Venoms-Based Discovery of Novel Modulators of Human Neuronal Α7 and Α3* Nicotinic Acetylcholine Receptors Rosa Esther Prahl Msc

Venoms-based discovery of novel modulators of human neuronal α7 and α3* nicotinic acetylcholine receptors Rosa Esther Prahl MSc. Cell Biology

A thesis submitted for the degree of Master of Philosophy at The University of Queensland in 2016 Institute for Molecular Bioscience

ABSTRACT

Nicotinic acetylcholine receptors (nAChRs) are pentameric ligand-gated channels that contain combinations of α and β subunits (heteromeric) or only five α subunits (homomeric). To date, 12 subunits have been cloned. Although different types of neuronal nAChRs are expected to occur in vivo, most structure-activity relationship studies have been carried out for just a few neuronal subtypes. Clinical and basic research has shown that cholinergic receptors play a role in several disorders of the nervous system such as chronic pain, Alzheimer’s disease and addiction to nicotine, alcohol and drugs. Unfortunately, the lack of selective modulators for each subtype of nAChR makes their pharmacological characterization difficult, which has slowed the development of therapeutic nAChR modulators with high selectivity and absence of off-target side effects.

Animal venoms have proven to be an excellent natural source of bioactive molecules with activity against ion channels. Venoms from different snakes and in particular molluscs of the Conus genus have been an important source of modulators of cholinergic receptors. However, spider and scorpion venoms have been studied less extensively and there are only two examples of cholinergic receptor modulators reported from arachnid venom. Thus, the primary goal of this thesis was to identify novel modulators of the human ganglionic α3* nAChR (where the asterisk indicates that β subunits form part of the heteropentameric receptor) and the homomeric neuronal α7 nAChR using spider and scorpion venoms.

“Hit” venoms from a selection of “modern” and “primitive” spiders were identified via high-throughput screens against the α3* and α7 nAChRs endogenously expressed in SH-SY5Y cells. A total of 71 spider venoms were screened, with representatives from nine different taxonomic families. The resulting data showed that 91% of the venoms presented some activity against the nAChR subtypes tested and most of these were non-selective blockers of the α3* and α7 subtypes. Toxins directed against vertebrate nAChRs likely evolved through frequent interactions between spiders and small vertebrate predators. Our data revealed that nAChR agonists are likely to be present in venom of the “primitive” spiders

and therefore were probably basally recruited in spider venom. In summary, the screening of spider venoms against vertebrate cholinergic receptors provided information about the origin and evolution of toxins against vertebrate receptors.

The search for nAChR modulators was extended to scorpion venoms that are recognized as sources of toxins that affect primarily the properties of voltage-gated sodium and potassium channels. In this work, venoms from Androctonus australis, Androctonus crassicauda, Leirurus quinquestriatus and Grosphus grandidieri (Buthidae) and Heterometus spinnifer (Scorpionidae) were screened against the α3* and α7 nAChRs using SH-SY5Y cells. The venom of Heterometrus spinnifer was found to contain an agonistic compound. Activity guided-fractionation allowed isolation of the “hit” toxin, and it was identified by high-resolution mass spectrometry as senecioylcholine. This non-peptide toxin has not previously been identified in arachnids but it was previously isolated from some marine gastropods (Muricidae). It shows high toxicity against vertebrates. Thus, this non-peptide toxin is likely to function in defence against vertebrate predators.

DECLARATION BY AUTHOR

This thesis is composed of my original work, and contains no material previously published or written by another person except where due reference has been made in the text. I have clearly stated the contribution by others to jointly-authored works that I have included in my thesis.

I have clearly stated the contribution of others to my thesis as a whole, including statistical assistance, survey design, data analysis, significant technical procedures, professional editorial advice, and any other original research work used or reported in my thesis. The content of my thesis is the result of work I have carried out since the commencement of my research higher degree candidature and does not include a substantial part of work that has been submitted to qualify for the award of any other degree or diploma in any university or other tertiary institution. I have clearly stated which parts of my thesis, if any, have been submitted to qualify for another award.

I acknowledge that an electronic copy of my thesis must be lodged with the University Library and, subject to the policy and procedures of The University of Queensland, the thesis be made available for research and study in accordance with the Copyright Act 1968 unless a period of embargo has been approved by the Dean of the Graduate School.

I acknowledge that copyright of all material contained in my thesis resides with the copyright holder(s) of that material. Where appropriate I have obtained copyright permission from the copyright holder to reproduce material in this thesis.

PUBLICATIONS DURING CANDIDATURE

No publications.

PUBLICATIONS INCLUDED IN THIS THESIS

No publications.

CONTRIBUTION BY OTHERS TO THE THESIS

Prof. Richard Lewis and Dr Irina Vetter (Institute for molecular Bioscience) supported during the screening of “hit” toxins of the 71 venoms with their expertise in FLIPR assays.

Chapter 1

Dr Volker Herzig (Institute for Molecular Bioscience) performed HPLC fractionation of crude venom from Macrothele gigas. Dr Irina Vetter (Institute for Molecular Bioscience) performed one FLIPR assay that was mentioned in section 2.3.4.

Chapter 2

HPLC fractionation of crude venom from Heterometrus spinnifer was performed by Dr Niraj Bende (Institute for Molecular Bioscience).

Dr Angela Salim identified the “hit” toxin from Heterometrus spinnifer as senecioylcholine using high-resolution electrospray ionization mass spectrometry. She also performed as well as the co-elution of the native compound with its synthetic form, and two of its synthetic isomers (tigloylcholine and angeloylcholine).

Pratik Neupane carried out the chemical synthesis of senecioylcholine, tigloylcholine and angeloylcholine.

Angela and Pratik were involved in the writing of the protocol for identification and chemical synthesis of senecioylcholine. Angela and Pratik are members of the group of Prof. Robert Capon at The Institute for Molecular Bioscience.

STATEMENT OF PARTS OF THE THESIS SUBMITTED TO QUALIFY FOR THE AWARD OF

ANOTHER DEGREE

None.

ACKNOWLEDGEMENTS

I thank Prof. Glenn King, Prof. Richard Lewis, Prof. Robert Capon, Dr Amanda Carozzi, Dr Irina Vetter, Dr Angela Salim, Pratik Neupane and Sassan Ranhama for their time, efforts and suggestions during the course of my thesis.

This work was funding and supported by the following institutions: Postgraduate studies of The University of Queensland (UQ), the UQ Institute for Molecular Bioscience, and a program grant from the Australian National Health & Medical Research Council.

I dedicate this thesis to the Holy Father, my husband Torben and my son Clouseau.

KEYWORDS spider venom, scorpion venom, ligand gated calcium channel, α7 nicotinic acetylcholine receptors, α3* nicotinic acetylcholine receptors, FLIPR, high throughput screening, choline ester, senecioylcholine

AUSTRALIAN AND NEW ZEALAND STANDARD RESEARCH CLASSIFICATIONS (ANZSRC)

ANZSRC code: 060101, Analytical Biochemistry, 80%

ANZSRC code: 060199, Biochemistry and Cell Biology not elsewhere classified, 20%

FIELDS OF RESEARCH (FOR) CLASSIFICATION

FoR code: 0601, Biochemistry and Cell Biology, 80%

FoR code: 0699, Other Biological Sciences, 20%

TABLE OF CONTENTS Abstract ...... 2

Declaration by author ...... 4 Publications during candidature ...... 5 Publications included in this thesis ...... 5 Contribution by others to the thesis ...... 6 Statement of parts of the thesis submitted to qualify for the award of another degree . 7 Acknowledgements ...... 8 Keywords ...... 9 Australian and New Zealand Standard Research Classifications (ANZSRC) ...... 9 Fields of Research (FoR) Classification ...... 9 Table of Contents ...... 10 List of figures and Tables ...... 12 List of abbreviations ...... 14 CHAPTER 1 — Introduction ...... 16 1.1 Nicotinic acetylcholine receptors ...... 17 1.1.1 Homopentameric/heteropentameric architecture of neuronal nAChRs ...... 17 1.1.2 The use of toxins to elucidate the molecular architecture of nAChRs ...... 19 1.1.3 Location and physiological function of nAChRs in vertebrates and invertebrates 21 1.1.4 Role of nAChRs in human disease ...... 24 1.2 Modulators of nAChRs ...... 26 1.2.1 Modulators of nAChRs as drugs and insecticides ...... 26 1.2.2 Novel subtype-selective modulators of vertebrate nAChRs are required to develop accuracy pharmacological tools and potential therapeutic leads ...... 28 1.3 Venom-derived modulators of nAChRs ...... 32 1.3.1 Important role played by subtype-selective α-conotoxins in study of vertebrate nAChRs ...... 32 1.3.2 nAChR modulators from other venomous animals ...... 34 1.3.3 Rationale for examining spider venoms as a source of novel nAChR modulators 34 1.4 Significance and aims of the current work ...... 35 1.4.1 Significance ...... 35 1.4.2 Aims ...... 36 CHAPTER 2 — nAChR modulators from spider venoms ...... 37 2.1 Introduction ...... 37 2.2 Methods and Materials ...... 39 2.2.1 Tissue culture reagents ...... 39 2.2.2 Spider venoms ...... 40 2.2.3 Cell lines and culture ...... 40 2.2.4 Preparation of the Calcium 4-dye ...... 41 2.2.5 High-throughput FLIPR calcium assay ...... 42 2.2.6 Activation of α3* nAChR by nicotine ...... 43 2.2.7 Activation of α7 nAChR by choline in the presence of the positive allosteric modulator PNU-120596 ...... 45

2.2.8 Identification of blockers and agonists for α3* and α7 nAChR using the FLIPRTETRA system ...... 46 2.2.9 Data analysis ...... 49 2.2.10 Statistical Analyses ...... 49 2.3 Results ...... 50 2.3.1 Screening of 71 spider venoms using the high-throughput FLIPRTETRA system .. 50 2.3.2 Screening of venoms from “modern” spiders (Infraorder Araneomorphae) ...... 50 2.3.3 Screening of venoms from “primitive” spiders (Infraorder Mygalomorphae) ...... 55 2.3.4 Identification of α7/α3* nAChR agonists in the venom of theraphosid spiders .... 55 2.3.5 Venoms that selectively inhibit vertebrate α3* nAChRs ...... 58 2.3.6 Venoms that non-selectively block both α3* and α7 nAChRs ...... 61 2.3.7 Theraphosid venoms with no apparent effect on nAChRs ...... 63 2.3.8 Effect of non-theraphosid mygalomorph venom on vertebrate nAChRs ...... 64 2.3.9 Taxonomic distribution of nAChR modulators within Araneae ...... 64 2.4 Discussion ...... 66 CHAPTER 3 — nAChR modulators from scorpion venom ...... 75 3.1 Introduction ...... 75 3.1.1 Scorpion toxins and their pharmacological action ...... 75 3.2 Methods ...... 77 3.2.1 Figure 3.1 shows an example of the protocol as applied to scorpion venoms ..... 77 3.2.2 Scorpion venoms ...... 80 3.2.3 Venom fractionation ...... 80 3.2.4 Discovery of a choline ester from the venom of H. spinnifer ...... 80 3.2.5 Determination of mass and purity ...... 81 3.2.6 Identification of active compound using HRMS and LCMS ...... 81 3.2.7 Chemical synthesis of choline esters ...... 82 3.3 Results ...... 84 3.3.1 High-throughput FLIPR assay ...... 84 3.3.2 Isolation of the active compound from H. spinnifer venom ...... 87 3.3.3 Identification of the active compound by high-resolution mass spectrometry ...... 90 3.4 Discussion ...... 92 CHAPTER 4 — Conclusions ...... 94 Summary and future directions ...... 95 References ...... 97 Appendix I ...... 125

LIST OF FIGURES AND TABLES

FIGURE 1.1. GENERAL ARCHITECTURE OF PENTAMERIC ACETYLCHOLINE RECEPTORS...... 18 FIGURE 1.2 α-CONOTOXINS AS RESEARCH TOOLS TO STUDY THE STRCUTURE AND FUNCTION OF NACHRS...... 21 FIGURE 1.3. STRUCTURES OF TOXINS AND TOXIN DERIVATIVES THAT TARGET NACHRS...... 31 TABLE 1.1 AMINO ACID SEQUENCES OF SELECTED α-CONOTOXINS. SI CONOTOXIN FROM CONUS STRIATUS THAT TARGETS THE NEUROMUSCULAR SUBTYPE CHOLINERGIC RECEPTOR; MII CONOTOXIN FROM CONUS MAGUS THAT TARGETS WITH HETEROMERIC NACHRS WITH HIGHER AFFINITY THAN THE HOMOMERIC SUBTYPE α7 NACHR; AUIA CONOTOXIN ISOLATED FROM CONUS AULICUS, WHICH TARGETS THE NEURONAL HETEROMERIC SUBTYPE α3β4. THE ASTERISKS (*) REPRESENT C-TERMINAL AMIDATION. TABLE ADAPTED FROM (LEWIS, 2012; MOLGÓ, 2013). LOOP 1 AND LOOP2 ARE THE TWO INTERCYSTINE LOOPS. 33 α-CONOTOXIN ...... 33 FIGURE 2.1. BASIC TWO-ADDITION FLIPR PROTOCOL TO MEASURE CALCIUM RESPONSES EVOKED BY ACTIVATION OF α7 OR α3* NACHRS...... 47 FIGURE 2.2. ‘TWO-ADDITION’ FLIPR PROTOCOL USED TO IDENTIFY NACHR AGONISTS...... 48 FIGURE 2.3. SCREENING OF VENOMS FROM “MODERN” SPIDERS AGAINST HUMAN α3* AND α7 NACHRS USING A HIGH-THROUGHPUT FLIPR ASSAY...... 52 TABLE 2.1. TAXONOMIC DISTRIBUTION OF NACHR MODULATORS FROM ARANEOMORPH SPIDER VENOMS. TWO SPECIES SPARASSIDAE SP. WERE ASSAYED; THE NUMBERS 1 AND 2 INDICATE THEY WERE COLLECTED IN TWO DIFFERENT AREAS IN PNG. VENOMS WITH HIGH BLOCKING EFFECT ARE INDICATED WITH TWO SYMBOLS DENOTED AS √√, WHILE THOSE WITH PARTIAL EFFECT ARE INDICATED WITH ONE √. THE GREEN AREA INDICATES THAT OTHER EFFECTS ON THE VENOM IS NOT OBSERVED...... 54 FIGURE 2.4. AGONISM OF HUMAN NACHRS BY THERAPHOSID VENOMS...... 57 TABLE 2.2. APPARENT AGONISTIC EFFECT CAUSED BY VENOMS FROM FOUR THERAPHOSID SPIDERS. AFTER VENOM ADDITION, THE FLUORESCENT CALCIUM RESPONSES EVOKED BY NACHRS IN THE ABSENCE OR PRESENCE OF PNU WERE CONVERTER INTO AREA UNDER CURVE USING SCREENWORKS SOFTWARE, VERSION 3.1. THE VALUES ARE AVERAGES OF THREE INDEPENDENT EXPERIMENTS WITH N = 3. THE MEAN OF THE CONTROL NICOTINE RESPONSE WAS 509 ± 50.40 WHILE THE CONTROL CALCIUM RESPONSE ELICITED BY CHOLINE WAS 1088 ± 38.31 ...... 58 TABLE 2.3. MYGALOMORPH SPIDERS WITH VENOMS THAT ARE LIKELY TO CONTAIN AGONISTS OF α7 NACHRS. N.D INDICATES THE EFFECT WAS NOT DETERMINED AND A FURTHER ANALYSIS IS REQUIRED TO CONFIRM THAT THE VENOM EFFECT IS VIA ACTIVATION OF α7 NACHRS IN THE PRESENCE OF PNU...... 58 FIGURE 2.5. THERAPHOSID SPIDER VENOMS THAT PREFERENTIALLY BLOCK α3 NACHRS...... 59 TABLE 2.4. THERAPHOSID SPIDERS WITH VENOMS THAT SELECTIVELY BLOCK α3* NACHRS...... 59 SPECIES WITH CHARACTERISTIC COLORS ARE INDICATED. (M) AND (F) REFER TO GENDER. THE NUMBERS BESIDE THE NAME OF EACH THERAPHOSID SPIDER INDICATES THAT THEY ARE INDIVIDUALS FROM THE SAME SPECIES, ALTHOUGH WERE COLLECTED FROM DISTINCT LOCALITIES...... 59 FIGURE 2.6. MANY THERAPHOSID VENOMS NON-SELECTIVELY BLOCK BOTH α3* AND α7 NACHRS...... 61 TABLE 2.5. THERAPHOSID SPIDERS WITH VENOMS THAT NON-SELECTIVELY BLOCK BOTH α3* AND α7 NACHRS. (M) AND (F) REFER TO GENDER. THE NUMBERS BESIDE THE NAME OF EACH THERAPHOSID SPIDER INDICATES THAT THEY ARE INDIVIDUALS FROM THE SAME SPECIES, ALTHOUGH WERE COLLECTED FROM DISTINCT LOCALITIES...... 62 FIGURE 2.7. FOUR TARANTULA VENOMS WITHOUT POTENT MODULATORS OF α3* OR α7 NACHRS...... 63 TABLE 2.6. VENOMS FROM AMERICAN THERAPHOSIDS THAT LACK OF MODULATORS OF HUMAN α3* AND α7 NACHRS...... 64 FIGURE 2.8. SUMMARISE OF THE HIGH-THROUGHPUT SCREENING OF SPIDER VENOMS...... 65

FIGURE 3.1. A. CRASSICAUDA VENOM CAN INDUCE CALCIUM RESPONSES ELICITED VIA ACTIVATION OF NAV CHANNELS...... 79 FIGURE 3.2. FLIPR SCREENING OF SCORPION VENOMS AGAINST α7 AND α3* NACHRS...... 85 FIGURE 3.3. NON-SELECTIVE ACTIVATION OF α7 AND α3* NACHRS EXPRESSED IN SH-SY5Y BY CRUDE VENOM FROM H. SPINNIFER...... 86 FIGURE 3.4. HPLC FRACTIONATION OF CRUDE VENOM FROM H. SPINNIFER. CHROMATOGRAM RESULTING FROM FRACTIONATION OF CRUDE VENOM VIA C18 RP-HPLC...... 87 FIGURE 3.5. ACTIVITY OF F10 FROM FRACTIONATION OF CRUDE VENOM FROM H. SPINNIFER...... 88

FIGURE 3.6. PURIFICATION OF THE “HIT’ TOXIN BY RP-HPLC...... 89 FIGURE 3.7. AGONISTIC ACTIVITY OF THE “HIT” TOXIN...... 89 FIGURE 3.8. CO-ELUTION OF BIOACTIVE COMPOUND FROM H. SPINNIFER VENOM AND SYNTHETIC CHOLINE ESTERS...... 91 TABLE I. LIST OF SPIDERS WHOSE VENOMS WERE USED IN THIS WORK ...... 125 TABLE II. LIST OF SCORPIONS WHOSE VENOMS WERE USED IN THIS WORK ...... 128

List of abbreviations

ACh Acetylcholine AChBP acetylcholine binding protein AuIB α-conotoxin AuIB AFU arbitrary fluorescent units BSA bovine serum albumin Ca2+ calcium ion Cav2.3 P/Q-type calcium channel CHCA α-cyano-4-hydroxycinnamic acid CNS central nervous system Da Dalton DRG dorsal root ganglion EDTA ethylenediaminetetraacetic acid ESIMS electrospray ionization-mass spectrometry FBS foetal bovine serum FITC fluorescein isothiocyanate FLIPR fluorescent imaging plate reader GABA γ-aminobutyric acid

GABAARs γ-aminobutyric acid type-A receptors HEPES 4-(2-hydroexyethyl)-1-piperazineethane sulfonic acid HILIC hydrophilic interaction liquid chromatography HPLC high-performance liquid chromatographic HRMS high- resolution mass spectrometry kDa Kilodalton LsIA α-conotoxin LsIA MALDI-TOF matrix assisted laser desorption/ ionization time-of- flight MS mass spectrometry NMR nuclear magnetic resonance NSAID non-steroidal anti-inflammatory drugs nAChRs nicotinic acetylcholine receptors PNS peripheral nervous system

PNU-120596 1-(5-chloro-2,4-dimethoxyphenyl)-3-(5-methylisoxazol- 3-yl) urea RLU relative light units PSS physiological salt solution RPMI Rosewell Park Memorial Institute RP-HPLC reversed phase high-performance liquid chromatography SAR structure-activity relationship TFA trifluoroacetic acid TRITC Tetramethylrhodamine TTX Tetrodotoxin

CHAPTER 1 — INTRODUCTION

Recent studies indicate that imbalance of the cholinergic system in human brain via nicotinic acetylcholine receptors (nAChRs) are linked to various neurological disorders (AhnAllen, 2012; Chu, 2005; Nides, 2006; Tregellas, 2011; Yu, 1996) while mutations in the nAChR can lead to frontal lobe epilepsy (for a review see (Steilein, 2001)). These findings have prompted research into the development of drugs that target neuronal cholinergic receptors. Nevertheless, most of the chemical components under trial lack selectivity, or hardly improve the medical condition (for reviews, see (Arneric, 2007; Jensen, 2005; Pammolli, 2011)). The discovery of new therapeutic agents is currently a challenge because of the complex arrangement of subunits of neuronal cholinergic receptors expressed in human brain (Gotti, 2007) and the fact that their roles under physiological and pathological conditions are not fully understood (for a review, see (Tuesta, 2011)).

To facilitate studies of the pharmacology of nAChRs, a search for new nAChR modulators in spider and scorpion venoms was carried out using a high- throughput FLIPR assay to rapidly identify “hit” venoms. The probability of finding native cholinergic modulators was expected to be high because arachnid venoms are estimated to contain about 1 million different components, and even though less than 0.01% of these molecules have been studied (for reviews, see (Escoubas, 2006b; King, 2014; King and Coaker, 2014; King and Hardy, 2013), many of them are peptidic toxins that act on diverse types of ionic channels (Meves, 1984; Quintero-Hernandez, 2013; Shahbazzadeh, 2007; Xie, 2012).

This chapter provides a review of what is known about the molecular structure of nAChRs and their physiological functions in vertebrates and invertebrates. Furthermore, I review the use of animal toxins as research tools to characterize nAChRs, as well as their roles in the development of therapeutic drugs and bioinsecticides.

1.1 NICOTINIC ACETYLCHOLINE RECEPTORS

1.1.1 Homopentameric/heteropentameric architecture of neuronal nAChRs nAChRs are members of the pentameric Cys-loop ligand-gated ion channel superfamily that are activated by the neurotransmitter acetylcholine (ACh) (for a review, see (Unwin, 2005)). Our current knowledge of the molecular structure of neuronal nAChRs began with determination of the N-terminal sequence of the nAChR isolated from the electric organ of the marbled ray Torpedo marmorata, followed by the cloning of its subunits. These subunits were subsequently shown to be closely related to cDNAs cloned from mammals that encoded 17 structurally homologous subunits, denoted α1–α10, β1–β4, γ, δ and ε. Each subunit comprises four α-helical transmembrane regions (M1–M4), of which M1-M3 are conserved and M4 has a variable extracellular C-terminal sequence; a conserved large N-terminal extracellular domain (ECD) of approximately 200 amino acid residues; a short intracellular C-terminal tail (see Figure 1.1A); and a variable cytoplasmic domain between the transmembrane regions M3 and M4 (Unwin, 2005). In addition, all subunits contain in the first extracellular domain a cysteine- loop (Cys–loop) that consists of a pair of cysteines, and in the case of mammals, the Cys-Cys pair is separated by 13 residues. This structural template describes the molecular backbone of different types of nAChRs.

A B Heteromeric Homomeric α3β4 α7

β4 α7

β4 α7 α3 α7

α3 β4 α7 α7

Figure 1.1. General architecture of pentameric acetylcholine receptors. (A) Ribbon representation of the three-dimensional structure of the pentameric nAChR from Torpedo marmorata determined using cryo-electron microscopy (Unwin, 2005). The three domains are indicated: extracellular domain (ECD), transmembrane domain and intracellular domain (ICD). The illustration was modified from the original in the review by Molgó et al. (Molgó, 2013). (B) Schematic representation of the arrangement of a heteromeric nAChR. The subtype α3β4 (left) contains two α3 and three β4 subunits. The heteromeric subtype has two binding sites for ACh (black circles) between α3 and β4 subunits. Homomeric subtypes like α7 (right) contains five identical subunits, and they have five bindings sites for ACh (black circles). The illustration was adapted from a review by Changeux (Changeux, 2010).

According to their localization and subunit composition, nAChRs are classified into two types, the so-called muscle-type nAChRs and neuronal nAChRs. Muscle-type nAChRs contain five classes of subunits denoted α1, β1, δ, and either γ or ε, which are arranged in a circular order of barrel-like staves around a central channel in the muscle sarcolemma, and with a stoichiometry of 2:1:1:1, which corresponds to two α1 subunits plus β1, δ, and γ/ε subunits (Miyazawa, 1999; Unwin, 1993).

In contrast to the muscle-type nAChRs, the neuronal-type receptors are more diverse. Neuronal nAChRs are encoded by nine α (α2–α7, α9, α8, α10) and three β (β2–β4) subunit genes. In the case of sensory and neuronal mammalian tissues, eleven subunits have been cloned, which are α2–α7, α9, α10, and β2–β4, but a homologue of the avian α8 subunit has not been described (reviewed in (Albuquerque, 2009; Nicke, 2004)). Neuronal nAChRs also contain five subunits that form a functional receptor via combinations of different α and β subunits (heteropentamers), such as α3β2 and α4β2. The ACh binding site is located at the interface between an α subunit and an adjacent β subunit (see Figure 1.1B). Also, a third subunit can be included in the receptor (α4)2(β2)2α5. In addition, the pentamers can consist of only α subunits, such as homopentamers of α7, α9 or α10 (Cooper, 1991); in these subtypes, the ACh binding site is located between α subunits (see Figure 1B).

The molecular architecture of heteropentamers and homopentamers is similar (see Figure 1.1B). Nevertheless, each subtype differs in function (Elgoyhen, 2001; Zhou, 2003). If novel and selective modulators for heteropentameric and homopentameric nAChRs were available some relevant questions in terms of the functional characterization of neuronal nAChRs could be answered. For example, the α6 and β3 subunits are co-expressed in brain areas such as the substantia nigra and the ventral tegmental area (Quik and McIntosh, 2006) which leads to two questions: firstly, why are these subunits co-expressed only in these areas of the brain, and secondly, is co-expression of these subunits related to the symptomatology of Parkinson’s disease or addiction to nicotine.

1.1.2 The use of toxins to elucidate the molecular architecture of nAChRs

Toxins derived from animal venoms have been a valuable source of pharmacological tools for studying structure-activity relationships of nAChRs. For example, the muscle-type nAChR was characterized for the first time by Changeux and colleagues in 1970 thanks to the use of α-bungarotoxin, which was purified from the venom of an elapid snake (Bungarus multicinctus), on membrane preparations derived from the electric organ of the eel Electrophorus electricus

(Changeux, 1970). In addition, the proposed binding site for ACh in nicotinic receptors, which lies in a cleft at the interface between the subunits, was determined from x-ray crystallographic structures of the acetylcholine binding protein (AChBP), a homolog of nAChRs, in complex with a variety of ligands (for a review, see (Tsetlin and Hucho, 2009)). AChBP is a soluble protein that shares sequence similarity to the ligand-binding site of α1 or α7 nAChRs. Following its discovery, several crystal structures have been determined of the complex formed AChBP and various α-conotoxins (see Figure 1.2A,B). For instance, co- crystallization of the complexes formed between AChBP and the α-conotoxins IMI and PnIA revealed that these peptide toxins bind at a similar spatial position in the ACh binding domain despite having divergent primary structures (see Figure 1.2C). Moreover, it was shown that hydrophobic interactions between the ACh ligand- binding domain, and a conserved proline residue in both conotoxins could lead to a differential selectivity to nAChRs (for reviews, see (Lebbe, 2014b; Lewis, 2012)).

