280. Michaelis, L.; Menten, M. L., Die Kinetik Der Invertinwirkung. Biochem. Z. 1913, 49, 333-369. 281. Vellom, D. C.; Radic, Z.; Li, Y.; Pickering, N. A.; Camp, S.; Taylor, P., Amino-Acid-Residues Controlling Acetylcholinesterase and Butylcholinesterase Specificity. Biochemistry 1993, 32, 12-17. 282. Hosea, N. A.; Berman, H. A.; Taylor, P., Specificity and Orientation of Trigonal Carboxyl Esters and Tetrahedral Alkylphosphonyl Esters in Cholinesterases. Biochemistry 1995, 34, 11528-11536. 283. Radic, Z.; Pickering, N. A.; Vellom, D. C.; Camp, S.; Taylor, P., 3 Distinct Domains in the Cholinesterase Molecule Confer Selectivity for Acetylcholinesterase and Butyrylcholinesterase Inhibitors. Biochemistry 1993, 32, 12074-12084. 284. Kaplan, D.; Ordentlich, A.; Barak, D.; Ariel, N.; Kronman, C.; Velan, B.; Shafferman, A., Does "Butyrylization" of Acetylcholinesterase through Substitution of the Six Divergent Aromatic Amino Acids in the Active Center

120

Gorge Generate an Enzyme Mimic of Butyrylcholinesterase? Biochemistry 2001, 40, 7433-7445. 285. Barak, D.; Kronman, C.; Ordentlich, A.; Ariel, N.; Bromberg, A.; Marcus, D.; Lazar, A.; Velan, B.; Shafferman, A., Acetylcholinesterase Peripheral Anionic Site Degeneracy Conferred by Amino-Acid Arrays Sharing a Common Core. J. Biol. Chem. 1994, 269, 6296-6305. 286. Ordentlich, A.; Barak, D.; Kronman, C.; Ariel, N.; Segall, Y.; Velan, B.; Shafferman, A., Contribution of Aromatic Moieties of Tyrosine-133 and of the Anionic Subsite Tryptophan-86 to Catalytic Efficiency and Allosteric Modulation of Acetylcholinesterase. J. Biol. Chem. 1995, 270, 2082-2091. 287. Shafferman, A.; Ordentlich, A.; Barak, D.; Kronman, C.; Ber, R.; Bino, T.; Ariel, N.; Osman, R.; Velan, B., Electrostatic Attraction by Surface Charge Does Not Contribute to the Catalytic Efficiency of Acetylcholinesterase. EMBO J. 1994, 13, 3448–3455. 288. Shafferman, A.; Velan, B.; Ordentlich, A.; Kronman, C.; Grosfeld, H.; Leitner, M.; Flashner, Y.; Cohen, S.; Barak, D.; Ariel, N., Substrate-Inhibition of Acetylcholinesterase - Residues Affecting Signal Transduction from the Surface to the Catalytic Center. EMBO J. 1992, 11, 3561-3568. 289. Kronman, C.; Ordentlich, A.; Barak, D.; Velan, B.; Shafferman, A., The Back Door Hypothesis for Product Clearance in Acetylcholinesterase Challenged by Site-Directed Mutagenesis. J. Biol. Chem. 1994, 269, 27819-27822. 290. Gnagey, A. L.; Forte, M.; Rosenberry, T. L., Isolation and Characterization of Acetylcholinesterase from Drosophila. J. Biol. Chem. 1987, 262, 13290-13298. 291. WHO, WHO Specifications and Evaluations for Public Health Pesticides. Propoxur., Available at http://www.who.int/whopes/quality/newspecif/en/ (accessed: 23 Feb. 2016). 292. Julian, P. L.; Pikl, J., Studies in the Indole Series. V. The Complete Synthesis of Physostigmine (Eserine). J. Am. Chem. Soc. 1935, 57, 755-757. 293. Sugimoto, H.; Yamanishi, Y.; Iimura, Y.; Kawakami, Y., Donepezil Hydrochloride (E2020) and Other Acetylcholinesterase Inhibitors. Curr. Med. Chem. 2000, 7, 303-339. 294. Saxena, A.; Redman, A. M. G.; Jiang, X. L.; Lockridge, O.; Doctor, B. P., Differences in Active-Site Gorge Dimensions of Cholinesterases Revealed by Binding of Inhibitors to Human Butyrylcholinesterase. Chem.-Biol. Interact. 1999, 119, 61-69. 295. Sugimoto, H.; Iimura, Y.; Yamanishi, Y.; Yamatsu, K., Synthesis and Structure-Activity-Relationships of Acetylcholinesterase Inhibitors - 1-Benzyl-4- (5,6-Dimethoxy-1-Oxoindan-2-Yl)Methyl Piperidine Hydrochloride and Related-Compounds. J. Med. Chem. 1995, 38, 4821-4829. 296. Blennow, K.; de Leon, M. J.; Zetterberg, H., Alzheimer's Disease. Lancet 2006, 368, 387-403. 297. Berg, L.; Andersson, C. D.; Artursson, E.; Hörnberg, A.; Tunemalm, A.-K.; Linusson, A.; Ekström, F., Targeting Acetylcholinesterase: Identification of Chemical Leads by High Throughput Screening, Structure Determination and Molecular Modeling. PLoS One 2011, 6.

121

298. Nicolet, Y.; Lockridge, O.; Masson, P.; Fontecilla-Camps, J. C.; Nachon, F., Crystal Structure of Human Butyrylcholinesterase and of Its Complexes with Substrate and Products. J. Biol. Chem. 2003, 278, 41141-41147. 299. Temeyer, K. B.; Tong, F.; Totrov, M. M.; Tuckow, A. P.; Chen, Q.-h.; Carlier, P. R.; de Leon, A. A. P.; Bloomquist, J. R., Acetylcholinesterase of the Sand Fly, Phlebotomus Papatasi (Scopoli): Construction, Expression and Biochemical Properties of the G119s Orthologous Mutant. Parasites Vectors 2014, 7. 300. Baell, J. B.; Holloway, G. A., New Substructure Filters for Removal of Pan Assay Interference Compounds (Pains) from Screening Libraries and for Their Exclusion in Bioassays. J. Med. Chem. 2010, 53, 2719-2740. 301. Saubern, S.; Guha, R.; Baell, J. B., Knime Workflow to Assess Pains Filters in Smarts Format. Comparison of Rdkit and Indigo Cheminformatics Libraries. Mol. Inf. 2011, 30, 847-850. 302. Congreve, M.; Carr, R.; Murray, C.; Jhoti, H., A Rule of Three for Fragment-Based Lead Discovery? Drug Discovery Today 2003, 8, 876-877. 303. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J., Experimental and Computational Approaches to Estimate Solubility and Permeability in Drug Discovery and Development Settings. Adv. Drug Delivery Rev. 1997, 23, 3-25. 304. Dvir, H.; Jiang, H. L.; Wong, D. M.; Harel, M.; Chetrit, M.; He, X. C.; Jin, G. Y.; Yu, G. L.; Tang, X. C.; Silman, I.; Bai, D. L.; Sussman, J. L., X-Ray Structures of Torpedo Californica Acetylcholinesterase Complexed with (+)-Huperzine a and (-)-Huperzine B: Structural Evidence for an Active Site Rearrangement. Biochemistry 2002, 41, 10810-10818. 305. Pang, Y.-P.; Brimijoin, S.; Ragsdale, D. W.; Zhu, K. Y.; Suranyi, R., Novel and Viable Acetylcholinesterase Target Site for Developing Effective and Environmentally Safe Insecticides. Curr. Drug Targets 2012, 13, 471-482. 306. Sellstedt, M.; Almqvist, F., Synthesis of a Novel Tricyclic Peptidomimetic Scaffold. Org. Lett. 2008, 10, 4005-4007. 307. Chorell, E.; Pinkner, J. S.; Phan, G.; Edvinsson, S.; Buelens, F.; Remaut, H.; Waksman, G.; Hultgren, S. J.; Almqvist, F., Design and Synthesis of C-2 Substituted Thiazolo and Dihydrothiazolo Ring-Fused 2-Pyridones: Pilicides with Increased Antivirulence Activity. J. Med. Chem. 2010, 53, 5690-5695. 308. Pang, Y.-P., Insect Acetylcholinesterase as a Target for Effective and Environmentally Safe Insecticides. Target Receptors in the Control of Insect Pests: Pt Ii 2014, 46, 435-494. 309. Choudhary, A.; Raines, R. T., An Evaluation of Peptide-Bond Isosteres. ChemBioChem 2011, 12, 1801-1807. 310. Tayeh, A.; Jalouk, L.; Al-Khiami, A. M., A Cutaneous Leishmaniasis Control Trial Using Pyrethroid-Impregnated Bednets in Villages near Aleppo, Syria / Ahmed Tayeh, Lama Jalouk, Aida Madani Al-Khiami. Geneva: World Health Organization 1997, (WHO/LEISH/97.41 Available at http://apps.who.int/iris/handle/10665/63792 (accessed: 28 Mar. 2016)). 311. Jalouk, L.; Al Ahmed, M.; Gradoni, L.; Maroli, M., Insecticide-Treated Bednets to Prevent Anthroponotic Cutaneous Leishmaniasis in Aleppo

