Conotoxins: Chemistry and Biology

Ai-Hua Jin1*, Markus Muttenthaler*1,2, Sebastien Dutertre3, Himaya Siddhihalu

Wickrama Hewage1, Quentin Kaas1, David J Craik1**, Richard J Lewis1** and Paul F

Alewood1**

1 Institute for Molecular Bioscience, The University of Queensland, Brisbane QLD 4072,

Australia

2Institute of Biological Chemistry, Faculty of Chemistry, University of Vienna, 1090 Vienna,

Austria.

3 Institut des Biomolécules Max Mousseron, Département des acides amines, Peptides et Protéines,

Unité Mixte de Recherche 5247, Université Montpellier 2 – Centre Nationale de la Recherche

Scientifique, Place Eugène Bataillon, 34095 Montpellier Cedex 5, France

Corresponding authors:

** Professor P F Alewood, E-mail: [email protected]

**Professor D Craik, E-mail: [email protected]

**Professor R J Lewis, E-mail: [email protected]

* Contributed equally to this manuscript

** Corresponding authors

Abstract

The venom of the marine predatory cone snails (genus Conus) has evolved for prey capture and defense, providing the basis for survival and rapid diversification of the now estimated 750+ species. A typical Conus venom contains hundreds to thousands of bioactive peptides known as conotoxins. These mostly disulfide-rich and well-structured peptides act on a wide range of targets such as ion channels, G protein-coupled receptors, transporters and enzymes. Conotoxins are of interest to neuroscientists as well as drug developers due to their exquisite potency and selectivity, not just against prey but also mammalian targets, thereby providing a rich source of molecular probes and therapeutic leads. The rise of integrated venomics has accelerated conotoxin discovery with now well over 10,000 conotoxin sequences published. However, their structural and pharmacological characterization lags considerably behind. In this review, we highlight the diversity of new conotoxins uncovered since 2014, their three-dimensional structures and folds, novel chemical approaches to their syntheses, and their value as pharmacological tools to unravel complex biology. Additionally, we discuss challenges and future directions for the field.

Table of Contents

1. Introduction ...... 87

2. Conotoxin classification and nomenclature ...... 98

3. Integrated venomics ...... 1211

3.1 Transcriptomics...... 1211

3.2 Proteomics ...... 1615

3.3 Bioinformatics tools ...... 1716

3.4 New sequences ...... 1817

4. Structural diversity ...... 3433

4.1 Structures of frameworks with four cysteines ...... 4342

4.2 Structures of Frameworks with six cysteines ...... 5049

4.3 Large peptides forming dimers ...... 5251

4.4 Disulfide-poor conotoxins ...... 5352

5. Conotoxin Synthesis ...... 5453

5.1 Oxidative folding strategies ...... 5554

5.2 Directed folding strategies ...... 5655

5.3 Disulfide bond isosteres / peptide mimetics ...... 6160

5.4 Cyclic conotoxins ...... 6362

5.5 Multivalent conotoxins ...... 6463

6. Pharmacological diversity ...... 6665

6.1 Novel conotoxins acting on ion channels and transporters ...... 6665

6.1.1. Calcium channel modulators ...... 6665

6.1.2 Calcium channel modulation via GABAB receptor ...... 6867

6.2 Novel conotoxins acting on nicotinic acetylcholine receptors ...... 7069

6.2.1 nAChR modulators ...... 7069

6.2.2 Atypical α-conotoxins targeting nAChRs ...... 7170

6.3 Sodium channel modulators ...... 7473

6.4 Potassium channel modulators ...... 7776

6.5 Activities beyond voltage- and ligand-gated ion channels ...... 7877

6.5.1 Conoinsulins ...... 7977

6.5.2 RF-amide peptides ...... 8079

6.5.3 Granulin-like conotoxins ...... 8281

7. Concluding remarks and outlook ...... 8483

8 References ...... 8685

List of Figures

Figure 1: Distribution pattern of the frameworks...... 1110 Figure 2: Alignment of selected framework I conotoxins...... 1918 Figure 3: Alignment of selected framework II conotoxins...... 1918 Figure 4: Alignment of selected framework III conotoxins...... 2019 Figure 5: Alignment of selected framework IV conotoxins...... 2120 Figure 6: Alignment of selected framework V conotoxins...... 2120 Figure 7: Alignment of selected framework VI/VII conotoxins...... 2221 Figure 8: Alignment of selected framework VIII conotoxins...... 2322 Figure 9: Alignment of selected framework IX conotoxins ...... 23 Figure 10: Alignment of selected framework X conotoxins...... 2423 Figure 11: Alignment of selected framework XI conotoxins ...... 2524 Figure 12: Alignment of selected framework XII conotoxins ...... 2625 Figure 13: Alignment of selected framework XIII conotoxins...... 2625 Figure 14: Alignment of selected framework XIV conotoxins...... 2726 Figure 15: Alignment of selected framework XV conotoxins ...... 2726 Figure 16: Alignment of selected framework XVI conotoxins ...... 2827 Figure 17: Alignment of selected framework XVII conotoxins ...... 2827 Figure 18: Alignment of selected framework XVIII conotoxins ...... 2928 Figure 19: Alignment of framework XIX conotoxins...... 2928 Figure 20: Alignment of selected framework XX conotoxins ...... 3029 Figure 21: Alignment of selected framework XXI conotoxins...... 3029 Figure 22: Alignment of selected framework XXII conotoxins ...... 3130 Figure 23: Alignment of selected framework XXIII conotoxins ...... 3130 Figure 24: Alignment of framework XXIV conotoxins ...... 3130 Figure 25: Alignment of framework XXV conotoxin ...... 3231 Figure 26: Alignment of framework XXVI conotoxins ...... 3231 Figure 27: Alignment of framework XXVII conotoxin ...... 3332 Figure 28: Alignment of framework XXVIII conotoxin ...... 3332 Figure 29: Structural variation and inclusion of new cysteine frameworks from the folds initially defined in Akondi et al., 2014...... 4544 Figure 30: Structures of new conotoxin folds...... 4746 Figure 31: Analysis of the conformation and energy of disulfides of the structure of Framework I...... 4847 Figure 32: Diversity of ω-conotoxins...... 6766 Figure 33: Chemically engineered selective analogs of Vc1.1 (A) and RgIA (B)...... 6968 Figure 34: N terminal mutants derived from αD-GeXXA and their functionality...... 7372 Figure 35: Recently discovered δ-conotoxins and their activities...... 7675 Figure 36: Cone snail venom derived conoinsulins and their activities...... 7978 Figure 37: Sequence alignment of all characterized conoRFamides with structurally similar human neuropeptide FF and Lymnaea stagnalis derived neuropeptide FMRFamide...... 8180 Figure 38: Comparison of the secondary structure between conotoxins and recombinant human granulin A...... 8382

List of Tables

Table 1: Cysteine frameworks and their pharmacological family in Conus venom...... 109 Table 2: Conus venom transcriptomes ...... 1413 Table 3: Overview of post-translational modifications ...... 16 Table 4: Classification of all known three-dimensional structures of wild-type and synthetic conotoxin into folds and sub-folds...... 3534 Table 5: Connectivities of cysteine frameworks with six cysteines and their fold...... 5250