A large number of α-conotoxins have been used to characterise many of the neuronal subtypes of nAChRs (for a review, see (Lebbe, 2014b)) and structural studies of the α-conotoxins : nAChR interaction have been invaluable (Bourne, 2005; Lewis, 2012). Nowadays, we can predict what amino acids residues can be modified to improve α-conotoxin affinity and selectivity. For example, it was shown that the selectivity of α-conotoxin, Lo1a from Conus longuriunis (see Figure 1.1D) could be switched from the neuronal α7 subtype to adult muscle-subtype nAChR by deleting an Asp residue from its C-terminus (Lebbe, 2014a). An α-conotoxin that can differentially target the muscle subtype over the neuronal subtype could be used to engineered probes for imaging studies of nAChRs in the peripheral nervous system. Because α marine cone snails represent the largest source of tools for studying vertebrate nAChRs, most of the modulators discovered to date are antagonists. In contrast, very few nAChRs agonists have been reported. Therefore, further research is needed to obtain new, subtype-selective nAChRs modulators that might be useful therapeutics and/or useful tools for addressing fundamental questions about nAChRs.

C D

Figure 1.2 α-Conotoxins as research tools to study the strcuture and function of nAChRs. (A) Top view of the X-ray structure of the complex between Aplysia californica AChBP (AcAChBP) and α-conotoxin PnIA (for a review see (Armishaw, 2010)). Each of the five subunits is shown in a different color (orange, green, blue, red and cyan). (B) A single subunit of AcAChBP interacting with an α-conotoxin , with the C-loop is an open position (cyan). The arrow indicates the difference in conformation between the C-loop in the open (cyan) and closed (red) conformations (for more details see review (Armishaw, 2010)). (C) Co-crystal structures between AcAChBP (transparent image, only two subunits shown for clarity) and α-conotoxin ImI. Residues Arg11 and Arg7 (labelled) that are the primary residues on ImI that interact with AcAChBP. Figure taken from Dutertre et al. (Dutertre, 2007). (D) NMR solution structure of α-conotoxin Lo1A. Disulfide bonds are shown in yellow. Image from Lebbe et al. (Lebbe, 2014a).

1.1.3 Location and physiological function of nAChRs in vertebrates and invertebrates

In vertebrates, nAChRs are expressed in most cells of the central and peripheral nervous system (Albuquerque, 2009; Cooper, 1991), and in non-neuronal cells (Macklin, 1998; Plummer, 2005). It seems that the role of ACh and nAChRs in the CNS is mainly modulatory because the cholinergic signal can be excitatory or

inhibitory of synaptic pathways in different cerebral areas (for a review, see (Picciotto, 2012)). The different roles of neuronal nAChRs are made possible by different combinations of α and β (Gotti, 2007). To date, the function of each nAChR subtype in each place of expression is still not fully understood.

Human imaging studies in vivo have allowed mapping of the distribution of nAChRs in the human brain (Kimes, 2006). Cholinergic projections are widely distributed in the brain. For example, neuronal nAChRs have been reported to be expressed in the white matter (Ding, 2004) and are likely to be involved in the regulation of neuronal excitability in the thalamocortical area (Bucher and Goaillard, 2011). Another cerebral region is the ventral tegmental area where the release of glutamate and dopamine are related to the reward responses generated by nicotine administration (Picciotto and Kenny, 2013). It contains GABAergic, glutamatergic, and dopaminergic neurons; and their activities are regulated by a combination of α4β2* nAChRs expressed mainly in GABAergic neurons and α7 nAChRs expressed in terminals of the glutamatergic neurons (Mansvelder, 2002). Neuronal nAChRs are not restricted to the postsynaptic membranes; they can also be expressed along the neuronal body, including axons and presynaptic terminals, and different subtypes of nAChRs can be co-expressed in a neuron (for a review, see (Picciotto, 2012)). Other cholinergic projections come from the basal forebrain complex to the cerebral cortex. Their terminals contain β2* nAChRs; and are distributed across all the cerebral cortex. Activation of the receptor lead to release of glutamate into the prefrontal cortex, and this mediates neuronal processes such as memory, attention and cue detection (Parikh, 2010).

To date, it is known that in humans the distribution of nAChRs is not limited to the nervous system. Cholinergic receptors are expressed in human aortic endothelial cells (α3, α5, α7, α10, β2-4) (Wada, 2007), vascular smooth muscle cells (α2-5, α7, α10) and immune cells, such as human platelets (α7) (Schedel, 2010); human mononuclear leukocytes (α7, β2) (Albaugh, 2004), human macrophages (α1, α7, α10), B-lymphocytes (α3-5, α7), T-lymphocytes (α3, α4, α7, β2, β4), (for a review, see (Bauwens, 2015)). The function of these receptors is linked to homeostasis of the inflammatory response by the cholinergic anti-inflammatory pathway. For

instance, when terminals of afferent fibres located in the vagus are stimulated by pro-inflammatory stimuli, action potentials are generated, and this causes efferent fibres to activate α7 expressed on macrophages by acetylcholine (for a review see (Egea, 2015)). In the end, α7 control the production of pro-inflammatory cytokines such as TNF-α and IL-6 (Zhang, 2015a) without affecting the levels of anti- inflammatory cytokines by inhibiting NF-kB nuclear translocation and activating several pathways including JAK2/STA3 (Li, 2015). However, the cholinergic anti- inflammatory pathway is not fully understood (for a review, see (Egea, 2015)).

In invertebrates, nAChRs have been found in the central nervous system, where they mediate the excitatory synapsis of ACh. Studies of insect brain nAChRs have been limited because there has only been limited success with heterologous expression of cloned insect nAChRs (Landsdell, 2012). The main problem is that cholinergic currents are evoked at high concentrations of ACh and the inhibitory currents are usually too small to be accurately detected with electrophysiology experiments using Xenopus oocytes, although co-expression of insect nAChRs with subunits of vertebrate nAChRs has partially facilitated the expression of insect nAChRs (Landsdell, 2012). There are still important nAChRs from insect species such as honeybee Apis mellifera that are important to study to design novel and safe insecticides. Today, it is known insect nAChRs have similar molecular structure to vertebrate nAChRs. For instance, in the genome of Drosophila melanogaster, 10 nAChR subunit genes of nAChRs have been identified: seven encoded the α subtype (Dα1-7), and three corresponded to β subtype (Dβ1-3) (Jones, 2007). The sequences of Dα5, Dα6 and Dα7 subunits are similar to each other, and to α7 subunit from vertebrate nAChRs (Grauso, 2002; Jones and Satelle, 2009). The Dα6 subunit is the most similar to the vertebrate α7 nAChR (Sattelle, 2005). This similarity may allow insect brains to serve as a model system for studying brain insect as an alternative animal model to study the role of nAChRs in different pathologies. For example, more than 40% of bee brain consists of Kenyon cells that are the major cells of the cerebral structure, known as mushroom bodies or corpora pedunculata. This structure processes multiple sensory information derived from the antennal lobe (Akalal, 2006; McGuire, 2001). It was shown that neonicotinoids cause up-regulation of nAChRs (Moffat, 2015);

and because nAChRs can cause mitochondrial depolarization it was suggested the effect of these insecticides may be mediated in part by up-regulation of nAChRs in Kenyon cells leading to mitochondrial dysfunction, memory deficits, and poor navigation (Fischer, 2014; Williamson, 2013).

To summarize, nAChRs engaged in different physiological functions that are guided by the variability of their subunits (McGehee and Role, 1995). Many efforts have been made to determine the role of various nAChR subtypes in the nervous system but most studies have focussed on the homomeric α7 and heteromeric α4 nAChRs (Koga, 2014; Schedel, 2010; Zhou, 2003). This is reflected in the increasing number of drugs that target α7 and α4β2, which are under development for treatment of several neurological conditions. However, the lack of specific modulators for other nAChR subtypes makes it challenging to elucidate the role of each subtype under both physiological and pathological conditions. Thus, there is still a significant need to discover or develop highly selective modulators of neuronal nAChRs.

1.1.4 Role of nAChRs in human disease

The role of ACh is as important as their molecular target, nAChRs. It is synthesised by the enzyme choline acetyltransferase that transfers an acetyl group from acetyl-CoA to choline (Korey, 1951). Its precursor, choline, is endogenously produced, although it is also absorbed in the small intestine to reach the levels required to accomplish several functions (Zeisel and da Costa, 2009). For example, choline is a precursor for the biosynthesis of constituents of plasma membrane such as phosphatidylcholine and sphingomyelin (Zeisel, 1992; Zeisel and da Costa, 2009). Furthermore, choline is a methyl group donor; and methylation is needed to regulate of expression of genes and mediate biosynthetic reactions (Zeisel and da Costa, 2009). A deficiency in choline has been related to liver damage (Buchman, 1995) and impairment of cognitive functions in older people (Fioravanti and Yanagi, 2005). Indeed, choline supplementation enhances cognitive functions in patients with schizophrenia (Knott, 2015). De-regulation of cholinergic function is linked to the function and expression of nAChRs. For

instance, chronic administration of nicotine can induce an increase in the number of neurons containing α4β2 nAChRs in the glutamatergic subthalamic nucleus, a cerebral area related to movement control (Xiao, 2015), or down-regulate the function of other subtype nAChRs. Below is discussed different pre-clinical and clinical evidence of the consequences of smoking, and links between de-regulation of nAChRs and major symptoms in neurodegenerative diseases.

Post-mortem studies in the brain of people who suffer from Alzheimer’s and Parkinson’s disease (Rinne, 1991), epilepsy and schizophrenia showed that the density of α4β2 nAChRs is extremely low. The degeneration of cholinergic circuits was confirmed in human brains after the development of nicotine radioligands for imaging studies in vivo. For example, in vivo studies of brains of people suffering from Alzheimer’s disease showed a decrease of α4β2 nAChRs (Kendziorra, 2010), and the density of nAChRs in arteries appeared to be affected as well (Bauwens, 2015).

In patients suffering from Alzheimer’s disease and Parkinson’s disease, there is a notable loss of cholinergic projections into the cerebral cortex, and this has been considered as part of the reason for the impaired cognition observed in these patients, although the loss of the cortical cholinergic activity in people with Parkinson’s disease is notable (for a review see (Müller and Bohnen, 2013)). In addition, depressive symptoms in patients with Parkinson’s disease is strongly correlated to loss of α4β2 nAChRs in the neuronal frontal corticomesostriatal circuitry, while those patients with mild cognitive impairment were associated with low levels of α4β2 in the neuronal frontal corticobasal ganglia-limbic-cerebellar circuitry that includes some areas of the former circuitry (Meyer, 2009). Furthermore, α4β2* nAChRs are widely distributed in the brain and are not restricted to cholinergic or dopaminergic neurons. Thus, down-regulation of this heteromeric subtype may lead to dysfunction of other neuronal circuitries (Emre, 2003). Cholinergic deregulation in the hippocampus may be associated with cognitive problems because this cerebral region is related to functions of learning and memory (Kutlu and Gould, 2015) and this correlation between dysfunction of

neuronal circuitry and cognitive functions is observed in various neuronal disorders (Drevets, 2008). nAChRs have also been identified in immune cells, and up-regulation of its activity in this cells has been linked to cardiovascular diseases such as atherosclerosis in smokers (Aicher, 2003; Egleton, 2009; Wada, 2007). Recent evidence suggests that the mechanism underlying this process could be activation of α7 nAChRs that trigger transcription of cell-adhesion proteins on endothelial cells (Alamanda, 2012). These proteins, such as E-selectin, favour the attachment of monocytes, and it might lead to abnormal formation of fibrous plaques in vascular smooth muscle cells (Alamanda, 2012). Notably, the duration and amount of nicotine administered via smoking are dependent variables, and when these reach critical values it may facilitate the activation of numerous nAChR subtypes (Koga, 2014; Wada, 2007) that are linked to signalling cascades (Wada, 2007) and cause abnormal proliferation of neuronal and non-neuronal cells (He, 2014; Wada, 2007).

1.2 MODULATORS OF NACHRS

1.2.1 Modulators of nAChRs as drugs and insecticides

Several clinical trials are being carried out to evaluate potential drugs that target nAChRs. For instance, some nicotine analogues are been used to facilitate quitting smoking (Rollema, 2007). Varenicline (trade name Chantix®) is currently being used for treatment of nicotine addiction and is a partial agonist of the α4β2 and α3* receptors. However, varenicline also activates the α7 nAChR (Mihalak, 2006), and it has been suggested that its mechanism of action involves is by modulation of several nAChRs (Bordia, 2012; Lotfipour, 2012) that contributes to decreasing the levels of dopamine that were previously increased by nicotine in the limbic system (Corrigall, 1992, 1994), thus, preventing the reward response induced by smoking (Mihalak, 2006; Rollema, 2007). Imaging studies in vivo of smokers under treatment with varenicline showed that α4β2* nAChRs are saturated after administration of the drug at low or high doses, and withdrawal effects were not detected. Unfortunately, varenicline can also cause adverse effects in smokers under treatment; nausea is very common and is one of the

primary reasons for withdrawal for treatment (Jorenby, 2006; Lam and Patel, 2007). The side-effect is likely to be mediated by activation of ganglionic nAChRs. This is likely to be a common problem when using therapeutic agents that target multiple subtypes of nAChRs distributed across the peripheral nervous system.

As mentioned previously, nAChRs are also important insecticide targets, and the so-called neonicotinoid insecticides that target insect nAChRs are one of the most successful classes of chemical insecticides. Nicotinoid insecticides are derived from nicotine, the major alkaloid derived from leaves and stems of the tobacco plant Nicotiana tabacum; nicotine is a secondary metabolite that is used by the plant to defend against herbivorous insects (Isman, 2006). Nicotine was used to control crop pests as far back as the 1700s (Siegwart, 2015) but its use was limited because of toxic effects on mammals and only moderate potency against insects. The neonicotinoids are chemically similar to nicotine but they have low toxicity to mammals as they were engineered to bind selectively to insect nAChRs (Brown, 2006). However, like the majority of chemical insecticides, most neonicotinoids are broadly active against a wide range of insect pests, including beneficial insects such bees (Déglise, 2002; Faucon, 2005). Honeybees are efficient pollinators that mediate plant reproduction, and therefore it would be highly desired to engineer insecticides that preferentially target nAChRs in insect pests but not beneficial insects such as pollinators or predators of the targeted pest.

Structure-activity relationship studies are helping to guide the development of more selective neonicotinoid insecticides. Crystal strictures of the complex formed between neonicotinoids and AChBP have been used to identify rational methods of improving the potency and/or selectivity of particular neonicotinoids (Matsuda, 2009). For instance, Matsuda et al. (Matsuda, 2009) compared the complex formed between imidacloprid, the most successful neonicotinoid, and α2β1 nAChRs from the honeybee Apis mellifera and the green peach aphid Myzus persicae (Matsuda, 2009). They found a remarkable structural difference between the binding sites for imidacloprid in the two nAChRs, with a hidden and broader binding groove in the aphid nAChR compared with the bee receptor. The authors

suggested that linking imidacloprid to a molecular fragment that fits the groove of the aphid nAChR could improve the selectivity for insect pests over bees.

To summarise, neonicotinoids are generally safe for humans but they can be toxic for small vertebrates such as birds and beneficial insects such as honeybees. Thus, further research is needed to develop more selective neonicotinoids with reduced ecological impact.

1.2.2 Novel subtype-selective modulators of vertebrate nAChRs are required to develop accuracy pharmacological tools and potential therapeutic leads

The selectivity of some toxins for nAChRs has facilitated the development of fluorescent binding probes and radioligands for use in biological sciences (Hannan, 2015; Schmaljohann, 2006) and nuclear medicine (Schmaljohann, 2005, 2006), respectively. It has helped to determine the distribution of nAChRs in the brain of rodents (Perry, 2002; Rueter, 2006) and humans (Ding, 2004; Kimes, 2006), including in people suffering from Alzheimer’s and Parkinson’s disease (Schmaljohann, 2005, 2006). For example, α-bungarotoxin a 74-residue peptide isolated from venom of Bungarus multicinctus, was used to affinity purify the nAChR from the electric organ of T. marmorata (Changeux, 1970). α-Bungarotoxin has also been useful for visualizing the distribution of nAChRs in the brain and the in the electric organ from Torpedo californica, and for identifying clones expressing epitopes from nAChR in Escherichia coli (Gershoni, 1987). Recently, it was shown that Alexa Fluor 555-labelled α-bungarotoxin labels GABAA-Rs expressed in HEK- 293 cells (Hannan, 2015), and this binding is inhibited by bicuculline, a competitive antagonist at GABAA-Rs, and d-tubocurarine, a non-selective nAChRs antagonist and a competitive antagonist at GABAA-Rs (Caputi, 2003) (see Figure 1.2A-C). This finding is not surprising because type-A GABAergic receptors are members of the pentameric Cys-loop ligand-gated ion channel family, like nAChRs (Corringer, 2012). The selectivity of α-bungarotoxin has been used to discover new functional subunits of nAChRs, such as the α7 subunit of nAChRs. The vertebrate neuronal subunit α7 was cloned for the first time in 1990 by Couturier et al. (Couturier,

1990). They found the subunit formed an oligomeric channel in the membrane of Xenopus laevis oocytes and it was activated by nicotine and choline, and desensitised rapidly. This was a receptor similar to the heteropentameric nAChR reported previously because α-bungarotoxin blocked its activation and was able to bind the subunit α7. Furthermore, iodinated α-bungarotoxin has been used in competition studies to evaluate the specificity of radioligands towards α7 nAChRs.

Nicotine radioligands are being used for imaging studies using positron emission tomography (PET) (for a review, see (Bauwens, 2015)). Note that radioligands are typically used for imaging studies at sub-therapeutic doses, and consequent safety rarely poses an issue. The first nicotine-based radioligand developed to label nAChR was [11C]-nicotine (Halldin, 1992) (see Figure 1.3E). It was synthesised at high purity but had low specificity when tested on human brain in vivo (Nybäck, 1994). Other radioligands that are chemically related to nicotine are still under development with some promising results, such as [18F]-ZW-104 and [18F]-AZAN-α (for a review, see (Bauwens, 2015)) (see Figures 1.3F-G). The first probe is a selective ligand for the β2 subunit of nAChR and has been tested pre-clinically. [18F]-AZAN-α has been assayed in human studies with promising results (for a review, see (Bauwens, 2015)). It reached the brain in less than 20 min after injection, and selectively bound the α4β2 subtype (Wong, 2013). In this study the specificity for α4β2 was evaluated through administration of varenicline, a selective partial agonist of α4β2 that is used to aid smoking cessation (Rollema, 2007). The specificity of the radiotracer for the receptor was confirmed after the drug blocked totally the uptake of [18F]-AZAN-α. Another radio-chemical tracer is [18F]-flubatine, which is an analogue of epibatidine, an alkaloid isolated from the skin of the frog Epipodobates tricolor (Fisher, 1994) (see Figures 1.3D,H).

A-85380 is another example of a radiotracer (Kimes, 2006; Sullivan, 1996) that has high affinity for α4β2 nAChRs (for a review, see (Lotfipour, 2011)). It allowed mapping of the distribution of nAChRs in the brain of smokers and non smokers (Kimes, 2006). Furthermore, it has been used to confirm that expression of α4β2 in the brain of patients with Alzheimer’s disease and dementia is significantly decreased, and is correlated with the degree of cognitive impairment (Colloby,

2010).

The development of agents to be used for human brain imaging in vivo has been extended to design of tracers selective for the α7 subtype. Some promising results has been obtained even where the homomeric α7 in the human brain is found in low concentrations compared to α4β2. For example, [18F]-NS-10743 and [18F]- ASEM (see Figure 1.3I-J) showed high affinity toward α7 and the distribution of these radioligands were found in brain areas where α7 is densely expressed, such as frontal cortex and hippocampus. To prove the specificity of the agents, the specific ligand SSR180771, which is a selective and partial agonist at human and rodent subtype α7, was used in competition studies. The tracers showed their specificity for α7 when these were displaced by SSR180771.

As discussed above, part of the problem with nAChR modulators as drug leads, insecticides, and pharmacological tools is their lack of selectivity. This has made it difficult to dissect the role of nAChRs in different physiological processes (García, 2015) as well as their role in pathophysiological processes such as cognitive dysfunction, depression, chronic anxiety, analgesia, inflammation and neurodegenerative diseases. Thus, there is still a great need to develop novel nAChR modulators that are both potent and selective. As discussed below, animal venoms have provided some of the most potent and selective nAChR modulators described to date, and represent a potential source of new nAChR modulators.

A) B) C) D)

N Cl

Epibatidine α-bungarotoxin Bicuculline Tubocurarine chloride 18F H) E) F) G) N 18F N 18F N O N

11 N H3[ C] N N N 11C-nicotine [18F]-ZW-104 [18F]-AZAN-α [18F]-Flubatine

I) J) O 18F O N 18F O S N N N N [18F]-NS-10743 [18F]-AZEM

Figure 1.3. Structures of toxins and toxin derivatives that target nAChRs. (A) α-bungarotoxin. Image adapted from https://en.wikipedia.org/wiki/Alpha- Bungarotoxin. (B) Bicuculline. Image from http://halofanon.wikia.com/wiki/Bicuculline. (C) Tubocurarine chloride. Image from https://en.wikipedia.org/wiki/Tubocurarine_chloride. (D) Epibatidine. (E) 11C- nicotine. (F) [18F]-ZW-184. Image from http://www.iphc.cnrs.fr/dirac/productReadHit.php?id=923. (G) [18F]-AZAN-α. Image adapted from (Bauwens, 2015). (H) [18F]-Flubatine. Image adapted from (Bauwens, 2015). (I) [18F]-NS-10743. Image adapted from (Bauwens, 2015). (J) [18F]-ASEM. Image from Horti et al. (Horti, 2014). The pyridine and the pyrrolidine rings are shown in orange and green, respectively. The chemical structure of nicotine is similar to that from epibatidine such as pyridine rings and nitrogen atoms (in orange color). The tracers are shown in blue.

1.3 VENOM-DERIVED MODULATORS OF NACHRS

1.3.1 Important role played by subtype-selective α-conotoxins in study of vertebrate nAChRs

Venoms from marine cone snails (genus Conus) have been extensively studied, although scarcely 0.1% of the predicted total number of venom compounds have been characterized so far (Lebbe, 2014b). Conus venoms are replete with disulfide-bridged peptides, named conopeptides or conotoxins that target a wide range of voltage and ligand-gated ion channels, including nAChRs (Lebbe, 2014b; Lewis, 2012; McIntosh, 1999). Indeed, the venoms of marine cone snails are the largest natural source of potent and selective nAChR blockers (for reviews, see (Lewis, 2012; McIntosh, 1999)). There are at least seven families α-conotoxins that bind specific muscle and neuronal nAChR subtypes with high selectivity (Wang, 2015). α-Conotoxins are the smallest conopeptides (generally 12–20 amino acid residues) that competitively binds to the ACh binding site of nAChRs. Their general framework is denoted CC-Xm-C-Xn-C, where C represents a cysteine residue and X corresponds to the two inter-cysteine loops that contain a variable region of amino acid residues. The number of amino acid residues is indicated as m/n and α-Conotoxins with a 3/5 loop size combination are usually purified from the venom of Conus species that feed on fish and are typically active on mammalian neuromuscular nAChRs, while other conopeptides with 4/3, 4/4, 4/5, 4/6 and 4/7 loop combinations preferentially act on mammalian neuronal nAChRs (for reviews, see (Lebbe, 2014b; Lewis, 2012; Terlau and Olivera, 2004)). See table 1.1 below.

Table 1.1 Amino acid sequences of selected α-conotoxins. SI conotoxin from Conus striatus that targets the neuromuscular subtype cholinergic receptor; MII conotoxin from Conus magus that targets with heteromeric nAChRs with higher affinity than the homomeric subtype α7 nAChR; AuIA conotoxin isolated from Conus aulicus, which targets the neuronal heteromeric subtype α3β4. The asterisks (*) represent C-terminal amidation. Table adapted from (Lewis, 2012; Molgó, 2013). Loop 1 and Loop2 are the two intercystine loops.

α-conotoxin Amino acid sequence Selectivity

C Loop 1 C Loop 2 C*

SI (3/5 loop) I CC NPA C GPKYS C* Muscle type α12βγδ

MII (4/7 G CC SNP V C HLEHSNL C* Neuronal subtype loop) α6α3β2β3>α3β2>α7

AuIA (4/7 G CC SYP P C FATNSDY C* Neuronal subtype loop) α3β4

Several studies of the interaction between α-conotoxins and AChBP have contributed to our understanding of how α-conotoxins can be selective for a particular nAChR (Dutertre, 2007). Basically, the different electrostatic interactions and hydrogen bonds between aromatic residues of α-conotoxins and amino acid residues outside of the ACh binding pocket are found in the conotoxins that block nAChRs while those conotoxins with fewer interaction are more likely to be agonists (Dutertre, 2004). These studies have also helped develop a structural model of the pharmacophore of α-conotoxins. It is delimitated by the two loops between the cysteine residues. One is conserved and determines the binding to the nAChR and the second loop consists of a variable sequence that determines their selectivity for a particular subunit of the nAChR (for reviews, see (Lebbe, 2014b; Lewis, 2012; Terlau and Olivera, 2004)).

1.3.2 nAChR modulators from other venomous animals

Snake venom is another large source of proteins and peptides that target nAChRs (Brahma, 2015; Utkin, 2015). The most abundant proteins described so far in elapid, colubrid and psammophid snakes are the three-finger toxins (3FTx) (Sunagar, 2013) that contain between 60 to 74 amino acid residues with molecular masses ranging between 6000 and 8000 Da (Manjunatha and Robin, 2010; Roly, 2014). Most 3FTxs target nAChRs. These toxins are divided into three types differentiated by the number of cysteines and the inter-cysteine spacing. Type I and type III α-neurotoxins with eight cysteine residues target neuromuscular nAChRs. Type II α-neurotoxins contain an extra pair of cysteine residues which form an additional disulfide bond that stabilises loop 2 (Fry, 2003; Sunagar, 2013). This additional disulfide bond facilitates interaction with neuronal α7 and α9-10 nAChRs, while maintaining affinity for the neuromuscular α1 nAChR (Fry, 2003). Another interesting group of 3FTxs are the so-called “non-conventional toxins” that have an additional disulfide bond that is not in the central loop II but instead in the N-terminal loop I (for a review see (Tsetlin, 2015)). Examples of non-conventional toxins that block nAChRs are candoxin from Bungarus candidus and the weak toxin (WTX) from Naja Kaouthia.

1.3.3 Rationale for examining spider venoms as a source of novel nAChR modulators

Spiders are very known to produce venoms that contain diverse chemical structures such as peptides up to 10 kDa, proteins, amino acids, and acylpolyamines (for a review, see (Estrada, 2007)). This complex and heterogeneous mixture of compounds allows spiders to prey on, and defend themselves against a wide range of prey and predators. The different repertoire of acylpolyamines appears to have evolved to allow predation of different species of invertebrates and small vertebrates.

Acylpolyamines can selectively block ionotropic glutamate receptors from insects and vertebrates including mammals (for a review see (Olsen, 2011)). Because glutamate is the main neurotransmitter at the neuromuscular junctions in insects,

block of glutamatergic receptors by acylpolyamines induces paralysis (Piek, 1971; Piek and Njio, 1975). However, acylpolyamines toxins are also reported to antagonise of nAChRs (Willians, 1997). The basic backbone of acylpolyamines is an aromatic acyl group and the polyamine chain that form the essential part of the molecule (Estrada, 2007). These two features are generally present in acylpolyamines derived from funnel-web, trap door and tarantula spiders (Strømgaard, 2005). The length of the polyamine chain and the degree of hydrophobicity determines the affinity for nAChRs.