122

Governorate, Syria: Results from Two Trials. Trans. R. Soc. Trop. Med. Hyg. 2007, 101, 360-367. 312. Ritmeijer, K.; Davies, C.; van Zorge, R.; Wang, S.-J.; Schorscher, J.; Dongu'du, S. I.; Davidson, R. N., Evaluation of a Mass Distribution Programme for Fine-Mesh Impregnated Bednets against Visceral Leishmaniasis in Eastern Sudan. Trop. Med. Int. Health 2007, 12, 404-414. 313. Ostyn, B.; Vanlerberghe, V.; Picado, A.; Dinesh, D. S.; Sundar, S.; Chappuis, F.; Rijal, S.; Dujardin, J.-C.; Coosemans, M.; Boelaert, M.; Davies, C., Vector Control by Insecticide-Treated Nets in the Fight against Visceral Leishmaniasis in the Indian Subcontinent, What Is the Evidence? Trop. Med. Int. Health 2008, 13, 1073-1085. 314. Picado, A.; Das, M. L.; Kumar, V.; Kesari, S.; Dinesh, D. S.; Roy, L.; Rijal, S.; Das, P.; Rowland, M.; Sundar, S.; Coosemans, M.; Boelaert, M.; Davies, C. R., Effect of Village-Wide Use of Long-Lasting Insecticidal Nets on Visceral Leishmaniasis Vectors in India and Nepal: A Cluster Randomized Trial. PLoS Neglected Trop. Dis. 2010, 4. 315. Bhattarai, A.; Ali, A. S.; Kachur, S. P.; Martensson, A.; Abbas, A. K.; Khatib, R.; Al-Mafazy, A.-w.; Ramsan, M.; Rotllant, G.; Gerstenmaier, J. F.; Molteni, F.; Abdulla, S.; Montgomery, S. M.; Kaneko, A.; Bjorkman, A., Impact of Artemisinin-Based Combination Therapy and Insecticide-Treated Nets on Malaria Burden in Zanzibar. PLoS Med. 2007, 4, 1784-1790. 316. Griffin, J. T.; Hollingsworth, T. D.; Okell, L. C.; Churcher, T. S.; White, M.; Hinsley, W.; Bousema, T.; Drakeley, C. J.; Ferguson, N. M.; Basanez, M.-G.; Ghani, A. C., Reducing Plasmodium Falciparum Malaria Transmission in Africa: A Model-Based Evaluation of Intervention Strategies. PLoS Med. 2010, 7. 317. Bockarie, M. J.; Pedersen, E. M.; White, G. B.; Michael, E., Role of Vector Control in the Global Program to Eliminate Lymphatic Filariasis. Annu. Rev. Entomol. 2009, 54, 469-487. 318. Eigege, A.; Kal, A.; Miri, E.; Sallau, A.; Umaru, J.; Mafuyai, H.; Chuwang, Y. S.; Danjuma, G.; Danboyi, J.; Adelamo, S. E.; Mancha, B. S.; Okoeguale, B.; Patterson, A. E.; Rakers, L.; Richards, F. O., Long-Lasting Insecticidal Nets Are Synergistic with Mass Drug Administration for Interruption of Lymphatic Filariasis Transmission in Nigeria. PLoS Neglected Trop. Dis. 2013, 7. 319. Reimer, L. J.; Thomsen, E. K.; Tisch, D. J.; Henry-Halldin, C. N.; Zimmerman, P. A.; Baea, M. E.; Dagoro, H.; Susapu, M.; Hetzel, M. W.; Bockarie, M. J.; Michael, E.; Siba, P. M.; Kazura, J. W., Insecticidal Bed Nets and Filariasis Transmission in Papua New Guinea. N. Engl. J. Med. 2013, 369, 745-753. 320. Sunish, I. P.; Rajendran, R.; Mani, T. R.; Munirathinam, A.; Dash, A. P.; Tyagi, B. K., Vector Control Complements Mass Drug Administration against Bancroftian Filariasis in Tirukoilur, India. Bull. W. H. O. 2007, 85, 138-145. 321. WHO, Handbook for Integrated Vector Management. Geneva: World Health Organization 2012, Available at: http://www.who.int/iris/handle/10665/44768 (accessed: 28 Mar. 2016). 322. Singh, J.; Petter, R. C.; Baillie, T. A.; Whitty, A., The Resurgence of Covalent Drugs. Nat. Rev. Drug Discovery 2011, 10, 307-317.

123

323. Kalgutkar, A. S.; Dalvie, D. K., Drug Discovery for a New Generation of Covalent Drugs. Expert Opin. Drug Discovery 2012, 7, 561-581. 324. Bauer, R. A., Covalent Inhibitors in Drug Discovery: From Accidental Discoveries to Avoided Liabilities and Designed Therapies. Drug Discovery Today 2015, 20, 1061-1073. 325. Hemingway, J.; Beaty, B. J.; Rowland, M.; Scott, T. W.; Sharp, B. L., The Innovative Vector Control Consortium: Improved Control of Mosquito-Borne Diseases. Trends Parasitol. 2006, 22, 308-312. 326. The Innovative Vector Control Consortium. http://www.ivcc.com/ (accessed: 28 Mar. 2016). 327. Eastman, R. T.; Fidock, D. A., Artemisinin-Based Combination Therapies: A Vital Tool in Efforts to Eliminate Malaria. Nat. Rev. Microbiol. 2009, 7, 864-874. 328. Keith, C. T.; Borisy, A. A.; Stockwell, B. R., Multicomponent Therapeutics for Networked Systems. Nat. Rev. Drug Discovery 2005, 4, 71-U10. 329. Dancey, J. E.; Chen, H. X., Strategies for Optimizing Combinations of Molecularly Targeted Anticancer Agents. Nat. Rev. Drug Discovery 2006, 5, 649-659. 330. Zimmermann, G. R.; Lehár, J.; Keith, C. T., Multi-Target Therapeutics: When the Whole Is Greater Than the Sum of the Parts. Drug Discovery Today 2007, 12, 34-42. 331. Purser, S.; Moore, P. R.; Swallow, S.; Gouverneur, V., Fluorine in Medicinal Chemistry. Chem. Soc. Rev. 2008, 37, 320-330. 332. Patrick , G. L., Drug Design: Optimizing Access to the Target. In An Introduction to Medicinal Chemistry Fourth Edition, Oxford University Press Inc., New York: the United States, 2009. 333. Tice, C. M., Selecting the Right Compounds for Screening: Does Lipinski's Rule of 5 for Pharmaceuticals Apply to Agrochemicals? Pest Manage. Sci. 2001, 57, 3-16. 334. Conley, J. M.; Meyer, J. M.; Nuss, A. B.; Doyle, T. B.; Savinov, S. N.; Hill, C. A.; Watts, V. J., Evaluation of Aadop2 Receptor Antagonists Reveals Antidepressants and Antipsychotics as Novel Lead Molecules for Control of the Yellow Fever Mosquito, Aedes Aegypti. J. Pharmacol. Exp. Ther. 2015, 352, 53-60. 335. Zhang, J. M.; Yang, P. L.; Gray, N. S., Targeting Cancer with Small Molecule Kinase Inhibitors. Nat. Rev. Cancer 2009, 9, 28-39. 336. Contreras, J. M.; Rival, Y. M.; Chayer, S.; Bourguignon, J. J.; Wermuth, C. G., Aminopyridazines as Acetylcholinesterase Inhibitors. J. Med. Chem. 1999, 42, 730-741. 337. Sheng, R.; Lin, X.; Li, J. Y.; Jiang, Y. K.; Shang, Z. C.; Hu, Y. Z., Design, Synthesis, and Evaluation of 2-Phenoxy-Indan-1-One Derivatives as Acetylcholinesterase Inhibitors. Bioorg. Med. Chem. Lett. 2005, 15, 3834-3837. 338. Andersson, C. D.; Hillgren, J. M.; Lindgren, C.; Qian, W.; Akfur, C.; Berg, L.; Ekström, F.; Linusson, A., Benefits of Statistical Molecular Design, Covariance Analysis, and Reference Models in Qsar: A Case Study on Acetylcholinesterase. J. Comput.-Aided Mol. Des. 2015, 29, 199-215.

1

Appendix I

Multiple alignment The multiple alignment of the AChE sequences (Table A1) was performed using Clustal Omega.1-3

Table A1. AChEs included in the sequence comparison. Species Protein Accession no Homo sapiens (human) AChE AAA68151.1 An. gambiae (mosquito) AChE1 XP_321792.2 An. gambiae (mosquito) AChE1-G119Sa Ae. aegypti (mosquito) AChE1 ABN09910.1 Culex pipiens (mosquito) AChE1 CAD33707.2 Bos taurus (European cattle) AChE NP_001069688.1 Mus musculus (mouse) AChE P21836 Torpedo californica (Pacific electric ray) AChE CAA27169.1 Caenorhabditis elegans (round worm) AChE1 CAA53080.1 Musca domestica (house fly) AChE CAC39209.1 Rattus norvegicus (brown rat) AChE NP_742006.1 Drosophila melanogaster (fruit fly) AChE CAA29325.1 Oryctolagus cuniculus (European rabbit) AChE AAA53235.1 Danio rerio (zebra fish) AChE NP_571921.1 Apis mellifera (honey bee) AChE2 NP_001035320.1 Gallus gallus (chicken) AChE NP_001035320.1 Blattella germanica (German cockroach) AChE1 ABB89946.1 aThe sequence of AgAChE1-G119S mutant was manually generated from the sequence

of AgAChE1.

2

Aligned sequences used to calculate sequence identity.

3

4

5

6

Sequence identity The overall sequence identity was calculated compared to hAChE after multiple alignment as described above. The overall sequences identity was calculated based hAChE residues 1-554. The more focused analyses were performed on the residues lining the active site gorge. The amino acids selected for this analysis are listed in Table A2. AgAChE1 was also compared to AaAChE1, D. melanogaster AChE2 and A. mellifera AChE2, using the same alignment and residues as described above.

Table A2. Residues selected for focused analysis of the active site gorge. The lower

part including the choline binding site (choline), the oxyanion hole (oxyanion), the

acylpocket (acyl) and the catalytic triad (CAT), and the top part including PAS.

Part of gorgea hAChE residue Part of gorgea hAChE residue

CAS Q71  PAS (PAS) Y72 

CAS V73  PAS (PAS) I74 

CAS T83  PAS (PAS) Y124 

CAS (choline) W86  PAS (PAS) W286 

CAS N87  PAS H287 

CAS P88  PAS V288 

CAS G120  PAS L289 

CAS (oxyanion) G121  PAS P290 

CAS (oxyanion) G122  PAS Q291 

CAS S125  PAS E292 

CAS (choline) Y133  PAS S293 

CAS E202  PAS V294 

CAS (CAT) S203  PAS R296 

CAS (oxyanion) A204  PAS (PAS) Y341 

CAS (acyl) F295  PAS G342 

CAS (acyl) F297   

CAS (CAT) E334   

CAS (choline) Y337   

CAS (acyl) F338   

CAS (CAT) H447   

CAS G448   

CAS I451   aAs referred to in Table 5 Chapter 1. Sub-sites or pockets are given in parentheses.

CAT = catalytic triad.

7

Phylogenetic tree The phylogenetic tree was generated using PHYLIP.4 A protein distance matrix was calculated with the protein distance algorithm using the Jones-Taylor-Thornton evolutionary model,5 one substitution rate category, interleaved sequences and the ANSI terminal type. The enzymes were clustered using the Neighbor-Joining method6 with randomized order. The clustering was visualized as un-rooted trees with the angle of arc set to 360 degrees.