List of Schemes

Scheme 1. Regioselective off-resin folding strategy for the synthesis of conoinsulin Con-Ins G1 using cysteine-to-selenocysteine replacement (Se-Mob) in combination with S-Trt and S-Acm protecting groups by Fmoc chemistry (Safavi-Hemami et al., 2015)...... 5756 Scheme 2. Regioselective off-resin folding strategy for the synthesis of the antiparallel dimeric N-terminal domain of αD-GeXXA using DTNB (5,5’-dithiobis-(2-nitrobenzoic acid)) in combination with S-Acm. DTNB activates the free thiol of the A-chain, thereby facilitating the desired regioselective interchain disulfide bond formation upon mixing it with the A- with the B-chain at pH 7.3. Iodine oxidation removes then the Acm protecting group forming the second intermolecular disulfide bond...... 5857 Scheme 3. Regioselective off-resin folding strategy for the synthesis of hydrophobic δ-conotoxins using an acid-cleavable solubility tag and Fmoc chemistry. Four lysine residues were coupled to the Rink Amide resin followed by acylation with the phenylacetamido (PAM) linker. Regioselective folding was achieved using pairs of S-Trt, S-Acm, and S-tBu orthogonal protecting groups. The solubility tag was in the end removed by treatment with HF...... 5958 Scheme 4. Regioselective on-resin folding strategy for the synthesis of α- LvIA using a three-glycine spaced Rink Amide ChemMatrix resin, Fmoc chemistry and (A) a combination of Allocam and Trityl protecting groups, or (B) a combination of Allocam and Mmt/StBu protecting groups. Removal of the Allocam protecting group is either achieved by palladium or iodine. Allocam, allyloxycarbonylaminomethyl; DMSO, dimethylsulfoxide; DTNP, 2,2’-dithiobis(5-nitropyridine); HS- (CH2)2-OH, 2-mercaptoethanol; Mmt, monomethoxy trityl; NMM, N-methyl morpholine; Npys, 2-(5- nitropyridyl); TFA, trifluoroacetic acid...... 6059 Scheme 5. Dicarba disulfide bond mimetics. A protected diaminodiacid building block (Fmoc, Alloc and Ally protecting groups) is used instead of a cysteine residue during the Fmoc-SPPS chain assembly of μ- SIIIA. Chain assembly is stopped before the paired cysteine residue. Cyclization is carried out after deprotection of the Alloc, Ally and Fmoc protecting groups. The rest of the peptide sequence is then assembled using standard Fmoc-SPPS. Following cleavage from resin, the remaining two disulfide bonds are formed in 0.1 M Tris buffer at pH 7.5...... 6261 Scheme 6. Two disulfide bonds from a single amino acid position. L-4,5-dithiolnorvaline (L-Dtn) is incorporated into α-ImI instead of the first cysteine residue, whereas the adjacent cysteine residue is replaced by alanine. The dithiol analog folds into a bioactive fold that is structurally highly similar to α- ImI (compared by NMR) and displays a 7.6-fold increased potency at the nAChR α7 receptor compared to α-ImI...... 6362 Scheme 7. Intramolecular native chemical ligation to produce N-C backbone cyclized conotoxins. A. The conotoxin sequence is assembled by SPPS including a short linker region (light grey) that bridges the distance between the N- and C-termini (generally determined by NMR). A C-terminal thioester and N- terminal cysteine are required for the intramolecular ligation reaction to occur, which first undergoes transesterification, followed by a S to N acyl shift, resulting in backbone cyclization via a peptide bond. B.

Example of the 3D NMR structure of a cyclized α-conotoxin (cyclic-MII with a 6-residue linker, PDB 2jaw)...... 6463

1. Introduction

Many organisms including snakes, spiders, scorpions, cone snails, anemones and some mammalian species use venom as either a defense mechanism or a weapon for prey capture (Prentis et al., 2018,

Pennington et al., 2018, Calderon-Celis et al., 2017, Dutertre et al., 2014). Conus venoms contain a diverse cocktail of bioactive disulfide-bond rich peptides, called conotoxins, that act on a wide range of targets including ion channels, G protein-coupled receptors (GPCRs), transporters and enzymes (Akondi et al.,

2014, Vetter and Lewis, 2012). Their disulfide bond frameworks stabilize compact loop structures that often contain protein-like secondary motifs such as  helices,  turns and  sheets, which are responsible for their high potency, exquisite receptor subtype selectivity and resistance to proteases, making them attractive neurological tools and leads for drug development.

Owing to their structural stability, relatively small size and target specificity, conotoxins are regarded as ideal molecular probes for target validation and peptide drug discovery. An early focus of conotoxin discovery has been to identify and characterize novel pain modulators, which was the logical first step considering that conotoxins often target ion channels involved in pain signaling. A translational outcome of these efforts is the FDA-approved drug Prialt (synthetic -MVIIA), an N-type calcium channel blocker identified from C. magus, used for the treatment of severe chronic pain (Pope and Deer, 2013).

Since this ground-breaking effort, -MrIA (Nielsen et al., 2005), -CVID (Lewis et al., 2000), contulakin

G (Barton et al., 2004) and α-Vc1.1 (Sandall et al., 2003) have entered clinical trials, but for various reasons their development was halted. Despite these setbacks, the number of conotoxin patents is increasing

(Starobova et al., 2018, Himaya and Lewis, 2018, Robinson et al., 2017c, Durek and Craik, 2015),

(Pennington et al., 2018) and conotoxin drug discovery research is accelerating due to recent innovations in sequencing, synthesis, and structural and pharmacological methods, further supporting therapeutic development opportunities. This review updates the recent advancements in conotoxin chemistry and biology since our last major review in this journal in 2014 (Akondi et al., 2014).

2. Conotoxin classification and nomenclature

Cone snails provide one of the highest venom diversities among venomous animals with current estimates of 1 million different bioactive conotoxins to exist (Dutertre and Lewis, 2012). Less than 1% of these have been sequenced (~10,000) and only a small fraction of these have been characterized pharmacologically.

Conotoxins are translated from mRNA as peptide precursors, typically with a conserved signal peptide region, a propeptide region and a mature peptide region (Terlau and Olivera, 2004). During the last ten years, most conotoxins have been discovered through next-generation sequencing (NGS) of the venom duct transcriptome. Conotoxin gene superfamily classify conotoxins based on similarities between their consensus signal sequences (Kaas et al., 2012). Although conotoxins within a superfamily typically share a similar signal peptide sequence, significant structural and functional diversity can be found for the encoded venom peptides (Robinson and Norton, 2014).

The sequence of the mature peptide region is highly diverse, in keeping with the high variety of conotoxins discovered in the venom. However, a key feature of conotoxins is their highly-conserved cysteine framework, which is also used to categorize them (Table 1) (Akondi et al., 2014). The resultant disulfide bonds often stabilize protein-like secondary structures and unique bioactive folds. There are currently 28 framework families (Figure 1) catalogued in ConoServer based on the number of cysteine residues, their loop size and distinct disulfide bond connectivity, among them only Framework XXVII

(2017) and XXVIII (2015) have been characterised at both the proteomic and transcriptomic level since our earlier review. Another 16 have been reported from recent transcriptome studies but remain uninvalidated at the peptide level (Lavergne et al., 2015), (i.e., not yet identified by proteomics in the collected venom of the cone snail).

Table 1: Cysteine frameworks and their pharmacological family in Conus venom.

Framework Cysteine pattern Disulfide bonds Pharmacological family I CC-C-C 2 α, ρ II CCC-C-C-C 3 α III CC-C-C-CC 3 α, ι, κ, μ IV CC-C-C-C-C 3 α, κ, μ V CC-CC 2 ε, μ, τ VI/VII C-C-CC-C-C 3 δ, γ, κ, μ, ω VIII C-C-C-C-C-C-C-C-C-C 5 α, σ IX C-C-C-C-C-C 3 N.D. X CC-CXOC 2 χ XI C-C-CC-CC-C-C 4 ι, κ XII C-C-C-C-CC-C-C 4 N.D. XIII C-C-C-CC-C-C-C 4 N.D. XIV C-C-C-C 4 α, κ XV C-C-CC-C-C-C-C 4 N.D. XVI C-C-CC 4 N.D. XVII C-C-CC-C-CC-C 4 N.D. XVIII C-C-CC-CC 3 N.D. XIX C-C-C-CCC-C-C-C-C 5 N.D. XX C-CC-C-CC-C-C-C-C 5 α XXI CC-C-C-C-CC-C-C-C 5 N.D. XXII C-C-C-C-C-C-C-C 4 N.D. XXIII C-C-C-CC-C 3 N.D. XXIV C-CC-C 2 α XXV C-C-C-C-CC 3 N.D. XXVI C-C-C-C-CC-CC 4 ω XXVII C-C-C-CCC-C-C 4 φ XXVIII C-CC-C-C-C 3 δ/κ X, any amino acid; O, hydroxyproline. N.D. not determined at molecular level. α, Nicotinic acetylcholine receptors; γ, Neuronal pacemaker cation currents (inward cation current); δ, Voltage-gated Na channels (agonist, delay inactivation); ε, Presynaptic

Ca channels or G protein-coupled presynaptic receptors; ι, Voltage-gated Na channels (agonist, no delayed inactivation); κ,

Voltage-gated K channels (blocker); μ, Voltage-gated Na channels (antagonist, blocker); ρ, Alpha1-adrenoceptors; σ,

Serotonin-gated ion channels 5-HT3; τ, Somatostatin receptor; χ, Neuronal noradrenaline transporter; ω, Voltage-gated Ca channels (blocker) and φ, granulin activity.