Recently, a toxin denoted VdTX-1 was purified from the venom of the Brazilian theraphosid spider Vitalius dubius (Rocha-E-Silva, 2013). This component has a molecular weight of 728 Da and it non-competitively blocks nAChRs. Surprisingly, this toxin was photosensitive, which might be why this toxin, and perhaps related toxins in other spider venoms, was not identified in previous studies. In any case, there is clearly some evidence for the presence of small-molecule nAChR modulators in spider venoms, suggesting that these venoms should be explored more systematically as a potential source of novel modulators.

1.4 SIGNIFICANCE AND AIMS OF THE CURRENT WORK

1.4.1 Significance nAChRs are considered a key molecular target for treatment of several neurological and vascular disorders (Bauwens, 2015; van Enkhuizen, 2015). Key knowledge about the pharmacology and distribution of nAChRs in the vertebrate nervous system has been achieved thanks to the use of animal toxins as research tools. However, we are still a long away from fully understanding of role of different subtypes of nAChRs in neurons and non-neuronal cells due largely to the lack of selective pharmacological modulators. Although animal venoms have proved to be an excellent source of nAChR modulators, most work has been carried out using venoms from snakes and cone snails, while spiders and scorpions remain largely unstudied as a potential source of nAChR modulators. Since spiders and scorpions use their venom to either hunt, or defend against, vertebrates, their venoms represent a potential source of compounds that can modulate the activity

of vertebrate nAChRs. Thus, in this thesis, I aimed to explore arachnid venoms for the presence of compounds that modulate the activity of vertebrate α3* and the neuronal α7 nAChRs.

1.4.2 Aims

Aim 1: Determine the taxonomic penetrance of nAChR modulators in venoms from the two major infraorders of spiders: the “primitive” spiders (Mygalomorphae) and “modern” spiders (Araneomorphae)

Aim 2: Isolate and pharmacologically characterize a non-peptide nAChR modulator identified in venom of the scorpion “Heterometrus spinnifer”.

CHAPTER 2 — NACHR MODULATORS FROM SPIDER VENOMS

2.1 INTRODUCTION

Venomous animals use their venom to hunt prey and/or protect themselves against predators (Billen, 2008; Olsen, 2011). Most animal venoms are a complex mixture of salts, small organic compounds, peptides and proteins (Escoubas, 2006b; Zhang, 2015b), with activity against a wide range of receptors (Inserra, 2013; Wang, 2015) and ion channels (Billen, 2008; DeBin, 1993; Nirthanan, 2002). Consequently, over the past 20 years, a number of venom molecules have been purified which have proved useful clinically, to agronomy, or as research tools (Harrison, 2014; King and Coaker, 2014; King and Hardy, 2013; Lewis, 2012; Nicke, 2004; Windley, 2012b).

One potentially rich source of compounds active against ligand-gated channels, which remains largely unexplored, is spider venoms (Nørager, 2014). It is predicted that all venoms from the estimated 100,000 extant species of spiders contains approximately 20 million bioactive molecules and scarcely 0.008% have been characterized so far (for reviews, see (Estrada, 2007; King and Coaker, 2014; King and Hardy, 2013; Windley, 2012b)). This has been due in part to the small amount of venom that can be obtained from arachnids, making traditional biochemical characterization challenging. However, the development of high- throughput functional assays has made it more feasible to functionally characterize venom from small venomous animals (for a review, see (Vetter, 2011)).

There are two reasons why spider venoms were considered to be a likely possible source of novel molecules that target human neuronal nAChRs. Firstly, each venom is estimated to contain several hundred to more than a thousand molecules in the desired molecular weight range of between 0.1 and 10 kDa (for reviews, see (Escoubas, 2000, 2006a, b; Estrada, 2007; King and Hardy, 2013)), with most of those already isolated having been shown to predominantly act on voltage- and ligand-gated ion channels (for reviews, see (Escoubas, 2000, 2006a, b; King and Hardy, 2013)). Secondly, even though spider toxins are designed to target mainly insect receptors (because insects are the main prey of spiders) (Corzo, 2002;

Vassilevski, 2008), the similarity of invertebrate receptors to their counterparts in vertebrates (Lansdell, 2012) suggests that modulators of human nAChR receptors might be found in spider venoms.

Some authors have divided spider toxins into two major groups: acylpolyamines and cysteine-rich peptide toxins (Liang, 2008; Vassilevski, 2009). The acylpolyamines were first isolated from theraphosid spiders (for reviews, see (Estrada, 2007; Strømgaard, 2001) and shown to cause fast paralysis in insects (Escoubas, 2000) but they are found in many spider venoms (Moore, 2009; Strømgaard and Mellor, 2004) They are known to block both ionotropic glutamatergic receptors (Nørager, 2014; Strømgaard and Mellor, 2004) and nAChR receptors in vertebrates and invertebrates (Mellor and Usherwood, 2004; Strømgaard, 2001). Their potential therapeutic and insecticidal use has led researchers to synthesise acylpolyamines (Strømgaard and Mellor, 2004) and study their structure-activity relationships in detail in order to improve their affinity for either glutamatergic (Kromann, 2002) or cholinergic receptors (Olsen, 2006). Recently, a small molecule of 728 Da was isolated from the venom of the theraphosid spider Vitalius dubius that blocked cholinergic receptors in vertebrate nerve-muscle preparations (Rocha-E-Silva, 2013). The structure of the molecule was not determined but its sensitivity to light and its small size suggests that it likely to be a polyamine.

In addition to polyamines, other small molecules can be found in spider venoms, such as amino acids (GABA, glutamate, acetylcholine), adenosine triphosphate (ATP) and adenosine monophosphate (ADP) (for reviews, see (Estrada, 2007; King and Hardy, 2013; Olsen, 2006, 2011). These molecules are endogenous modulators of ion channels and receptors both in invertebrates and vertebrates but their lack of selectivity makes these molecules unsuitable as drug leads (for a review, see (King and Hardy, 2013)).

Most spider venoms are dominated by high concentrations of small cysteine-rich peptides (King and Hardy, 2013). Most of these peptide toxins conform to a structural scaffold known as an inhibitor cysteine knot (Nicke) or "knottin" fold (for a review, see (King and Hardy, 2013)). Numerous spider-venom ICK peptides that

target ion channels and receptors have been identified in the venom of both mygalomorph (Ayroza, 2012; Corzo, 2008) and araneomorph (Trachsel, 2012; Vassilevski, 2013) spiders. In contrast, there is a small group of spiders whose venoms contain cytolytic peptides that consist of linear amphipathic domains (for reviews, see (King and Hardy, 2013; Windley, 2012a). These peptides, which are devoid of disulphide bridges, have been identified in venoms of araneomorph spiders from the families Ctenidae, Lycosidae and Oxyopidae (for reviews, see (King and Hardy, 2013; Kuhn-Nentwig, 2003).

Apart from the discovery of huwentoxin-I (HWTX-I), which blocks vertebrate cholinergic transmission, from venom of the theraphosid Haplopelma schmidti (formerly Selenocosmia huwen Selenocosmia (now Ornithoctonus huwena) (Zhou, 1997), there is no evidence of spider-venom peptides with effects on cholinergic receptors. In this work, a systematic approach was undertaken to screen 71 spider venoms, extracted from representative species of both “primitive” and “modern” spiders, for modulators of vertebrate neuronal nAChRs, using a high-throughput FLIPRTETRA screening approach. It was hoped the data derived from the screening of spider venoms would identify lead peptide toxins that could be further characterised as potential therapeutic agents. However, no “hit” peptide toxin could be identified. Nevertheless, cholinergic modulatory activity was identified in venoms from a number of species and this data was used to map the taxonomic distribution of modulators that target human neuronal nAChRs.

2.2 METHODS AND MATERIALS

This chapter describes the strategy followed to identify modulators of nAChRs in crude spider venoms using a high-throughput FLIPR assay. The protocol used for the high-throughput FLIPR assay was based on that described by Vetter and Lewis (Vetter, 2012; Vetter and Lewis, 2010).

2.2.1 Tissue culture reagents

RPMI (Roswell Park Memorial Institute) medium, (+)-L glutamine; DPBS

(Dulbecco’s Phosphate Buffered Saline) without CaCl2 and MgCl2; 0.25 %

Trypsin/EDTA were from Gibco Life Technologies (Mulgrave, Vic, Australia). Fetal Bovine Serum (FBS) were from Scientific Life (Clayton, Vic, Australia). PNU- 120596 (1-(5-chloro-2,4 -dimethoxyphenyl)-3-(5-methylisoxazol-3-yl) urea) was from Sigma-Aldrich. FLIPRTETRA pipet tips, and Calcium-4 no-wash kit were from Molecular Devices (Sunnyvale, CA, USA).

384-Well cell/imaging microplates (flat-well, clear-bottom black polystyrene) and 384-Well reagent microplates (clear-well, round-bottom polypropylene) were from Corning® Incorporation (Corning, Lowell, MA, USA). Nicotine and choline were from Sigma-Aldrich (Castle Hill, New South Wales, Australia), and were prepared as a stock solution of 100 mM in MilliQ water. The composition of the Physiological Salt Solution (PSS) used in all the assays was NaCl 140 mM, glucose 11.5 mM,

KCl 5.9 mM, MgCl2 1.4 mM, NaH2PO4 1.2 mM, NaHCO3 5 mM, CaCl2 1.8 mM, HEPES 10 mM, and pH adjusted at 7.4 using NaOH (Vetter 2012).

2.2.2 Spider venoms

Crude venoms were extracted from spiders by Dr Volker Herzig (Institute for Molecular Bioscience, The University of Queensland) via electrical stimulation of the chelicerae. Venoms were then lyophilized and stored at –20°C. Before starting fractionation of venoms using reverse-phase high-performance liquid chromatography (RP-HPLC), crude venom (1 mg) was dissolved in 5% Buffer B

(90% CH3CN, 0.05% TFA in H2O).

Crude venom from 71 spiders was used in this research (see Appendix I for the list of spiders whose venom was analysed for nAChR activity using the FLIPRTETRA system). For the FLIPR screens, four different concentrations of spider venom were prepared using MilliQ water as diluent. All venoms and toxins were dissolved in MilliQ water before running the FLIPR calcium assay.

2.2.3 Cell lines and culture

In this project SH-SY5Y cells were used for all FLIPR screens because they are

reported to endogenously express the neuronal α7 nAChR and the “ganglionic” nAChRs, named α3β2 and α3β4 (Dajas-Bailador, 2002). To facilitate the description of these heteromeric receptors, an asterisk (*) is included to denote the possibility of various β subunits in combinations with α3. The calcium responses evoked by these subtypes after stimulation with choline and PNU-120596, and nicotine, respectively, were characterized previously by Vetter and Lewis (Vetter and Lewis, 2010).

SH-SY5Y human neuroblastoma cells (a kind gift from Victor Diaz, Max Planck Institute for Experimental Medicine, Göttingen, Germany) were maintained at 37°C with 5% (v/v) CO2 in RPMI medium which contained 2 mM L-glutamine and 15% (v/v) FBS. Cells were passaged every 3–5 days using 0.25% trypsin/EDTA at a dilution of 1:7. The SH-SY5Y cells were passaged a maximum of 10–20 times in antibiotic-free medium, prior to use for assay. The SH-SY5Y cells used in this project did not show irregular expression of α3* and α7 nAChRs as the control calcium responses, which were obtained in each high-throughput screening, were comparable to the control responses obtained by Vetter and Lewis (Vetter and Lewis, 2010) during the characterization of α7 and α3* nAChRs expressed in SH- SY5Y cells.

Two days prior to assay, a confluent culture of SH-SY5Y cells was collected by trypsinization and resuspended in RPMI with 15% (v/v) FBS, then 20,000 SH- SY5Y cells per well were seed in uncoated black–walled 384-well imaging plates and maintained at 37°C and 5% CO2 for 48 h before running the experiment (Vetter and Lewis, 2010). This time is the optimum to obtain a homogenous monolayer of cells, thus avoiding problems associated with cellular differentiation and changes in the levels of endogenous expression of receptors and ion channels (Vetter, 2012).

2.2.4 Preparation of the Calcium 4-dye

A stock solution of calcium 4-dye was prepared according to the instructions of Molecular Devices. Briefly, 10 ml of PSS is added to the component A to make a

10 x concentrated stock solution, then the solution is protected from light with aluminium foil and stored in aliquots of 100 µL at –20°C. Aliquots can be reused, at least once, if refrozen at –20°C immediately post-experiment.

2.2.5 High-throughput FLIPR calcium assay

The FLIPR-based calcium-influx assay has been shown to be a good high- throughput approach to identify and characterize toxins derived from animal venoms (Inserra, 2013; Jin, 2014; Wang, 2015). In this research project, a FLIPR- based calcium assay was used to expedite the discovery of bioactive molecules from complex mixtures of spider venoms.

71 spider venoms were screened using the fluorescent calcium FLIPR assay, and either agonistic or antagonistic function at α7 and α3* nAChRs was measured according to the methodology described by Vetter and Lewis (Vetter and Lewis, 2010). Briefly, 20,000 SH-SY5Y cells per well were seeded on uncoated black– walled-384-well walled imaging plates and maintained at 37°C and 5% CO2 for 48 h before running the experiment. Once a 95% confluent monolayer was obtained, the RPMI medium was removed from the cell plate by flicking it, the cells were immediately loaded with Calcium 4-no-wash-dye (Molecular Devices), prepared as a 1 x solution in PSS as indicated above, for 30 min at 37°C and 5% CO2. A washing step after medium removal and after dye loading was not required as a no-wash calcium dye was used and all the control calcium responses elicited by cholinergic receptors stimulated by nicotine and choline were comparable to the calcium responses obtained in previous work (Vetter and Lewis, 2010).

It is noteworthy that the 30 min of incubation time is optimal for dye-uptake, and de-esterification of the acetoxymethyl ester (AM) by endogenous esterases (Vetter, 2012; Vetter and Lewis, 2010). The cell preparation was always protected from the light during the incubation to prevent photobleaching of the fluorophore.

After dye loading, the cell plate was transferred to the FLIPRTETRA fluorescence plate reader and a protocol test signal was executed to obtain an appropriate level

of fluorescent baseline that is approximately 1,000 arbitrary fluorescence units with a standard deviation less than 5%. In general, the excitation intensity was adjusted in 80% and camera gain around 150. All fluorescent calcium responses were recorded using a ‘two-addition’ protocol and a cooled CCD camera with excitation wavelength of 470–495 nm and an emission wavelength of 515–575 nm for over a period of 10 min.

After baseline fluorescence readings were recorded every second for 5 s, variable amounts of crude venom (2 µg; 1 µg; 0.2 µg; 0.02 µg), venom fractions (1% of venom fraction in relation to 1 mg of crude venom fractionated by RP-HPLC was tested) or pure toxin (1% of pure toxin in relation to 1 mg of crude venom dissolved in MilliQ water was tested) were added to the cell plate (addition 1) and the calcium response was measured every second for 300 s. At the end of this time, cells were subsequently stimulated with nicotine (30 µM) or choline (30 µM) + PNU-120596 (10 µM) (addition 2), and readings continued every second for an additional 300 s.

From the stock solution of nicotine (100 mM) and choline (100 mM), a working solution for each agonist was prepared at 120 µM in MilliQ water. The final concentration of nicotine or choline in each well of the plate that contains the SH- SY5Y cells was 30 µM.

2.2.6 Activation of α3* nAChR by nicotine

The activation of the α3* nAChR by nicotine (30 µM) was measured using the ‘two-addition’ protocol mentioned above. This protocol is described in Section 2.2.8 and a schematic of the protocol is shown in Figure 2.1A.

The reagent plate with venoms or venom fractions takes a long time to prepare when testing 71 venoms at different concentrations. To avoid loss of incubation time once cells were loaded with calcium 4-dye, the reagent plate with venoms or venom fractions was prepared before loading the cells.

After loading the SH-SY5Y cells, a fluorescent baseline was recorded for 5 s then PSS was added to the cells (addition 1), followed by readings each second for 300s. During this recording time, calcium responses are not expected to be different from the baseline. Finally, nicotine was added (addition 2) to yield a final concentration in the assay of 30 µM, and the fluorescence was measured for 300 s. At the end of this time there is a transient increase in the fluorescent calcium signal that corresponds to activation of α3* by the agonist nicotine.

To confirm that the observed calcium responses were elicited by activation of nAChRs, d-tubocurarine, a non-selective blocker of nAChRs (Wenningmann and Dilger, 2001), was added to the SH-SY5Y cells before stimulation of α3* by nicotine. A stock solution of d-tubocurarine was prepared at 100 mM in MilliQ water and its final concentration in the assay was 30 µM. All FLIPR experiments included a positive control (calcium response elicited by activation of α3* by nicotine) and a negative control that corresponds to inhibition of these responses by d-tubocurarine.

Figure 2.1B shows the ‘two-addition’ protocol that was employed to detect blockers of the calcium responses elicited by α3* nAChRs. After recording the fluorescent baseline for 5 s, d-tubocurarine was added to the cells (addition 1), followed readings each second for 300 s. No calcium responses are expected to occur during this time. Then, a second fluorescent baseline was read for 5 s, followed stimulation of α3* by nicotine (addition 2), and the fluorescent calcium responses were measured for another 300 s. As expected, no increase in the fluorescent calcium response was detected after nicotine addition. Figure 2.1D shows a typical recording of the inhibition of nicotine-induced calcium responses by d-tubocurarine.

The calcium responses caused by stimulation of α3* with nicotine were generally smaller than those induced by choline. Furthermore, the kinetics of activation and inactivation were slow and they can be detected on the FLIPR assay without incubation of the SH-SY5Y cells with PNU-120596 (Figure 2.1C).

2.2.7 Activation of α7 nAChR by choline in the presence of the positive allosteric modulator PNU-120596

The α7 nAChRs have fast kinetics of deactivation and long-lasting desensitization that makes it difficult to measure its activity (Lopez, 1998). An experimental strategy to detect its activity in SH-SY5Y cells was the addition of PNU-120596 (10 µM) in the Calcium 4-no-wash-dye solution, followed by incubation in SH-SY5Y cells for 30 min at 37°C and 5% CO2 (Vetter and Lewis, 2010). The stock solution of PNU-120596 was prepared at 100 mM in MilliQ water and 5% (v/v) DMSO, and stored in aliquots of 20 µL at –20°C. Aliquots can be reused if refrozen at –20°C immediately post-experiment.

The activation of the α7 nAChRs by choline (30 µM final concentration) was measured basically using the same protocol as that for the study of the α3* subtype. Briefly, after reading the fluorescent baseline for 5 s, PSS was added (addition 1) to the cells and fluorescence measurements were made for a further 300s. Then, the agonist choline was added (addition 2) to stimulate the α7 nAChRs, followed by fluorescence measurements for another 300 s. As expected, the activation of fluorescence nAChRs by choline elicited an increase in the fluorescent calcium response that corresponds to an increase in the concentration of intracellular calcium.

A typical trace of increase of intracellular calcium after stimulation of α7 nAChR by choline is shown in Figure 2.1E. The positive allosteric modulator PNU-120596 (10 µM) was included in assays to delay the deactivation kinetics of α7 nAChR and allow recording of the fluorescent Ca2+ responses (Vetter and Lewis, 2010).The calcium responses are mostly caused by activation of α7 when PNU and choline were used, although subtypes α3β2/α3β4/α5 are also activated and they were part of the calcium response (Vetter and Lewis, 2010).

The same ‘two-addition’ protocol was used to confirm whether the calcium responses were elicited by α7 nAChR. Briefly, d-tubocurarine (30 µM final concentration) was added to the cells (addition 1) and the fluorescent calcium

response was read for 300 s, then choline was added (addition 2) to stimulate the receptor, and a final reading of the fluorescence was carried out for another 300 s. As expected, d-tubocurarine blocked the calcium response elicited by activation of α7 nAChR (Figure 2.1F). The following Section 2.2.8 provides an explanation of how the FLIPR assays were used to identify nAChR blockers and agonists in arachnid venoms.

2.2.8 Identification of blockers and agonists for α3* and α7 nAChR using the FLIPRTETRA system

A ‘two-addition’ protocol was used to detect the presence of α3* and α7 blockers in venoms or venom fractions. Crude venoms, venom fractions, or pure toxin were added first to the cells, and then the fluorescence was measured for 300 s. Those venoms or venom fractions that contain blockers inhibited the calcium responses elicited by either α3* or α7 when they were stimulated by nicotine or choline (addition 2), respectively.

On the other hand, an agonist compound contained in the venoms or venom fractions caused an increase in the fluorescent calcium response before stimulation of the receptors by agonists (addition 2). Inhibition by d-tubocurarine (30 µM) was used to confirm that the observed calcium response was caused by activation of the receptor rather than activation of voltage-gated calcium channels (Sousa, 2013) or other endogenously expressed GPCRs or ion channels coupled to an increase in intracellular Ca2+.

A)# B)#

PSS+ agonist+ d(tubocurarine+ agonist+

addition+1+ addition+2+ Ca2+ Ca2++ addition+1+ addition+2+ dye+ dye+

SH(Sy5Y+cells+ SH(Sy5Y+cells+ C)( D)( 4" 4" addi5on(1( addi5on(2( addi5on(1( addi5on(2( (PSS)( (Nico5ne)( (d=tubocurarine)( (Nico5ne)( 3" 3"

2" 2" (RLU)( (RLU)( 1" 1"

0" 0" Maximun(calcium(response( 1" 151" 301" 451" 601" Maximun(calcium(response( 1" 151" 301" 451" 601" E)( 5me((seconds)( F)( 5me((seconds)( 8" addi5on(1( addi5on(2( 8" addi5on(1( addi5on(2( (PSS)( (Choline)( (d=tubocurarine)( (Choline)( 6" 6"

4" 4" (RLU)( (RLU)( 2" 2"

Maximun(calcium(response( 0" 1" 151" 301" 451" 601" Maximun(calcium(response( 1" 151" 301" 451" 601" 5me((seconds)( 5me((seconds)(

Figure 2.1. Basic two-addition FLIPR protocol to measure calcium responses evoked by activation of α7 or α3* nAChRs. (A) Schematic of the protocol to measure calcium responses caused by activation of the heteromeric α3* nAChR: SH-SY5Y cells were plated onto a 384-well plate and loaded with calcium 4-dye after discarding the RPMI medium. The method is based on two additions, and control experiments were always included in every assay. PSS or test compound was added first to the SH-SY5Y cells and the fluorescence measured for 300 s, then the α3* nAChR or α7 nAChR were stimulated by nicotine or choline (addition 2), respectively, and the resulting calcium responses were measured for another 300 s. (B) Schematic of protocol used to confirm that the calcium responses were elicited by activation of nAChRs. If the calcium responses observed in (A) were elicited by activation of α3* or α7 nAChRs they should be inhibited when d-tubocurarine (addition 1) was added before agonist addition, nicotine or choline, respectively. (C) Calcium responses recorded by the FLIPRTETRA system when α3* nAChRs were stimulated by nicotine (addition 2). (D) The calcium response was elicited by the activation of α3* because the addition of d-tubocurarine (addition 1) inhibited the calcium response elicited by the receptor when it was stimulated by nicotine (addition 2). (E) A typical calcium response elicited by activation of α7 nAChR using choline as agonist (addition 2). (F) The calcium response elicited by activation of subtype α7 nAChR using choline was inhibited when d-tubocurarine was added before the addition of choline (addition 2). The traces of control calcium responses are an average of one experiment with n = 4.

A)( B)(

Venom/( fraction(venom( agonist( Venom/( fraction(venom( agonist(

addition(1( addition(2( Ca2+( addition(1( Ca2+d addition(2( dye(( ye( +bloc ker( SH+Sy5Y(cells(

C) crude venom D)

H. hainanum 1 µg crude venom addition 1 H. hainanum 1 µg addition 1 choline 30 µM 3 choline 30 µM 3 addition 2 addition 2

2 2

1 1

Maximun calcium response (RLU) response calcium Maximun 0 (RLU) response calcium Maximun 0 1 151 301 451 601 1 151 301 451 601 time (seconds) time (seconds)

Figure 2.2. ‘Two-addition’ FLIPR protocol used to identify nAChR agonists. (A) Venom was added (addition 1), then fluorescent responses were measured for 300 s. After the cells were stimulated by nicotine or choline (addition 2), fluorescence measurements were continued for another 300 s. If any venom compound contains a blocker, the fluorescent calcium response induced by nicotine or choline (addition 2) will be inhibited. (B) If any venom compound contains an agonist there will be an increase in the fluorescence after addition 1. d-tubocurarine (30 µM) is added during the loading process, then venom is added (addition 1), if the following response is inhibited then the response was caused by activation of nAChRs. Finally, addition 2 occurs, followed a second reading for 300 s. (C) 1 µg of crude venom from H. hainanum was added (addition 1) to the SH- SY5Y cells, then the calcium fluorescent response was measured for 300 s. After choline was added (addition 2), the fluorescent signal was measured for other 300 s. (D) To confirm that the increase in fluorescence after venom addition was caused by cholinergic activity, d-tubocurarine (30 µM) was included in the Calcium 4-dye solution, and added to the cells. Then, the crude venom (H. hainanum) was added (addition 1) and the fluorescent response was measured. The response was inhibited, and after choline addition (addition 2) the response remained inhibited. The venom was assayed in the presence of PNU-120596 (10 µM). The traces shown in panels (C) and (D) are averages with n = 3.

2.2.9 Data analysis

The 71 spider venoms were tested in triplicate in at least three independent experiments. All data was analysed as followed: Raw fluorescence values obtained after venom addition were subtracted from the baseline fluorescence values measured during the reads between 0 and 5 s after starting the assay. In addition, the raw data readings obtained after stimulation of the receptors by nicotine or choline were subtracted from the baseline raw fluorescent values registered between 300 and 305 s. Then, the resulting raw fluorescent reads were converted to maximum calcium responses by calculating in relative light units the maximum values reached over the baseline reads using the ScreenWorks, software, version 3.1. Once the data was corrected using the baseline fluorescent reads, it was exported and analysed using the software Prism, version 5.0.

Data are expressed as mean ± standard error of the mean (SEM) from n = 3 independent experiments or as the mean ± SEM of the values obtained in the three independent experiments with a total n = 9.

FLIPR data were normalized to the maximum fluorescence calcium response evoked by activation of α3* nAChR with nicotine (30 µM) or by activation of α7 nAChR with choline (30 µM) and indicated as FMax (%)

The fluorescent traces from the FLIPR data are reported as calcium response over baseline that is corrected previously by subtracting the average of the fluorescence elicited by the negative control (only components of buffer PPS) using ScreenWorks software. This eliminates the artefacts caused by buffer addition.

2.2.10 Statistical Analyses

Comparative analysis of the mean of calcium responses induced by crude venoms of each species on α3* and α7 nAChRs versus control responses was carried out

through a standard one-way ANOVA and non-parametric statistical analysis, using the analysis tool of GraphPad Prism®, version 5.0.

2.3 RESULTS

A high-throughput FLIPR screen was carried out to examine the effect of 12 araneomorph and 59 mygalomorph venoms (see Appendix I, Table 1-2 for a list of species) on the calcium responses evoked by activation of α3* and α7 nAChRs endogenously expressed in SH-SY5Y cells. Following the initial screen, at least two additional screens were performed to enhance the statistical significance of the data, although the data for each species was highly reproducible, indicating that the venom composition of each species is constant across different pools of venom. To facilitate analysis of the 71 spider venoms, the resulting data from araneomorph spiders is presented first in accordance with an Araneae phylogenetic tree. The second group corresponds to mygalomorph spiders, and the data was compared in accordance with the geographical origin (state and country) of each species. Then, common and different elements between both suborders were described and later discussed (see discussion section) to determine possible mechanisms of evolution of modulators for vertebrate neuronal cholinergic receptors.