References 1. Sievers, F.; Wilm, A.; Dineen, D.; Gibson, T. J.; Karplus, K.; Li, W.; Lopez, R.; McWilliam, H.; Remmert, M.; Soeding, J.; Thompson, J. D.; Higgins, D. G., Fast, Scalable Generation of High-Quality Protein Multiple Sequence Alignments Using Clustal Omega. Mol. Syst. Biol. 2011, 7. 2. Goujon, M.; McWilliam, H.; Li, W.; Valentin, F.; Squizzato, S.; Paern, J.; Lopez, R., A New Bioinformatics Analysis Tools Framework at Embl-Ebi. Nucleic Acids Res. 2010, 38, W695-W699. 3. McWilliam, H.; Li, W.; Uludag, M.; Squizzato, S.; Park, Y. M.; Buso, N.; Cowley, A. P.; Lopez, R., Analysis Tool Web Services from the Embl-Ebi. Nucleic Acids Res. 2013, 41, W597-W600. 4. Felsenstein, J., Phylip (Phylogeny Inference Package) Version 3.6. Distributed by the Author. Department of Genome Sciences, University of Washington, Seattle, Available at: http://evolution.genetics.washington.edu/phylip.html (accessed: 24 Apr. 2016). 2005. 5. Jones, D. T.; Taylor, W. R.; Thornton, J. M., The Rapid Generation of Mutation Data Matrices from Protein Sequences. Comput. Appl. Biosci. 1992, 8, 275-282. 6. Saitou, N.; Nei, M., The Neighbor-Joining Method - a New Method for Reconstructing Phylogenetic Trees. Mol. Biol. Evol. 1987, 4, 406-425.

8

Appendix II Table A2. Methods used for per plate normalization and hit identification.

% of control

 

Robust % of control

 

% of sample

 

Robust % of sample

 

NPI

 

Where   

 

raw measurement   mean of all samples within plate   median of all samples within plate   standard deviation of all samples within plate   mean of negative controls within plate (maximum activity)   mean of positive controls within plate (minimum activity)   median of negative controls within plate (maximum activity)

9

Table A2. Results from different normalization and hit identification methods showing inhibition data. Methoda Meanb Medianc Stdd Cut-offe Hitsf AChE1-hitsg

PoC Ag 3 2 10 33 235

338 Aa 3 2 9 31 286

RPoC Ag 2 1 11 34 212

319 Aa 3 2 9 31 286

PoS Ag 0 -1 10 30 232

332 Aa 0 -1 9 28 297

RPoS Ag 1 0 10 31 229

333 Aa 1 0 9 29 299

NPI Ag 3 2 10 33 222

338 Aa 3 2 9 31 282

aPoC: % of control, RPoC: robust % of control, PoS: % of sample, RPoS: robust % of

sample, NPI: normalized % inhibition; bMean of all samples (17.500) after

normalization; cMedian of all samples after normalization; dStandard deviation of

mean of all samples after normalization; eMean + 3 standard deviations of all samples

after normalization; fNumber of hits; eNumber of unique hits of either AgAChE1,

AaAChE1, or both.

Table A3. Comparison of the results from different hit methods. Methoda Total no. hits Hits of both methods (%)

PoC vs RPoC 339 94

PoS vs RPoS 338 96

PoC vs PoS 352 90

PoC vs NIP 338 100

aPoC: % of control, RPoC: robust % of control, PoS: % of sample, RPoS: robust % of

sample, NPI: normalized % inhibition.

10

a)

b)

Figure A1. Plate order effect plots (activity vs plate order) from the HTS of AgAChE1 (a) and AaAChE1 (b). Raw data measurements are shown as grey dots and plate medians as black dots.

11

Identification of PAINS The hits were searched for pan-assay interference structures (PAINS) using a PAINS filter implemented in KNIME 2.10.21 using Indigo nodes as described by Saubern et al.2 The filter is based on the reported PAINS filter3 converted to SMARTS.4 According to this workflow, 16 of the 335 AChE1-hits contained PAINS. The identified PAINS consisted of five tertiary anilines, four phenolic Mannich bases, 3 alkylidene barbiturates, 2 hydroxyphenolhydrazones, one catechol, and one aminoacridine. Most of the PAINS belonged to hit set B (four hits belonged to set A, ten set B and two to set D). Analysis of the complete AChE1-hAChE hit set (425 compounds) showed that 25 hits contained PAINS. Based on the HTS data, eight of these were classified as potentially AChE1-selective, nine as potentially hAChE-selective, five as non-selective, and three as miscellaneous.

References 1. Berthold, M. R.; Cebron, N.; Dill, F.; Gabriel, T. R.; Kötter, T.; Meinl, T.; Ohl, P.; Sieb, C.; Thiel, K.; Wiswedel, B., Studies in Classification, Data Analysis, and Knowledge Organization (Gfkl 2007). Springer: Heidelberg, 2007. 2. Saubern, S.; Guha, R.; Baell, J. B., Knime Workflow to Assess Pains Filters in Smarts Format. Comparison of Rdkit and Indigo Cheminformatics Libraries. Mol. Inf. 2011, 30, 847-850. 3. Baell, J. B.; Holloway, G. A., New Substructure Filters for Removal of Pan Assay Interference Compounds (Pains) from Screening Libraries and for Their Exclusion in Bioassays. J. Med. Chem. 2010, 53, 2719-2740. 4. Guha, R., Pains Substructure Filters as Smarts. http://blog.rguha.net/?p=850 (accessed: 29 Feb. 2016).

12

Appendix III

Synthesis General. All reactions were carried out under inert atmosphere (N2) unless otherwise stated. THF and DMF were dried in a solvent drying system and freshly collected prior to reaction (THF was passed through neutral alumina; DMF was passed thorough activated molecular sieves followed by an isocyanate scrubber). All microwave reactions were carried out in a monomode reactor using Smith process vials sealed with a Teflon septum and an aluminum crimp top. The temperature was measured with an IR sensor, and reaction times refer to the irradiation time at the target temperature. Reactions were monitored using TLC (silica gel matrix, layer thickness 200 µm, particle size 25 µm) with UV-detection (254 nm) or developed using KMnO4 solution. Flash column chromatography (eluents given in brackets) was performed on normal phase silica gel (Merck, 60 Å, 40-63 µm). 1H and 13C NMR spectra were recorded on a Bruker DRX-400 instrument at 298 K in CDCl3 using residual CHCl3 (δH = 7.26 ppm) or CDCl3 (δC = 77.16 ppm) as an internal standard, (CD3)2SO using residual (CD3)(CD2H)SO (δH = 2.50 ppm) or (CD3)2SO (δC =39.52 ppm) as an internal standard, or CD3OD using residual CD2HOD (δH = 3.31 ppm) or CD3OD (δC = 49.0 ppm) as an internal standard. When CDCl3:CD3OD mixtures were used, CD3OD was used as the internal standard, and when CD2Cl2:CD3OD mixtures were used, CD2Cl2 was used as the internal standard using residual CDHCl2 (δH = 5.32 ppm) or CDCl3 (δC = 53.84 ppm). LC-MS analyses were performed on a Waters LC system using a Xterra MS C18 18.5 µm 4.6x50 mm column and an acetonitrile:water eluent system containing 0.2% formic acid. Eluting compounds were detected by monitoring the eluent’s absorption (254 nm) and mass spectrometry was performed in positive ion mode using a Waters micromass ZG 2000 electrospray instrument.

3-(Boc-amino)propyl bromide. 3-Bromopropylamine hydrobromide (2.10 g, 9.60mmol) and Boc2O (1.84 ml, 8.00 mmol) were suspended in CH2Cl2 (70 ml) and TEA (3.91 ml, 28.0 mmol) was added while stirring. The suspension cleared, and the reaction was allowed to stir at rt for 3h, before being concentrated under reduced pressure. The resulting crude was taken up in EtOAc and washed three times with KHSO4, followed by aq. NaHCO3 (sat.). The organic layer was dried over Na2SO4, filtered and concentrated under reduced pressure to give 3-(Boc-amino)propyl bromide as a colorless oil (1.79 g, 94%), which slowly crystallized at 4 °C and was used in the next step without further purification. 1H-NMR (400 MHz, CDCl3): δ 4.80 (br s, 1H), 3.39 (t, J = 6.6 Hz, 2H), 3.24-3.19 (m , 2H), 2.03-1.97 8m, 2H), 1.39 (s, 9H).

13

1-(3-tert-butyloxycarbonylaminopropyl)-4-ethylpiperazine.  1-Ethylpiperazine (2.00 ml, 15.8 mmol) followed by TBAI (139 mg, 0.38 mmol) was added to 3-(Boc-amino)propyl bromide (1.79 g, 7.52 mmol) dissolved in CH2Cl2 (50 ml) while stirring. The reaction mixture was stirred at rt overnight. Analysis by LC-MS indicated that the reaction had not gone to completion and so the mixture was concentrated under reduced pressure, re-dissolved MeCN (30 ml), and stirred for an additional 2 days before being concentrated under reduced pressure. The resulting crude was taken up in EtOAc, washed three times with aq. NaHCO3 (sat.) and once with brine. The organic layer was dried over Na2SO4, filtered and concentrated under reduced pressure to give 1-(3-tert-butyloxycarbonylaminopropyl)-4-ethylpiperazine (1.24 g, 61% yield) as a clear oil. 1H-NMR (400 MHz, CDCl3): δ 5.57 (s, 1H), 3.15-3.02 (m, 2H), 2.53-2.22 (m, 12H), 1.61-1.50 (m, 2H), 1.33 (s, 9H), 1.04-0.94 (m, 3H);13C-NMR (100 MHz, CDCl3): δ 156.1, 78.6, 56.9, 53.1, 52.8, 52.3, 40.0, 28.4, 26.3, 11.9.