Three frameworks dominate the landscape (Figure 1): Framework VI/VII conotoxins that have the

10 stable inhibitory cystine knot (ICK) motif (Norton and Pallaghy, 1998) rank number one with 719 members identified. Framework I, the α-conotoxins, ranks second with 691 members, followed by Framework III, the μ-conotoxins, with 430 members.

Figure 1: Distribution pattern of the frameworks. Representative structures of Framework I and X are shown. The current numbers of identified conotoxins are labeled on top of each framework bar.

A IUPHAR (International Union of Basic and Clinical Pharmacology) nomenclature that is based on conotoxin framework and gene superfamily is used to describe conotoxins (Akondi et al., 2014, Kaas et al., 2008). There are currently 28 gene superfamilies where each superfamily is subdivided based on the framework employed (Akondi et al., 2014, Kaas et al., 2012, Kaas et al., 2008, Robinson and Norton, 2014).

A simple example is that of the A superfamily which now has the six frameworks, I, II, IV, VI/VII, XIV

11 and XXII. As their pharmacology is uncovered a Greek letter is added to the nomenclature, e.g., α-GI, where α indicates that the conotoxin targets the nicotinic acetylcholine receptor (nAChR), G indicates its origin C. geographus, and I indicates its Framework I; a letter is often used to define variants (Akondi et al., 2014, Kaas et al., 2012, Kaas et al., 2008). When the target is unknown, the Greek letter is omitted and the species letter is written in lower case with an Arabic number for the framework and small letters for conotoxin variants (e.g., μ-reg3b). An alternative nomenclature for peptide toxins proposed by King et al Formatted: Highlight in 2008 (King et al., 2008) that divides the toxin name into three parts (activity, biological source and relationship to other toxins) has not been widely adopted and the original scheme of McIntosh et al in 1999

(McIntosh et al., 1999) continues to be used to name conotoxins.

3. Integrated venomics

Transcriptomic analyses of the cone snails’ venom ducts have led to a dramatically growing number of conotoxin transcript sequences. The term 'integrated venomics' describes the systematic study of the whole toxin profile of venom and venom ducts by integrating transcriptomes, proteomes and bioinformatics

(Oldrati et al., 2016, King, 2011, Calvete, 2017). It is a powerful approach, as next-generation sequencing can detect precursors as well as rare variants not possible through classical proteomics (Dutt et al., 2019,

Robinson et al., 2017b, Li et al., 2017, Peng et al., 2016, Lavergne et al., 2015, Jin et al., 2015b, Himaya et al., 2015, Barghi et al., 2015b). Combined with high-sensitivity mass spectrometry, this approach has revealed a new level of conotoxin diversity (Himaya et al., 2018, Rodriguez et al., 2015, Lavergne et al.,

2015, Gao et al., 2017).

3.1 Transcriptomics.

The advent of high-throughput sequencing technologies such as the 454 GS FLX Titanium (Roche),

Solexa GAII (Illumina), APG SOLiD 3 (Life) or HeliScope (Helicos Biosciences) platforms revolutionized transcriptomic research. To date, the venom duct transcriptomes of 30 species (5 fish-, 4 mollusk-, and 20

12 worm-hunters, plus the generalist C. californicus), have been sequenced (Table 2). They form a representative dataset of the whole cone snail world (30 pieces of 750 the whole jigsaw WoRMS: http://www.marinespecies.org/).

The Roche 454 pyrosequencing platform was extensively used from 2008-2013 (Mardis, 2017) and advanced our understanding on venom diversity, post-translational processes and the mechanisms governing the observed remarkable conotoxin diversity. In a comparative example of analyzing the venom duct of C. striatus, the 454 platform generated up to 750,000 expressed sequence tags (ESTs) in a single run compared to only 897 ESTs generated by Sanger sequencing (Pi et al., 2006). The long reads (350 base pairs (bp) on average, up to 700 bp, single-end) were valuable as they provided complete sequence and annotation information for species where no reference genome sequence was available. With the discontinuation of 454 technology in 2013, Illumina is now the most widely used system and provides higher quality reads compared to the 454 platform due to the shorter and more accurate sequencing results; nevertheless, the data require a more sophisticated data assembly (Mardis, 2017). The classic assembly concept derives from early genome sequencing, combining short-insert, paired-end and long-insert sequences to maximize coverage. Although most assembly software have incorporates carefully optimized parameters and validation procedures to remove artifacts (Allam et al., 2015) there remains a problematic bias with the Conus venom duct assembly: most algorithms (such as ABySS, Mira, Trinity, Velvet and

Oases) are designed to reduce substitution, deletion and insertion events, which, in the case of hypervariable genes such as the ones of conotoxins, can eliminate many of the minor, yet important, true biological variations (Jin et al., 2019b, Lavergne et al., 2015).

Alternative sequencing platforms include the more recently developed PacBio (Rhoads and Au,

2015) system, which generates longer reads (1500 bp), although also with a high error rate that makes this system suboptimal for Conus transcriptome studies. The recent Thermo fisher ABI 3730 platform generates intermediate length reads (500 bp) with low error rates, which makes itthis a promising alternative considering that the average conotoxin precursor is ~70 amino acids (~210 bp) long.

Table 2: Conus venom transcriptomes

Species Platform Raw reads Assembly Accession No. Proteome Reference C. bullatus Illumina 102,278,116 Abyss N/A (Hu et al., 454 848,394 Roche 2011) C. consors 454 213,561 SeqMan Pro N/A (Terrat et al., 2012) C. geographus 454 P, 167,211 MIRA NCBI (Hu et al., SRR503413-416 PC, 238,682 2012) DC, 186,398 D, 199,680 C. pulicarius 454 359,213 Forge-G N/A (Lluisma et al., 2012) C. marmoreus 454 179,843 Newbler DDBJ Yes (Dutertre et AB850695-852 al., 2013, Lavergne et al., 2013) C. miles 454 255,829 Newbler Genbank Yes (Jin et al., KP216816-863 2013) C. victoriae 454 701,536 MIRA NCBI Yes (Robinson SRR833564 et al., 2014) C. planorbis 454 93,184 Newbler N/A Yes (Jin et al., 2015b) C. episcopatus Illumina 20,885,730 SOAP, DDBJ Yes (Lavergne Oases, DRA003531, et al., 2015)

Trinity PRJDB3896, SAMD00029744, DRX030964, DRR034331, SAMD00029745, DRX030965, DRR034332, SAMD00029746, DRX030966, DRR034333 C. catus 454 136,495 Newbler N/A Yes (Himaya et al., 2015) C. miliaris Illumina 131,525,916 Trinity NCBI (Genomic PRJNA257931, 130,566,380 Resources SRP045405 Developme 87,201,606 nt et al., 2015) C. tribblei 454 121,139 Newbler SRR1803937, (Barghi et Illumina 25,825,187 Trinity SRR1803938, al., 2015b) C. lenavati Illumina 32,335,757 SRR1803939, (Barghi et SRR1803940, 28,637,629 al., 2015a) 33,027,986 SRR1803941,