2.3.1 Screening of 71 spider venoms using the high-throughput FLIPRTETRA system

In all cases, unless specified, 2 µg of crude venom from each spider was tested against α3* and α7 nAChRs endogenously expressed in human SH-SY5Y cells. For the screening of these 71 spider venoms, a two-addition protocol was used (for further details see the methods section).

2.3.2 Screening of venoms from “modern” spiders (Infraorder Araneomorphae)

A phylogenetic tree of the infraorder Araneomorphae is shown in Figure 2.3, each superfamily is indicated on the main branches of the tree, and the name of each

representative spider considered in this study is indicated beside the superfamily name.

To describe the results obtained with araneomorph spider venoms, the spiders were group in three groups, and are described in order of evolutionary appearance. The first group corresponds to the superfamily Austrochiloidea, and the one species from this taxon is Hickmania troglodytes, found in Tasmania. The spider makes and hangs beneath extensive sheet webs, and occasionally wraps prey after biting. The crude venom of H. troglodytes at 2 µg did not modify the calcium responses evoked in SH-SY5Y cells by the agonists nicotine and choline (Figure 2.3A).

According to the tree, the superfamily Araneoidea appeared in the Cretaceous 145 million years ago after the family Austrochiloidea. Crude venoms from four exemplars of the superfamily Araneoidea were used in this work: Cyrtophora moluccensis, Eriophora transmarina, Nephila pilipes and Nephila plumipes. This is a large taxon that consists of approximately 29% of the spiders described so far (Griswold, 2005). Venom from E. transmarina and the two Nephila species blocked both the α3* and α7 nAChRs with high potency whereas C. moluccensis specifically blocked only the heteromeric α3* nAChRs (Figure 2.3A). This venom had no effect on the choline response (Figure 2.3B), indicating that blockers are likely to target mainly the α3*. In contrast, N. pilipes and N. plumipes venom had equally potent inhibitory effects on α3* and α7 nAChRs, see Figures 2.3A and 2.3B, respectively.

A) Jurassic( Cretaceous( 150 RTA Clade 145. 200. 65. 2. ma. Partial effect No effect on nAChRs on nAChRs 100 Austrochiloidea. ** H.#troglodytes# ** * 50 ** **** C.#Moluccensis# *** ******** # 0 **** E.#transmarina# Normalized calcium response (%) Nephilidae. N.#Pilipes#. Araneoidea. N.#plumites# . 1 . 2

N. pilipes Therididae. N. plumites H. jugulans H. troglodytes Lycosidae sp. C. moluccensisE. transmarina M. australianus SparassidaeSparassidae sp sp AncylometesV. sylvestriformis sp.

d-tubocurarine (30 µM) 2 µg crude venom / araneomorph spider

B) PNU-120596 (10 µM) RTA Clade Sparassidae. Sparassidae#sp.1# 150 ENTELEGYNAE( Sparassidae#sp.2# Partial No effect on ARANEOMORPHAE(( effect on

(“modern”(SPIDERS)( nAChRs nAChRs 100

* Dionychans. H.#Jugulans# 50 ****

*** Ctenidae. **** **** Ancylometes#sp.# **** ****

V.#sylvestriformis# Normalized calcium response (%) 0

. 1 . 2 Lycosidae. Lycosidae#sp.## N. pilipes RTA.clade. N. plumites H. jugulans H. troglodytes Lycosidae sp. C. moluccensisE. transmarina AncylometesM. australianussp. Pisauridae. M.#australianus# SparassidaeSparassidae sp sp V. sylvestriformis

d-tubocurarine (30 µM) Agelenidae. 2 µg crude venom / araneomorph spider Figure 2.3. Screening of venoms from “modern” spiders against human α3* and α7 nAChRs using a high-throughput FLIPR assay. The left panel shows a phylogenetic tree of the infraorder Araneomorphae. The name of the spider species whose venoms were used in this study is denoted next to the branch of each family. The tree was adapted from (Coddington and Levi, 1991); ma = millions of years. On the right panel is shown: A) 2 µg of crude venoms of the twelve araneomorph spiders were assayed in the absence of PNU- 120596 (10 µM). B) 2 µg of venoms from the same araneomorph spiders were assayed in the presence of PNU-120596 (10 µM). Venom from H. troglodytes (Family Austrochiloidea) had no effect on α3* and α7 nAChRs in the absence and presence of PNU-120596, respectively. Venoms from E. transmarina, N. pilipes, and N. plumites (Superfamily araneoidea) had a potent blocking effect on α3* and α7 subtypes with the exception of venom from C. moluccensis that only had potent blocking effect on α3* nAChRs in the absence of PNU-120596. As seen on the phylogenetic tree (left panel), spiders from the RTA clade evolved after the araneoid spiders; and venoms from Sparassidae sp. (1) and (2), and Lycosidae sp. showed a partial blocking effect on α3* and α7 subtypes without and with PNU-120596, respectively, while venoms from H. jugulans, and Ancylometes sp. had no effect on nAChRs, although venom from V. sylvestriformis only had a partial blocking effect on α3* nAChRs in the absence of PNU-120596. Venom from M. australianus blocked completely α3* and α7 nAChRs in the absence and presence of PNU-120596, respectively. The data on panel A was normalized in relation to the maximum calcium response elicited with nicotine addition (30 µM) (positive control) in the absence of PNU-120596 (10 µM) and expressed in percentage while data from panel B was normalised in relation to the maximum

calcium response elicited with choline addition (30 µM) (positive control) in the presence of PNU-120596. The negative control corresponds to the inhibition of the calcium response of nAChRs by d-tubocurarine addition at 30 µM in the absence or presence of PNU-120596 (10 µM). The values are averages of three independent experiments with n = 3. Dotted lines represent the control response in the absence of venoms. Using the unpaired t-test, the means between the effect of each venom and the control response without venom were compared. The number of (*) indicates the degree of significance at P < 0.05 probability level. Those values that lack of significance do not have the symbol (*).

The third group of spiders used herein belong to the Retrolateral Tibial Apophysis (RTA) clade, and venoms from six exemplars of this group were assayed. Table 2.1 list these species and their geographic distribution. Most species are from Australasia, although the two ctenid species (Ancylometes sp. and Viridasius sylvestriformis) are from Central and South America. Of the six venoms tested, the two sparassid venoms (Sparassidae sp. 1 and Sparassidae sp. 2), and the one lycosid venom (Lycosidae sp.) showed to have a partial blocking effect on α3* and α7, while one sparassid venom (H. jugulans), and the two ctenid venoms had no effect on vertebrate α3* and α7 nAChRs, although venom from V. sylvestriformis had a partial blocking effect on α3*. In striking contrast, venom from the giant water spider Megadolomedes australianus (family Pisauridae) blocked both nAChR subtypes with high potency.

Table 2.1. Taxonomic distribution of nAChR modulators from araneomorph spider venoms. Two species Sparassidae sp. were assayed; the numbers 1 and 2 indicate they were collected in two different areas in PNG. Venoms with high blocking effect are indicated with two symbols denoted as √√, while those with partial effect are indicated with one √. The green area indicates that other effects on the venom is not observed. Spider Effect on human neuronal α3* and α7 nAChRs Block of α3* Block of No Location & α7 α3* effect Infraorder Araneomorphae Family Austrochiloidea Hickmania √ Tasmania, Australia troglodytes Family Araneoidea Cyrtophora √√ Gardens across South moluccensis East and coast from Queensland, Northern and western Australia Eriophora √√ Gardens across transmarina eastern and southern Australia Nephila pilipes √√ Gardens Northern Australia, PNG, China, Japan Nephila plumites √√ Sydney region RTA Clade Family Sparassidae Heteropoda √ Queensland, Australia jugulans Sparassidae sp. √ Papua New Guinea 1 Sparassidae sp. √ Papua New Guinea 2 Family Ctenidae Ancylometes sp. √ Central and South America Viridasius √ Amazon, South sylvestriformis America Family Lycosidae Lycosidae sp. √ Western Australia Family Pisauridae Megadolomedes √√ South Eastern australianus Australia

2.3.3 Screening of venoms from “primitive” spiders (Infraorder Mygalomorphae)

Mygalomorphs are the oldest group of spiders. They are known as “primitive” spiders because they retain some ancestral morphological features, such as downward pointing chelicera (Coddington and Levi, 1991). In this study, venoms were derived from 59 mygalomorph spiders. The infraorder Mygalomorphae is comprised of 15 taxonomic families (Bond, 2012), of which two were considered here, namely Hexathelidae (1 species) and Theraphosidae (58 species).

Four mygalomorph venoms were found to contain nAChR agonists (Section 2.9.4), 24 venoms selectively blocked α3* nAChRs (Section 2.9.5), 26 venoms inhibited both α3* and α7 nAChRs (Section 2.9.6), and four mygalomorph venoms were found to have no effect on α3* or α7 nAChRs (Section 2.9.7). Each of these four groups is discussed in the following sections.

2.3.4 Identification of α7/α3* nAChR agonists in the venom of theraphosid spiders

The two-addition protocol was used to identify potential agonists of vertebrate nAChRs. Addition of venom (1 or 2 µg per well) from Haplopelma hainanum, Pamphobeteus sp., Sericopelma rubronites and Xenesthis sp. caused an increased calcium response in the FLIPR assay both in the absence and presence of PNU (Table 2.2). Based in a previous FLIPR assay performed by Dr Vetter (data not showed), the fluorescent responses evoked by 1 µg of venom from H. hainanum, Pamphobeteus sp., Sericopelma rubronites and in the absence of PNU were not abolished with addition of d-tubocurarine, indicating that these venoms lack of agonists of α3* nAChRs. Interestingly, the same venoms at 1 µg were potentiated in the presence of PNU, although were no assayed in the presence of d-tubocurarine (data not showed). Based on these previous results, I evaluated if the calcium responses evoked for the same venoms at 1 µg and in the presence of PNU were related to activation of the α7 nAChRs (Table 2.3). The venom responses from H. hainanum, Pamphobeteus sp., S. rubronites, and Xenesthis sp.

were assayed with the previous addition of d-tubocurarine (Figure 2.4A-D). The traces derived from the FLIPRTETRA system are presented to show the strategy used to identify agonists in these venoms. After crude venom addition (addition 1) at 1 µg, there was an increase of the calcium response that was extinguished at the end of the 300 s (Figure 2.4B-D) although the venom from H. hainanum caused an increase of the fluorescence that was sustained during this period of time (0–300 s) (Figure 2.4A). Then, after choline addition (addition 2), the calcium response was not different from that evoked by stimulation of the α7 nAChRs by choline (positive control) (Figure 2.4B-D). Interestingly, venom from H. hainanum caused complete inhibition of the calcium response evoked by choline stimulation of the α7 nAChR (see Figure 2.4A). A further analysis was not carried out to determine if the differences of calcium responses in the presence of PNU and addition of d-tubocurarine were statistically significant due to the limitation of amount of these venoms, although the venom from H. hainanum is likely to have an agonist compound.

crude venom H. hainanum 1 µg + crude venom Pamphobeteus sp . 1 µg + PNU-120596 (10 µM) PNU-120596 (10 µM) crude venom Pamphobeteus sp. 1 µg + crude venom H. hainanum 1 µg B)# 4 A)# + PNU-120596 (10 µM) + d-tubocurarine (30 µM) PNU-120596 (10 µM) + d-tubocurarine (30 µM) 3 3

2 2

(RLU) 1 1 (RLU) 0 0 1 151 301 451 601 Maximun calcium response calciumresponse Maximun 1 151 301 451 601 time (seconds)

time (seconds) response calcium Maximun time (seconds)

crude venom S. rubronitens 1 µg + PNU-120596 (10 µM) crude venom S. rubronitens 1 µg + crude venom Xenesthis sp. 1 µg +

PNU-120596 (10 µM) + d-tubocurarine (30 µM) C)# 3 D)# PNU-120596 (10 µM) crude venom Xenesthis sp. 1 µg + PNU-120596 (10 µM) 3 + d-tubocurarine (30 µM)

2 2

1 1 (RLU)

0 0 1 151 301 451 601 1 151 301 451 601 Maximun calcium response (RLU) response calciumMaximun

time (seconds) response calcium Maximun time (seconds) Figure 2.4. Agonism of human nAChRs by theraphosid venoms. All traces are averages of the maximum calcium responses over baseline (with n = 3) elicited by crude venom from (A) Haplopelma hainanum, (B) Pamphobeteus sp., (C) Sericopelma rubronites, and (D) Xenesthis sp.. All responses elicited between 5 s and 300 s are caused by venom addition. The responses after choline addition were measured between 300 s and 600 s. All crude venoms were tested at 1 µg/well in the presence of PNU-120596 (10 µM). Dotted lines of each graphic correspond to averages of calcium responses over baseline (with n=3) elicited by the venoms in the presence of d-tubocurarine (30 µM) and PNU-120596 (10 µM).

Table 2.2. Apparent agonistic effect caused by venoms from four theraphosid spiders. After venom addition, the fluorescent calcium responses evoked by nAChRs in the absence or presence of PNU were converter into area under curve using ScreenWorks software, version 3.1. The values are averages of three independent experiments with n = 3. The mean of the control nicotine response was 509 ± 50.40 while the control calcium response elicited by choline was 1088 ± 38.31 Spider Calcium response Calcium response over over baseline after baseline after venom addition venom addition at at 2 µg when PNU-120596 (10 2 µg µM) was included in the assay Haplopelma 345.6 ± 136 822.2 ± 127 hainanum Pamphobeteus sp. 542.2 ± 164.2 517 ± 179 Sericopelma 241.1 ± 101.5 696 ± 137 rubronites Xenesthis sp. 325.5 ± 151.4 678.8 ± 80

Table 2.3. Mygalomorph spiders with venoms that are likely to contain agonists of α7 nAChRs. N.D indicates the effect was not determined and a further analysis is required to confirm that the venom effect is via activation of α7 nAChRs in the presence of PNU. Location and affect on human α7 nAChRs Spider Agonistic effect Location Haplopelma hainanum √ Southern China and Vietnam Pamphobeteus (platyomma) N.D Ecuador sp. (F) Sericopelma rubronites (F) N.D Panama Xenesthis sp. (F) “white N.D Colombia colour”

2.3.5 Venoms that selectively inhibit vertebrate α3* nAChRs

24 theraphosid venoms were found to selectively block human α3* nAChRs (Figure 2.5A). The 24 species are listed in Table 2.4 along with their geographic distribution. These venoms did not have a substantial effect on calcium responses in the FLIPR assay when PNU-120596 was included (Figure 2.5B).

150 A) 150 B) PNU-120596 (10 µM)

100 100 *** **** **** ** *** *** 50 **** **** *** **** **** 50 *** **** **** **** **** *** **** **** *** ** **** **** **** **** **** **** **** ** Normalized calcium response (%) Normalized calcium response (%) 0 0 9-I 9-I 6-F 5-E 2-B 3-C 6-F 1-A 4-D 7-G 8-H 5-E 2-B 3-C 1-A 4-D 7-G 8-H 10-J 16-P 19-S 10-J 12-L 20-T 18-R 11-K 14-N 15-O 17-Q 21-U 22-V 24-X 16-P 19-S 12-L 20-T 13-M 18-R 11-K 23-W 14-N 15-O 17-Q 21-U 22-V 24-X 13-M 23-W -tubocurarine d -tubocurarine d

Figure 2.5. Theraphosid spider venoms that preferentially block α3 nAChRs. First screening of venom from “primitive” spiders against human α3* and α7 nAChRs was carried out by using a high-throughout calcium assay. The calcium responses elicited by crude venom was calculated as the area under fluorescence curve normalized to the maximum α3* response evoked by nicotine (A) or stimulation of α7 by choline (B). Each value corresponds to averages of normalized calcium responses over baseline of at least three independent experiments and n = 3. All venoms were tested at 2 µg/well, in the absence (A) and presence (B) of PNU-120596. For a species identification see Table 2.4. Dotted lines represent the control response in the absence of venoms. On graphic A and B, the control responses corresponds to stimulation of α3* with nicotine (30 µM), and α7 choline (30 µM), respectively. Using one-way ANOVA, the means between each venom and the control responses (nicotine response were compared). The number of (*) indicates the degree of significance at P < 0.05 probability level. Those values that lack of significance do not have the symbol (*).

Table 2.4. Theraphosid spiders with venoms that selectively block α3* nAChRs. Species with characteristic colors are indicated. (M) and (F) refer to gender. The numbers beside the name of each theraphosid spider indicates that they are individuals from the same species, although were collected from distinct localities. Number Theraphosid Geographic distribution Avicularia metallica Guyana (coast), Suriname 1-A (F) (coast) Avicularia ulrichea 2-B All Brazil (F) Avicularia sp. 1 3-C Peru-Amazonas (M) “purple” Coremiocnemis 4-D Australia tropix

5-E Chilobrachys sp. 1 Malaysia-Penang island Pampas from Argentina and Grammostola 6-F Paraguay. Brazil south and grossa 8 (F) north of Paraguay Grammostola 7-G Brazilian south, Entre Rios iheringi (M) Euathlus sp. 1 (F) 8-H Chile “green” Euathlus sp. 2 (F) 9-1 Chile “red” Euathlus vulpinus 10-J Chile highland (F) “crème” Orphnaecus 11-K Philippines philippinus 12-L Orphnaecus sp. Philippines-Paqnai-island Paraphysa scrofa 13-M Chile (F) West Papua New-Guinea 14-N Pelinobius muticus (PNG) Poecilotheria 15-O Sri Lanka, India striatus Pterinuchilus Central, Eastern, Southern 16-P murinus Africa Selenocosmia Australia-East Coast 17-Q crassipes Queensland Selenocosmia 18-R East PNG effera Selenocosmia sp. 19-S PNG 1 Selenocosmia sp. 20-T PNG 2 Selenotholus 21-U Australia foelschei Thrixopelma 22-V Peru cyaneolum (F) 23-W Thrixopelma sp. (F) Peru – Cajamarca Thrixopelma 24-X Peru – Lagunas lagunas sp. (F)

2.3.6 Venoms that non-selectively block both α3* and α7 nAChRs

26 theraphosid venoms were found to non-selectively block both α3* (Figure 2.6A) and α7 nAChRs (Figure 2.6B). The 26 species from which these venoms were obtained are listed in Table 2.5 along with their geographic distribution.

150 150 A) B) PNU-120596 (10 µM)

100 100 * * ** **** ********* ** 50 50 *** **** **** **** *** *** *** * **** **** **** *** *** *** *** *** **** **** **** **** **** *** *** *** *** *** **** **** ******** 0 *** *** ************ *** ****************** ****** 0 Normalized calcium response (%) Normalized calcium response (%)

1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

-tubocurarine -tubocurarine d d Figure 2.6. Many theraphosid venoms non-selectively block both α3* and α7 nAChRs. Normalised calcium responses obtained from FLIPR screen of venoms in the absence (A) and presence (B) of PNU-120596. Each value corresponds to averages of normalized calcium responses over baseline of at least three independent experiments and n = 3. For a list of species see Table 2.5. Dotted lines represent the control response in the absence of venoms. On graphic A and B, the control responses corresponds to stimulation of α3* with nicotine (30 µM), and α7 choline (30 µM), respectively. Using one-way ANOVA, the means between each venom and the control responses (nicotine response were compared). The number of (*) indicates the degree of significance at P < 0.05 probability level. Those values that lack of significance do not have the symbol (*).

Table 2.5. Theraphosid spiders with venoms that non-selectively block both α3* and α7 nAChRs. (M) and (F) refer to gender. The numbers beside the name of each theraphosid spider indicates that they are individuals from the same species, although were collected from distinct localities. Numb Theraphosid species Location er Acanthoscurria cf 1 Peru, Bolivia insubtilis (F) Namibia, Zimbabwe, Ceratogyrus 2 Mozambique, South bechunanicus Africa 3 Ceratogyrus ezendami Mozambique Zimbabwe, 4 Ceratogyrus marshalli Mozambique North India-Meghalaya 5 Chilobrachys sp. 2 (State) Eupalaestrus Argentina, Brazil, 6 campestratus (F) Paraguay Argentina, Brazil, 7 Ephebopus uatuman (F) Paraguay 8 Grammostola porteri (F) All Chilean coast Bolivia, Argentina, and 9 Grammostola rosea (M) all Chilean coast 10 Grammostola sp. 1 (F) (Chilean North) 11 Grammostola sp. 2 (F) Concepcion, Chile 12 Grammostola sp.3 (F) Maule River, Chile 13 Holothele sp.1 (F) Carabobo, Colombia 14 Holothele sp. 2 (F) Peru 15 Idiothele nigrofulva South Africa 16 Lasiodorides striatus (F) Brazil 17 Lasiodora striatipes (F) Brazil Savannahs from Brazil 18 Nhandu chromatus 5 (F) and Paraguay 19 Nhandu tripepii (M) State Para, Brazil 20 Nhandu vulpinus Brazilian South Phormictopus Dominican Republic, 21 cancerides (F) Cuba, Haiti 22 Pamphobeteus cf petersi Ecuador, Peru Psalmopoeus 23 Trinidad and Tobago cambridgei (F) Psalmopoeus 24 Venezuela langenbucheri (F) 25 Psalmopoeus pulcher Panama

(F) 26 Theraphosinae sp. (F) Chile

2.3.7 Theraphosid venoms with no apparent effect on nAChRs

Figure 2.7 summarizes FLIPR data obtained with venom from four American theraphosids (Aphonopelma crinitum, Avicularia sp., Homoeomma sp. and Pseudhapalopus sp.; Table 2.6). These venoms had only a minor effect on α3* nAChRs (Figure 2.7A) and no effect on α7 nAChRs (Figure 2.7B).

150 150 A) B) PNU-120596 10 µM

* 100 100 **** *** ****

50 50 ****

**** Normalized response (%) response Normalized 0 (%) response Normalized 0

cularia sp. 2 Avicularia Homoeommasp. 2 sp. Avi Homoeomma sp. Pseudaphalopus sp. Pseudhapalopus sp. d-tubocurarineAphonopelma (30 µM) crinitum d-tubocurarineAphonopelma (30 µM) crinitum Figure 2.7. Four tarantula venoms without potent modulators of α3* or α7 nAChRs. Normalized FLIPR responses resulting from a screen of crude venom against human α3* (A) and α7 nAChRs (B). All venoms were tested at 2 µg/well. Each value corresponds to averages of normalized calcium responses over baseline of at least three independent experiments and with n = 3. For legend see table 2.5. Dotted lines represent the control response in the absence of venoms. On graphic A and B, the control responses corresponds to stimulation of α3* with nicotine (30 µM), and α7 choline (30 µM), respectively. Using one-way ANOVA, the means between each venom and the control responses (nicotine response were compared. The number of (*) indicates the degree of significance at P < 0.05 probability level. Those values that lack of significance do not have the symbol (*).

Table 2.6. Venoms from American theraphosids that lack of modulators of human α3* and α7 nAChRs. Theraphosid spider Geographic location Aphonopelma crinitum Northern Mexico (F) Avicularia sp. (F) 2 Bolivia-Rio Madre Homoeomma sp. (blue) Peru Pseudhapalopus sp. South America (F)

2.3.8 Effect of non-theraphosid mygalomorph venom on vertebrate nAChRs

Unfortunately, I only had access to one non-theraphosid mygalomorph venom for this study, namely Macrothele gigas from the family Hexathelidae (see condensed phylogenetic tree in Figure 2.8). M. gigas venom blocked both the α3* and α7 nAChR subtypes. At 2 µg, M. gigas venom reduced the nicotine-evoked calcium response in the FLIPR assay by 28 ± 22% and the response evoked by choline in the presence of PNU-120596 by 46 ± 27%.

2.3.9 Taxonomic distribution of nAChR modulators within Araneae

Figure 2.8 summarizes the data obtained from the high-throughput FLIPR-based screen of spider venoms against human nAChRs. Figure 2.8A shows the distribution of blockers, agonists and inactive venoms within the infra-orders Mygalomorphae and Araneomorphae, while Figure 2.8B shows the distribution of nAChR modulators mapped onto a condensed phylogenetic tree of spiders (order Araneae).

A) B)

Mygalomorphae Agonists no effect PALAEOZOIC MESOZOIC CAINOZOIC 6% 7% CARBONIFEROUS PERMIAN TRIASSIC JURASSIC CRETACEOUS PALEOGENE NGENE 359 299 251 200 145 65 23 2

MESOTHELAE NOT VENOMOUS NS MESOTHELAE Selective Non- ATYPIDAE block of α3* NS selective ARANEAE HEXATHELIDAE 41% block of α3* MYGLAMORPHAE and α7 THERAPHOSIDINA (primitive spiders) RASTELLOIDINA 46% NS OPISTHO- PALEOCRIBELLATE THELAE NS HAPLOGYNAE NS ARCHAEIDAE & HUTTONIDAE NS LAGONOMEGOPIDAE NEO- Araneomorphae Agonists Selective CRIBELLATE AUSTROCHILDAE 0% block of α3* ERESOIDAE 8% NS ENTELE- DEINOPIDAE NS no effect GYNAE ULOBORIDAE NOT VENOMOUS NS ARANEOMORPHAE 34% (modern spiders) ARANEOIDAE DIVIDED CRIBELLUM NS CLADE RTA CLADE

K No effect on human 3* or 7 nAChRs Selective inhibitor of human 3* nAChRs E No Y Non-selective inhibitor of human 3* and 7 nAChRs Non-selective agonist of human 3* and 7 nAChRs substancial non- Potent non- selective selective block of α3* block of α3* and α7 and α7 24% 34%

Figure 2.8. Summarise of the high-throughput screening of spider venoms. (A) Pie charts summarising the different types of nAChR modulators found in the venoms of mygalomorph spiders and araneomorph spiders. (B) Distribution of nAChRs modulators found in spider venoms mapped onto a condensed phylogenetic tree. Circles to the right of the tree denoting biological activities are sized according to the dominance of that pharmacology in the venom. NS denotes "not sampled".

2.4 DISCUSSION

Today, it is recognized that an imbalance of the cholinergic pathways in humans may lead to several pathologies (Picciotto, 2012). Therefore, nAChRs are now molecular targets; and several studies have been carrying out to understand their mechanisms of regulation, but few advances have been made (Bencherif, 2014; Pammolli, 2011; Russo, 2014). In the field of agronomy, invertebrate nAChRs are also targets of neonicotinoids that are used to control insect pests (Fischer, 2014; Matsuda, 2009); and these are one of the most successful classes of chemical insecticides ever developed, although it is no selective for insect pests, it has been found to be lethal to other insects of importance in agriculture such as pollinators (Moffat, 2015). Part of the problem is that many studies of cholinergic receptors are limited by the number and type of selective modulators for nAChRs (Bauwens, 2015; Gotti, 2007).

Since spiders are the most successful insect predators on the planet (for a review see (King and Hardy, 2013)), it would be surprising if they had not developed nAChR modulators as part of their venom arsenal. To date, however, there are very few reports of this pharmacological activity in spider venoms. We therefore decided to screen spider venoms for nAChR modulators. Since cloned insect nAChRs were not available, we performed screens against two subtypes of vertebrate nAChRs (α3* and α7) based on the hypothesis that nAChR modulators might also be useful for deterring vertebrate predators.

Several high-throughput FLIPR screenings were carried out to identify “hit” venoms that were to contain promising lead compounds to further use as research tools, or lead drugs. However, the data indicated that most of the spider venoms contained predominantly blockers highly potent for subtype α3* nAChRs, and blockers for α7 nAChRs; and a non-selective modulator was not considered to isolate because adverse effects of some agents under current study are associated with low selectivity for cholinergic receptors (Kalayci, 2013; Koga, 2014; Mihalak, 2006). Even when a toxin lead, which was expected not being acylpolyamine type but selective for either α3* or α7 nAChRs, was not identified,

the resulting data was suitable to further analyse the distribution of modulators across the phylogenetic tree of spiders.