1-(3-Aminopropyl)-4-ethylpiperazine tri(trifluoroacetic acid) salt. TFA (7.64 ml, 99.5 mmol) was added dropwise to 1-(3-tert-butyloxycarbonylaminopropyl)-4-ethylpiperazine (1.08 g, 3.98 mmol) dissolved in CH2Cl2 (30 ml) at 0 °C while stirring. The reaction was then stirred at rt for 18 h, before being concentrated under reduced pressure to give an oil. The oil was stirred in 50 ml of Et2O with occasional scraping to give an off-white powder. The liquid was decanted and the resulting solids were triturated with fresh Et2O four times. The residual solvent was removed under reduced pressure to give the 1-(3-aminopropyl)-4-ethylpiperazine tri(trifluoroacetic acid) salt (1.98 g, 97% yield) as an off-white powder. 1H-NMR (600 MHz, 333 K, (CD3)2SO): δ 7.88 (br s, 3H), 6.61-5.02 (br s, 4H), 3.11 (q, J = 7.2 Hz, 2H), 3.01-2.82 (m, 5H), 2.72-2.70 (m, 2H), 1.84-1.79 (m, 2H), 1.22 (t, J = 7.2 Hz, 3H); 13C-NMR (150 MHz, 333 K, (CD3)2SO): δ 158.9 (q, J = 34 Hz), 116.9 (q, J = 296 Hz), 53.7, 51.0 49.8, 49.3, 37.4, 23.3, 9.2

(Biphenyl-4-yloxy)-acetic acid (44a). Compound 44 was prepared according to previously described methods.1

Sodium 2-([1,1’-Biphenyl]-4-yloxy)acetate (44b). Chloroacetic acid (1.12 g, 11.8 mmol) and 4-phenylphenol (13) (400 mg, 2.35 mmol) were dissolved in MeOH (10 ml). NaOH (676 mg, 16.9 mmol) dissolved in MeOH (25 ml) was added and the reaction was heated at reflux for 18 h. The formed precipitate was filtered and washed with 1 M HCl and CHCl3. Trituration with CHCl3 gave 44b (405 mg, 75% yield) as a white solid. 1H NMR (600 MHz, CD3OD) δ 7.55 (dd, J1 = 8.2 Hz, J2 = 1.2 Hz, 2H), 7.51 (dt, J1 = 8.8 Hz, J2 = 2.5 Hz, 2H), 7.38 (t, J = 7.7 Hz, 2H), 7.26 (t, J = 7.4 Hz, 1H), 7.01 (dt, J1

14

= 8.8 Hz, J2 = 2.5 Hz, 2H), 4.41 (s, 2H); 13C NMR (150 MHz, CD3OD) δ 176.5, 159.8, 142.2, 135.0, 129.7, 128.8, 127.6, 127.5, 116.1, 68.5.

2-Chloro-1-(4-morpholinopiperidin-1-yl)ethan-1-one (46). TEA (53 μl, 0.38 mmol) was added dropwise to chloroacetyl chloride (30 μl, 0.38 mmol) in CH2Cl2 (4 ml) at 0 °C. 4-Morpholinopiperidine (45) (50 mg, 0.29 mmol) was added in portions, and the reaction mixture was kept at 0 °C for 10 min before stirring was continued at rt for 1.5 h. The mixture was washed with NaHCO3 (aq., sat.) followed by brine. Each aqueous layer was back-extracted once with CH2Cl2. The combined organic layers were dried over Na2SO4, filtered and concentrated to give compound 46 (38 mg, 52% yield), which was used in the following step without further purification. 1H NMR (400 MHz, CDCl3) δ 4.56-4.55 (m, 1H), 4.09 (d, J = 12.0 Hz, 1H), 4.04 (d, J = 12.0 Hz, 1H), 3.92-3.88 (m, 1H), 3.73-3.71 (m, 4H), 3.16-3.09 (m, 1H), 2.73-2.66 (m, 1H), 2.56-2.54 (m, 4H), 2.47-2.40 (m, 1H), 1.95-1.88 (m, 2H), 1.59-150 (m, 1H), 1.50-1.39 (m, 1H). 13C NMR (150 MHz, CDCl3) δ 165.0, 67.3, 61.7, 49.9, 45.7, 41.2, 28.8, 27.9.

2-([1,1'-Biphenyl]-4-yloxy)-1-(4-morpholinopiperidin-1-yl)ethan-1-one. Amide 46, (38 mg, 0.15 mmol), 4-phenylphenol (39 mg, 0.23 mmol), K2CO3 (43 mg, 0.31 mmol), and KI (1 mg, 0.008 mmol) were mixed in DMF (1.5 ml) and stirred at rt for 3 days. The reaction mixture was diluted with EtOAc, and washed with NaHCO3 (aq., sat.) and three times with brine. The organic layer was dried over Na2SO4, filtered and concentrated to give an oil that solidified upon addition of EtOAc. The resulting solid was recrystallized from EtOAc to give 2-([1,1'-biphenyl]-4-yloxy)-1-(4-morpholinopiperidin-1-yl)ethan-1-one as light yellow crystals (28 mg, 48% yield). The product contains ca. 4% of 4-phenylphenol. 1H NMR (400 MHz, CDCl3) δ 7.54-7.50 (m, 4H), 7.43-7.39 (m, 2H), 7.32-7.29 (m, 1H), 7.02 (d, J = 8.8 Hz, 2H), 4.75 (d, J = 13.2 Hz, 1H), 4.70 (d, J = 13.2 Hz, 1H), 4.58 (d, J = 13.2 Hz, 1H), 4.07 (d, J = 13.2 Hz, 1H), 3.71-3.69 (m, 4H), 3.11-3.04 (m, 1H), 2.72-2.65 (m, 2H), 2.52-2.50 (m, 4H), 2.45-2.38 (m, 1H), 1.91-1.86 (m, 2H), 1.50-1.35 (m, 2H); 13C NMR (10 MHz, CDCl3) δ 166.3, 157.6, 140.7, 134.8, 128.9, 128.4, 126.95, 126.87, 115.0, 68.0, 67.3, 61.8, 49.8, 44.6, 41.6, 28.9, 28.0.

2-([1,1'-Biphenyl]-4-yloxy)-1-(4-morpholinopiperidin-1-yl)ethan-1-one hydrochloride (47). 2-([1,1'-Biphenyl]-4-yloxy)-1-(4-morpholinopiperidin-1-yl)ethan-1-one (24 mg, 0.06 mmol) was treated with 5 ml ½ sat. HCl in CH2Cl2 while stirring at 0 °C. Stirring was continued on ice for 4h before the mixture was concentrated under reduced pressure. The resulting solid was recrystallized from a mixture of MeOH and EtOAc to give compound 47 (26 mg, 91% yield). 1H NMR (600 MHz, (CD3)2SO) δ 11.56 (br s, 1H), 7.61-7.58 (m, 4H), 7.44-7.41 (m, 2H), 7.32-7.29 (m, 1H), 7.02 (d, J =

15

8.8 Hz, 2H), 4.91 (d, J = 14.5 Hz, 1H), 4.86 (d, J = 14.5 Hz, 1H), 4.47 (d, J = 13.0 Hz, 1H), 4.01 (d, J = 13.0 Hz, 1H), 3.94-3.90 (m, 4H), 3.44-3.38 (m, 3H), 3.08-3.04 (m, 3H), 2.63-2.59 (m, 1H), 2.19-2.15 (m, 2H), 1.79-1.73 (m, 1H), 1.59-1.53 (m, 1H); 13C NMR (150 MHz, (CD3)2SO) δ 165.7, 157.7, 139.8, 132.9, 128.9, 127.7, 126.8, 126.2, 115.1, 65.9, 63.3, 62.2, 48.2, 42.6, 40.1, 26.1, 25.4.

General procedure for the amide coupling. The carboxylate 44b or carboxylic acid 50 (1 eq.) was suspended in THF with a drop of DMF, and cooled to 0 °C. Oxalyl chloride (5 eq.) was added dropwise over 5 min before the mixture was stirred at rt for 1 h. The solvent and excess oxalyl chloride were removed under reduced pressure. The resulting residue was dissolved in THF and cooled to 0 °C. The corresponding amine (2.2 equiv.) was added and the reaction mixture was stirred at rt overnight before being concentrated under reduced pressure.

General procedure for the amide coupling workup. The reaction residue was dissolved in 50 ml of dichloromethane. For non-basic compounds 48a and 48b, the organic phase was washed with 1 M HCl (100 ml), aq. NaHCO3 (sat., 50 ml), and brine (50 ml); the organic phase was dried over MgSO4 and the solvent was removed under reduced pressure. For basic compounds 48c, 49a-d, and 51a-b, the organic phase was extracted twice with 1 M HCl (50 ml); the combined aqueous phases were made basic by the addition of Na2CO3 (sat.), extracted twice with CH2Cl2, dried over MgSO4, filtered, and concentrated under reduced pressure.

General procedure for preparation of HCl salts. Small amounts of the corresponding HCl salts of the amines were prepared for biochemical evaluation by dissolving the amine in CH2Cl2 (1-2 ml/ 100 mg) and treating it with half-sat. HCl in CH2Cl2 (~4-8 ml/100 mg) at rt for 10 min. The mixture was concentrated and then co-concentrated from CH2Cl2 twice. If needed, the resulting HCl salts were purified by precipitation from MeOH:Et2O or CH2Cl2:Et2O mixtures and careful removal of the solvent.

1-[([1,1’-Biphenyl]-4-yloxy)acetyl]piperidine (49a). Starting from 44b and following the general procedure for amide coupling, compound 49a was prepared as an orange solid (42 mg, 68% yield). 1H NMR (400 MHz, CDCl3) δ 7.57-7.50 (m, 4H), 7.45-7.38 (m, 2H), 7.34-7.28 (m, 1H), 7.05-7.01 (m, 2H), 4.72 (s, 2H), 3.61-3.55 (m, 2H), 3.54-3.48 (m, 2H), 1.71-1.52 (m, 6H); 13C NMR (100 MHz, CDCl3) δ 166.2, 157.7, 140.8, 134.7, 128.8, 128.4, 126.90, 126.89, 115.0, 68.0, 46.6, 43.4, 26.6, 25.7, 24.6.

16

4-[([1,1’-Biphenyl]-4-yloxy)acetyl]morpholine (48b). Starting from 44b and following the general procedure for amide coupling, compound 48b was prepared as a colorless solid (46 mg, 74% yield). 1H NMR (400 MHz, CDCl3) δ 7.57-7.50 (m, 4H), 7.45-7.38 (m, 2H), 7.35-7.28 (m, 1H), 7.05-7.00 (m, 2H), 4.74 (s, 2H), 3.71-3.59 (m, 8H); 13C NMR (100 MHz, CDCl3) δ 166.7, 157.3, 140.6, 135.0, 128.9, 128.5, 127.0, 126.9, 115.0, 67.9, 66.94, 66.87, 46.1, 42.6.