SRR1803942

C. vexillum 454 138,641 Newbler SRR2890189 Yes (Prashanth et al., 2016) C. arenatus Illumina 27,258,786 Trinity SRX1323883-894 (Phuong et C. coronatus 24,688,386 al., 2016) C. ebraeus 22,790,706 C. imperialis 24,788,128 C. lividus 26,203,686 C. marmoreus 26,116,312 C. quercinus 22,956,900 C. rattus 28,843,008 C. sponsalis 28,987,708 C. varius 28,480,236 C. virgo 21,635,650 C. californicus 26,639,152 C. betulinus Illumina S, 50,926,032 Trinity, N/A (Peng et al., M, 93,247,084 SOAP 2016) B, 51,896,884 denovo- Trans ABI 3730 11,026

C. gloriamaris Illumina 42,602,912 Trinity NCBI Yes (Robinson SRR5499408 et al., 2017b) C. andremenezi Illumina S1: 58, 834, 536 Trinity GenBank (Li et al., S2: 54, 177, 324 MF576542-988. 2017) C. praecellens S1: 70, 826, 570 S2: 32, 695, 570 C. tulipa 454 S1: 100,564 Newbler N/A Yes (Dutt et al., S2: 33,516 2019) C. imperialis 454 S1: 249,349 Newbler and GenBank Yes (Jin et al., S2: 379,998 Trinity KT377395-426 2019b)

The venom duct transcriptomes of many cone snail species have been studied (Table 2), including the deadliest C. geographus (Dutertre et al., 2014, Hu et al., 2012) to the most beautiful C. gloriamaris

(Robinson et al., 2017b), from piscivores C. tulipa (Dutt et al., 2019), molluscivores C. marmoreus

(Dutertre et al., 2013), vermivores C. imperialis (Jin et al., 2019b) to the early diverging taxa scavenger C. californicus (Mardis, 2017). On average, >100 conotoxin sequences at the gene sequence level and >1,000 conotoxins at the peptide level are observed for each species (Himaya and Lewis, 2018, Robinson et al.,

2017c, Gao et al., 2017, Oldrati et al., 2016), providing a valuable resource for future drug discovery efforts.

A recent highlight of Conus venom transcriptomic research included the parallel analysis of two C. tulipa 15 specimens revealing striking differences in conotoxin expression levels between individuals with broad overlap at the superfamily level but dramatic variation at the individual sequence level (Dutt et al., 2019).

3.2 Proteomics

Proteomics has become an integral part of the venom peptide discovery pipeline that benefits from the rapid technological advances of mass spectrometry, its powerful software and rapidly expanding databases (Table 2). It is thus not surprising that the observed Conus peptide diversity drastically increased

(Prashanth et al., 2014), including the discovery of enzymes and hormones (Robinson et al., 2017a).

Intriguingly, insulin analogs are also part of the injected venom cocktail. Insulins from C. geographus, tulipa and kinoshitai exhibit quite diverse conoinsulin sequences that are active on mollusk, fish, but also on the human insulin receptors (Ahorukomeye et al., 2019, Safavi-Hemami et al., 2016).

Proteomics also played a major role in the discovery that cone snails can switch between different venom compositions depending if they prey or defend themselves (Calvete, 2017, Dutertre et al., 2014).

Proteomics was furthermore used to demonstrate that venoms of different specimens of the same species displayed a considerable intraspecific variation (Himaya et al., 2018, Rodriguez et al., 2015, Himaya et al.,

2015, Violette et al., 2012). For example, a study of predation-evoked venom of two specimens of C. imperialis revealed a strong correlation between transcription and translation of highly expressed conotoxins but not those expressed at lower levels (Jin et al., 2019b). A combination of bioanalytical techniques to uncover the extent of venom expression variability in C. purpurascens, demonstrated pronounced intraspecific venom variability, as well as dramatic difference between dissected and injected venoms (Rodriguez et al., 2015).

In addition to the intraspecific variation, post-translation modifications (PTMs) have contributed to the diversity of Conus peptides. Table 3 provides an overview of PTMs discovered to date. Individual examples have been sourced from VenomZone < https://venomzone.expasy.org/>.

Table 3: Overview of post-translational modifications

PTMs Entries Common Position Mass shift Amidation 450 C-terminal -0.9840 Bromination 47 Trp 77.9465 D-amino acid 58 Trp, Leu, Phe, Val 0 Disulfide bond 1226 Cys -2.0156 Gamma-Carboxy 118 Glu 43.9898 Glycosylation 14 Thr - Hydroxylation 293 Pro, Val, Lys 15.9949 Oxidation 13 Trp, His, Met 15.9949 Pyroglutamic 67 N-term Gln -17.0265 S-cysteinyl cysteine 1 Cys 119.0041 Sulfation 8 Tyr 79.9568

3.3 Bioinformatics tools

The high throughput technologies used in venomics (Escoubas and King, 2009, Vetter et al., 2011), in particular transcriptomics, generate large datasets that require bioinformatics support to fully explore their potential (Prashanth and Lewis, 2015). Besides the tools available through ConoServer (Kaas et al.,

2012), ConoSorter (Lavergne et al., 2013) and ConoDictor (Koua et al., 2012), several other methodologies have recently emerged to improve the accuracy of sequence identification and to automate classification into superfamilies (Prashanth et al., 2014).

Early attempts using in silico approaches to classify conotoxin superfamilies used a mathematical representation of a peptide, taking into account its pseudo amino acid composition (Mondal et al., 2006).

Of the methods tested (Blast, ISort predictor, least Hamming distance algorithm, least Euclidean distance algorithm and multi-class support vector machines), the multi-class support vector machines (SVM) performed the best, with an overall accuracy of 88% correct predictions from a test batch containing A, M,

O and T superfamilies and a negative set of unrelated cysteine rich peptides from different eukaryotes.

Later, an integrated feature-based approach (PredCSF) gave 91% accuracy (Fan et al., 2011), predictions using diffusion map dimensionality reduction and a subspace classifier (dHKNN) gave 92% accuracy (Yin et al., 2011), a binomial distribution and radial basis function network approach gave 86% (Yuan et al.,

2013), while a sequence-based predictor (iCTX-Type) that incorporated dipeptide occurrence frequencies into general pseudoamino acid composition gave 91% accurate predictions (Ding et al., 2014). Overall, it 17 seems that in silico approaches can successfully rival manual curation and annotation, though the complicated mathematics involved likely prevented widespread use; more recently some became available on user friendly web-servers (Ding et al., 2014).

Further, mMachine learning tools have been also used to identify and classify conotoxins into superfamilies (Dao et al., 2017, Mansbach et al., 2019). First, a random forest-based predictor (ICTCPred) with hybrid features and the SMOTE technique gave 91% accuracy when adopting the Relief-IFS method

(Zhang et al., 2016). Next, Wu et al. (2016) incorporated three new pseudo-amino acid properties to support vector machine (SVM) classifier (Mondal et al., 2006) to achieve 95% accuracy (Wu et al., 2016).

Combining an analysis of variance and correlation (AVC) in the SVM model produced an overall accuracy of 92% (Xianfang et al., 2017). Finally, ConusPipe, a new tool that integrates three machine learning models

(logistic regression, semi-supervised learning and an artificial neural network) retrieved conotoxin sequences from the venom gland transcriptomes of ten different Conus species with an overall accuracy of

96–98% (Li et al., 2018). Alternatively, nuclear magnetic resonance (NMR) has been used to identify short sequence tags that can be used to rapidly recover the full-length peptide sequences from transcriptome database interrogation (Wilson and Daly, 2018). Importantly, this method is sample sparing and does not require any prior manipulation of the sample (e.g. protease digestion).

3.4 New sequences

Given that the functions of most of the emerging new sequences areis undetermined, we have chosen to describe the function of selected new sequences based on their cysteine rich frameworks.