Because spiders are venomous animals with more than 45,000 extant species spread across 114 taxonomic families (see http://www.wsc.nmbe.ch/statistics/), it was impossible in this study to sample broadly across this immense phylogenetic tree, so we instead chose 71 species spread across the major suborder Opisthothelae, which is comprised of the infraorders Araneomorphae ("modern" spiders") and Mygalomorphae ("primitive spiders"). Note that Mesothelae, the other taxonomic suborder with Araneae, comprises only ~100 extant species that are thought to be non-venomous (for a review, see (Xu, 2015)).

With the caveat that it is difficult to make firm conclusions in the absence of deeper taxonomic sampling across the phylogenetic tree, two trends emerge from the data summary shown in Figure 2.8; and the discussion is based on the trends:

1) Although nAChR modulators can be found in distantly related species across the phylogenetic tree, they were absent from 3 of the 6 venoms that we sampled from the RTA clade as well as the single member of the family Austrochilidae that we sampled. In addition, the 3 left venoms from the RTA clade had a partial blocking effect on nAChRs. This indicates that modulation of vertebrate nAChRs was either a basal pharmacology in spider venoms that was secondarily lost in some species, or that it arose independently numerous times during Araneae evolution. Deeper phylogenetic sampling in combination with biochemical characterisation of the nAChR modulators found in spider venom will be required to discriminate between these possibilities.

2) In this study, nAChR agonists, and specific blockers of the α3* subtype were only found in the venom of theraphosid spiders, the largest family of mygalomorph spiders with ~1000 species. Thus, mygalomorph venoms appear to be richer in nAChR modulators than araneomorph venoms.

From the total of 71 spider venoms used in the present work, 12 belonged to spiders of the suborder Araneomorphae. The resulting data indicated that

modulators for vertebrate nAChRs have evolved differently that were leading by driven forces of natural selection such as geographical isolation and interactions between preys and predators (Morgenstern, 2011; Zelanis, 2012). Probably, the reason of absence of modulators in the venom of H. troglodytes is because it lives in caves and underground drainages (Doran, 2001), and it is unlikely to encounter vertebrate predators. Furthermore, they do not forage for insects instead they use their long webs up to 1 metre of diameter to trap insects such as cave crickets.

The cheliceral structure of “modern” spiders evolved into an improved tool that can be moved sideways, unlike fangs from “primitive” spiders that are moved up and down 1. This feature in araneomorph spiders allows them grapping easily big preys such as small vertebrates without using their venoms (Nyffeler and Pusey, 2014). They attack their prey by crushing their fangs on the neck of preys that could be small fish and lizards (Nyffeler and Pusey, 2014). In contrast, in mygalomorph spiders, the position and movement of their cheliceraes do not allow mygalomorphs to hunt preys bigger than big insects. Instead, they preferentially use their venoms to immobilise them.

We could hypothesise that appearance of strong, and robust cheliceraes that are moved sideways as forceps and the reduction of the length of “modern’ spiders could lead to reduce the synthesis of modulators for vertebrate nAChRs in spider venom. For instance, spider venoms with no potent blocking effect or no effect on nAChRs belonged to spiders from the phylogenetic branch RTA clade, although in the same phylogenetic clade, modulators for nAChRs could randomly appear on other species. It can be evidenced when no substantial blocking effect on nAChRs was observed in venoms from Ancylometes sp., V. sylvestriformis (Ctenidae), and the lack of no modulators in venom from H. jugulans (Sparassidae), while venom from Megadolomedes australianus contain potent blockers for nAChRs. The two ctenid spiders studied in this research work have strong and long fangs that may be used as defence. Moreover, they are found at night at the edge of water sources in remote areas of the forest or the jungle (Nyffeler and Pusey, 2014)

1 Most information about the habitat, biology, and behaviour of spiders was obtained from web pages of The Australian Museum and http://www.arachne.org.au..

where they may have few contact to humans. Interestingly, M. australianus have strong and large fangs and also potent blockers for vertebrate nAChRs. We could speculate that toxins against nAChRs allowed that the spider could prey on fish, which can be double of its size, tadpoles and insects (Nyffeler and Pusey, 2014). An important physical adaptation in the spider is the presence of long legs that can sense the vibrations of water when fish are nearby. Nevertheless, M. australianus is relatively large because the length of its legs; and it may cause being seen easily by natural enemies, and toxins against vertebrate nAChRs may be an advantage.

Interestingly, our data showed that Araneomorph spiders from the araneoidea family produced venoms with potent antagonists for vertebrate α3* and α7 nAChRs. For instance, two Australasian species of the genus Nephila, N. pilipes and N. plumites showed to have in their venoms blockers for both α3* and α7 nAChRs. These compounds are likely to be acylpolyamines that have previously been reported in other Nephila species such as Nephila madagascariencis (for a review, see (Estrada, 2007)). Their preys are commonly insects, reviewed in (Harvey, 2007) that are caught in their webs; because they produce a very strong web between trees, on occasions small birds can get trap. Their natural enemies are small vertebrates such as hummibirds (Vollrath, 1985). It seems modulators against vertebrates could be evolved to protect themselves from them. Also, humans could be considered as predators. Some native tribes from villages and settlements from Thailand consider spiders of genus Nephila as food (for a review, see (Harvey, 2007). Sometimes N. pilipes is found in house gardens and N. plumipes can build their webs against buildings (for a review, see (Harvey, 2007)) and it makes these spiders more exposed to be grape from people; and it might explain why venoms from N. pilipes and N. plumites contain potent blockers for human nAChRs.

Mygalomorphae or “primitive” spiders

Because the suborder Mygalomorphae is considered an ancient monophyletic group (Coddington and Levi, 1991), and its oldest fossil was dated to the Triassic

(Selden and Gall, 1992), this phylogenetic group can be particularly useful to estimate the origin of modulators for vertebrate nAChRs.

From the 71 spider species used to initially searching for lead compounds, 59 venoms belonged to spiders of the suborder Mygalomorphae 2. The data showed that most of the venoms blocked both subtype of mammalian nAChRs, and it is likely the toxins were type acylpolyamines. Indeed the first acylpolyamine was discovered in theraphosid spiders (for a review, see (Estrada, 2007)).

The only hexathelid species assayed herein was Macrothele gigas. Its venom contained blockers for vertebrate α3* and α7 nAChRs. It also might contain an agonist for α3* but an extra analysis of the active fraction may be carried out to confirm the agonistic effect. It is likely blockers type acylpolyamines existed in the venom of primitive spiders since the Triassic period because the ancestor of M. gigas has been estimated to appear on that era. Interestingly, venoms from Hexathelidae and Mygalomoprhae contain agonists and antagonists. This similarity is not surprising because hexathelid and mygalomorph spiders are closely related. It also suggests that a common ancestor could contain modulators for vertebrate nAChRs in its venom. They are typically much larger than araneomorphs, and they are consequently more prone to predation from vertebrate predators such as birds, rodents and small reptiles.

On the other hand, a potential agonistic activity on nAChR was detected in venom from Haplopelma hainanum. This spider lives in burrows that can reach up to 1 meter of depth. Its venom is categorized as highly venomous. The data herein indicates this venom is likely to contain an agonistic component for nAChRs. Probably, native acetylcholine may be present in the venom in large amounts because there is a report of small molecules of a related species. For example, in the venom of Haplopelma lividum was detected the polyamine spermine, histamine, adenosine; and glutamic acid that was abundant in the venom fraction of low molecular weight molecules (Moore, 2009). The plasticity of toxins to evolve in accordance to environmental pressures, such as type of prey and type of

2 The data is summarised in Figure 2.8.

predators have been reported recently in the venom of Haplopelma hainanum (Zhang, 2015b). It was shown that its intraspecific venom variations of blockers for vertebrate nAChRs could be attributed to interaction gene-environment. In addition, the presence of agonist for α7 nAChRs in venoms from Pamphobeteus sp., Sericoplema rubronites and Xenesthis sp. remain to be confirmed. These tarantulas are usually trade as pets; and if the presence of agonists for α7 nAChRs is due to the constant interaction with humans when these are in captivity need to be evaluated. It would be interesting to further study venoms from these tarantulas that naturally inhabit in the forest and the Amazon in South America.

According to the results of the present thesis, spider venoms from the family theraphosidae, such as Selenocosmia crassipes, Selenocosmia effera, Selenocosmia sp (Papua New Guinea), Coremiocnemis tropix, Selenotholus foelschei contain potent toxins that block preferentially human neuronal α3* nAChRs. In Australia, Selenocosmia sp. is known to cause severe pain in humans and death in canines while venoms from Coremiocnemis tropix (spread out in North Australia) and Selenotholus foelschei (inhabit only in Western Australia) are shown to have a pre synaptic action at the neuromuscular junction in chicks (Herzig, 2008). The analysis of our data indicates that these venoms may contain compounds that preferentially act on the autonomic nervous system where the “ganglionic” nAChR subtypes are expressed, and mediate the pain observed in people bite by these spiders instead of targeting the nAChR muscle subtype.

On the other hand, theraphosid spiders from the new world studied herein showed to contain blockers for vertebrate cholinergic receptors. Scarcely information is available about venomic, diet and habitat of spiders from the new world, although, some works have been carried out in some spiders assayed in this work (Bertani and Fukushima, 2009).

In this study, the data showed that venom from Aphonopelma crinitum had some blocking effect on α3* nAChRs but it was not significant. The presence of acylpolyamines cannot be discarded because in a previous work two acylpolyamines were discovered in the venom of Aphonopelma chalcodes

(Skinner, 1990). Aphonopelma crinitum, Pseudhapalopus sp. and Homoeomma sp. may produce few blockers for α3* nAChRs with no effect on α7 nAChRs because these are found in remote areas where there is little human activity . For example, Aphonopelma crinitum lives in dry areas or desserts of North Mexico.

The plasticity of toxins to evolve in accordance to environmental pressures have been reported recently in venoms of several species from Avicularia sp., Grammostola, sp., Holothele sp., Nhandu sp., and Psalmopoeus sp. (Zhang, 2015b). The diversity of antagonists for vertebrate nAChRs could be attributed to interaction gene-environment. The Spiders Euathlus sp. and Thrixopelma sp. inhabit along the western coast of Chile, and Thrixopelma cyaneolum inhabit the coast along western Peru, all seemed to contain blockers for the subtype α3* nAChRs. However, one venom was extracted from a Grammostola grossa species that inhabits the Eastern Coast of Argentina. Its venom also had a blocking effect on the subtype α3* nAChRs but the presence of modulators in its venom cannot be attributed to the type of prey or predators that are found on the geographical place where this species inhabits (the coast). G. grossa may be the same group of spiders that originally inhabit the Amazons in Brazil that are spread around adjacent regions of Brazil. At some point Grammostola species could evolved into other specie that produce blockers for both subtypes of α3* and α7 nAChRs. This is the case for Grammosotla rosea that is found in Bolivia, then, there could be a spread of this species heading down to south America like the case of Grammostola iheringi and Grammostola porteri that can be found in Chile and Argentina. In general, few research are carried out about tarantula venoms, in particular for those species found in Brazil, such as Acanthoscurria paulensis (Mourão, 2013). Arboreal species Avicularia genus consist of 13 species and these are found in Central and South America. Scarce studies about venoms from Avicularia juruensis exist in the scientific literature (Ayroza, 2012). Avicularia species are threatened for severe deforestation in the rainforest in Brazil and the illegal trade to collectors (Bertani and Fukushima, 2009). It could explain why several species of Avicularia has been spread across other regions around Brazil. Interestingly, all species studied herein that are found in Brazil (in the forests of the Amazonas) contain blockers for α3* nAChRs.

The two tarantulas Holothele sp. were collected originally from Colombia and Peru. Their venoms contain blockers for both subtype nAChRs. Probably both species are the same and these were spread around Colombia and northern Peru. These tarantulas live in burrows, and blockers for cholinergic receptors could be used for protection against small vertebrate predators. Like Holothele sp., the theraphosid Nhandu vulpinus, Nhandu tripepii and Nhandu chromatus live in burrows in the savannahs and tropical forests of South Brazil. Their venoms have low toxicity and could be comparable to bee sting. Their diet is based on crickets, mealworms and houseflies. It is likely the blocking effect on nAChRs is for protection against small predators.

It is known from previous studies that nicotinic receptors are densely expressed in insects and these share similar pharmacological profiles to counterparts expressed in vertebrates (Sattelle, 2005). It suggests that the effect observed in human nicotinic receptors caused by some spider venoms from the two major lineages Mygalomorphae and Araneomorphae could share some similitude to neuronal nicotinic cholinergic receptors from invertebrates, which known to be expressed in neurons and cell bodies in insects. It is likely that selective toxins for invertebrate nAChRs may be found because spiders based most of their diet on insects.

The mechanisms of toxin evolution and diversity in spiders are currently not well understood (Zhang, 2015b), although, this thesis provides with important insights of the evolutionary role of toxins against mammalian cholinergic receptors. The number of different spider venoms used on the present thesis is small to make generalised conclusions. However, the results so far indicated that blockers for vertebrate cholinergic receptors were likely to be present in old ancestor of mygalomorph spiders. Inhibitors for human neuronal α3* and α7 nAChRs are the major compounds in “primitive” spiders. These are likely to be acylpolyamines that have previously been reported in “primitive” spiders (Skinner, 1990; Zhang, 2015b); and it implies that acylpolyamines have existed about 400 million years ago, and are the simplest toxin to synthesise with high potency, unlike synthesis of peptides or proteins. Furthermore, the data suggests that modulators for

vertebrate nAChRs were decreased notoriously when araneomorph spiders appeared.

Our data also implies that primitive spiders produce a wide range of modulators for vertebrate nAChRs that could include agonistic modulators on α7. It seems the large size of mygalomorph spiders and their cheliceral position to subdue their preys was no advantageous with the appearance of new vertebrate animals. It is likely that in “primitive” spiders the role of toxins for both subtypes of mammalian nAChRs is for defence. We could speculate also that in “modern” spiders synthesis of agonists for α7 is energetically more expensive than no selective antagonists for α3* and α7 nAChRs. After all, no selective blockers may lead to the same outcome that warns vertebrate predators away.

To summarize, we can suggest that “modern” spiders (Araneopmorphae) appeared with new physical characteristics to adapt them to new ecosystems; and in parallel their venoms evolved with the appearance of different vertebrate species, including small mammals. Araneomorph spiders retain the presence of cheliceraes but it seems these were adapted to pinch the preys and avoid they scape, instead of rely on their venoms to immobilise their preys, like in the case of Mygalomorph spiders.

The resulting data of this thesis is a started point to further selection of venoms with potential effect on cholinergic receptors of invertebrates. Extra screenings using the FLIPR system, and including as positive control neonicotinoids may help to choose “hit” venoms.

This is the first study of distribution of toxins from spider venoms against human neuronal nAChRs, and it is hoped that it can serve to the discovery of novel toxins against invertebrates as well as help to reveal the evolutional process of spider toxins.

CHAPTER 3 — NACHR MODULATORS FROM SCORPION VENOM

3.1 INTRODUCTION

Like spiders, scorpion venoms contain a vast number of uncharacterized compounds, and therefore they represent an untapped source of novel toxins with potential applications in medicine, agronomy and as research tools (Harrison, 2014; King, 2014; King and Coaker, 2014; Wu, 2014). It has been estimated that the molecules currently described from scorpion venoms represent less than 1% of all components predicted to be in their venoms (Quintero-Hernandez, 2013). Thus, we extended the search for modulators of vertebrate α7 and α3* nAChRs to five scorpion venoms. As outlined below, this work led to the isolation of a non- selective agonist of α7 and α3* nAChRs, from the venom of the giant forest scorpion Heterometrus spinnifer.

3.1.1 Scorpion toxins and their pharmacological action

Scorpions (Arthropoda: Chelicerata: Arachnida: Scorpiones) use their venoms to immobilise prey and/or to defend against predators or competitors. The common biological function of these venoms is their action on a variety of voltage-gated ion channels in invertebrates and small vertebrates. In particular, scorpion venoms are rich in modulators of both insect and vertebrate voltage-gated sodium (for a review, see (Egea, 2015)) and Potassium (Kv) (Chugunov, 2013; Morales-Lázaro, 2015; Quintero-Hernandez, 2013; Swartz, 2013). The most toxic scorpion venoms for mammals are from Androctonus sp. and Leirurus quinquestriatus (paleotropical especies) and Centruroides sp. and Tityus sp. (neotropical especies), and these venoms have been extensively studied due to their clinical relevance (DeBin, 1993; Diaz, 2009; Kularatne and Senanayake, 2014; Quintero-Hernandez, 2013; Rjeibi, 2011). Most of the components purified and identified from scorpions to date are toxins that act on Nav (Forsyth, 2012; Meves, 1984; Zhu, 2012), Kv

(Camargos, 2011; Nirthanan, 2005; Xie, 2012), Cav (Norton and McDonough, 2008; Shahbazzadeh, 2007) and chloride channels (DeBin, 1993; Nie, 2012;

Rjeibi, 2011; Zheng, 2000). In particular, toxins that target Nav channels that are largely responsible for the symptomatology of mammalian envenomation

(Caliskan, 2012; Clot-Faybesse, 2000; Quintero-Hernandez, 2013) and paralysis in insects (Billen, 2008; DeBin, 1993; Forsyth, 2012). However, the effect of scorpion venom on nAChRs has been scarcely documented.

Androctonin, a disulphide-bridged peptide isolated from the hemolymph, but not the venom of Androctonus australis, shows some similarity to a nAChR named α- conotoxin SII, but its effect on nAChRs has not been determined (Ehret-Sabatier, 1996).

A non-toxic venom extract from the scorpion Buthus occitanus tunetanus contains a component that depolarises muscle cells via an effect on nAChRs (Cheikh, 2007), but the authors did not isolate the active component.

Venoms from scorpions reported as non-threatening to humans have not been extensively studied. Interestingly, venom from the giant forest scorpion Heterometrus spinnifer was reported to contain high concentration of acetylcholine and norepinephrine (Nirthanan, 2002). The authors suggested that acetylcholine induces muscle contraction in rat nerve-smooth muscle preparations through activation of muscarinic receptors. However, all of the bioassays were carried out with a venom fraction that contained all molecules from the venom with molecular masses less than 3 kDa, and the activity was not compared with pure acetylcholine.

From our previous screening of 71 arachnid venoms for searching modulators for nAChR using a high-throughput FLIPR calcium assay, we detected an agonistic activity elicited from crude venom of H. spinnifer. We continued further to isolate the “hit” toxin responsible for the activity because venom compounds seem to have promising applications in therapy (Nie, 2012; Yang, 2011) and likely in agronomy (Morales-Lázaro, 2015 #7), as mentioned before the venom is devoid of harm in humans (Kularatne and Senanayake, 2014) and has not been extensively studied due to the lack of clinical interest (Cheikh, 2007; Nirthanan, 2002).

The agonistic compound was isolated, and identified as a choline ester, named senecioylcholine. It activates human neuronal nAChRs endogenously expressed

in SH-SY5Y cells and the synthetic form did not show insecticidal effect. It is known ester of choline occurs naturally (Erspamer and Glässer, 1957; Erspamer, 1953; Florey, 1963; Keyl, 1957), and mainly purified from hypobranchial gland of gastropods (Keyl and Whittaker, 1958; Roseghini, 1996). Even when acetylcholine is a choline ester present in invertebrate nervous system (Florey, 1963), senecioylcholine has not been before identified in scorpion venoms. Few pharmacological studies have been done for senecioylcholine and its isomer named tygloylcholine (Roseghini, 1996). Furthermore, there are discrepancies on the function of choline ester in the venom of gastropods as mechanism of defence (Erspamer and Glässer, 1957; Roseghini, 1996) and in the correct identification of senecioylcholine or tygloylcholine (Baker and Duke, 1976, 1978).

The toxicity of senecioylcholine reported in vertebrates (Roseghini, 1996) and its non-selective agonistic effect on human neuronal nAChRs suggest its role in the venom is highly for defence against small vertebrates. The discovery of senecioylcholine in the venom of a scorpion, which uses its chelae to subdue its preys, makes a significant contribution in the evolution of venom compounds as defensive weapons, and it is the first evidence of a convergent evolution between scorpions and molluscs. To provide with a complete base for future research works in interactions prey-predator and evolution, this study was extended up to the pharmacological characterization of senecioylcholine and its isomers tygloylcholine and angeloylcholine, which could be present in other arachnid venoms.

3.2 METHODS

We used the high-throughput FLIPR assay outlined in Chapter 2 to screen five scorpion venoms for modulators of vertebrate α7 and α3* nAChRs. For details of the procedures used to identify “hit” venoms and purify bioactive venom components see the Section 2.5 and 3.3.2, respectively.

3.2.1 Figure 3.1 shows an example of the protocol as applied to scorpion venoms

Figure 3.1A, shows the fluorescent response obtained by stimulation of α3* nAChRs with nicotine, and inhibition of this response by d-tubocurarine. Figure 3.1B shows that the calcium fluorescence response is evoked mainly by activation of α3* nAChRs, which was inhibited by d-tubocurarine, and this nicotine response is independent of activation of Nav channels. On Figure 3.1C is shown the fluorescent response evoked by crude venom from A. crassicauda, which induces a robust increase in fluorescence prior to the addition of nicotine. The venom- induced increase in fluorescence was not substantially inhibited by d-tubocurarine

(Figure 3.1D) but it was notably by TTX (Figure 3.1 E), a non-selective Nav channel inhibitor (Figure 3.1E), Thus, the robust increase in fluorescence induced by A. crassicauda venom is likely due to the presence of an agonist of Nav channels endogenously expressed in SH-SY5Y cells.

nicotine 30 µM A) 3 B) 3 nicotine 30 µM d-tubocurarine 30 µM TTX (2 µM)

2 2 RLU) (RLU) ( calcium response response calcium

1 1 Maximun calcium response response calcium Maximun Maximun 0 0 1 151 301 451 601 1 151 301 451 601 time (seconds) time (seconds) 4 4 D) C) A. crassicauda venom 2 µg A. crassicauda venom 2 µg d-tubocurarine 30 µM + 3 3 nicotine 30 µM A. crassicauda venom 2 µg

2 2 (RLU) RLU) ( calcium response response calcium

1 1 Maximun calcium response response calcium Maximun

Maximun 0 0 1 151 301 451 601 1 151 301 451 601

E) 4 time (seconds) time (seconds) A. crassicauda venom 2 µg 3 TTX 2 µM + 2 A. crassicauda venom 2 µg calcium response (RLU) response calcium

Maximun 0 1 151 301 451 601 time (seconds) Figure 3.1. A. crassicauda venom can induce calcium responses elicited via activation of Nav channels. (A) Fluorescent responses elicited by stimulation of α3* nAChRs with nicotine, and inhibition of this response by d- tubocurarine. (B) The calcium fluorescent response observed is mediated mostly by activation of α3* nAChRs with nicotine (30 µM). The inhibitor for Nav channels, TTX, does not inhibit the calcium response. (C) Addition of crude venom (2 µg) from A. crassicauda elicited fluorescent response (black trace), prior to the activation of α3* nAChRs with nicotine (blue trace). The venom-induced increase in fluorescence was not substantially inhibited by d-tubocurarine (panel D) but it was significantly inhibited by 2 µM TTX (panel E).

3.2.2 Scorpion venoms

Dr Volker Herzig kindly provided five scorpion venoms for screening against vertebrate α7 and α3* nAChRs. The list of scorpion species used in this study is given in Appendix I.

3.2.3 Venom fractionation

Crude scorpion venom (1 mg) was dissolved in 1 ml of 5% of buffer B (90% acetonitrile containing 0.05% TFA) and centrifuged at 5,000 g for 15 min at 4°C. After centrifugation, the supernatant was loaded onto a Phenomenex C18 reverse- phase HPLC (RP-HPLC) column Jupiter 250 x 4.60 mm; 5 µm particle size; 300 Å pore size and fractionation performed using Shimadzu HPLC system. Venom fractionation was performed using a flow rate of 1 ml/min and the following gradient of Buffer B in Buffer A (Milli-Q water + 0.05% TFA): 0–5 min: 5% Buffer B; 5–10 min: 5–20% Buffer B; 10–50 min: 20–40% Buffer B; 50–55 min: 40–80% Buffer B; 55–57 min: 80% Buffer B.

The elution profile was recorded at 214 nm and 280 nm, and fractions were collected manually based on 214 nm absorbance. At the end of the fractionation the samples were frozen, lyophilized, dissolved in Milli-Q water, and stored at – 20°C until required.

3.2.4 Discovery of a choline ester from the venom of H. spinnifer

We isolated the active fraction from H. spinnifer venom using two chromatographic steps of purification using the FLIPR high-throughput calcium assay described in the previous section. The active fraction from the initial venom fractionation using a Jupiter C18 RP-HPLC column was further fractionated on a Grace Discovery Science Vydac® 218TP C18 RP-HPLC column 250 x 4.60 mm; 5 µm particle size; and 300 Å pore size using a flow rate of 1 ml/min and the following gradient: 0–5 min: 5% Buffer B; 5–10 min: 5–10% Buffer B; 10–50 min: 10–30% Buffer B; 50–55 min: 30–80% Buffer B; 55–57 min: 80% Buffer B.

3.2.5 Determination of mass and purity

The active component isolated from H. spinnifer venom was dissolved in 5% of acetonitrile (grade HPLC) and analysed via matrix-assisted laser desorption/ionization (MALDI) mass spectrometry (MS) performed on an Applied Biosystems mass spectrometer operating in the positive ion linear mode. The sample (0.4 µL) was mixed on a MALDI plate with 0.4 µl of a-cyano-4- hydroxycinnamic acid (CHCA) matrix (6 mg/mL) dissolved in acetonitrile/ MilliQ water /formic Acid (50:45:5, v/v).

The MALDI-TOF spectra did not reveal any masses higher than 1 kDa. Thus, the active fraction was further analysed via LCMS using an Agilent 1100 Series separations module equipped with diode array detector and Agilent 1100 Series LC/MSD mass detector in both positive and negative ion modes. The sample was run on an Agilent Zorbax-C8 column 4.6 mm x 150 mm, 5 µm particle size using a gradient of 10–100% methanol/H2O (with constant 0.05% formic acid modifier) over 15 min, using a flow rate of 1 mL/min.

3.2.6 Identification of active compound using HRMS and LCMS

High-resolution (HR) electrospray ionisation (ESI) MS measurements were obtained on a Bruker micrOTOF with an ESI probe by direct infusion in acetonitrile at 3 µL/min using sodium formate clusters as an internal calibrant. The active + venom yielded m/z 186.1504 [M] (calculated for C10H20NO2, 186.1489).

LC/MS was performed on native compound co-injected with synthetic choline esters using an Agilent 1100 series LC/MS with an Agilent Zorbax-C18 aqua column 4.6 x 150 mm, 5 µm particle size. Isocratic elution was performed using

4% methanol/H2O (with constant 0.02% formic acid modifier) at a flow rate of 1 mL/min, with single ion extraction at 186 [M+].

3.2.7 Chemical synthesis of choline esters

3.2.7.1 Synthesis of chlorocholine chloride

SOCl2 NMe3 NMe3 HO o Cl Cl 70 C, 1h Cl Choline chloride (279 mg, 2 mmol) was dried under high vacuum at 200°C for 2 h.

SOCl2 (2 mL) was added and the solution was heated under reflux for 1 h. After cooling, solvent was evaporated to dryness. Solid residue was redissolved in methanol (3 x 2 mL), and re-evaporated to dryness. The resultant white solid was dried under high vacuum at 80°C for 2 h.