2-([1,1'-Biphenyl]-4-yloxy)-1-(4-ethylpiperazin-1-yl)ethan-1-one. Sodium 2-([1,1'-biphenyl]-4-yloxy)acetate (44b) (830 mg, 0.12 mmol) was suspended in 2 ml of CH2Cl2. Oxalyl chloride (21 μl, 0.24 mmol) was added dropwise, followed by one drop of DMF, while stirring at rt. The reaction mixture was stirred at rt for 1 h, after which it was concentrated and co-concentrated twice from CH2Cl2. The residue was dissolved in 2 ml of CH2Cl2, and NaHCO3 (11 mg, 0.13 mmol) and 1-ethylpiperazine (17 μl, 0.13 mmol) dissolved in 0.5 ml H2O was added. Stirred at rt overnight. The reaction mixture was diluted with EtOAc, washed with NaHCO3 (aq., sat.), and the EtOAc phase was extracted three times with 2 M HCl. The aqueous layer was made basic by the addition of NaOH (s) and extracted three times with EtOAc. The combined organic layers were washed with brine, dried over Na2SO4, filtered and concentrated under reduced pressure to give 2-([1,1'-biphenyl]-4-yloxy)-1-(4-ethylpiperazin-1-yl)ethan-1-one (27 mg, 69% yield). 1H NMR (400 MHz, CDCl3) δ 7.55-7.51 (m, 4H), 7.43-7.39 (m, 2H), 7.33-7.29 (m, 1H), 7.03-7.01 (m, 2H), 4.73 (s, 2H), 3.69-3.62 (m, 4H), 2.47-2.44 (m, 4H), 2.43 (q, J = 7.2 Hz, 2H), 1.09 (t, J = 7.2 Hz, 3H); 13C NMR (100 MHz, CDCl3) δ 166.4, 157.5, 140.7, 134.9, 128.9, 128.4, 127.0, 126.9, 115.1, 67.9, 53.1, 52.5, 52.3, 45.4, 42.2, 12.0.

2-([1,1'-Biphenyl]-4-yloxy)-1-(4-ethylpiperazin-1-yl)ethan-1-one hydrochloride (48c). 2-([1,1'-Biphenyl]-4-yloxy)-1-(4-ethylpiperazin-1-yl)ethan-1-one (15 mg, 0.05 mmol) was dissolved in 1 ml CH2Cl2, and 2 ml 1M HCl in Et2O was added dropwise at rt while stirring. Stirred at rt for 10 min before the formed white precipitate was filtered and washed with Et2O to give 48c (17 mg, 90% yield) as white solid. 1H NMR (400 MHz, CDCl3:CD3OD 1:1) δ 7.53-7.50 (m, 4H), 7.40-7.39 (m, 2H), 7.29-7.26 (m, 1H), 7.02-7.00 (m, 2H), 4.87-4.77 (m, 2H), 4.67-4.64 (m, 1H), 4.29-4.22 (m, 1H), 3.79 (m, 1H), 3.58-3.55 (m, 2H), 3.35-3.27 (m, 1H), 3.17 (q, J = 7.2 Hz, 2H), 3.02-2.86 (m, 2H), 1.09 (t, J = 7.2 Hz, 3H); 13C NMR (100 MHz, CDCl3: CD3OD 1:1) δ 168.0, 157.7, 140.9, 135.7, 129.3, 128.8, 127.4, 127.1, 115.4, 67.5, 52.9, 51.7, 51.4, 42.6, 39.4, 9.2.

17

2-([1,1’-Biphenyl]-4-yloxy)-N-[3-(dimethylamino)-propyl]acetamide hydrochloride (49a). Starting from 44b and following the general procedure for amide coupling, 2-([1,1’-biphenyl]-4-yloxy)-N-[3-(dimethylamino)-propyl]acetamide was prepared as a colorless solid (44 mg, 67% yield). This material was converted to the HCl salt using the general procedure to give 49a as a white solid. 1H NMR (400 MHz, CDCl3) δ 12.20 (br s, 1H), 7.69 (br s, 1H), 7.59-7.47 (m, 4H), 7.44-7.36 (m, 2H), 7.30 (t, J = 7.3 Hz, 1H), 7.06 (d, J = 8.5 Hz, 2H), 4.54 (s, 2H), 3.59-3.48 (m, 2H), 3.06-2.92 (m, 2H), 2.74 (br d, J = 3.6 Hz, 6H), 2.20-2.06 (m, 2H); 13C NMR (100 MHz, CDCl3) δ 169.3, 156.9, 140.4, 135.1, 128.9, 128.4, 127.1, 126.8, 115.2, 67.3, 55.2, 43.0, 35.8, 24.7.

2-(4-Biphenylyloxy)-N-[3-(1-piperidinyl)propyl]-acetamide hydrochloride (49b). Starting from 44b and following the general procedure for amide coupling and purification by column chromatography (CHCl3:MeOH 9:1), 2-(4-biphenylyloxy)-N-[3-(1-piperidinyl)propyl]-acetamide was prepared as an orange solid (15 mg, 20% yield). The material was converted to the HCl salt using the general procedure to give 49b as a white solid. 1H NMR (400 MHz, CDCl3) δ 11.76 (br s, 1H), 7.74 (br s, 1H), 7.61-7.46 (m, 4H), 7.46-7.35 (m, 2H), 7.35-7.26 (m, 1H), 7.15-6.98 (m, 2H), 4.54 (s, 2H), 3.60-3.36 (m, 4H), 3.00-2.79 (m, 2H), 2.62-2.43 (m, 2H), 2.34-2.08 (m, 4H), 1.91-1.72 (m, 3H), 1.45-1.26 (m, 1H); 13C NMR ( 100 MHz, CDCl3) δ 169.2, 156.9, 140.5, 135.1, 128.9, 128.4, 127.0, 126.8, 115.3, 67.3, 54.5, 53.5, 36.0, 23.8, 22.6, 22.2.

2-([1,1’Biphenyl]-4-yloxy)-N-(3-(4-methylpiperazin-1-yl)propyl)acetamide dihydrochloride (49c). Starting from 44b and following the general procedure for amide coupling, 2-([1,1’biphenyl]-4-yloxy)-N-(3-(4-methylpiperazin-1-yl)propyl)acetamide was prepared as a yellow solid (330 mg, 68% yield). 1H NMR (600 MHz, CDCl3) δ 7.57-7.53 (m, 4H), 7.50 (br s, 1H), 7.43 (t, J = 7.6 Hz, 2H), 7.32 (t, J = 7.4 Hz, 1H), 7.02 (d, J = 8.8 Hz, 2H), 4.53 (s, 2H), 3.46 (q, J = 6.1 Hz, 2H), 2.69-2.26 (m, 6H), 2.43 (t, J = 6.4 Hz, 2H), 2.24 (s, 3H), 1.73 (quint, J = 6.4 Hz, 2H), 1.69-1.55 (m, 2H); 13C NMR (150 MHz, CDCl3) δ 168.3, 157.1, 140.5, 135.4, 128.9, 128.5, 127.2, 126.9, 115.5, 68.2, 57.4, 55.1, 53.5, 46.1, 38.9, 25.7. Following the general procedure a small amount of the HCl salt 49c was prepared for biological evaluations. 1H NMR (400 MHz, CDCl3:CD3OD 1:1) δ 7.60-7.50 (m, 4H), 7.39 (t, J = 7.6, 2H), 7.28 (t, J = 7.4 Hz, 1H), 7.06 (d, J = 8.7 Hz, 2H), 4.57 (s, 2H), 3.86-3.47 (m, 8H), 3.43 (t, J = 6.4 Hz, 2H), 3.26-3.17 (m, 2H), 2.96 (s, 3H), 2.10-2.00 (m, 2H); 13C NMR (100 MHz, CDCl3:CD3OD 1:1) δ 170.5, 157.9, 141.0, 135.8, 129.4, 128.9, 127.6, 127.2, 115.9, 68.0, 55.4, 53.6, 50.8, 43.8, 37.5, 26.1.

18

2-([1,1'-Biphenyl]-4-yloxy)-N-(3-(4-ethylpiperazin-1-yl)propyl)acetamide hydrochloride (49d). Sodium 2-([1,1'-biphenyl]-4-yloxy)acetate (44b) (40 mg, 0.16 mmol) was suspended in 2 ml of CH2Cl2. Oxalyl chloride (28 μl, 0.32 mmol) was added dropwise, followed by one drop of DMF, while stirring at rt. The reaction mixture was stirred at rt for 1 h, after which it was concentrated and co-concentrated twice from CH2Cl2. The residue was dissolved in 2 ml of CH2Cl2, and NaHCO3 (60 mg, 0.72 mmol) and 3-(4-ethylpiperazin-1-yl)propan-1-amine (90 mg, 0.18 mmol) dissolved in 0.5 ml H2O was added. Stirred at rt overnight. The reaction mixture was diluted with EtOAc, washed with NaHCO3 (aq., sat.), and the EtOAc phase was extracted three times with 1 M HCl. The aqueous layer was made basic by the addition of NaOH (s) and extracted three times with EtOAc. The combined organic layers were washed with brine, dried over Na2SO4, filtered and concentrated under reduced pressure. The solid was dissolved in 1 ml CH2Cl2 and treated with 1M HCl in Et2O at r.t. for 10 minutes. The mixture was concentrated and then co-concentrated from CH2Cl2 twice. The resulting solid was dissolved in MeOH followed by precipitation by the addition of Et2O and careful removal of the solvent to give compound 49d (8 mg, 11% yield). 1H NMR (600 MHz, CD3OD) δ 7.61-7.56 (m, 4H), 7.42-7.40 (m, 2H), 7.31-7.28 (m, 1H), 7.11-7.09 (m, 2H), 4.61 (s, 2H), 3.85 (bs, 4H), 3.50 (bs, 4H), 3.45 (t, J = 6.5 Hz, 2H), 3.34 (q, J = 7.2 Hz, 2H), 3.28 (t, J = 7.4 Hz, 2H), 2.07 (dt, J = 15.0, 6.8 Hz, 2H), 1.40 (t, J = 7.3 Hz. 3H); 13C NMR (100 MHz, CD3OD) δ 172.1, 158.6, 141.7, 136.2, 129.9, 129.2, 128.0, 127.6, 116.2, 68.2, 55.8, 53.8, 50.0, 49.6, 36.8, 25.4, 9.5.

2-(4-Iodophenoxy)-N-[3-(1-piperidinyl)propyl]acetamide. Starting from 50 and following the general procedure for amide coupling, 2-(4-iodophenoxy)-N-[3-(1-piperidinyl)propyl]acetamide was prepared as an orange solid (105 mg, 76% yield). 1H NMR (400 MHz, CDCl3): δ 7.77 (br s, 1H), 7.58 (dt, J1 = 9.0 Hz, J2 = 2.7 Hz, 2H), 6.70 (dt, J1 = 9.0 Hz, J2 = 2.7 Hz, 2H), 4.44 (s, 2H), 3.47-3.38 (m, 2H), 2.41-2.26 (m, 6H), 1.72-1.64 (m, 2H), 1.58-1.50 (m, 4H), 1.45-1.36 (m, 2H); 13C NMR (100 MHz, CDCl3) δ 167.8, 157.6, 138.7, 117.3, 84.5, 68.0, 58.4, 54.9, 39.4, 25.9, 25.4, 24.4.