Framework I CC-C-C

Framework I describes the common cysteine framework of α-conotoxins, observed in most Conus venoms. Two interesting examples are α-RegIIA and α-MrIC (Figure 2). α-RegIIA is the most potent (46 nM) antagonist at the human α3β4 nicotinic acetylcholine receptor (nAChR), though with poor selectivity

18 as it is also active against the α3β2 and α7 subtypes (Cuny et al., 2016). α-MrIC is the first co-agonist for the nAChR, which exclusively activates α7 if modulated by type II allosteric modulators including

PNU12059, TQS and SB-206553 (Mueller et al., 2015, Jin et al., 2014).

Figure 2: Alignment of selected framework I conotoxins. Sequences are extracted from the ConoServer database. Individual references are listed in ConoServer in each peptide page. The same applies to Figure

3-28.

Framework II CCC-C-C-C

Only four conotoxins have been identified from this framework II (Figure 3) with three consecutive cysteine residues. VxII, which was initially isolated in 2002 by venom purification of C. vexillum (Jiang et al., 2006), has been rediscovered in the venom gland transcriptome and proteome of C. vexillum recently

(Prashanth et al., 2016). Cp2-DD02 was discovered at the nucleic acid level (Zhou et al., 2013) and

ConoServer predicted the mature sequence of Cp2-DD02 from its protein precursor. Synthetic VxII (Jiang et al., 2006) was injected intracranially into 4-week-old mice, where it resulted in a series of symptoms, such as being sedative, tail stiffening and inducing twisted jumping.

Figure 3: Alignment of selected framework II conotoxins.

Framework III CC-C-C-CC

μ-Conotoxins have been very valuable tools to study sodium channel subtypes (Tosti et al., 2017,

Israel et al., 2017, Green et al., 2014, Tietze et al., 2012, Norton, 2010). 10 μ-conotoxins have been discovered/revisited during the last 5 years, including μ-Mi040 (Jin et al., 2013), μ-SmIIIA, μ-SIIIA, μ-

GIIIA, μ-PIIIA, μ-SxIIIA, μ-BuIIIB, μ-Vx3-VV01, μ-reg3b (Franco et al., 2018), and μ-KIIIA. Depending on the number of residues located in the last loop (between Cys4 and Cys5), this framework has been divided into sub branches, namely M-1, -2, -3 and -4 (Corpuz et al., 2005)(Figure 4). Different disulfide connectivity had been reported between and within the sub-branch conotoxins (Du et al., 2007). The subtypes show little sequence homology, and their loop sizes vary (Franco et al., 2018).

Figure 4: Alignment of selected framework III conotoxins. Listed in brackets are the subtypes.

Framework IV CC-C-C-C-C

The majority of the 58 framework IV conotoxins (Kaas et al., 2012) have been characterized from the venom of fish-hunters and belong to the κA-family of conopeptides. Some recent analogs of this group include the three C4.1-C4.3 peptides from C. catus (Himaya et al., 2015) and CcTx, which is a glycopeptide from C. consors (Hocking et al., 2013)(Figure 5).

Figure 5: Alignment of selected framework IV conotoxins.

Framework V CC-CC

Framework V peptides are small (8-17 amino acids) and found in a variety of species independent of diet (Remigio and Duda, 2008). Their pharmacology is unknown and unlike α-conotoxins their sequences are widely divergent. They also exhibited many post-translational modifications such as gamma- carboxyglutamate (γ), bromotryptophan (W), hydroxyproline (O), glycosylated threonine (T) and C- terminal amidation (*), examples are TxVA, Gla-MrIII and Gla-MrIV (Walker et al., 1999, Hansson et al.,

2004) (Figure 6). 78 framework V sequences have been predicted from transcriptomic work in the last five years, Mr5.4a, Mr5.4b, Mr5.5 and MrVA had been identified at the proteomic level (Dutertre et al., 2013), no further PTM characterization was discussed.

Figure 6: Alignment of selected framework V conotoxins.

Framework VI/VII C-C-CC-C-C

The FDA approved drug ω-MVIIA is a VI/VII framework. Framework VI/VII contains the largest group of conotoxins characterized so far with 719 entries in ConoServer. Among the newly characterized framework V/VII peptides (Figure 7), δ-SuVIA is a potent vertebrate-active δ-conotoxin characterized from a vermivorous cone snail C. suturatus. -SuVIA is equipotent at hNav1.3, hNav1.4 and hNav1.6 with EC50 in the low nanomolar range (Jin et al., 2015a). Another interesting peptide, μO§-GVIIJ from C. geographus, has a unique posttranslational modification, S-cysteinylated cysteine, which makes possible formation of a covalent tether of peptide to its target Na channels at a distinct ligand-binding site (Gajewiak et al., 2014). 21

ω-MoVIA and ω-MoVIB were isolated and characterized from a vermivorous cone species C. moncuri, and they potently inhibited human Cav2.2 in fluorimetric assays and rat Cav2.2 in patch clamp studies

(Sousa et al., 2018b). Interestingly, the arginine at position 13 in ω-MoVIA and ω-MoVIB was determined to be critical for activity, even though this position iswas usually a functionally critical tyrosine in piscivorous ω-conotoxins (Sousa et al., 2018b).

Figure 7: Alignment of selected framework VI/VII conotoxins.

Framework VIII C-C-C-C-C-C-C-C-C-C

Framework VIII is of pharmacological interest with a total of 18 entries in ConoServer (Figure 8).

The first discovered member σ-GVIIIA (England et al., 1998) was purified from the venom of C. geographus and blocks the 5-hydroxytryptamine (5-HT3) receptor, an excitatory serotonin-gated ion channel. σ-GVIIIA has a brominated tryptophan and a hydroxylated derivative of tryptophan. Another member of the framework VIII family, RVIIIA from C. radiatus , on the other hand has an unusually broad targeting specificity for nicotinic acetylcholine receptor (nAChR) subtypes (Teichert et al., 2005). No new sequences have been discovered in the past 5 years although serotonin receptors remain interesting molecular targets for neurotoxins.

Figure 8: Alignment of selected framework VIII conotoxins.

Framework IX C-C-C-C-C-C

With 24 entries in ConoServer TxIXA (Figure 9) was the first framework IX peptide purified from the venom of C. textile venom and it elicited a distinctive spasmodic symptomatology in mice (Lirazan et al., 2000). A highly homologous peptide gm9a was cloned from C. gloriamaris (Miles et al., 2002). These peptides share a distinctive cysteine pattern to framework VI/VII, but they have been reported having a similar cystine knot motif (Miles et al., 2002). In fact, two P-superfamily conotoxins, gm9a and bru9a, had been backbone cyclized by joining the N- and C-termini with short peptide linkers (Akcan et al., 2015).

The cyclized derivatives maintained their potency and had conformations similar to the ICK cystine knot motif (Akcan et al., 2015).

Figure 9: Alignment of selected framework IX conotoxins 23

Framework X CC-CXOC

χ-MrIA and χ-MrIB are two peptides discovered from the venom of C. marmoreus (Figure 10).

They are selective inhibitors of the human norepinephrine transporter (hNET) and therefore are potentiala drug candidates for attenuating chronic neuropathic pain (Sharpe et al., 2001). χ-MrIA analogs with different truncations of the pro-peptide that contains portions of the χ-MrIA molecule have been found by variable processing mechanism through integrated venomics techniques (Dutertre et al., 2013). Follow-up investigations confirmed that analogs with comparable inhibitory activity to χ-MrIA all maintained the three pharmacophore residues identified previously (Brust et al., 2009), along with all four cysteine residues, indicating the importance of maintaining the framework integrity (Ziegman et al., 2019).

Figure 10: Alignment of selected framework X conotoxins.