3.2.7.2 Synthesis of tigloylcholine

O O i) K2CO3, dioxane/H2O, 3 1' rt 2 NMe3 OH 1 O ii) 4 2' NMe3 Cl Cl 5 Cl DMSO, 60 oC, 48 h

To a solution of tiglic acid (200 mg, 2 mmol) in a dioxane–water (3:2) mixture (10 mL), a solution of K2CO3 (138 mg, 1 mmol in water 10 mL) was added. After stirring for 10 min, the mixture was evaporated to dryness and again dried under high vacuum at 70 °C for 1 h. Chlorocholine chloride and anhydrous DMSO were added and the mixture heated at 60 °C for 48 h. DMSO was removed by vacuum distillation and the residue was purified by reverse phase column chromatography to afford tigloylcholine as a light brown hygroscopic solid. 1H NMR (600 MHz,

CD3OD): δ 6.92 (qq, 7.1, 1.3 Hz, H-3), 4.59 (m, H2-2ʹ), 3.75 (m, H2-1ʹ), 3.22 (s, 13 NMe3), 1.85 (dd, 1.3, 1.1 Hz, H3-5), 1.82 (dq, J = 7.1, 1.1 Hz, H3-4). C NMR (150

MHz, CD3OD): δ 168.4 (C-1), 140.3 (C-3), 129.1 (C-2), 66.4 (C-2ʹ), 59.1 (C-1ʹ), 54.60 (NMe), 54.58 (NMe), 54.55 (NMe), 14.6 (C-4), 12.3 (C-5). HRMS calc. for + C10H20NO2 186.1489, found m/z = 186.1491.

3.2.7.3 Synthesis of senecioylcholine

4 O O i) K2CO3, dioxane/H2O, 1' rt 2 NMe3 OH 3 1 O ii) 5 2' NMe3 Cl Cl Cl o DMSO, 60 C, 48 h

As described for the synthesis of tigloylcholine, starting with choline chloride (2 mmol, 279 mg) and 2-methyl butenoic acid (2 mmol, 200 mg), senecioylcholine was synthesized and purified by reverse phase column chromatography, to afford 1 senecioylcholine as a white hygroscopic solid. H NMR (600 MHz, CD3OD): δ 5.76

(qq, 1.2, 1.2 Hz, H-2), 4.54 (m, H2-2ʹ), 3.73 (m, H2-1ʹ), 3.22 (s, NMe3), 2.18 (d, 1.2 13 Hz, H3-4), 1.93 (d, 1.2 Hz, H3-5), C NMR (150 MHz, CD3OD): δ 166.8 (C-1), 161.2 (C-3), 115.8 (C-2), 66.4 (C-2ʹ), 58.1 (C-1ʹ), 54.66 (NMe), 54.63 (NMe), 54.61 + (NMe), 27.6 (C-4), 20.6 (C-5). HRMS calc. for C10H20NO2 186.1489, found m/z = 186.1485.

3.2.7.4 Synthesis of angeloylcholine

4 O O i) K2CO3, dioxane/H2O, 1' rt 3 2 NMe3 OH 1 O ii) 2' NMe3 Cl Cl Cl 5 o DMSO, 60 C, 48 h

As described for the synthesis of tigloylcholine, starting with choline chloride (2 mmol, 279 mg) and angelic acid (2 mmol, 200 mg), angeloylcholine was synthesized and purified by reverse phase column chromatography, to afford angeloylcholine as a white hygroscopic solid. 1H NMR (600 MHz, δ 6.23 (qq, 7.2,

1.3 Hz, H-3), 4.60 (m, H2-2ʹ), 3.75 (m, H2-1ʹ), 3.22 (s, NMe3), 2.00 (dq, 7.2, 1.3 Hz, 13 H3-4), 1.91 (dq, J = 1.3, 1.3 Hz, H3-5). C NMR (150 MHz, CD3OD): δ 167.9 (C-1), 141.7 (C-3), 128.0 (C-2), 66.4 (C-2ʹ), 58.6 (C-1ʹ), 54.59 (NMe), 54.56 (NMe), 54.54 + (NMe), 20.8 (C-4), 16.2 (C-5). HRMS calc. for C10H20NO2 186.1489, found m/z = 186.1483.

3.3 RESULTS

3.3.1 High-throughput FLIPR assay

To speed the venom-based discovery of “hit” biomolecules from five scorpion venoms a high-throughput FLIPR assay was performed (King, 2014; Vetter, 2011). Five venoms derived from scorpions of the family Buthidae were screened using the high-throughput FLIPR assay described in Chapter 2. At 2 µg, none of the venoms had a significant effect on vertebrate α7 or α3* nAChRs (Figure 3.2). However, crude venom (1 or 2 µg) from H. spinnifer elicited a transient calcium response that was potentiated by 50% in the presence of the positive allosteric modulator PNU-120596 (Figure 3.3A). This response was completely blocked by d-tubocurarine (Figure 3.3B) indicating that the venom activates neuronal nAChRs. The total elimination of the calcium response in the presence of PNU- 120596 by d-tubocurarine suggests that the agonistic effect observed is mediated totally by α3* and α7 nAChRs. Two other concentrations of venoms were tested at 0.02 µg or at 0.2 µg (data not shown), and no effect was observed. The agonistic effect of this venom is therefore dose-dependent.

crude venom A. australis 1 µg + A)# B)# PNU-120596 (10 µM) crude venom L. quinquestriatus 1 µg + crude venom A. australis 1 µg + C)# 3 PNU-120596 (10 µM) + 3 PNU-120596 (10 µM) d-tubocurarine (30 µM) 4 crude venom L. quinquestriatus 1 µg + C) A. crassicauda PNU-120596 (10 µM) + crude venom 2 µg d-tubocurarine 30 µM+ 3 2 d-tubocurarine (30 µM) 2 venom A. crassicauda 2 µg

(RLU) 1 1 1

0 0 0 1 151 301 451 601 Maximun calcium response Maximunresponse calcium 1 151 301 451 601 (RLU) response calcium Maximun 1 151 301 451 601 time (seconds) time (seconds)

Maximun calcium response (RLU) response calcium Maximun time (seconds)

D)# crude venom G.grandidieri 1 µg +

PNU-120596 (10 µM) E)# venom H. spinnifer 1 µg + PNU 4 crude venom G. grandidieri 1 µg + 3

d-tubocurarine 30 µM PNU-120596 (10 µM) + d-tubocurarine (30 µM) 3 2 2

1 1

0 (RLU) response Calcium 0 1 151 301 451 601 1 151 301 451 601 Maximun calcium response (RLU) response calcium Maximun time (seconds) time (seconds)

Figure 3.2. FLIPR screening of scorpion venoms against α7 and α3* nAChRs. 1 µg of crude venoms from (A) L. quinquestriatus, (B) A. australis, (C) A. crassicauda, (D) G. grandidieri and (E) H. spinnifer was added to loaded- SH- SY5Y cells in the presence of PNU-120596. Calcium responses were then measured for 300 s, followed by choline addition, and recording of calcium response for a further another 300 s. To show the participation of cholinergic receptors in the calcium responses following venom addition, the cells were treated by d-tubocurarine before starting the assay. There was no effect of d- tubocurarine in the responses caused by venom from (A) L. quinquestriatus (trace in dashed lines and orange colour), (B) A. australis (trace in dashed lines and dark blue colour), and (C) A. crassicauda (trace in red colour). However, d-tubocurarine caused significant inhibition of the calcium response elicited by venoms from (D) G. grandidieri (trace in dashed lines and black colour) and (E) H. spinnifer (trace in blue colour). Traces are the average of at least three wells.

2 µg venom A) 6 5 2 µg venom + PNU 4 3 2 1 0

Calcium response (RLU) 1 51 101 151 201 251 me (seconds)

1 µg venom + PNU120596 3 B) 1 µg venom + PNU120596 + d- 2 tubocurarine

1 Calcium response (RLU) 0 1 51 101 151 201 251 301 me (seconds)

Figure 3.3. Non-selective activation of α7 and α3* nAChRs expressed in SH- SY5Y by crude venom from H. spinnifer. (A) 2 µg of crude venom from H. spinnifer induced an increase of intracellular calcium in SH-SY5Y cells (blue trace) that was potentiated by PNU-120596 (red trace). (B) The calcium response elicited by 1 µg of crude venom from H. spinnifer in the presence of PNU-120596 (red trace) was totally blocked by d-tubocurarine (black trace in black colour). Traces are the average of at least three wells.

3.3.2 Isolation of the active compound from H. spinnifer venom

1 mg of crude venom from H. spinnifer was separated by RP-HPLC (Figure 3.4) and fractions were manually collected. Each resulting fraction at 0.42 µg of total crude venom was evaluated using the FLIPR assay to evaluate its impact on activation of nAChRs using nicotine or choline plus the positive allosteric modulator PNU-120596 to stimulate the heteromeric subtypes α3β2/α3β4 and the homomeric subtype α7 nAChR, respectively. Only fraction 10, which eluted at 9.49% acetonitrile in the C18 RP-HPLC separation (inset, Figure 3.4), activated the α3* subtype as its activity was substantially inhibited (66%) by d-tubocurarine (30 µM) (Figure 3.5). Furthermore, the increased response in the presence of PNU-120596 (Figure 3.5) indicates that α7 nAChRs are also activated by fraction 10. Thus, fraction 10 contains the agonistic molecule that is responsible for the calcium response elicited by crude H. spinnifer venom.

3 2 Fraction 10

1 Absorbance 214 nm 2 0 12 14 16 Time (min)

1 Absorbance 214 nm 0 20 40 60 Time (min)

Figure 3.4. HPLC fractionation of crude venom from H. spinnifer. Chromatogram resulting from fractionation of crude venom via C18 RP- HPLC. The fraction highlighted in black (F10) activated α7 and α3* nAChRs in the FLIPR assay. The inset shows an expansion of the region containing F10, which eluted at ∼ 13.6 min.

300

200 ****

Max **** % F 100 ****

F10 (4.2 µg) F10 (4.2 µg) F10 (4.2 µg)

PNU120596 (10 µM) d-tubocurarine (30 µM)

Figure 3.5. Activity of F10 from fractionation of crude venom from H. spinnifer. Active fraction 10 (F10) (1% from initial fractionation of 1 mg of crude venom) was assayed against vertebrate nAChRs using the FLIPR. F10 induced a maximum fluorescent response similar to that evoked by nicotine. The response was inhibited by d-tubocurarine and potentiated by PNU-120596. The difference are significant with a P < 0.05. One-way ANOVA was used to compare the means of each experimental treatment in relation to the control response (nicotine response). The values are averages of three independent experiments with n = 3. Dotted lines represent the control response in the absence of venoms.

Fraction 10 (72 µg) was further fractionated using an analytical RP-HPLC (GRACE-Vydac C18 column RP-HPLC) (Figure 3.6) and the resulting fractions collected manually. Each of the active fractions were then analysed using the FLIPR assay, for activity against vertebrate nAChRs. Only two fractions were active in the FLIPR assay, which eluted at 11.1 min (fraction 4) and 13.6 min (fraction 7) (Figure 3.6). Unfortunately, further study of fraction 7 was not possible due to its low abundance as evidenced by its very low absorbance at 214 nm. Thus, further studies focused on fraction 4. Assay of fraction 4 (77.35 µM) in the FLIPR assay revealed that it induced a fluorescent response that was completely inhibited by 30 µM d-tubocurarine and potentiated by 10 µM PNU-120596 (Figure 3.7). We therefore focussed attention on identification of the active component in this fraction.

300

200

100 Absorbance (214 nm) Fraction 4 Fraction 7 0 0 11.1 20 40 Time (min)

Figure 3.6. Purification of the “hit’ toxin by RP-HPLC. Chromatogram resulting from separation of fraction 10 (72 µg) on an analytical C18 RP-HPLC. The “hit” toxin (fraction 4) eluted at 11.1 min (10.35% Buffer B), and fraction 7 eluted at 13.6 min (12.9% Buffer B).

250 ****

200 ****

150 Max

% F 100

**** 0 ****

Native

-tubocurarine (30 µM) d senecioylcholine (77 µM) PNU120596 (10 µM) d-tubocurarine (30 µM) Native senecioylcholine Native(77 µM) senecioylcholine (77 µM)

Figure 3.7. Agonistic activity of the “hit” toxin. Fraction 4 (77 µM) from analytical HPLC induced a fluorescent response that was totally blocked by d-tubocurarine and by PNU-120596. This suggests that fraction 4 contains the “hit” toxin that is primarily responsible for the non-selective agonism of α3* and α7 nAChRs by H. spinnifer venom. The difference are significant with a P < 0.05. One-way ANOVA was used to compare the means of each experimental treatment in relation to the control response (nicotine response). The values are averages of three independent experiments with n = 3. Dotted lines represent the control response in the absence of venoms.

3.3.3 Identification of the active compound by high-resolution mass spectrometry

Analysis of fraction 4 using MALDI-TOF-MS revealed no masses larger than 1 kDa; it is difficult to detect masses smaller than this via MALDI-TOF-MS due to the noise caused by the matrix (CHCA). Thus, the sample was further analysed using HR-MS; this revealed a single compound with m/z 186.1489. Database searching revealed that the mass formula of C10H20N1O2 corresponds to a previously described choline ester named senecioylcholine (Keyl, 1957; Roseghini, 1996; Whittaker, 1960). However, there are two isomers of senecioylcholine, tigloylcholine and angeloylcholine, with identical mass formula and it is impossible to determine, which of these isomers corresponds to the native venom-derived compound based on HR-MS data alone. Thus, all three isomers were chemical synthesized and LC-MS co-elution studies were performed to identify the native compound. As shown in Figure 3.8, senecioylcholine but neither of the two isomers, co-elutes with the active nAChR agonist purified from H. spinnifer venom. Thus, we conclude that the α3*/α7 nAChR agonist in H. spinnifer venom is senecioylcholine.

synthetic angeloylcholine 15000

10000

5000

12500 synthetic tigloylcholine 10000 7500 5000 2500 0

10000 synthetic senecioylcholine 8000 6000 4000

total ion count total 2000 0

native senecioylcholine 5000 4000 3000 2000 1000 0

15000 co-injection of native + synthetic senecioylcholine

10000

5000

0 1 2 3 4 5 6 7 8 9 Retention time (min)

Figure 3.8. Co-elution of bioactive compound from H. spinnifer venom and synthetic choline esters. LC-MS chromatograms of synthetic choline esters (top three panels), the α3*/α7 nAChR agonist from H. spinnifer venom (bottom panel). LC was performed using a Zorbax C18-aqua column (4.6 × 150 mm), 5 µm particle size with a gradient of methanol + 0.02% formic acid at a flow rate of 1 ml/min.

3.4 DISCUSSION

H. spinnifer is a scorpion from the Scorpionidae family. It is widely distributed in South-eastern Asia but its venom has been poorly studied due to its innocuous effect on humans. However, it is still a rich source of mostly uncharacterised compounds that could be useful as novel modulators of nAChRs with clinical or agronomic relevance. In this study, we used activity-guided fractionation to isolate a novel agonist of vertebrate nAChRs from the venom of H. spinnifer. By using HR-MS in combination with chemical synthesis, we identified this agonist as senecioylcholine, a choline ester. We found that senecioylcholine is a non- selective agonist of both α3β2/α3β4 and α7 nAChR subtypes endogenously expressed in SH-SY5Y cells.

Choline esters are secreted by the hypobranchial gland of predatory marine gastropods from the family Muricidae (phylum Mollusca), for a review, see (Benkendorff, 2013) where they are postulated to play a role in predation and/or defence against predators. They have also been detected in the ventral nerve cord of the American lobster Homarus americanus (phylum Arthropoda, class Crustacea) (Keyl, 1957). The main choline ester identified in Mollusks is murexine (Erspamer and Glässer, 1957) although senecioylcholine has also been detected in significant quantities in the hypobranchial gland of Dicathais orbita. A more recent study identified tigloylcholine, an isomer of senecioylcholine, muricids of the genus Thais (Shiomi, 1998). Murexine and tigloylcholine are both lethal to mice, with intravenous administration leading to LD50 values of 8.5 mg/kg and 0.92 mg/kg (Shiomi, 1998), respectively, for a review, see (Benkendorff, 2013). Like murexine, senecioylcholine is also a neuromuscular blocking compound with nicotine action on vertebrates (Roseghini, 1996). Interestingly, however, even high doses of senecioylcholine have no effect on the fruit fly Drosophila melanogaster (data not shown). Thus, it is likely that senecioylcholine is used by H. spinnifer for defence against vertebrate predators.

A few bioactive compounds have previously been isolated from H. spinnifer venom k-KTx1.3, a non-toxic peptidic modulator of Kv channels (Nirthanan, 2005), and six novel antimicrobial peptides (Nie, 2012; Wu, 2014). H. spinnifer venom was also

shown to contain high concentrations of acetylcholine (∼80 µM) and this was proposed to be responsible for the reversible venom-evoked contracture of chick biventer cervicis muscle that was blocked by d-tubocurarine (Nirthanan, 2002). However, all the bioassays were carried out using a venom fraction that contained all molecules with molecular masses less than 3 kDa. Thus, it is likely that senecioylcholine, which was identified in H. spinnifer venom in the current study, mighty be at least partially responsible for the observed contracture of chick biventer cervicis muscle. The discovery of senecioylcholine in both scorpion venom (phylum Arthropoda) and predatory marine gastropods (phylum Mollusca) suggests that these phylogenetically unrelated organisms have convergently evolved the use of choline esters to defend themselves against vertebrate predators. It remains to be seen whether choline esters are ubiquitous in scorpion venoms or restricted to specific families or genera. However, the fact that only 2 of the scorpion venoms tested in this study appear to contain agonists of human α3β2/α3β4 and α7 nAChRs suggests that senecioylcholine and related choline esters are not widely distributed in scorpion venoms.

CHAPTER 4 — CONCLUSIONS

Venomous animals use their venom to hunt preys and/or protect themselves against predators or competitors (Olsen, 2011). Arachnid venoms are complex mixtures of salts, small organic molecules, peptides and proteins that target a wide range of neuronal receptors and ion channels (Billen, 2008; Estrada, 2007; Morales-Lázaro, 2015; Quintero-Hernandez, 2013). A large number of modulators of voltage-gated ion channels have been isolated from the venom of spiders and scorpions, but there are very few reports of arachnid venom components modulating the activity of nAChRs. This is surprising given that neonicotinoid insecticides are one of the most successful classes of chemical insecticides, which suggests that targeting of insect nAChRs would be a useful pharmacology for arachnid venoms. Thus, the present work was carried out to survey the prevalence of nAChR modulators in arachnid venoms.

Until recently, most characterizations of arachnid venoms used traditional biological assays that require large amounts of venom or venom fraction, which hampered studies of small venomous animals such as arachnids. In the current study we used a high-throughput FLIPR assay that enabled screens to be rapidly performed against human α3* and α7 nAChRs using only a few micrograms of venom. This approach allowed simultaneous screening of 76 arachnid venoms at four different concentrations.

The data were reproducible even when different pools of venoms were used. This suggests that nAChR modulators in arachnid venoms are maintained even when they are kept in captivity. Our data revealed that spider venoms are a rich source of nAChR modulators, and that at least some scorpion venoms contain choline esters that agonize vertebrate nAChRs.

We propose that these nAChR modulators are primarily used for defence against vertebrate predators. It would be of great interest in future to perform high- throughput screens of arachnid venoms against insect nAChRs. Predation of insects is the primary role of arachnid venoms and therefore we might expect to find an even greater abundance and variety of compounds in these venoms that

modulate the activity of invertebrate nAChRs. Nevertheless, the current results strongly support the view that arachnid venoms are a novel source of modulators of ligand-gated ion channels with potential biotechnological applications.

SUMMARY AND FUTURE DIRECTIONS

The current work revealed, for the first time, that many spider venoms (65 of 71 tested in this study) contain modulators of vertebrate nAChRs. Moreover, 31 of the 65 active venoms contained non-selective modulators that affected both α3* and α7 nAChRs, whereas a smaller number (25) were selective modulators of vertebrate α3* nAChRs. This suggests that during the course of evolution, non- selective modulators of vertebrate nAChRs were selected for the purposes of defence against vertebrate predators. However, future studies aimed at screening a taxonomically broader range of spiders against a larger panel of vertebrate nAChR subtypes will be required to address this hypothesis.

The current study also identified a non-selective agonist of vertebrate nAChRs in the venom of the giant forest scorpion Heterometrus spinnifer (Scorpionidae). In contrast with most channel modulators found in arachnid venoms, this nAChR modulator was identified as senecioylcholine, a choline ester. Senecioylcholine was previously identified in the hypobranchial gland of some predatory marine gastropods (family Muricidae) but this is the first report of it being present in scorpion venom. H. spinnifer is one of the largest scorpion species but its venom is no lethal to humans. However, envenomation of humans by H. spinnifer leads to intense pain and inflammation, and senecioycholine may contribute to these symptoms. Thus, this study provides the first evidence for convergent evolution of a non-peptide toxin directed against vertebrate nAChRs. It would be interesting in future studies to examine whether senecioylcholine and related choline ester are found more broadly in scorpion venoms, particularly in species that are not relevant in clinic.

While the current study focussed on vertebrate nAChRs, it would be of great interest in future to examine whether arachnid venoms also target nAChR

subtypes found in their major prey, namely insects. Compounds that specifically targeted insect, but not vertebrate, nAChRs would be of significant interest as bioinsecticide leads. In this respect, it might therefore be interesting to examine venoms that were shown in this study to be inactive against vertebrate nAChRs, such as those from the araneomorph spider Heteropoda jugulans and the mygalomorph spider Aphonopelma crinitum.

Most venom research to date has focussed on animals of medical relevance, but this study has highlighted the value of studying venoms with non-lethal effects on humans when searching for novel compounds. The continued development of high-throughput screening methods and analytical tools that require only small amounts of venom will provide more opportunities in future for exploring the vast pharmacological repertoire encoded within the venoms of small venomous animals such as spiders and scorpions.

REFERENCES

1. AhnAllen, C.G., 2012. The role of the α7 nicotinic receptor in cognitive processing of persons with schizophrenia. Curr. Opin. Psychiatry 25, 103 –108.

2. Aicher, A., 2003. Nicotine strongly activates dendritic cell-mediated adaptive immunity: potential role for progression of atherosclerosis. Circulation 107, 604–611.

3. Akalal, D.B., Wilson, C., Zong, L., Tanaka, N., Ito, K. and Davis, R., 2006. Roles for Drosophila mushroom body neurons in olfactory learning and memory. Learning and Memory 13, 659–668.

4. Alamanda, V., Singh, S., Lawrence, N.J. and Chellapan, S.P., 2012. Nicotine-mediated induction of E-selectin in aortic endothelial cells requires Src kinase and E2F1 transcriptional activity. Biochem. Biophys. Res. Commun. 418, 56–61.

5. Albaugh, G., Bellavance, E., Strande, L., Heinburger, S., Hewitt, C.W. and Alenxander, J.B., 2004. Nicotine induces mononuclear leukocyte adhesion and expression of adhesion molecules, VCAM and ICAM, in endothelial cells in vitro. Ann. Vasc. Surg. 18, 302–307.

6. Albuquerque, E.X., Pereira, E.F.R., Alkondon, M. and Rogers, S.W., 2009. Mammalian nicotinic acetylcholine receptors: from structure to function. Physiol. Rev. 89, 73–120.

7. Armishaw, C.J., 2010. Synthetic α-conotoxin mutants as for studying nicotinic acetylcholine receptors and in the development of novel drug leads. Toxins 2, 1471–1499.

8. Arneric, S.P., Holladay, M. and Williams, M., 2007. Neuronal nicotinic receptors: a perspective on two decades of drug discovery research. Biochem. Pharmacol. 74, 1092 –1101.

9. Ayroza, G., Ferreira, I.L., Sayegh, R.S., Tashima, A.K. and da Silva Junior P.I., 2012. Juruin: an antifungal peptide from the venom of the amazonian pink toe spider, Avicularia juruensis, which contains the inhibitory cystine knot motif. Front. Microbiol. 3, 324.

10. Baker, J.T., Duke, C.C., 1976. Isolation of choline and choline ester salts of tyrindoxyl sulphate from the marine molluscs Dicathais orbita and Mancinella keineri. Tetrahedon Lett. 17, 1233–1234.

11. Baker, J.T., Duke, C.C., 1978. N-methylmurexine a naturally occuring marine compound? Tetrahedon Lett. 19, 5047–5048.

12. Bauwens, M., Mottaghy, F.M. and Bucerius, J., 2015. PET imaging of the human nicotinic cholinergic pathway in atherosclerosis. Curr. Cardiol. Rep. 17, 67.

13. Bencherif, M., Narla, S.T. and Stachowiak, M.S., 2014. Alpha7 neuronal nicotinic receptor: a pluripotent target for diseases of the central nervous system. CNS Neurol. Disord. Drug Targets 13, 836– 845.

14. Benkendorff, K., 2013. Natural product research in the Australian marine invertebrate Dicathais orbita. Mar. Drugs 11, 1370–1398.

15. Bertani, R., Fukushima, C.S., 2009. Description of two new species of Avicularia Lamarck 1818 and redescription of Avicularia diversipes (C.L. Koch 1842) (Araneae, Theraphosidae, Aviculariinae)-three possibly threatened Brazilian species. Zootaxa 2223, 25–47.

16. Billen, B., Bosmans, F. and Tytgat, J., 2008. Animal peptides targeting voltage-activated sodium channels. Curr. Pharm. Des. 14, 2492–2502.

17. Bond, J.E., Hendrixson, B.E., Hamilton, C.A. and Hedin, M., 2012. A reconsideration of the classification of the spider infraorder Mygalomorphae (Arachnida: Araneae) based on three nuclear genes and morphology. PLos One 7, e38753.

18. Bordia, T., Hrachova, M., Chin, M., McIntosh, J.M. and Quik, M., 2012. Varenicline is a potent partial agonist at α6β2* nicotinic acetylcholine receptors in rat and monkey striatum. Neuropharmacology 342, 327– 334.

19. Bourne, Y., Talley, T. T., Hansen, S. B., Taylor, P. and Marchot, P. , 2005. Crystal structure of a Cbtx-AChBP complex reveals essential interactions between snake α-neurotoxins and nicotinic receptors. EMBO J. 24, 1512–1522.

20. Brahma, R.K., McCleary, R.J., Kini, R.M. and Doley, R., 2015. Venom gland transcriptomics for identifying, cataloging, and characterizing venom proteins in snakes. Toxicon 93, 1–10.

21. Brown, L.A., Ihara, M., Buc kingham, S.D., Matsuda, K. and Sattelle, D.B., 2006. Neonicotinoid insecticides display partial and super agonist actions on native insect nicotinic acetylcholine receptors. J. Neurochem. 99, 608–615.

22. Bucher, D., Goaillard, J.M., 2011. Beyond faithful conduction: short- term dynamics, neuromodulation, and long-term regulation of spike propagation in the axon. Prog. Neurobiol. 94, 307–346.

23. Buchman, A.L., Dubin, M., Moukarzel, A., Jenden, D., Roch, M.H., Rice, K., Gorbein, J. and Ament, M.E., 1995. Choline defficiency: a cause of hepatic steatosis during parenteral nutrition that can be reversed with intravenous choline supplementation. Hepatology 22, 1399–1403.

24. Caliskan, F., Quintero-Hernandez, V., Restano-Cassulini, R., Baptista, C.V., Zamudio, F.Z., Coronas, F.I. and Possani, L., 2012. Turkish scorpion Buthacus macrocentrus: general characterization of the venom and description of Bu1, a potent mammalian Na+-channel α- toxin. Toxicon 59, 408–415.

25. Camargos, T.S., Restano-Cassulini, R., Possani, L.D., Peigneur, S., Tytgat, J., Schwartz, C.A. Alves, E.M., de Freitas, S.M., and Schwartz, E.F., 2011. The new kappa-KTx2.5 from the scorpion Opisthacanthus cayaporum. Peptides 32, 1509–1517.