2-(4-Iodophenoxy)-N-[3-(1-piperidinyl)propyl]acetamide hydrochloride (51a). 2-(4-Iodophenoxy)-N-[3-(1-piperidinyl)propyl]acetamide (71 mg, 0.18 mmol) was dissolved in 1 ml of CH2Cl2 and treated with ½ sat HCl in CH2Cl2 (3.5 ml) at rt while stirring. Stirred at rt for 10 min before being concentrated under reduced pressure. The resulting oil was co-concentrated twice from CH2Cl2 to remove excess HCl and the crude was precipitated from a mixture of MeOH and Et2O to give 51a (66 mg, 83% yield). 1H NMR (400 MHz, CDCl3) δ 11.67 (br s, 1H), 7.87 (br t, J = 5.5 Hz, 1H), 7.55 (dt, J1 = 8.9 Hz, J2 = 2.5 Hz, 2H), 6.79 (dt, J1

19

= 8.9 Hz, J2 = 2.5 Hz, 2H), 4.45 (s, 2H), 3.51-3.36 (m, 4H), 2.88 (t, J = 7.0 Hz, 2H), 2.64-2.45 (m, 2H), 2.31-2.06 (m, 4H), 1.94-1.74 (m, 3H), 1.49-1.30 (m, 1H); 13C NMR (100 MHz, CDCl3) δ 168.8, 157.3, 138.5, 117.4, 84.3, 67.1, 54.3, 53.4, 35.9, 23.7, 22.6, 22.2.

2-(4-Iodophenoxy)-N-(3-(4-methylpiperazin-1-yl)-propyl)acetamide. Starting from 50 and following the general procedure for amide coupling, 2-(4-iodophenoxy)-N-(3-(4-methylpiperazin-1-yl)-propyl)acetamide was prepared as an orange oil (385 mg, 85% yield). 1H NMR (600 MHz, CDCl3) δ 7.60 (d, J = 8.9 Hz, 2H), 7.45 (br s, 1H), 6.73 (d, J = 8.9 Hz, 2H), 4.45 (s, 2H), 3.44 (q, J = 6.1 Hz, 2H), 2.67-2.20 (m, 8H), 2.42 (t, J = 6.4 Hz, 2H), 2.25 (s, 3H), 1.71 (quint, J = 6.4 Hz, 2H); 13C NMR (150 MHz, CDCl3) δ 167.8, 157.5, 138.7, 117.5, 84.6, 68.1, 57.4, 55.1, 53.5, 46.1, 39.0, 25.6.

2-(4-Iodophenoxy)-N-(3-(4-methylpiperazin-1-yl)-propyl)acetamide dihydrochloride (51b). 2-(4-Iodophenoxy)-N-(3-(4-methylpiperazin-1-yl)-propyl)acetamide (221 mg, 0.53 mmol) was dissolved in 2 ml of CH2Cl2 and treated with ½ sat HCl in CH2Cl2 (10 ml) at rt while stirring. Stirred at rt for 10 min before being concentrated under reduced pressure. The resulting crude oil was dissolved in MeOH (1 ml) and treated again with ½ sat HCl in CH2Cl2 (8 ml) at rt while stirring for 10 min before being concentrated under reduced pressure and co-concentrated twice from CH2Cl2 to remove excess HCl to give 51b (quantitative yield) as a light brown foam. 1H NMR (600 MHz, CDCl3) δ 7.58 (d, J = 9.0 Hz, 2H), 6.76 (d, J = 9.0 Hz, 2H), 4.46 (s, 2H), 3.77-3.64 (m, 6H), 3.64-3.52 (m, 2H), 3.40 (t, J = 6.5 Hz, 2H), 3.21-3.12 (m, 2H), 2.94 (s, 3H), 2.08-1.99 (m, 2H); 13C NMR (150 MHz, CDCl3) δ 170.3, 157.9, 139.0, 117.6, 84.6, 67.5, 55.0, 50.6, 43.1, 36.3, 24.4.

N-(3-(4-ethylpiperazin-1-yl)propyl)-2-(4-iodophenoxy)acetamide hydrochloride (51c). 2-(4-Iodophenoxy)acetic acid (50) (100 mg, 0.36 mmol) was suspended in 2 ml of CH2Cl2. Oxalyl chloride (62 μl, 0.71 mmol) was added dropwise, followed by one drop of DMF, while stirring at rt. The reaction mixture was stirred at rt for 1 h, after which it was concentrated and co-concentrated twice from CH2Cl2. The residue was dissolved in 2 ml of CH2Cl2, and NaHCO3 (136 mg, 1.62 mmol) and 3-(4-ethylpiperazin-1-yl)propan-1-amine (203 mg, 0.40 mmol) dissolved in 0.5 ml H2O was added. Stirred at r.t. overnight. The reaction mixture was diluted with EtOAc, washed with NaHCO3 (aq., sat.), and the EtOAc phase was extracted three times with 1 M HCl. The aqueous layer was made basic by the addition of NaOH (s) and extracted three times with EtOAc. The combined organic layers were washed with brine, dried over Na2SO4, filtered and concentrated

20

under reduced pressure. The solid was dissolved in 1 ml CH2Cl2 and treated with 1M HCl in Et2O at r.t. for 10 minutes. The mixture was concentrated and then co-concentrated from CH2Cl2 twice. The resulting solid was dissolved in MeOH followed by precipitation by the addition of Et2O and careful removal of the solvent to give compound 51c (17 mg, 9% yield). 1H NMR (600 MHz, CD3OD) δ 7.64-7.61 (m, 2H), 6.86-6.83 (m, 2H), 4.55 (s, 2H), 3.89 (bs, 4H), 3.55 (bs, 4H), 3.43 (t, J = 6.5 Hz, 2H), 3.36 (q, J = 7.0 Hz, 2H), 3.29 (t, J = 8.0 Hz, 2H), 2.06 (dt, J = 15.0, 6.9 Hz, 2H), 1.42 (t, J = 7.3 Hz, 3H); 13C NMR (100 MHz, CD3OD) δ 171.6, 159.0, 139.6, 118.4, 84.7, 68.1, 55.7, 53.4, 50.0, 49.6, 36.8, 25.3, 9.5.

2-((4'-Methoxy-[1,1'-biphenyl]-4-yl)oxy)-N-(3-(piperidin-1-yl)propyl)acetamide. 2-(4-Iodophenoxy)-N-(3-(piperidin-1-yl)propyl)acetamide hydrochloride (51a) (30 mg, 0.07 mmol) was dissolved in 3 ml of DMF. Na2CO3 (1.5 ml, 2 M) was added, followed by 4-methoxyphenylboronic acid (23 mg, 0.15 mmol) and PEPPSI-iPr (2.5 mg, 0.0004 mmol). N2 was bubbled through the mixture while stirring at rt for 10 min. The reaction mixture was heated by MWI to 110 °C for 10 min. The reaction mixture was poured into half-sat. brine and extracted three times with EtOAc. The combined organic layers were washed with brine, and then extracted twice with 2 M HCl. The combined aqueous layers were made basic by the addition of NaOH (s) and extracted three times with EtOAc. The combined organic layers were washed with brine, dried over Na2SO4, filtered and concentrated. Precipitation from CH2Cl2/heptane gave 2-((4'-Methoxy-[1,1'-biphenyl]-4-yl)oxy)-N-(3-(piperidin-1-yl)propyl)acetamide as a white solid (28 mg, 98% yield). 1H NMR (400 MHz, CDCl3) δ 7.76 (br s, 1H), 7.50-7.46 (m, 4H), 6.98-6.95 (m, 4H), 4.52 (s, 2H), 3.84 (s, 3H), 3.45 (q, J = 5.8 Hz, 2H), 2.37 (t, J = 6.4 Hz, 2H), 2.40-2.28 (m, 4H), 1.71 (quint, J = 6.3 Hz, 2H), 1.60-1.52 (m, 4H), 1.46-1.37 (m, 2H); 13C NMR (100 MHz, CDCl3) δ 168.3, 159.0, 156.7, 134.9, 133.2, 128.1, 127.9, 115.2, 114.3, 68.1, 58.3, 55.5, 54.9, 39.2, 25.9, 25.6, 24.5.

2-((4'-Methoxy-[1,1'-biphenyl]-4-yl)oxy)-N-(3-(piperidin-1-yl)propyl)acetamide hydrochloride (52b). 2-((4'-Methoxy-[1,1'-biphenyl]-4-yl)oxy)-N-(3-(piperidin-1-yl)propyl)acetamide (21 mg, 0.05 mmol) was dissolved in 1 ml of CH2Cl2, and 2 ml of 1 M HCl in Et2O was added dropwise at rt while stirring. The vessel contents were stirred for 15 minutes, then concentrated and co-concentrated twice from EtOAc. The resulting white solid was triturated with hot EtOAc, then dissolved in MeOH and precipitated by the addition of Et2O. The precipitate was filtered off and washed with Et2O to give 52b (13 mg, 57% yield). 1H NMR (400 MHz, CD3OD) δ 7.55 (d, J = 8.8 Hz, 2H), 7.50 (d, J = 8.8 Hz, 2H), 7.06 (d, J = 8.8 Hz, 2H), 6.97 (d, J = 8.8 Hz, 2H), 4.60 (s, 2H), 3.82 (s, 3H), 3.49-3.42 (m,

21

2H), 3.42-3.37(m, 2H), 3.03-2.96 (m, 2H), 2.87-2.77 (m, 2H), 2.01-1.85 (m, 4H), 1.85-1.65 (m, 3H), 1.50-1.36 (m, 1H); 13C NMR (100 MHz, CD3OD) δ 172.2, 160.5, 158.1, 136.0, 134.1, 128.7, 128.6, 116.2, 115.3, 68.2, 55.74, 55.71, 54.3, 49.8, 36.8, 25.4, 24.3, 22.6.