Framework XI C-C-CC-CC-C-C

The first group of framework XI peptides are r11a, r11b, r11c, r11d and r11e which were purified from C. radiatus venom (Jimenez et al., 2003). These peptides elicit action potentials on amphibian peripheral axons and were grouped into the 'lightning-strike cabal' of toxins that produce instant immobilization of fish prey (Jimenez et al., 2003). Framework XI conotoxins are abundant (Dutertre et al.,

2014, Jin et al., 2013, Dutertre et al., 2013, Hu et al., 2012, Lluisma et al., 2012) with 67 ConoServer entries and they are valuable probes for dissecting the molecular components of axons. The most recently discovered member, Xm11a (Figure 11), from the venom of C. ximenes is the first conotoxin that is able to

24 inhibit the growth of M. tuberculosis at a low micromolar concentration with a potency similar to that of two other currentlt available clinical drugs (Figueroa-Montiel et al., 2018).

Figure 11: Alignment of selected framework XI conotoxins

Framework XII C-C-C-C-CC-C-C

The first framework XII peptide was discovered from C. gloriamaris Gla-MrII (Hansson et al.,

2004) and contains 50 amino acids long, four4 disulfide bonds and five5 γ-carboxyglutamic acid residues.

Gla-MrII was recently observed in a transcriptome and proteome study of C. marmoreus (Dutertre et al.,

2013) (Figure 12). No pharmacology and or structure determination have been reportedcarried out on it so far.

Figure 12: Alignment of selected framework XII conotoxins

Framework XIII C-C-C-CC-C-C-C

Displaying an uncommon framework, de13a iswas a major component in the venom of C. delessertii and has a high content of PTMs, including hydroxylation of four residues, bromination of one residue and amidation of the C-terminus (Aguilar et al., 2005)(Figure 13). De13b iswas a close analog (88% sequence identity) to de13a and was discovered by cDNA cloning. With the high density of PTMs, the synthesis and correct folding of the de13 conotoxins are crucial before any structure-function studies can commence.

Mi044 from C. miles with different loop sizes was uncovered using an integrated venomics approach (Jin et al., 2013). Their pharmacology has yet to be determined.

Figure 13: Alignment of selected framework XIII conotoxins. K hydroxylysine.

Framework XIV C-C-C-C (79)

Framework XIV was defined with the discovery of FlfXIVA-C from the venom of C. floridanus floridensis, and VilXIVa from the venom of C. villepinii (Moller et al., 2005) (Figure 14). It was described as a four-cysteine and three-loop conotoxin (Moller et al., 2005). Nevertheless, a peptide with Framework

XIV, the scratcher peptide was discovered in 1990 (Olivera et al., 1990). In 2016, Cal14.1a from the sea

26 snail C. californicus was found to decrease cell viability, activate caspases, and reduce expression of the prosurvival protein NFκB-1(Oroz-Parra et al., 2016), making it the first reported conotoxin having apoptotic activity in human lung cancer cell lines (Oroz-Parra et al., 2016).

Figure 14: Alignment of selected framework XIV conotoxins.

Framework XV C-C-CC-C-C-C-C (28)

ViXVA was the first characterized Framework XV peptide from the venom of a worm-hunting species C. virgo (Peng et al., 2008). ItViXVA was assumed to adopt an "ICK+1" disulfide bond connectivity, but has yet to be characterized at the structural and pharmacological levels (Figure 15).

Figure 15: Alignment of selected framework XV conotoxins

Framework XVI C-C-CC

Qc16a with Fframework XVI (Figure 16) was identified from the venom of vermivorous C. quercinus (Ye et al., 2011). The disulfide bond connectivity was determined to be CysI-CysIV and CysII-

CysIII, similar to Framework V. The nuclear magnetic resonance structure of Qc16a adopts a ribbon conformation with a simple β-turn motif (Ye et al., 2011) similar to χ-MrIA. Qc16a causes symptoms of

27 depression in mice when injected intracranially (Ye et al., 2011).

Figure 16: Alignment of selected framework XVI conotoxins

Framework XVII C-C-CC-C-CC-C

ca16a peptide was reassigned by ConoServer as Framework XVII (Figure 17) having been previously described as Framework XVI. ca16a was purified, sequenced, and cloned from a worm-hunting cone snail, C. caracteristicus (Yuan et al., 2008). It was described as an extremely hydrophilic peptide which was rich in polar residues of Gly, Ser, and Thr as well as a hydroxylated Pro residue. Its pharmacology is yet to be determined.

Figure 17: Alignment of selected framework XVII conotoxins

Framework XVIII C-C-CC-CC

BeTXIIa, with Framework XVIII (Figure 18) from C. betulinus, was isolated and sequenced in 1999

(Chen et al., 1999). S18.1 is a predicted sequence from an early EST cDNA library from the venom ducts of C. striatus (Pi et al., 2006). This framework is uncommon and their three-dimensional (3D) structure, synthesis and pharmacology remain to be investigated.

Figure 18: Alignment of selected framework XVIII conotoxins

Framework XIX C-C-C-CCC-C-C-C-C

DiXIXA, with the uncommon Framework XIX (Figure 19), was purified from the venom of C. distans (Chen et al., 2008). It has five5 PTMs including one γ-carboxyglutamate and four hydroxyproline residues. Synthetic DiXIXA and native di19a co-eluted, although the exact disulfide bond connectivity has not been determined. DiXIXA caused a hyper excitable phenotype in mice greater than three weeks of age at lower doses, and lethargy at higher doses. Pu19.1 is a predicted sequence from the transcriptome work of the venom ducts of C. pulicarius (Lluisma et al., 2012).

Figure 19: Alignment of framework XIX conotoxins.

Framework XX C-CC-C-CC-C-C-C-C

This framework was initially assigned to be XII but was later moved to Framework XX (Loughnan et al., 2006) (Figure 20). Mature D superfamily peptides form hetero-, homo- and pseudohomodimers. Formatted: Highlight

Framework XX peptides from C. vexillum were recently revisited via a transcriptomics. αD-conotoxins

(VxXXA-VxXXC) were identified as the major transcripts and exclusively found in the defensive venom

(Prashanth et al., 2016). Another Framework XX peptide discovered within the last five years is αD-

GeXXA from C. generalis (Xu et al., 2015). An SAR study was carried out based on the crystal structure of αD-GeXXA and the newly identified binding site on nAChRs for this peptide provides a valuable basis for the rational design of new compounds that target the nAChR (Xu et al., 2015).

Figure 20: Alignment of selected framework XX conotoxins

Framework XXI CC-C-C-C-CC-C-C-C

P21a was isolated from the venom of C. purpurascens and has a 10-cysteine and 7-loop framework

(Moller and Mari, 2011). Despite having a 48% sequence homology with con-ikot-ikot from C. striatus, it does not form a dimer. P21a provides evidence that the Conus venom arsenal includes larger molecules that are directly injected into the prey. Therefore, cone snails can utilize toxins that are comparable in size to thosee ones commonly found in other venomous animals (Moller and Mari, 2011). Vc21.1 is a predicted sequence from the C. victoriae transcriptome (Robinson et al., 2014). Interestingly, G21.1 was discovered with high sequence identity but with one extra cysteine residue close to the C-terminus (Dutertre et al.,

2014) (Figure 21).

Figure 21: Alignment of selected framework XXI conotoxins.

Framework XXII C-C-C-C-C-C-C-C (9)

Framework XXII was defined from a cDNA analysis of C. californicus (Elliger et al., 2011) with six predicted peptide examples (Cal22a-f) (Figure 22). Interestingly, Framework XXII peptides identified from other species (i.e., Mr22.1 Vc22.1 and cl10.1) have dramatically different loop sizes and variable length. The wide range and the divergence between the sequences could be due to large phylogenetic distance between C. californicus and Indo-Pacific species (Elliger et al., 2011, Biggs et al., 2010).

Figure 22: Alignment of selected framework XXII conotoxins

Framework XXIII C-C-C-CC-C

NMR analysis of Im23a, from Framework XXIII (Figure 23), demonstrated this peptidewas determined by NMR analysis of Im23a, to have a I-II, III-IV, V-VI pattern of disulfide bridges and a novel Commented [DC1]: This correction is important as it is the peptide not the framework that has a disulfide connectity. helical hairpin fold. Intracranial injection of Im23a or Im23b into mice induced excitatory symptoms (Ye et al., 2012).