26. Caputi, L., Bengtson, C.P., Guatteo, E., Bernardi, G. and Mercuri, N.B., 2003. D-tubocurarine reduces GABA responses in rat substantia nigra dopamine neurons. Synapse 47, 236–239.

27. Changeux, J., Kasai, M. and Lee, C. , 1970. Use of a snake venom toxin to characterize the cholinergic receptor protein. Proc. Natl. Acad. Sci. U. S. A. 67, 1241–1247.

28. Changeux, J.P., 2010. Nicotine addiction and nicotinic receptors: lessons from genetically modified mice. Nat. Rev. Neurosci. 11, 389– 401.

29. Cheikh, A., Benkhalifa, R., Potreau, D., Raymond, G., El Ayeb, M. and Cognard, C., 2007. A novel depolarizing activity of scorpion venom fraction M1 due to activation of skeletal muscle nicotinic receptors. Toxicon 49, 117–122.

30. Chu, L.W., Ma, E.S., Lam, K.K., Chan, M.F. and Lee, D.H., 2005. Increased α7 nicotinic acetylcholine receptor protein in Alzheimer' s disease patients. Dement. Geriatr. Cogn. Disord. 19, 106–112.

31. Chugunov, A.O., Koromyslova, A.D., Berkut, A.A., Peigneur, S., Tytgat, J., Polyansky, A.A., Pentkovsky, V.M., Vassilevski, A.A., Grishin, E.V. and Efremov, R.G., 2013. Modular organization of α- toxins from scorpion venom mirrors domain structure of their targets, sodium channels. J. Biol. Chem. 288, 19014–19027.

32. Clot-Faybesse, O., Guieu, R., Rochat, H. and Devaux, C., 2000. Toxicity during early development of the mouse nervous system of a scorpion neurotoxin active on sodium channels. Life Sci. 66, 185–192.

100

33. Coddington, J.A., Levi, H.W., 1991. Systematics and evolution of spiders (Araneae). Annu. Rev. Ecol. Evol. Syst. 22, 565–592.

34. Colloby, S.J., Perry, E.K., Pakrasi, S., Pimlott, S.L., Wyper, D.J., McKeith, I.G., Williams, E.D. and O'Brien, J.T., 2010. Nicotinic 123I- 5IA-85380 single photon emission computed tomography as a predictor of cognitive progression in Alzheimer's disease and dementia with Lewy bodies. Am. J. Geriatr. Psychiatry 18, 86–90.

35. Cooper, E., Couturier, S. and Ballivet, M. , 1991. Pentameric structure and subunit stoichiometry of a neuronal nicotinic acetylcholine receptor. Nature 350, 235–238.

36. Corrigall, W.A., Coen, K.M. and Adamson, K.L., 1994. Self- administered nicotine activates the mesolimbic dopamine system through the ventral tegmental area. Brain Res. 653, 278–284.

37. Corrigall, W.A., Franklin, K.B., Coen, K.M. and Clarke, P.B., 1992. The mesolimbic dopaminergic system is implicated in the reinforcing effects of nicotine. Psychopharmacology (Berl) 107, 285–289.

38. Corringer, P.J., Poitevin, F., Prevost, M.S., Sauguet, L., Delarue, M. and Changeux, J.P., 2012. Structure and pharmacology of pentameric receptor channels: from bacteria to brain. Structure 20, 941–956.

39. Corzo, G., Diego-Garcia, E., Clement, H., Peigneur, S., Odell, G., Tytgat, J., Possani, L.D. and Alagon, A., 2008. An insecticidal peptide from the theraphosid Breachypelma smithi spider venom reveals common molecular features among spider species from different genera. Peptides 29, 1901–1908.

40. Corzo, G., Villegas, E., Gómez-Lagunas, F., Possani, L.D., Belokoneva, O.S., and Nakajima, T., 2002. Oxyopinins, large amphipathic peptides isolated from the venom of the wolf spider Oxyopes kitabensis with cytolytic properties and positive insecticidal

101

cooperativity with spider neurotoxins. J. Biol. Chem. 277, 23627– 23637.

41. Couturier, S., Bertrand, D., Matter, J.M., Hernandez, M. C., Bertrand, S., Millar, N., Valera, S., Barkas, T. and Ballivet, M., 1990. A neuronal nciotinic acetylcholine receptor subunit (α 7) is developmentally regulated and forms a homo-oligomeric channel blocked by α-BTX. Neuron 5, 847–856.

42. Dajas-Bailador, F.A., Mogg, A.J. and Wonnacott, S., 2002. Intracellular Ca2+ signals evoked by stimulation of nicotinic acetylcholine receptors in SH-SY5Y cells: contribution of voltage-operated Ca2+ channels and Ca2+ stores. J. Neurochem. 81, 606–614.

43. DeBin, J.A., Maggio, J.E. and Strichartz, G.R., 1993. Purification and characterization of chloro-toxin, a chloride channel ligand from the venom of the scorpion. Am. J. Physiol. 264, C361–C369.

44. Déglise, P., Grünewald, B. and Gauthier, M., 2002. The insecticide imidacloprid is a partial agonist of the nicotinic receptor of homneybee kenyon cells. Neurosci. Lett. 321, 13–16.

45. Diaz, P., D'Suze, G., Salazar, V., Sevcik, C., Shannon, J.D., Sherman, N.E. and Fox, J.W., 2009. Antibacterial activity of six novel peptides from Tityus discrepans scorpion venom. A fluorescent probe study of microbial membrane Na+ permeability changes. Toxicon 54, 802–817.

46. Ding, Y.S., Fowler, J.S., Logan, J., Wang, G.J., Telang, F., Garza, V., Biegon, A., Pareto, D., Rooney, W., Shea, C., Alexoff, D., Volkow, N.D. and Vocci, F. , 2004. 6-[18F]Fluoro-A-85380, a new PET tracer for the nicotinic acetylcholine receptor: studies in the human brain and in vivo demonstration of specific binding in white matter. Synapse 53, 184– 189.

102

47. Doran, N.E., Richardson, A.M.M. and Swain, R., 2001. The reproductive behaviour of Hickmania troglodytes, the tasmanian cave spider (Araneae, Austrochilidae). J. Zool. 253, 405–418.

48. Drevets, W.C., Price, J.L. and Furey, M.L., 2008. Brain structural and functional abnormalities in mood disorders: implications for neurocircuitry models of depression. Brain Struct. Funct. 213, 93–118.

49. Dutertre, S., Nicke, A., Tyndall, J.D. and Lewis, J.R., 2004. Determination of α-conotoxin binding modes on neuronal nicotinic acetylcholine receptors. J. Mol. Recognit. 17, 339–347.

50. Dutertre, S., Ulens, C., Buttner, R., Fish, A., van Elk, R., Kendel, Y., Hopping, G., Alewood, P.F., Schroeder, C., Nicke, A., Smit, A.B., Sixma, T.K. and Lewis, J.R., 2007. AChBP-targeted α-conotoxin correlates distinct binding orientations with nAChR subtype selectivity. EMBO J. 26, 3858–3867.

51. Egea, J., Buendia, I., Parada, E., Navarro, E., León , R. and Lopez, M.G., 2015. Anti-inflammatory role of microglial alpha7 nAChRs and its role in neuroprotection. Biochem. Pharmacol. 97, 463–472.

52. Egleton, R.D., Brown, K.C. and Dasgupta, P., 2009. Angiogenic activity of nicotinic acetylcholine receptors: implications in tobacco-related vascular diseases. Pharmacol. Ther. 121, 205–223.

53. Ehret-Sabatier, L., Loew, D., Goyffon, M., Fehlbaum, P., Hoffmann, J.A., van Dorsselaer, A. and Bulet, P., 1996. Characterization of novel cysteine-rich antimicrobial peptides from scorpion blood. J. Biol. Chem. 271, 29537–29544.

54. Elgoyhen, A.B., Vetter, D.E., Katz, E., Rothlin, C.V., Heineman, S.F. and Boulter, J., 2001. α10: a determinant of nicotinic cholinergic receptor function in mammalian vestibular and cochlear mechanosensory hair cells. Proc. Natl. Acad. Sci. U. S. A. 98, 3501– 3506.

103

55. Emre, M., 2003. Dementia associated with Parkinsons's disease. Lancet Neurol. 2, 229–237.

56. Erspamer, V., Glässer, A., 1957. The pharmacological actions of murexine (urocanylcholine). Brit. J. Pharmacol. 12, 176–184.

57. Erspamer, V.a.B., O., 1953. Isolation of murexine from extracts of hypobranchial glands of Murex trunculus and its identification as beta- [imidazolyl-4(5)]-acryl-choline]. Biochem. Z. 324, 66–73.

58. Escoubas, P., 2006a. Mass spectrometry in toxinology: a 21st-century technology for the study of biopolymers from venoms. Toxicon 47, 609–613.

59. Escoubas, P., Diochot, S. and Corzo, G., 2000. Structure and pharmacology of spider venom neurotoxins. Biochimie 82, 893–907.

60. Escoubas, P., Sollod, B. and King, G.F., 2006b. Venom landscapes: minig the complexity of spider venoms via a combined cDNA and mass spectrometric approach. Toxicon 47, 650–663.

61. Estrada, G., Villegas, E. and Corzo, G., 2007. Spider venoms: a rich source of acylpolyamines and peptides as new leads for CNS drugs. Nat. Prod. Rep. 24, 145–161.

62. Faucon, J.P., Aurières, C., Drajnudel, P., Mathieu, L., Ribière, M., Martel, A.C., Zeggane, S., Chauzat, M.P. and Aubert, M.F., 2005. Experimental study on the toxicity of imidacloprid given in syrup to honey bee (Apis mellifera) colonies. Pest. Manag. Sci. 61, 111–125.

63. Fioravanti, M., Yanagi, M., 2005. Cytidinediphosphocholine (CDP- choline) for cognitive and behavioural disturbances associated with chronic cerebral disorders in the elderly. Cochrane Database Syst. Rev. 2, CD000269.

104

64. Fischer, J., Müller, T., Spatz, A.K., Greggers, U., Grünewald, B. and Menzel, R., 2014. Neonicotinoids interfere with specific components of navigation in honeybees. PLos One 9, e91364.

65. Fisher, M., Huangfu,D., Shen, T.Y. and Guyenet, P.G., 1994. Epibatidine: an alkaloid from the poison frog Epipedobates tricolor is a powerful ganglionic depolarizing agent. J. Pharmacol. Exp. Ther. 270, 702–707.

66. Florey, E., 1963. Acetylcholine in invertebrate nervous system. Can. J. Biochem. Physiol. 41, 2619–2626.

67. Forsyth, P., Sevcik, C., Martinez, R., Castillo, C. and D'Suze, G., 2012. Bactridine's effects on DUM cricket neurons under voltage clamp conditions. J. Insect. Physiol. 58, 1676–1685.

68. Fry, B.G., Wüster, W., Kini, R.M., Brusic, V., Khan, A., Venkataraman, D. and Rooney, A.P., 2003. Molecular evolution and phylogeny of elapid snake venom three-finger toxins. J. Mol. Evol. 57, 110–129.

69. García, A.P., Aitta,-Aho, T., Schaaf, L., Heeley, N., Heuschmid, L., Bai, Y., Barrantes, F.J. and Apergis-Schoute, J., 2015. Nicotinic α4 receptor-mediated cholinergic influences on food intake and activity patterns in hypothalamic circuits. PLos One 10, e0133327.

70. Gershoni, J.M., 1987. Expression of the α-bungarotoxin binding site of the nicotinic acetylcholine receptor by Escherichia coli transformants. Proc. Natl. Acad. Sci. U. S. A. 84, 4318–4321.

71. Gotti, C., Moretti, M., Gaimarri, A., Zanardi, A., Clementi, F. and Zoli, M., 2007. Heterogeneity and complexity of native brain nicotinic receptors. Biochem. Pharmacol. 74, 1102 –1111.

72. Grauso, M., Reenan, R.A., Culetto, E. and Sattelle, D.B., 2002. Novel putative nicotinic acetylcholine receptor subunit genes, Dα5, Dα6 and Dα7, in Drosophila melanogaster identify a new and highly conserved

105

target of adenosine deaminase acting on RNA-mediated A-to-I pre- mRNA editing. Genetics 160, 1519–1533.

73. Griswold, C.E., Ramirez, M.J., Coddington, J.A. and Norman, I.P., 2005. Atlas of phylogenetic data for entelegyne spiders (araneae: araneomorphae: entelegynae) with comments on their phylogeny. Proc. Calif. Acad. Sci. 56, 1–324.

74. Halldin, C., Nagren, K., Swahn, C-G., Langström, B. and Nybäck, H., 1992. (S)- and (R)- (11C) nicotine and the metabolite (R/S)-[11C] nicotine. Preparation, metabolite studies and in vivo distribution in the human brain using PET. Nucl. Med. Biol. 19, 871–880.

75. Hannan, S., Mortensen, M. and Smart, T.G., 2015. Snake neurotoxin

α-bungarotoxin is an antagonist at native GABAA receptors. Neuropharmacology 93, 28–40.

76. Harrison, P., Abdel-Rahman, M.A. and Miller, K., 2014. Antimicrobial peptides from scorpion venoms. Toxicon 88, 115–137.

77. Harvey, M.S., Austin, A.D. and Adams, M., 2007. The systematics and biology of the spider genus Nephila (Araneae: Nephilidae) in the australasian region. Invertebr. Syst. 21, 407–451.

78. He, F., Li, B., Zhao, Z., Zhou, Y., Hu, G., Zou, W., Hong, W., Zou, Y., Jiang, C., Zhao, D. and Ran, P., 2014. The pro-proliferative effects of nicotine and its underlying mechanism on rat airway smooth muscle cells. PLos One 9, e93508.

79. Herzig, V.a.H., W.C., 2008. Neurotoxic and insecticidal properties of venom from the australian theraphosid spider Selenotholus foelschei. NeuroToxicology 29, 471–475.

80. Horti, A.G., Gao, Y., Kuwabara, H., Wang, Y., Abazyan, S., Yasuda, R.P., Tran, T., Xiao, Y., Sahibzada, N., Holt, D.P., Kellar, K.J., Pletnikov, M.V., Pomper, M.G., Wong, D.F. and Dannals, R.F., 2014.

106

18F-ASEM, a radiolabeled antagonist for imaging the α7-nicotinic acetylcholine receptor with PET. J. Nucl. Med. 55, 672–677.

81. Inserra, M., Kompella, S.N., Vetter, I., Brust, A., Daly, N.L., Cuny, H., Craik, D.J., Alewood, P.F., Adams, D.J. and Lewis, R.J., 2013. Isolation and characterization of α -conotoxin LsIA with potent activity at nicotinic acetylcholine receptors. Biochem. Pharmacol. 86, 791–799.

82. Isman, M., 2006. Botanical insecticides, deterrents, and repellents in modern agriculture and an increasingly regulated world. Annu. Rev. Entomol. 51, 45–66.

83. Jensen, A.A., Frølund, B., Liljefors, T. and Krogsgaard-Larsen, P., 2005. Neuronal nicotinic acetylcholine receptors: structural revelations, target identifications, and therapeutic inspirations. J. Med. Chem. 48, 4705–4745.

84. Jin, A.-H., Vetter, I., Dutertre, S., Abraham, N., Emidio, N.B., Inserra, M., Murali, S.S., Christie, M. J., Alewood, P. and Lewis, R., 2014. MrIC, a novel α-conotoxin agonist in the presence of PNU at endogenous α7 nicotinic acetylcholine receptors. Biochemistry 53, 1–3.

85. Jones, A.K., Brown, A.M. and Satelle, D.B., 2007. Insect nicotinic acetylcholine receptor gene families: from egentic model organisms to vector, pest and beneficial species. Invert. Neurosci. 7, 67–73.

86. Jones, A.K., Satelle, D.B., 2009. Diversity of insect nicotinic acetylcholine receptor subunits. Adv. Exp. Med. Biol. 683, 25–43.

87. Jorenby, D.E., Hays, J.T., Rigotti, N.A., Azoulay, S., Watsky, E.J., Williamns, K.E., Billing, C.B., Gong, J., Reeves, K.R. and Vareniclne phase 3 study group, 2006. Efficacy of varenicline, and alpha4beta2 nicotinic acetylcholine receptor partial agonist, vs placebo or sustained- release bupropion for smoking cessation: a randomized controlled trial. JAMA 296, 56–63.

107

88. Kalayci, A., Eren, A., Kocabay, G., Karabay, C.Y., Gecmen, C. and Kirma, C., 2013. Varenicline-induced coronary thrombosis. Ann. Pharmacother. 47, 1727–1729.

89. Kendziorra, K., Wolf, H., Meyer, P.M., Barthel, H., Hesse, S., Becker, G.A., luthardt, J., Schildan, A., Patt, M., Sorger, D., Seese, A., Gertz, H.J. and Sabri, O., 2010. Decreased cerebral alpha4beta2* nicotinic acetylcholine receptor availability in patients with mild cognitive impairment and Alzheimer's disease assessed with positron emission tomography. Eur. J. Nucl. Med. Mol. Imaging 38, 515–525.

90. Keyl, M.J., Michaelson, I.A. and Whittaker, V.P., 1957. Physiologically active choline esters in certain marine gastropods and other invertebrates. J. Physiol. 139, 434–454.

91. Keyl, M.J., Whittaker, V.P., 1958. Some pharmacological properties of murexine (urocanoylcholine). Br. J. Pharmacol. Chemother. 13, 103– 106.

92. Kimes, A.S., Horti, A.G., London, E.D., Chefer, S.I., Contoreggi, C., Ernst, M., Friello, P., Koren, A.O., Kurian, V., Matochik, J.A., Pavlova, O., Vaupel, D.B. and Mukhin, A.G., 2006. 2-[18F]F-A-85380: PET imaging of brain nicotinic actylcholine receptors and whole body distribution in humans. FASEB J. 17, 1331–1333.

93. King, G.F., 2014. The future of venoms-based drug discovery: an interview with Glenn King. Future Medicinal Chemistry 16, 1613–1615.

94. King, G.F., Coaker, H., 2014. The future of venoms-based drug discovery: an interview with Glenn King. Future Med. Chem. 6, 1613– 1615.

95. King, G.F., Hardy, M.C., 2013. Spider-venom peptides: structure, pharmacology, and potential for control of insects pests. Annu. Rev. Entomol. 58, 475–496.

108

96. Knott, V., de la Salle, S., Choueiry, J., Impey, D., Smith, D., Smith, M., Beaudry, E., Saghir, S., LLivitsky, V. and Labelle, A. , 2015. Neurocognitive effects of acute choline supplementation in low, medium and high performer. Pharmacol. Biochem. Behav. 131, 119– 129.

97. Koga, M., Kanaoka, Y., Ohkido, Y., Kubo, N., Ohishi, K., Sugiyama, K., Yamauchi, A. and Kataoka, Y., 2014. Varenicline aggravates plaque formation through α7 nicotinic acetylcholine receptors in ApoE KO mice. Biochem. Biophys. Res. Commun. 455, 194–197.

98. Korey, S.R., de Braganza, B. and Nachmansohn, D., 1951. Choline acetylase. V. Esterifications and transacetylations. J. Biol. Chem. 189, 705–715.

99. Kromann, H., Krikstolaityte, S., Andersen, A.J., Andersen, K., Krogsgaard-Larsen, P., Jaroszewski, J.W., Egebjeerg, J. and Strømgaard, K., 2002. Solid-phase synthesis of polyamine toxin analogues: potent and selective antagonists of Ca2+-permeable AMPA receptors. J. Med. Chem. 45, 5745–5754.

100. Kuhn-Nentwig, L., 2003. Antimicrobial and cytolytic peptides of venomous arthropods. Cell Mol. Life Sci. 60, 2651–2668.

101. Kutlu, M.G., Gould, T.J., 2015. Nicotinic receptors, memory, and hippocampus. Curr. Top. Behav. Neurosci. 23, 137–163.

102. Lam, S., Patel, P.N., 2007. Verenicline: a selective alpha4beta2 nicotinic acetylcholine receptor partial agonist approved for smoking cessation. Cardiol. Rev. 15, 154–161.

103. Landsdell, S.J., Collins, T., Goodchild, J. and Millar, N.S., 2012. The Drosophila nicotinic acetylcholine receptor subunits Dα5 and Dα7 form functional homomeric and heteromeric ion channels BMC Neurosci. 13, 73.

109

104. Lansdell, S.J., Collins, T., Goodchild, J. and Millar, N.S., 2012. The drosophila nicotinic acetylcholine receptor subunits Dα5 and Dα7 form functional homomeric and heteromeric ion channels BMC Neurosci. 13, 73.

105. Lebbe, E.K., Peigneur, S., Maiti, M., Devi, P., Ravichandran, S., Lescrinier, E., Ulens, C., Waelkens, E., D'Souza, L., Herdewijn, P. and Tytgat, J., 2014a. Structure-function elucidation of a new α-conotoxin, Lo1a, from Conus longurionis. J. Biol. Chem. 289, 9573–9583.

106. Lebbe, E.K.M., Peigneur, S., Wijesekara, I. and Tytgat, J., 2014b. Conotoxins targeting nicotinic acetylcholine receptors: an overview. Mar. Drugs 12, 2970–3004.

107. Lewis, R.J., Dutertre, S., Vetter, I. and Christie, M.J. , 2012. Conus venom peptide pharmacology. Pharmacol. Rev. 64, 259-–298.

108. Li, T., Wu, S., Zhang, H., Wang, Y., Luo, H., Zuo, X. and Xiao, X., 2015. Activation of nicotinic receptors inhibits TNF-α-induced production of pro-inflammatory mediators through the JAK2/STAT3 signaling pathway in fibroblast-like synoviocytes. Inflammation 38, 1424–1433.

109. Liang, S., 2008. Proteome and peptidome profiling of spider venoms. Expert review of proteomics 5, 731–746.

110. Lopez, M.G., Montiel, C., Herrero, C.J., Garcia-Palomero, E., Mayorgas, I., Hernandez-Guijo, J.M., Villaroya, M., Olivares, R., Gandia, L., McInstosh, J.M., Olivera, B.M. and Garcia, A.G., 1998. Unmasking the functions of the chromaffin cell α7 nicotinic receptor by using short pulses of acetylcholine and selective blockers. Proc. Natl. Acad. Sci. U. S. A. 95, 14184–14189.

111. Lotfipour, S., Mandelkem, M. and Brody, A.L., 2011. Quantitative molecular imaging of neuronal nicotinic acetylcholine

110

receptors in the human brain with A-85380 radiotracers. Curr. Med. Imaging Rev. 7, 107–112.

112. Lotfipour, S., Mandelkem, M., Alvarez-Estrada, M. and Brody, A.L., 2012. A single administration of low-dose varenicline saturates α4β2* nicotinic acetylcholine receptors in the human brain. Neuropsychopharmacology 37, 1738–1748.

113. Macklin, K.D., Maus, A.D., Pereira, E.F., Albuquerque, E.X. and Conti-Fine, B.M., 1998. Human vascular endothelial cells express functional nicotinic acetylcholine receptors. J. Pharmacol. Exp. Ther. 287, 435–439.

114. Manjunatha, K.R., Robin, D., 2010. Structure, function and evolution of three-finger toxins: mini proteins with multiple targets. Toxicon 56, 855–867.

115. Mansvelder, H.D., Keath, J.R. and McGehee, D.S., 2002. Synaptic mechanisms underlie nicotine-induced excitability of brain reward areas. Neuron 33, 905–919.

116. Matsuda, K., Kanaoka, S., Akamatsu, M. and Sattelle, D.B., 2009. Diverse actions and target-site selectivity of neonicotinoids: structural insights. Mol. Pharmacol. 76, 1–10.

117. McGehee, D.S., Role, L.W., 1995. Physiological diversity of nicotinic acetylcholine receptors expressed by vertebrate neurons. Annu. Rev. Physiol. 57, 521–546.

118. McGuire, S., Le, P. and Davis, R., 2001. The role of Drosophila mushroom body signalling in olfactory memory. Science 17, 1330– 1333.

119. McIntosh, J.M., Santos, A.D. and Olivera, B.M., 1999. Conus peptides targeted to specific nicotinic acetylcholine receptor subtypes. Annu. Rev. Biochem. 68, 59–88.

111

120. Mellor, I.R., Usherwood, P.N., 2004. Targeting ionotropic receptors with polyamine-containing toxins. Toxicon 43, 493–508.

121. Meves, H., Simard, J.M. and Watt, D.D., 1984. Voltage- dependent effect of a scorpion toxin on sodium current inactivation. Eur. J. Physiol. Pflugers Arch. 402, 24–33.

122. Meyer, P.M., Strecker, K., Kendziorra, K., Becker, G., Hesse, S., Woelpl, D., Hensel, A., Patt, M., Sorger, D., Wegner, F., Lobsien, D., Barthel, H., Brust, P., Gertz, H.J. Sabri, O. and Schwartz, J., 2009. Reduced alhpa4beta2*-nicotinic acetylcholine receptor binding and its relationship to mild cognitive and depressive symptoms in Parkinson's disease. Arch. Gen. Psychiatry. 66, 866–877.

123. Mihalak, K.B., Carroll, F.I. and Luetje, C.W., 2006. Varenicline is a partial agonist at alpha4beta2 and a full agonist at alpha7 neuronal nicotinic receptors. Mol. Pharmacol. 70, 801–805.

124. Miyazawa, A., Fujiyoshi, Y., Stowell, M. and Unwin, N. , 1999. Nicotinic acetylcholine receptor at 4.6Å resolution: transverse tunnels in the channel wall. J. Mol. Biol. 288, 765–786.

125. Moffat, C., Pacheco, J.G., Sharp, S., Samson, A.J., Bollan, K.A., Huang, J., Buckland, S.T. and Connolly, C.N., 2015. Chronic exposure to neonicotinoids increases neuronal vulnerability to mitochondrial dysfunction in the blumblebee (Bombus terrestris). FASEB J. 29, 2112–2119.

126. Molgó, J., Rómulo, A., Benoit, E. and Iorga, B.I., 2013. Physical and virtual screening methods for marine toxins and drug discovery targeting nicotinic acetylcholine receptors. Expert. Opin. Drug Discov. 8, 1203–1223.

127. Moore, S., Smyth, W.F., Gault, V.A., O'Kane, E. and McClean, S., 2009. Mass spectrometric characterization and quantitation of selected low molecular mass compounds from the venom of

112

Haplopelma lividum (Theraphosidae). Rapid Commun. Mass Spectrom. 23, 1747–1755.

128. Morales-Lázaro, S.L., Hernández-García, E., Serrano-Flores, B. and Rosenbaum, T. , 2015. Organic toxins as tools to understand ion channels mechanisms and structure. Curr. Top. Med. Chem. 15, 581– 603.

129. Morgenstern, D., Rohde, B.H., King, G.F., Tal, T., Sher,D. and Zlotkin, E., 2011. The tale of a resting gland: transcriptome of a replete venom gland from the scorpion Hottentotta judaicus. Toxicon 57, 695– 703.

130. Mourão, C.B., Heghinian, M.D., Barbosa, E.A., Marí, F., Bloch, C. Jr., Restano-Cassulini, R., Possani, L.D. and Schwartz, E.F., 2013. Characterization of a novel peptide toxin from Acanthoscurria paulensis spider venom: a distinct cysteine assignment to the HWTX-II family. Biochemistry 52, 2440–2452.

131. Müller, M.L.T.M., Bohnen, N., 2013. Cholinergic dysfunction in Parkison's disease. Curr. Neurol. Neurosci. Rep. 13, 377.

132. Nicke, A., Wonnacott, S. and Lewis, R., 2004. α-conotoxins as tools for the elucidation of structure and function of neuronal nicotinic acetylcholine receptor subtypes. Eur. J. Biochem. 271, 2305–2319.