2-((4'-Methoxy-[1,1'-biphenyl]-4-yl)oxy)-N-(3-(4-methylpiperazin-1-yl)propyl)acetamide. 2-(4-Iodophenoxy)-N-(3-(4-methylpiperazin-1-yl)propyl)acetamide dihydrochloride (51b) (20 mg, 0.04 mmol) was dissolved in 1.6 ml of DMF. Na2CO3 (0.8 ml, 2 M) was added followed by 4-methoxyphenylboronic acid (12 mg, 0.08 mmol) and PEPPSI-iPr (1.4 mg, 0.002 mmol). N2 was bubbled through the mixture while stirring at rt for 10 min. The reaction mixture was heated by MWI to 110 °C for 10 min, and then poured into half-sat. brine and extracted three times with EtOAc. The combined organic layers were washed with brine, and then extracted three times with 2 M HCl. The combined aqueous layers were made basic by addition of NaOH (s) and extracted three times with EtOAc. The combined organic layers were washed with brine, dried over Na2SO4, filtered and concentrated. Precipitation from CH2Cl2/heptane gave 2-((4'-methoxy-[1,1'-biphenyl]-4-yl)oxy)-N-(3-(4-methylpiperazin-1-yl)propyl)acetamide as a white solid (14 mg, 86% yield). 1H NMR (400 MHz, CDCl3) δ 7.50-7.45 (m, 5H), 7.00-6.95 (m, 4H), 4.51 (s, 2H), 3.84 (s, 3H), 3.46 (q, J = 6.0 Hz, 2H), 2.68-2.34 (m, 8H), 2.44 (t, J = 6.5 Hz, 2H), 2.28 (s, 3H), 1.73 (quint, J = 6.4 Hz, 2H); 13C NMR (100 MHz, CDCl3) δ 168.4, 159.1, 156.6, 135.1, 133.0, 128.1, 127.9, 115.5, 114.4, 68.2, 57.1, 55.5, 54.8, 53.1, 45.8, 38.7, 25.6.

2-((4'-Methoxy-[1,1'-biphenyl]-4-yl)oxy)-N-(3-(4-methylpiperazin-1-yl)propyl)acetamide dihydrochloride (52b). 2-((4'-Methoxy-[1,1'-biphenyl]-4-yl)oxy)-N-(3-(4-methylpiperazin-1-yl)propyl)acetamide (11 mg, 0.03 mmol) was dissolved in 0.5 ml CH2Cl2, and 1M HCl in Et2O (1 ml) was added dropwise at rt while stirring. After 15 min, the suspension was concentrated, then co-concentrated from chloroform three times to give the dihydrochloride salt 52b (11 mg, 85% yield). 1H NMR (400 MHz, CDCl3: CD3OD 1:1) δ 7.49 (d, J = 8.7 Hz, 2H), 7.46 (d, J = 8.8 Hz, 2H), 7.01 (d, J = 8.7 Hz, 2H), 6.94 (d, J = 8.8 Hz, 2H), 4.53 (s, 2H), 3.81 (s, 3H), 3.76-3.54 (m, 8H), 3.47-3.32 (m, 2H), 3.23-3.15 (m, 2H), 2.94 (s, 3H), 2.11-2.01 (m, 2H); 13C NMR (100 MHz, CDCl3:CD3OD 1:1) δ 170.9, 159.5, 157.1, 135.3, 133.5, 128.3, 128.2, 115.6, 114.8, 67.7, 55.6, 55.1, 50.5, 43.1, 36.3, 24.4.

Methyl 4-iodophenoxyacetate (53). 4-Iodophenoxyacetic acid 50 (416 mg, 1.50 mmol) was dissolved in 3 ml of MeOH, and 3 drops of H2SO4 (conc.) was added while stirring at rt. The reaction vessel was sealed, heated to 85 °C, and stirred for 24 h. After cooling, the reaction mixture was poured into NaHCO3 (aq., sat.) and extracted three times with Et2O. The combined

22

organic layers were washed with brine, dried over Na2SO4, and filtered. The resulting light yellow solid was triturated with heptane five times to give compound 53 as white needles (320 mg, 73% yield). 1H NMR (400 MHz, CDCl3) δ 7.57 (d, J = 9.0 Hz, 2H), 6.69 (d, J = 9.0 Hz, 2H), 4.60 (s, 2H), 3.80 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 169.1, 157.8, 138.5, 117.1, 84.3, 65.4, 52.5.

Methyl 2-((4'-methoxy-[1,1'-biphenyl]-4-yl)oxy)acetate. The procedure was adapted from a previously developed method.2-3 Methyl 4-iodophenoxyacetate (53) (152 mg, 0.52 mmol), 4-methoxybenzeneboronic acid (158 mg, 1.04 mmol), KF (57 mg, 0.99 mmol), and PEPPSI-iPr (3.5 mg, 0.005 mmol) were dissolved in 2 ml of MeOH. N2 was bubbled through the mixture while stirring at rt for 10 min. The vessel contents were heated by MWI to 110 °C for 10 min. The reaction mixture was diluted with EtOAc and brine and extracted three times with EtOAc. The combined organic layers were washed with brine, dried with Na2SO4, filtered and concentrated. The crude material was purified with flash column chromatography (Heptane:EtOAc, 7:3) to yield methyl 2-((4'-methoxy-[1,1'-biphenyl]-4-yl)oxy)acetate as a white solid (127 mg, 90% yield). 1H NMR (600 MHz, CDCl3) δ 7.49-7.46 (m, 4H), 6.97-6-95 (m, 4H), 4.67 (s, 2H), 3.84 (s, 3H), 3.83 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 169.6, 159.0, 157.0, 134.8, 133.3, 128.0, 127.9, 115.0, 114.3, 65.6, 55.5, 52.4.

2-((4'-methoxy-[1,1'-biphenyl]-4-yl)oxy)acetic acid (54). Methyl 2-((4'-methoxy-[1,1'-biphenyl]-4-yl)oxy)acetate (117 mg, 0.43 mmol) and LiOH (0.65 ml, 1M, 0.65 mmol) were suspended in THF and allowed to stir at r.t. overnight. The mixture was diluted with EtOAc and 1 M HCl. The white precipitate that was formed was filtered off and dried under vacuum to yield compound 54 as a white solid (110 mg, 99% yield). 1H NMR (600 MHz, CD3OD) δ 7.49.7.47 (m, 4H), 6.99-6.95 (m, 4H), 4.68 (s, 2H), 3.81 (s, 3H); 13C NMR (100 MHz, CD3OD) δ 172.7, 160.3, 158.5, 135.6, 134.5, 128.6, 128.6, 116.0, 115.2, 66.0, 55.7.

1-(4-(4-Ethylpiperazin-1-yl)piperidin-1-yl)-2-((4'-methoxy-[1,1'-biphenyl]-4-yl)oxy)ethanone dihydrochloride (55). 2-((4'-methoxy-[1,1'-biphenyl]-4-yl)oxy)acetic acid (54) (40 mg, 0.15 mmol), 1-ethyl-4-(4-piperidinyl) piperazine (31 mg, 0.15 mmol), and TBTU (60 mg, 0.19 mmol) were suspended in DMF (1 ml) followed by the addition of TEA (63 µl, 0.45 mmol). The reaction mixture was briefly heated with a heat gun and thereafter allowed to stir at r.t. for 3 days. NaHCO3 (sat.) was added to the mixture which resulted in a pale yellow precipitate. The precipitate was collected and dissolved in EtOAc/H2O. The water phase was extracted three times with EtOAc. The combined organic layers were washed with brine

23

three times, dried over Na2SO4, filtered and concentrated. The solid was dissolved in 1 ml CH2Cl2 and treated with 1M HCl in Et2O at r.t. for 20 minutes. The mixture was concentrated and then co-concentrated from CH2Cl2 twice. The resulting solid was dissolved in MeOH followed by precipitation by the addition of Et2O and careful removal of the solvent to give compound 55 (10 mg, 13% yield). 1H NMR (600 MHz, CD2Cl2:CD3OD 1:1) δ 7.45-7.42 (m, 4H), 6.96-6.89 (m, 4H), 4.78-4.73 (m, 2H), 4.70-4.66 (m, 1H), 4.14-4.12 (m, 1H), 3.76 (s, 3H), 3.66 (bs, 4H), 3.60 (bs, 4H), 3.51-3.47 (m, 1H), 3.22 (q, J = 7.1 Hz, 2H), 3.14-3.10 (m, 1H), 2.68-2.64 (m, 1H), 2.22-2.18 (m, 2H), 1.78-1.74 (m, 1H), 1.66-1.62 (m, 1H), 1.35 (t, J = 7.1 Hz, 3H); 13C NMR (100 MHz, CD2Cl2:CD3OD 1:1) δ 168.8, 159.6, 157.6, 135.1, 133.6, 128.2, 128.1, 115.5, 114.7, 67.5, 64.1, 55.5, 52.7, 49.4, 46.2, 43.9 and 41.0 (1C), 27.3 and 26.8 (1C), 9.2.

N-(3-(4-ethylpiperazin-1-yl)propyl)-2-((4'-methoxy-[1,1'-biphenyl]-4-yl)oxy)acetamide hydrochloride (56). 2-((4'-methoxy-[1,1'-biphenyl]-4-yl)oxy)acetic acid (54) (30 mg, 0.12 mmol) was suspended in 2 ml of THF. Oxalyl chloride (20 μl, 0.23 mmol) was added dropwise, followed by one drop of DMF, while stirring at rt. The reaction mixture was stirred at rt for 1 h, after which it was concentrated and co-concentrated twice from CH2Cl2. The residue was dissolved in 2 ml of CH2Cl2, and Na2CO3 (98 mg, 0.92 mmol) and 3-(4-ethylpiperazin-1-yl)propan-1-amine (119 mg, 0.23 mmol) dissolved in 0.5 ml H2O was added. Stirred at rt for 3 days. The reaction mixture was diluted with EtOAc, washed three times with NaHCO3 (aq., sat.), and once with brine. The organic phase was dried over Na2SO4, filtered and concentrated under reduced pressure. The solid was dissolved in 1 ml CH2Cl2 and treated with 1M HCl in Et2O at r.t. for 20 minutes. The mixture was concentrated and then co-concentrated from CH2Cl2 twice. The resulting solid was dissolved in MeOH followed by precipitation by the addition of Et2O and careful removal of the solvent to give compound 56 (15 mg, 27% yield). 1H NMR (400 MHz, CD2Cl2:CD3OD 1:1) δ 7.49-7.43 (m, 4H), 7.00-6.90 (m, 4H), 5.50 (s, 2H), 3.77 (s, 3H), 3.67 (bs, 4H), 3.60 (bs, 4H), 3.39 (t, J = 7.0 Hz, 2H), 3.22 (q, J = 7.4 Hz, 2H), 3.16 (t, J = 7.9 Hz, 2H), 2.02 (dt, J = 15.1, 7.0 Hz, 2H), 1.36 (t, J = 7.2 Hz. 3H); 13C NMR (100 MHz, CD2Cl2:CD3OD 1:1) δ 170.8, 159.6, 157.2, 135.2, 133.5, 128.3, 128.1, 115.6, 114.7, 67.7, 55.6, 55.2, 52.8, 49.06, 48.95, 36.3, 24.4, 9.1.