Figure 23: Alignment of selected framework XXIII conotoxins

Framework XXIV C-CC-C

FThe uncommon framework XXIV is uncommon, with only two ConoServer entries (Figure 24).,

α-VxXXIVA was cloned and characterized from C. vexillum (Luo et al., 2013). It preferentially inhibits the

α9α10 subtype nAChR. Another peptide Mi041 from C. miles contains only 11 amino acid residues with three prolines in this rather short sequence (Jin et al., 2013).

Figure 24: Alignment of framework XXIV conotoxins

Framework XXV C-C-C-C-CC

The sole entry in ConoServer for Framework XXV, as25a, was isolated from the venom of C. cancellatus (Aguilar et al., 2013) (Figure 25). Upon intracranial injection in mice, the purified as25a provokes paralysis of the hind limbs and death atwith a low dose of 240 pmol (Aguilar et al., 2013). No further pharmacology was characterized for this framework peptide.

Figure 25: Alignment of framework XXV conotoxin

Framework XXVI C-C-C-C-CC-CC

A single peptide from this framework, RsXXIVA, was isolated from the venom duct of C. regularis for this frameworkin 2013 (Bernaldez et al., 2013) (Figure 26). This conotoxin contains 40 amino acids and at the time exhibitedexhibits a novel arrangement of eight cysteine residues (C-C-C-C-CC-CC).

Surprisingly, two loops of thise novel peptide are highly identical similar to the amino acids sequence of

ω-MVIIA. The total length and disulfide pairing of both peptides are quite different, although the two most important residues for the described function of ω-MVIIA (Lys2 and Tyr13) are also present in

RsXXIVAthe peptide reported here. Electrophysiological analysis using superior cervical ganglion (SCG) neurons indicates that RsXXIVA inhibits Cav2.2 channel current in a dose-dependent manner with an EC50 of 2.8 μM, whose effect is partially reversed after washing. Furthermore, RsXXIVA was tested in hot-plate assays to measure the potential anti-nociceptive effect to an acute thermal stimulus, showing an analgesic effect in acute thermal pain and formalin pain models. However, the low affinity for CaV2.2 suggests that the primary target of the peptide could be different from that of ω-MVIIA.

Figure 26: Alignment of framework XXVI conotoxins

Framework XXVII C-C-C-CCC-C-C

φ-MiXXVIIA was discovered from C. miles with a cysteine framework comprising three consecutive cysteine residues and four disulfide bonds (Jin et al., 2017) (Figure 27). Regioselective chemical synthesis helped decipher the disulfide bond connectivity and the structure of φ-MiXXVIIA was determined by NMR spectroscopy. Only mild granulin-like activity has been discovered for this peptide.

Figure 27: Alignment of framework XXVII conotoxin

Framework XXVIII C-CC-C-C-C

The sole member of Framework XXVIII is κ-Mo3964 (Figure 28) from C. monile. (Kancherla et al., 2015). The expressed folded peptide has a β-sandwich structure that is stabilized by intersheet cross disulfide bonds. This toxin inhibited voltage-gated potassium channels in dorsal root ganglion (DRG) neurons.

Figure 28: Alignment of framework XXVIII conotoxin

In summary, the integration of omics is a powerful approach to study venoms in great detail.

Systematic data mining, high-throughput peptide production and bioassays, as well as rational structure- activity relationship (SAR) design need to become an integral part of the pipeline to effectively translate the venomics output into relevant (pre-)clinical outcomes.

4. Structural diversity

There are currently 218 three-dimensional structures of conotoxins reported in ConoServer (Kaas et al., 2012). Since the publication of our general classification of conotoxin folds in 2014 (Akondi et al.,

2014), 66 additional experimental structures have been determined, including 52 NMR solution structures and 14 structures solved by X-ray crystallography. These additional structures are indicated by a superscript asterisk in Table 4 so that readers may easily identify progress in the field since our earlier review. Amongst these new entries are 30 structures of natural conotoxins and collectively these have considerably expanded our knowledge of the structural diversity of this class of venom-derived peptides. In addition to these native sequences, SAR studies have resulted in the determination of 19 structures of mutated or engineered (e.g., backbone cyclized) conotoxins. Finally, 11 structures of conotoxins that had been published before 2014 were re-determined, in some instances using different conditions such as a different solvent or in complex with a molecular target. In this section, we focus on these new structures and readers are referred to the previous review for data on the earlier structures.

Conotoxins displaying cysteine Framework I (a four-cysteine framework) remain the most common grouping that has been structurally studied, with the publication of 39 structures of 25 wild-type toxins in the last 5 years. Next most common are Frameworks VI/VII and III (both comprising six cysteines), with

26 and 20 structures of 18 and 15 wild-type toxins, respectively. These three cysteine frameworks, i.e., I,

III and VI/VII, represent about half of the currently available conotoxin three-dimensional structures of wild-type toxins. The high frequency of structural studies of these frameworks parallels the current knowledge at the sequence level as ~40% of known conotoxins (either from transcriptomics and/or proteomics data) display one of these three frameworks.

As part of our update on structures, we refine the nomenclature associated with conotoxin folds that we proposed in 2014 (Akondi et al., 2014). Details of the updated fold and sub-fold for each new conotoxin three-dimensional structure are given in Table 4, with changes from the initial publication of the

34 classification indicated by a superscript asterisk following new peptide names, database identifiers or fold or sub-fold names.

Table 4: Classification of all known three-dimensional structures of wild-type and synthetic conotoxin into folds and sub-folds.

Cys Gene Cono Size # Loop Method BMRB PDB Namea Speciesb frame super Server (aa) Cysd sizee f IDg IDg workc family IDg Fold A: four cysteines, globular [connectivity 1-3, 2-4] Sub-fold A1 (one turn of helix in first loop) α-ImI C. imperialis I 12 4 4/3 A NMR 1G2G, 5, 24, 1IMI, 25, 27 1CNL, 1IM1 X-ray 2BYPh, 34, 35 2C9Th α-ImI [D5N] I 12 4 4/3 NMR 4847 1E76 10 α-ImI [R7L] I 12 4 4/3 NMR 4846 1E75 9 α-ImI I 12 4 4/3 NMR 20107 131 [A9L,W10Y,R11ABA] α-ImI [R11E] I 12 4 4/3 NMR 4845 1E74 8 α-ImI [C2Agl,C8Agl] I 12 2 4/3 NMR 20033 128 α-ImI [C2U,C8U] I 12 4 4/3 NMR 6897 2BC7 97 α-ImI I 12 4 4/3 NMR 6896 2BC8 98 [C2U,C3U,C8U,C12U] α-RgIA C. regius I 12 4 4/3 A NMR 20002, 2JUT, 118, 15435 2JUS 123, 191 α-RgIA [D5E] I 12 4 4/3 NMR 15367 2JUR 119 α-RgIA [P6V] I 12 4 4/3 NMR 15436 2JUQ 121 α-RgIA [Y10W] I 12 4 4/3 NMR 15368 2JUS 120 [2,8]-cis dicarba RgIA* I 12 4 4/3 NMR 25174* 2MTO* 178* [3,12]-cis dicarba I 12 4 4/3 NMR 25186* 2MTT* 179* RgIA* [3,12]-trans dicarba I 12 4 4/3 NMR 25187* 2MTU* 180* RgIA* α-BuIA C. bullatus I 13 4 4/4 A NMR 15031 2I28 7 35