133. Nides, M., Oncken, C., Gonzales, D., Rennard, S., Watsky, E.J., Anziano, R. and Reeves, K.R., 2006. Smoking cessation with varenicline, a selective α4β2 nicotinic receptor partial agonist: results from a 7-week, randomized, placebo- and bupropion-controlled trial with 1-year follow-up. Arch. Intern. Med. 166, 1561–1568.

134. Nie, Y., Zeng, X.C., Yang, Y., Luo, F., Luo, X., Wu, S., Zhang, L. and Zhou, J., 2012. A novel class of antimicrobial peptides from the scorpion Heterometrus spinnifer. Peptides 38, 389–394.

113

135. Nirthanan, S., Pil, J., Abdel,-Mottaleb, Y., Gopalakrishnakone, P., Joseph, J.S., Sato, K. and Tytgat, J., 2005. Assignment of voltage- gated potassium channel blocking activity to kappa-KTx1.3, a non-toxic homologue of kappa-hefutoxin-1, from Heterometrus spinifer venom. Biochem. Pharmacol. 69, 669–678.

136. Nirthanan, S., Pil, J., Joseph, J.S., Gopalakrishnakone, P., Joseph, J.S., Khoo, H.E., Cheah, L.S. and Gwee, M.C.E., 2002. Biochemical and pharmacological characterization of the venom of the black scorpion Heterometrus spinnifer. Biochem. Pharmacol. 63, 49– 55.

137. Nørager, N.G., Poulsen, M.H., Jensen, A.G., Jeppesen, N.S., Kristensen, A.S. and Strørmgaard, K., 2014. Structure-activity relationships studies of N-methylated and N-hydroxylated spider polyamine toxins as inhibitors of ionotropic glutamate receptors. J. Med. Chem. 57, 4940–4949.

138. Norton, R.S., McDonough, S.I., 2008. Peptides targeting voltage-gated calcium channels. Curr. Pharm. Des. 14, 2480–2491.

139. Nybäck, H., Halldin, C., Ahlin, A., Curvall, M. and Eriksson, L. , 1994. PET studies of the uptake of (S)- and (R)-11C-nicotine in the human brain: difficulties in visualizing specific receptor binding in vivo. Psychopharmacology 115, 31–36.

140. Nyffeler, M., Pusey, B.J., 2014. Fish predation by semi-auqatic spiders: a global pattern. PLos One 9, e99459.

141. Olsen, C.A., Kristensen, A.S. and Strømgaard, K., 2011. Small molecules from spiders used as chemical probes. Angew. Chem. Int. Ed. 50, 11296–11311.

142. Olsen, C.A., Mellor, I.R., Wellendorph, P., Usherwood, P.N., Witt, M., Franzyk, H. and Jaroszewski, J.W., 2006. Turning wasp toxin structure for nicotinic receptor antagonism: cyclohexylalanine-

114

containing analogues as potent and voltage-dependent blockers. Chem. Med. Chem. 1, 303–305.

143. Pammolli, F., Magazzini, L. and Riccaboni, M. , 2011. The productivity crisis in pharmaceutical R & D. . Nat. Rev. Drug Discov. 10, 428–438.

144. Parikh, V., Ji, J., Decker, M.W. and Sarter, M., 2010. Prefrontal beta2 subunit-containing and alpha7 nicotinic acetylcholine receptors differentially control glutamatergic and cholinergic signaling. J. Neurosci. 30, 3518–3530.

145. Perry, D.C., Xiao, Y., Nguyen, H.N., Musachio, J.L., Dávila- García, M.I. and Kellar, K.J., 2002. Measuring nicotinic receptors with characteristics of alpha4beta2, alpha3beta2 and alpha3beta4 subtypes in rat tissues by autoradiography. J. Neurochem. 82, 468–481.

146. Picciotto, M.R., Higley, M.J. and Mineur, Y.S., 2012. Acetylcholine as a neuromodulator: cholinergic signaling shapes nervous system function and behavior. Neuron 76, 116–129.

147. Picciotto, M.R., Kenny, P.J., 2013. Molecular mechanisms underlying behaviors related to nicotine addiction. Cold Spring Harb. Perspect. Med. 3, a012112.

148. Piek, T., Mantel, P. and Engels, E., 1971. Neuromuscular block in insects caused by the venom of the digger wasp Philanthus triangulum F. Comp. Gen. Pharmacol. 2, 317–331.

149. Piek, T., Njio, K.D., 1975. Neuromuscular block in honeybees by the venom of the bee wolf wasp (Philanthus triangulum F.). Toxicon 13, 199–201.

150. Plummer, H.K.r., Dhar, M. and Schuller, H.M., 2005. Expression of the alpha7 nicotinic acetylcholine receptor in human lung cells. Respir. Res. 6, 29.

115

151. Quik, M., McIntosh, J.M., 2006. Striatal α-6* nicotinic acetylcholine receptors: potential targets for Parkinson's disease therapy. J. Pharmacol. Exp. Ther. 316, 481–489.

152. Quintero-Hernandez, V., Jimenez-Vargas, J.M., Gurrola, G.B., Valdivia, H.H.F. and Possani, L., 2013. Scorpion venom components that affect ion-channels function. Toxicon 76, 328–342.

153. Rinne, J.O., Myllykylä., Lönnberg, P. and Marjamäki, P., 1991. A postmortem study of brain nicotinic receptors in Parkinson's and Alzheimer's disease. Brain. Res. 547, 167–170.

154. Rjeibi, I., Mabrouk, K., Mosrati, H., Berenguer, C., Mejdoub, H., Villard, C., Laffitte, D., Ouafik, L., Luis, J., Elayeb, M. and Srairi-Abid, N., 2011. Purification, synthesis and characterization of AaCtx, the first chlorotoxin-like peptide from Androctonus australis scorpion venom. Peptides 32, 656–663.

155. Rocha-E-Silva, T.A., Rostelato-Ferreira, S., Leite, G.B., DaSilva P.I J.R., Hyslop, S. and Rodrigues-Simioni, L. , 2013. VdTX1, a reversible nicotinic receptor antagonist isolated from venom of the spider Vitalius dubius (Theraphosidae). Toxicon 70, 135–141.

156. Rollema, H., Coe, J.W., Chambers, L.K., Hurst, R.S., Stahl, S.M. and Williams, K.E., 2007. Rationale, pharmacology and clinical efficacy of partial agonists of alpha4beta2 nACh receptors for smoking cessation. Trends Pharmacol. Sci. 28, 316–325.

157. Roly, Z.Y., Islam, M.M. and Reza, M.A., 2014. A comparative in silico characterization of functional and physicochemical properties of 3FTx (three finger toxin) proteins from four venomous snakes. Bioinformation 10, 281–287.

158. Roseghini, M., Severini, C., Erspamer, G.F. and Erspamer, V., 1996. Choline esters and biogenic amines in the hypobranchial gland

116

of 55 molluscan species of the neogastropod muricoidea superfamily. Toxicon 34, 33–35.

159. Rueter, L.E., Donnelly-Roberts, D.L., Curzon, P., Briggs, C.A., Anderson, D.J. and Bitner, R.S., 2006. A-85380: a pharmacological probe for the preclinical and clinical investigation of the alphabeta neuronal nicotinic acetylcholine receptor. CNS Drug Rev. 12, 100–112.

160. Russo, P., Del Bufalo, A., Milic, M., Salinaro, G., Fini, M. and Cesario, M., 2014. Cholinergic receptors as target for cancer therapy in a systems medicine perspective. Curr. Mol. Med. 14, 1126–1138.

161. Sattelle, D.B., Jones, A.K., Sattelle, B.M., Matsuda, K., Reenan, R. and Biggin, P.C., 2005. Edit, cut and paste in the nicotinic acetylcholine receptor gene family of Drosophila melanogaster. Bioessays 27, 366–376.

162. Schedel, A., Thornton, S., Schloss, P., Kluter, H. and Bugert, P., 2010. Human platelets express functional alpha7-nicotinic acetylcholine receptors. Arterioscler. Thromb. Vasc. Biol. 31, 928–934.

163. Schmaljohann, J., Gündisch, D., Minnerop. M., Bucerius, J., Joe, A., Guhlke, S., Biersack, H.J. and Wüllner, U., 2006. In vitro evaluation of nicotinic acetylcholine receptors with 2-[18F]F-A85380 in Parkinson's disease. Nucl. Med. Biol. 33, 305–309.

164. Schmaljohann, J., Gündisch, D., Minnerop. M., Joe, A., Bucerius, J., Dittmar, C., Jessen, F., Guhlke, S. and Wüllner, U., 2005. A simple and fast method for the preparation of n.c.a. 2-[18F]F-A85380 for human use. Appli. Radiat. Isot. 63, 433–435.

165. Selden, P.A., Gall, J.C., 1992. A triassic Mygalomorph spider from the northern Vosges, France. Paleontology 35, 211–235.

166. Shahbazzadeh, D., Srairi-Abid, N., Feng, W., Ram, N., Borchani, L., Ronjat, M., Akbari, A., Pessah, I.N., De Waard, M. and El Ayeb, M., 2007. Hemicalcin, a new toxin from the Iranian scorpion

117

Hemiscorpius lepturus which is active on ryanodine-sensitive Ca2+ channels. Biochem. J. 404, 89–96.

167. Shiomi, K., Ishii, M., Shimakura, K., Nagashima, Y. and Chino, M., 1998. Tygloylcholine: a new choline ester toxin from the hypobranchial gland of two spsecies of muricid gastropods (Thais clavigera and Thais bronni). Toxicon 36, 795–798.

168. Siegwart, M., Graillot, B., Blachere Lopez, C., Besse, S., Bardin, M., Nicot, P.C. and Lopez-Ferber, M., 2015. Resistance to bioinsecticides or how to enhance their sustainability: a review. Front. Plant. Sci. 6, 381.

169. Skinner, W.S., Dennis, P.A., Lui, A., Carney, R.L. and Quistad, G.B., 1990. Chemical characterization of acylpolyamine toxins from venom of a trap-door and two tarantulas. Toxicon 28, 541–546.

170. Sousa, S.R., Vetter, I., Ragnarsson, L. and Lewis, R., 2013. Expression and pharmacology of endogenous Cav channels in SH- SY5Y human neuroblastoma cells. PLos One 8, e59293.

171. Steilein, O.K., 2001. Genes and mutations in idiopathic epilepsy. Am. J. Med. Genet. 106, 139–145.

172. Strømgaard, K., Andersen, K., Krogsgaard-Larsen, P. and Jaroszewski, J.W., 2001. Recent advances in the medicinal chemistry of polyamine toxins. Mini Rev. Med. Chem. 1, 317–338.

173. Strømgaard, K., Jensen, L.S. and Vogensen, S.B., 2005. Polyamine toxins: development of selective ligands for ionotropic receptors. Toxicon 45, 249–254.

174. Strømgaard, K., Mellor, I., 2004. AMPA receptor ligands: synthetic and pharmacological studies of polyamines and polyamine toxins. Med. Res. Rev. 24, 589–620.

118

175. Sullivan, J.P., Donnelly-Roberts, D., Briggs, C.A., Anderson, D.J., Gopalakrishnan, M., Piattoni-Kaplan, M., Campbell, J.E., McKenna, D.G., Molinari, E., Hettinger, A.M., Garvey, D.S., Wasicak, J.T., Holladay, M.W., Williams, M. and Americ, S.P., 1996. A-85380 [3- 2(S)-azetidinylmethoxy) pyridine]: in vitro pharmacological properties of a novel, high affinity alpha4 beta 2 nicotinic acetylcholine receptor ligand Neuoropharmacology 35, 725–734.

176. Sunagar, K., Jackson, T.N.W, Undheim, E.A.B., Ali, S.A., Antunes, A. and Fry, B.G., 2013. Three-fingered RAVERs: rapid accumulation of variations in exposed residues of snake venom toxins. Toxins (Basel) 5, 2172–2208.

177. Swartz, K.J., 2013. The scorpion toxin and the potassium channel. elife 2.

178. Terlau, H., Olivera, B.M., 2004. Conus venoms: a rich source of novel ion channel-targeted peptides. Physiol. Rev. 84, 41–68.

179. Trachsel, C., Siegemund, D., Kämpfer, U., Kopp, L.S., Bühr, C., Grossmann, J., Lüthi, C., Cunningham, M., Nentwig, W., Kuhn- Nentwig, L., Schurch, S. and Schaller, J., 2012. Multicomponent venoms of the spider Cupiennius salei: a bioanalytical investigation applying different strategies. FEBS J. 279, 2683–2694.

180. Tregellas, J.R., Tanabe, J., Rojas, D.C., Shatti, S., Olincy, A., Johnson, L., Martin, L.F., Soti, F., Kem, W.R., Leonard, S. and Freedman, R., 2011. Effects of an α7 -nicotinic agonist on default network activity in schizophrenia. Biol. Psychiatry. 69, 7–11.

181. Tsetlin, V., Hucho, F., 2009. Nicotinic acetylcholine receptors at atomic resolution. Curr. Opin. Pharmacol. 9, 306–310.

182. Tsetlin, V.I., 2015. Three-finger snake neurotoxins and Ly6 proteins targeting nicotinic acetylcholine receptors: pharmacological

119

tools and endogenous modulators. Trends Pharmacol. Sci. 36, 109– 123.

183. Tuesta, L., Fowler, C.D. and Kenny, P.J., 2011. Recent advances in understanding nicotinic receptor signalling mechanisms that regulate drug self-administration behavior. Biochem. Pharmacol. 82, 984–995.

184. Unwin, N., 1993. Nicotinic acetylcholine receptor at 9Å resolution J. Mol. Biol. 229, 1101–1124.

185. Unwin, N., 2005. Refined structure of the nicotinic acetylcholine receptor at 4Å resolution. J. Mol. Biol. 346, 967–989.

186. Utkin, Y.N., 2015. Animal venom studies: current benefits and future developments. World J. Biol. Chem. 6, 28-–33.

187. van Enkhuizen, J., Janowsky, D.S., Olivier, B., Minassian, A., Perry, W., Young, J.W. and Geyer, M.A., 2015. The catecholaminergic- cholinergic balance hypothesis of bipolar disorder revisited. Eur. J. Biochem. 753, 114–126.

188. Vassilevski, A.A., Kozlov, S.A. and Grishin, E.V., 2009. Molecular diversity of spider venom. Biochemistry (Moscow) 74, 1505– 1534.

189. Vassilevski, A.A., Kozlov, S.A., Samsonova, O.V., Egorova, N.S., Karpunin, D.V., Pluzhnikov, K.A., Feofanov, A.V. and Grishin, E.V., 2008. Cyto-insectotoxins, a novel class of cytolytic and insecticidal peptides from spider venom. Biochem. J. 411, 687–696.

190. Vassilevski, A.A., Sachkova, M.Y., Ignatova, A.A., Kozlov, S.A., Feofanov, A.V. and Grishin, E.V., 2013. Spider toxins comprising disulfide-rich and linear amphipathic domains: a new class of molecules identified in the lynx spider Oxyopes takobius. FEBS J. 280, 6247–6261.

120

191. Vetter, I., 2012. Development and optimization of FLIPR high throughput calcium assays for ion channels and GPCRs. Adv. Exp. Med. Biol. 740, 45–82.

192. Vetter, I., Davis, J.L., Rash, L.D., Anagi, R., Mobli, M., Alewood, P.F., Lewis, R. and King, G.F., 2011. Venomics: a new paradigm for natural products-based drug discovery. Amino acids 40, 15–28.

193. Vetter, I., Lewis, R.J., 2010. Characterization of endogenous calcium responses in neuronal cell lines. Biochem. Pharmacol. 79, 908–920.

194. Vollrath, F., 1985. Web spider's dilemma: a risky move or site dependent growth. Oecologia 68, 69–72.

195. Wada, T., Naito, M., Kenmochi, H. and Sasaoka, T., 2007. Chronic nicotine exposure enhances insulin-induced mitogenic signaling via up-regulation of alpha7 nicotinic receptors in isolated rat aortic smooth muscle cells. Endocrinology 148, 790–799.

196. Wang, S., Zhao, C., Liu, Z., Wang, X., Liu, N., Du, W. and Dai, Q., 2015. Structural and functional characterization of a novel α- conotoxin Mr1.7 from Conus marmoreus targeting neuronal nAChR α3β2, α9 α10 and α6/ α3β2β3 subtypes. Mar. Drugs 13, 3259–3275.

197. Wenningmann, I., Dilger, J.P., 2001. The kinetics of inhibition of nicotinic acetylcholine receptors by (+)-tubocurarine and pancuronium. Mol. Pharmacol. 60, 790–796.

198. Whittaker, V.P., 1960. Pharmacologically active choline esters in marine gastropods. Ann. N. Y. Acad. Sci. 90, 695–705.

199. Williamson, S.M.a.W., G.A., 2013. Exposure to multiple cholinergic pesticides imparirs olfactory learning and memory in honeybees. J. Exp. Biol. 216, 1799–1807.

121

200. Willians, K., 1997. Interactions of polyamines with ion-channels. Biochem. J. 325, 289–297.

201. Windley, M., Herzig, V., Dziemborowicz, S.A., Hardy, M., King, G.F. and Nicholson, G.M., 2012a. Spider-venom as bioinsecticides. Toxins (Basel) 4, 191–227.

202. Windley, M.J., Herzig, V.,Dziemborow, S.A., Hardy, M., King, G.F. and Nicholson, G.M., 2012b. Spider-venom peptides as bioinsecticides. Toxins 4, 191–227.

203. Wong, D.F., Kuwabara, H., Kim, J., Brasic, J.R., Chamroonrat, W., Gao, Y., Valentine, H., Willis, W., Mathur, A., McCaul, M.E., Wand, G., Gean, E.G., Dannals, R.F. and Horti, A.G., 2013. PET imaging of high-affinity alpha4beta2 nicotinic acetylcholine receptors in humans with 18F-AZAN, a radioligand with optimal brain kinetics. J. Nucl. Med. 54, 1308–1314.

204. Wu, S., Nie, Y., Zeng, X.C., Cao, H., Zhang, L., Zhou, L., Yang, Y., Luo, X. and Liu, Y., 2014. Genomic and functional characterization of three new venom peptides from the scorpion Heterometrus spinifer. Peptides 53, 30–41.

205. Xiao, C., Miwa, J.M., Henderson, B.J., Wang, Y., Deshpande, P., McKinney, S.L. and Lester, H.A., 2015. Nicotinic receptor subtype- selective circuit patterns in the subthalamic nucleus. J. Neurosci. 35, 3734–3746.

206. Xie, S., Feng, J., Yu, C., Li, Z., Wu, Y., Cao, Z., Li, W., He, X., Xiang, M and Han, S., 2012. Identification of a new specific Kv1.3 channel blocker Ctri9577, from the scorpion Chaerilus tricostatus. Peptides 36, 94–99.

207. Xu, X., Liu, F., Chen, J., Ono, H., Li, D. and Kuntner, M., 2015. A genus-level taxonomic review of primitively segmented spiders (Mesothelae, Liphistiidae). Zookeys 488, 121–151.

122

208. Yang, R., Zhao, Z., He, Y., Wu, L., Cai, D., Hong, W., Wu, Y., Cao, Z., Zheng, C. and Li, W., 2011. A new natural α-helical peptide from the venom of the scorpion Heterometrus petersii kills HCV. Peptides 32, 11–19.

209. Yu, D., Zhang, L., Eisele, J.L., Bertrand, D., Changeux, J.P. and Weight, F.F., 1996. Ethanol inhibition of nicotinic acetylcholine type α7 receptors involves the amino-terminal domain of the receptor. Mol. Pharmacol. 50, 1010–1016.

210. Zeisel, S.H., 1992. Choline: an important nutrient in brain development, liver, function and carcinogenesis. J. Am. Coll. Nutr. 11, 473–481.

211. Zeisel, S.H., da Costa, K.A., 2009. Choline: an essential nutrient for public health. Nutr. Rev. 67, 615–623.

212. Zelanis, A., Andrade-Silva, D., Rocha, M.M., Furtado, M.F., Serrano, S.M., Junqueira-de-Azevedo, I.L. and Ho, P.L., 2012. A transcriptomic view of the proetome variability of newborn and adult Bothrops jararaca snake venoms. PLoS Negl. Trop. Dis. 6.

213. Zhang, W., Liu, Y., Hou, B., Gu, X. and Ma, Z., 2015a. Activation of spinal alpha-7 nicotinic acetylcholine receptor attenuates remifentanil-induced postoperative hyperalgesia. Int. J. Clin. Exp. Med. 8, 1871–1879.

214. Zhang, Y.Y., Huang, Y., He, Q.Z., Luo, J., Zhu, L., Lu, S.S., Liu, J.Y., Huang, P.F., Zeng, X.Z. and Liang, S.P., 2015b. Structural and functional diversity of peptide toxins from tarantula Haplopelma hainanum (Ornithoctonus hainana) venom revealed by transcriptomic, peptidomic and patch-clamp approaches. J. Biol. Chem. 290, 14192– 14207.

215. Zheng, X.C., Li, W.X., Zhu, S.Y., Peng, F., Zhu, Z.H., Wu, K.L. and Yiang, F.H., 2000. Cloning and characterization of a cDNA

123

sequence encoding the precursor of a chlorotoxin-like peptide from the Chinese scorpion Buthus martensii Karsch. Toxicon 38, 1009–1014.

216. Zhou, P.A., Xie, X.J., Li, M., Yang, D.M., Xie, Z.P., Zong, X. and Liang, S.P., 1997. Blockade of neuromuscular transmission by huwentoxin-I, purified from the venom of the chinese bird spider Selenocosmia huwena. Toxicon 35, 39–45.

217. Zhou, Y., Nelson, M.E., Kuryatov, A., Choi, C., Cooper, J. and Lindstrom, J., 2003. Human α4β2 acetylcholine receptors formed from linked subunits. J. Neurosci. 23, 9004–9015.

218. Zhu, S., Peigneur, S., Gao, B., Lu, X., Cao, C. and Tytgat, J., 2012. Evolutionary diversification of Mesobuthus α- scorpion toxins affecting sodium channels. Mol. Cell Proteomics 11, 1–18.

219. Kularatne, S.A.M., Senanayake, N., 2014. Venomous snake bites, scorpions, and spiders, in: Ferro, J.B.a.J.M. (Ed.), Handbook of Clinical Neurology. Elsevier B.V., Sri Lanka.

124

APPENDIX I

Table I. List of spiders whose venoms were used in this work Araneomorphae (“modern” spiders) Abbreviations No effect on α3* and α7 nAChR Austrochiloidea family 1. Hickmania troglodytes H. troglodytes Blocking effect on α3* nAChR Araneoidea family 2. Cyrtophora moluccensis C. moluccensis Blocking effect on α3* and α7 nAChR Araneoidea family 3. Eriophora transmarina E. transmarina Genus Nephila 4. Nephila pilipes N. pilipes 5. Nephila plumipes N. plumites No effect on α3* and α7 nAChR RTA Clade Superfamily Ctenoidea 6. Ancylometes sp. Ancylometes sp. 7. Viridasius sp. (sylvestriformis) Viridasius sp. Sparassidae family 8. Heteropoda jugulans H. jugulans Partial blocking effect on α3* and α7 nAChR RTA Clade Superfamily Lycosoidea 9. Lycosidae sp. Lycosidae sp. Sparassidae family 10. Sparassidae sp. 1 Sparassidae sp. 1 11. Sparassidae sp. 2 Sparassidae sp. 2

125

Blocking effect on α3* and α7 nAChR RTA Clade Pisauridae family 12. Megadolomedes M. australianus Mygalomorphae (“primitive” spiders) Abbreviations Blocking effect on neuronal α3* and α7 nAChR 1. Macrothele gigas M. gigas 2. Acanthoscurria cf insubtilis A. insubtilis 3. Ceratogyrus bechunanicus C. bechunanicus 4. Ceratogyrus marshalli C. marshalli 5. Ceratogyrus ezendami C. ezendami 6. Eupalaestrus campestratus E. campestratus 7. Ephebopus uatuman E. uatuman 8. Chilobrachys sp. 2 Chilobrachys sp. 2 9. Grammostola porteri G. porteri 10. Grammostola rosea G. rosea 11. Grammostola sp. (Chilean Grammostola sp. 1 North) 12. Grammostola sp. (Concepcion, Grammostola sp. 2 Chile) 13. Grammostola sp. (Maule River, Grammostola sp. 3 Chile) 14. Holothele sp. (Carabobo, Holothele sp. 1 Colombia) 15. Holothele sp. (Peru) Holothele sp. 2 16. Idiothele nigrofulva I. nigrofulva 17. Lasiodorides striatus L. striatus 18. Lasiodora striatipes L. striatipes 19. Nhandu chromatus N. chromatus 20. Nhandu tripepii N. tripepii

126

21. Nhandu vulpinus N. vulpinus 22. Phormictopus cancerides P. cancerides 23. Pamphobeteus cf petersi Pamphobeteus cf petersi 24. Psalmopoeus cambridgei P. cambridgei 25. Psalmopoeus pulcher P. pulcher 26. Psalmopoeus langenbucheri P. langenbucheri 27. Theraphosinae sp. (Chile) Theraphosinae sp. Potent blocking effect on α3* nAChR 28. Avicularia metallica A. metallica 29. Avicularia ulrichea A. ulrichea 30. Avicularia sp. (Amazonas Avicularia sp. 1 Purple, Peru) 31. Coremiocnemis tropix C. tropix 32. Chilobrachys sp. (Penang, Chilobrachys sp. 1 Malaysia) 33. Euathlus sp. (green, chile) Euathlus sp. 1 34. Euathlus sp. (“red”=”fire”, Euathlus sp. 2 Chile) 35. Euathlus vulpinus (crème, E. vulpinus Chile highland) 36. Grammostola grossa G. grossa 37. Grammostola iheringi G. iheringi 38. Orphnaecus philippinus O. philippinus 39. Orphnaecus sp. (‘treedweller’, Orphnaecus sp. Paqnai-Island, Phillippines) 40. Paraphysa scrofa P. scrofa 41. Pelinobius muticus P. muticus 42. Poecilotheria striatus P. striatus 43. Pterinochilus murinus P. murinus 44. Selenocosmia crassipes S. crassipes 45. Selenocosmia effera S. effera 46. Selenocosmia sp. (Papua Selenocosmia sp. 1

127

New-Guinea) 47. Selenocosmia sp. Selenocosmia sp. 2 48. Selenotholus foelschei S. foelschei 49. Thrixopelma cyaneolum T. cyaneolum (PERU) 50. Thrixopelma sp. (Chile) Thrixopelma sp. 51. Thrixopelma lagunas sp. T. lagunas (Peru) Potential agonistic effect on α7 nAChR. Further analysis is required to confirm the venom effect on subtype α7. 52. Haplopelma hainanum H. hainanum 53. Pamphobeteus sp. Pamphobeteus sp. (platyomma) 54. Sericopelma rubronites S. rubronites 55. Xenesthis sp. (‘white’, Xenesthis sp. Colombia) No effect on α3* and α7 nAChR 56. Aphonopelma crinitum A. crinitum 57. Avicularia sp. (Rio Madre, Avicularia sp. 2 Bolivia) 58. Homoeomma sp. (blue) Homoeomma sp. 59. Pseudhapalopus sp. Pseudhapalopus sp. (langhaarig)

Table II. List of scorpions whose venoms were used in this work Abbreviations Buthidae family 1. Androctonus australis A. australis 2. Androctonus crassicauda A. crassicauda 3. Leirurus quinquestriatus L. quinquestriatus 4. Grosphus grandidieri G. grandidieri Scorpionidae family

128

Agonistic effect on α3* and α7 nAChR 1. Heterometrus spinnifer H. spinnifer

129