IC50 determinations The IC50 values for AgAChE1, AaAChE1 and hAChE were determined according to the following procedure. Freshly prepared stock solutions of the compounds were prepared from solid material in DMSO at a concentration of 100 mM. Working dilutions thereof were prepared in either 0.1 M sodium phosphate buffer (pH 7.4) or MilliQ water, depending on the solubility of the

24

compounds. Compound solutions of eight different concentrations up to a maximum of 1 mM were used. The activity measurements were performed using secreted non-purified proteins in growth medium, and enzymatic activity was measured using the Ellman assay4 adapted to a 96-well format. Liquid handling was performed manually or using a QIAgility robotic benchtop instrument (Qiagen). Assays were performed at 30 °C in a final assay volume of 200 µl of 0.1 M phosphate buffer (pH 7.4) containing 0.2 mM 5,5′-dithiobis(2-nitrobenzoic acid) and 1 mM acetylthiocholine iodide. The enzymatic reaction was measured by monitoring changes in the absorbance of individual wells at 412 nM over 60 s in a FlexStation 3 Multi-Mode Microplate Reader (Molecular Devices) or the PowerWave HT Microplate Spectrophotometer (BioTek). The average slope determined for eight positive (uninhibited) controls on each plate was taken to represent 100% activity and the activity observed in the sample wells were quantified in relation to this value. IC50 values were calculated using non-linear regression (curve fitting) in GraphPad Prism5 and the log [inhibitor] vs. response variable slope equation was fitted using four parameters.

Generation, collection and refinement of crystal structures The crystallization of mAChE was performed as previously described.6 Small amounts of 49b or 49c were added to a soaking solution consisting of 30% (v/v) polyethylene glycol 750 monomethylether in 100 mM HEPES buffer, pH 7.0 until saturation was reached. The soaking solution was then added to a crystal of mAChE. Soaking was performed over approximately five minutes and the crystals were incubated for an additional few minutes prior to flash-freezing in liquid nitrogen. X-ray diffraction data were collected at the MAX-lab synchrotron (Lund, Sweden) using beam line I911-3 equipped with MAR Research CCD detectors. Images were collected using an oscillation angle of 1.0° per exposure. The intensity data were indexed and integrated using XDS7 and scaled using Scala.8 The structures of 49b•mAChE and 49c•mAChE were determined using rigid-body refinement starting with a modified apo structure of mAChE (pdb code: 1J069). The presence of the ligand in the binding site of the mAChE crystals was estimated based on the initial 2|Fo|−|Fc| and |Fo|−|Fc| omit maps, and only protein-ligand complexes where ligand occupancies were detected were subjected to further refinement. Further crystallographic refinement, as well as evaluation of the final model, was performed using the Phenix software suite10 (Table A1). Several rounds of refinement were performed, alternating with manual rebuilding of the model after visualizing the 2|Fo|−|Fc| and |Fo|−|Fc| electron density maps using COOT.11

25

Table A1. Data collection and refinement statistics.

49b•mAChEa 49c•mAChEa

Resolution range (Å) 39.58 - 2.7 (2.796 - 2.7) 46.1 - 2.5 (2.589 - 2.5) Space group P 21 21 21 P 21 21 21 Unit cell (Å) 78.0 x 110.3 x 227.3 78.9 x 111.6 x 227.3 Total reflections 294665 (29226) 452566 (45150) Unique reflections 52051 (5211) 70281 (6942) Multiplicity 5.7 (5.6) 6.4 (6.5) Completeness (%) 95.12 (97.15) 99.98 (100.00) Mean I/sigma(I) 18.33 (4.09) 18.17 (4.48) Wilson B-factor 47.97 44.78 R-merge 0.08195 (0.5547) 0.07405 (0.476) R-meas 0.08967 0.08079 CC1/2 0.998 (0.96) 0.998 (0.974) CC* 1 (0.99) 1 (0.993) R-work 0.1819 (0.2869) 0.1812 (0.2338) R-free 0.2132 (0.3216) 0.2280 (0.2524) No. of non-hydrogen atoms: 8524 8536 macromolecules 8336 8336 ligands 52 53 water 136 147 Protein residues 1068 1068 RMS(bonds) 0.005 0.012 RMS(angles) 0.89 1.35 Ramachandran favored (%) 95 95 Ramachandran outliers (%) 0.28 0.75 Clashscore 11.88 10.91 Average B-factor: 54.10 54.60 macromolecules 53.80 54.40 ligands 111.20 92.80 solvent 50.50 52.20

aStatistics for the highest-resolution shell are shown in parentheses.

Modelling of 34 and 49c in AaAChE1 Based on the binding mode in mAChE, the ligand 49c was manually mutated into the active site of the homology model of AaAChE1.1 In Maestro,12 the complex was prepared by adding hydrogens, assigning bond orders and creating disulfide bonds. A constrained energy minimization was performed of the complex using the Polak-Ribiere Conjugate Gradient method13 with a gradient convergence threshold of 0.05. The ligand was allowed to move freely while the positions of all protein heavy atoms were only allowed to move +/- 0.2 Å from their initial positions before being subjected to a resistive force of 100 kJ/mol Å2.13The constrained energy minimization was made using Macromodel14 and the OPLS2005 force field15

26

with water as the solvent (using a constant electrostatic treatment with a dielectric constant of 1) using the Generalized-Born/Surface-Area model. Charges were calculated using the OPLS2005 force field and non-bonded interactions were considered within distance of 8.0, 12.0, and 4.0 Å for van der Waals, electrostatic, and hydrogen bond interactions, respectively.

Conformations of 34 aligned on the modeled pose of 49c in AaAChE1 were generated using the flexible alignment tool implemented in MOE.16-17 Partial charges were calculated using the MMFF94x force field in MOE16 and stochastically generated conformations of 34 were aligned to 49c. The alignments were evaluated both in terms of the internal energy of the ligand, and the similarity of the configuration based on molecular features such as hydrogen bond acceptors and donors, size, and hydrophobic moieties. This yielded 12 conformation of 34, which were visually inspected and the conformation with the smallest average strain of the molecules was selected for further modeling. The selected conformation was mutated into the active site of the AaAChE1 homology model and the resulting complex was prepared and subjected to a constrained energy minimization as described above for 49c with the difference that all heavy atoms in the complex were only allowed to move +/- 0.2 Å from their initial positions before being subjected to a resistive force of 100 kJ/mol Å2.

References 1. Engdahl, C.; Knutsson, S.; Ekström, F.; Linusson, A., Discovery of Selective Inhibitors Targeting Acetylcholinesterase 1 from Disease-Transmitting Mosquitoes. (Submitted) 2016. 2. Sellstedt, M.; Almqvist, F., Synthesis of a Novel Tricyclic Peptidomimetic Scaffold. Org. Lett. 2008, 10, 4005-4007. 3. Chorell, E.; Pinkner, J. S.; Phan, G.; Edvinsson, S.; Buelens, F.; Remaut, H.; Waksman, G.; Hultgren, S. J.; Almqvist, F., Design and Synthesis of C-2 Substituted Thiazolo and Dihydrothiazolo Ring-Fused 2-Pyridones: Pilicides with Increased Antivirulence Activity. J. Med. Chem. 2010, 53, 5690-5695. 4. Ellman, G. L.; Courtney, K. D.; Andres, V.; Featherstone, R. M., A New and Rapid Colorimetric Determination of Acetylcholinesterase Activity. Biochem. Pharmacol. 1961, 7, 88-95. 5. Graphpad Prism Version 6.04 for Windows, Graphpad Software, La Jola, Ca, USA, www.graphpad.com. 6. Ekström, F.; Akfur, C.; Tunemalm, A. K.; Lundberg, S., Structural Changes of Phenylalanine 338 and Histidine 447 Revealed by the Crystal Structures of Tabun-Inhibited Murine Acetylcholinesterase. Biochemistry 2006, 45, 74-81. 7. Kabsch, W., Automatic Processing of Rotation Diffraction Data from Crystals of Initially Unknown Symmetry and Cell Constants. J. Appl. Crystallogr. 1993, 26, 795-800.

27

8. Murshudov, G. N.; Vagin, A. A.; Dodson, E. J., Refinement of Macromolecular Structures by the Maximum-Likelihood Method. Acta Crystallogr., Sect. D: Biol. Crystallogr. 1997, 53, 240-255. 9. Bourne, Y.; Taylor, P.; Radic, Z.; Marchot, P., Structural Insights into Ligand Interactions at the Acetylcholinesterase Peripheral Anionic Site. EMBO J. 2003, 22, 1-12. 10. Adams, P. D.; Grosse-Kunstleve, R. W.; Hung, L. W.; Ioerger, T. R.; McCoy, A. J.; Moriarty, N. W.; Read, R. J.; Sacchettini, J. C.; Sauter, N. K.; Terwilliger, T. C., Phenix: Building New Software for Automated Crystallographic Structure Determination. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2002, 58, 1948-1954. 11. Emsley, P.; Cowtan, K., Coot: Model-Building Tools for Molecular Graphics. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2004, 60, 2126-2132. 12. Maestro, Version 10.3, Schrödinger, Llc, New York, NY, USA. 13. Polak, E.; Ribiere, G., Note on the Convergence of Methods of Conjugate Directions. Revue Francaise d'Informatique et de Recherche Operationnelle 1969, 3, 35-43. 14. Macromodel, Version 10.9, Schrödinger, Llc, New York, NY, USA. 15. Kaminski, G. A.; Friesner, R. A.; Tirado-Rives, J.; Jorgensen, W. L., Evaluation and Reparametrization of the Opls-Aa Force Field for Proteins Via Comparison with Accurate Quantum Chemical Calculations on Peptides. J. Phys. Chem. B 2001, 105, 6474-6487. 16. Molecular Operating Environment (MOE), 2014.10; Chemical Computing Group Inc., 1010 Sherbooke St. West, Suite #910, Montreal, QC, Canada, H3A 2R7, 2011. 17. Labute, P.; Williams, C.; Feher, M.; Sourial, E.; Schmidt, J. M., Flexible Alignment of Small Molecules. J. Med. Chem. 2001, 44, 1483-1490.