X-ray 4EZ1h* 161h* α-AuIB C. aulicus I 15 4 4/6 A NMR 1MXN, 31, 13 1DG2 cyclic-AuIB-4 (GGAA) I 19 4 4/6 NMR 142 cyclic-AuIB-5 I 20 4 4/6 NMR 143 (AGAGA) cyclic-AuIB-6 I 21 4 4/6 NMR 144 (GGAAGG) Bt1.8* C. betulinus I 16 4 4/7 NMR 25954* 2NAY* 200* α-TxID* C. textile I 15 4 4/6 NMR 18964* 2M3I* 181* α-EI C. ermineus I 18 4 4/7 A NMR 1K64 18 α-Epi [sTy15>Y] I 16 4 4/7 X-ray 1A0M 20 α-GIC C. I 16 4 4/7 A NMR 5985 1UL2 26 geographus X-ray 5CO5h* 202h* α-GID C. I 18 4 4/7 A NMR 5585 1MTQ 15 geographus α-GID [A10V] * I 19 4 4/7 NMR 30225* 5UG3* 205* α-GID [V13Y] * I 19 4 4/7 NMR 30226* 5UG5* 206* α-Lo1a* C. I 18 4 4/7 NMR 19476* 2MD6* 173* longurionis α-LsIA* C. limpusi I 17 4 4/7 X-ray 5T90h* 204h* α-LvIA* C. lividus I 17 4 4/7 NMR 19501* 2MDQ* 172* X-ray 5XGLh* 209h* α-MII C. magus I 16 4 4/7 A NMR 1M2C, 21, 29 1MII α-MII [E11A] I 16 4 4/7 NMR 145 cyclic-MII-6 I 22 4 4/7 NMR 6818 2AJW 32 cyclic-MII-7 I 23 4 4/7 NMR 6817 2AK0 33 α-OmIA C. omaria I 17 4 4/7 A NMR 6237 2GCZ 5 α-PeIA C. pergrandis I 16 4 4/7 A NMR 139 X-ray 5JMEh* 203h* α-PIA C. I 18 4 4/7 A NMR 6720 1ZLC 36 purpurascens α-Pni1 I 16 4 4/7 X-ray 1PEN 12 α-PnIA I 16 4 4/7 X-ray 2BR8h 30 [A10L,D14K,sTy15Y] α-PnIB C. pennaceus I 16 4 4/7 A X-ray 1AKG 16 α-RegIIA* C. regius I 16 4 4/7 A NMR 157*

ρ-TIA C. tulipa I 19 4 4/7 A NMR 1IEN, 65, 194* 2LR9* α-TxIA C. textile I 16 4 4/7 A X-ray 2UZ6h 110 α-TxIB* C. textile I 16 4 4/7 NMR 18736 2LZ5 156 α-Vc1.1 C. victoriae I 16 4 4/7 NMR 7177 2H8S 4 [2,8]-cis dicarba Vc1.1* I 16 4 4/7 NMR 19577* 2MFX* 169* [2,8]-trans dicarba I 16 4 4/7 NMR 19578* 2MFY* 170* Vc1.1* [3,16]-trans dicarba I 16 4 4/7 NMR 19587* 2MG6* 171* Vc1.1* cyclic-Vc1.1 I 22 4 4/7 NMR 149 Cyclic-Vc1.1 I 16 4 4/7 NMR 30412* 6CGX* 222* [D11E,E14A] * Cyclic-Vc1.1 16 2 NMR 25516* 2N07* 197* [C2H,C8F] * α-Vc1.2 C. victoriae I 16 4 4/7 A NMR 20126 141 Sub-fold A2 (no turn of helix in first loop) α-CnIA C. consors I 14 4 3/5 NMR 1B45 53 α-GI C. I 13 4 3/5 A NMR 1XGA 22 geographus X-ray 1NOT 11 α-GI [N4Benzoyl- I 13 4 3/5 NMR 2FRB 3 phenylalanine] α-GI [S12Benzoyl- I 13 4 3/5 NMR 2FR9 2 phenylalanine] α-SI C. striatus I 13 4 3/5 A NMR 4503 1QMW 1 X-ray 1HJE 17 α-LtXIVA C. litteratus XIV 13 4 3/3/2 L NMR 21014 148 Sub-fold A3 (no turn of helix in first loop, second loop similar to sub-fold A1) χ-CMrVIA [K6P] X 11 4 4/2 NMR 2IH6 111 χ-CMrVIA [K6P] X 11 4 4/2 NMR 2IH7 112 amidated Fold B: six cysteines, three disulfide bonds not in a knotted arrangement [connectivity 1-4, 2-5, 3-6] Sub-fold B1 (one turn of helix in second loop, two turns of helix overall) μ-BuIIIB* C. bullatus III 24 6 6/3/5 M NMR 18203* 2LO9* 152* μ-BuIIIB [G2a] * III 24 6 6/3/5 NMR 18206* 2LOC* 153* μ-CnIIIC C. consors III 22 6 5/4/5 NMR 2YEN 150 μ-GIIIA C. III 22 6 5/4/4 M NMR 1664, 1TCG, 82, 84, geographus 1665 1TCJ 134, 135

μ-GIIIA [R13A] III 22 6 5/4/4 NMR 1TCH , 83, 85 1TCK μ-GIIIB C. III 22 6 5/4/4 M NMR 1GIB 64 geographus μ-KIIIAi C. kinoshitai III 16 6 5/4/4 M NMR 20048 2LXG* 129, 186* μ-PIIIA C. III 22 6 5/4/4 M NMR 6027 1R9I 79 purpurascens μ-RIIIK [T24A] III 24 6 6/4/4 NMR 146 μ-SIIIA C. striatus III 20 6 1/4/5 M NMR 20025 125 μ-SmIIIA C. III 22 6 5/4/5 M NMR 5881, 1Q2J 77 stercusmusca rum μ-TIIIA C. tulipa III 22 6 5/4/4 M NMR 20024 126 Sub-fold B2 (no turn of helix in second loop, one turn of helix overall) α-PIIIE C. III 24 6 4/5/4 M NMR 5113 1AS5, 51, 68 purpurascens 1JLO α-PIIIF C. III 24 6 4/5/4 M NMR 5112 1JLP 69 purpurascens Fold C: six cysteines, three disulfide bonds forming a cystine knot [connectivity 1-4, 2-5, 3-6] Sub-fold C1 (six residues in first loop) δ-Am2766 C. amadis VI/VII 26 6 6/6/3/ O1 NMR 1YZ2 94 3 ω-CVID C. catus VI/VII 27 6 6/6/3/ O1 NMR 138 6 δ-EVIA C. ermineus VI/VII 32 6 6/9/3/ O1 NMR 1G1P, 62, 63 3 1G1Z ω-FVIA C. fulmen VI/VII 25 6 6/6/3/ NMR 2KM9 137 4 μ-conotoxin-GS C. VI/VII 27 6 6/3/4/ O1 NMR 1AG7 50 geographus 7 ω-GVIA C. VI/VII 27 6 6/6/2/ O1 NMR 2CCO ,1 72, 89, geographus 6 TTL,1O 100 MC ω-GVIA [O10>K] VI/VII 27 6 6/6/2/ NMR 1TR6 86 6 μ-GVIIJ [C24S] * C. VI/VII 35 6 6/6/3/ O1 NMR 26674* 2N8H* 182* geographus 6 μ-MrVIB C. VI/VII 31 6 6/9/4/ O1 NMR 6135 1RMK 80 marmoreus 4 ω-MVIIA C. magus VI/VII 25 6 6/6/3/ O1 NMR 1DW4, 55, 56, 4 1DW5, 70, 73, 1MVI, 88 38

1OMG, 1TTK ω-MVIIA with C- VI/VII 26 6 6/6/3/ NMR 1FEO 59 terminal Gly 4 ω-MVIIA [R10>K] VI/VII 25 6 6/6/3/ NMR 1TT3 87 4 ω-MVIIC C. magus VI/VII 26 6 6/6/3/ NMR 4500 1CNN, 54, 74 5 1OMN ω-MVIIC VI/VII 26 6 6/6/3/ NMR 1V4Q 90 [S17K,S19R,K25R] 5 μ-MfVIA* C. magnificus VI/VII 32 6 6/9/4/ NMR 25804* 2N7F* 198* 4 ω-MoVIB* C. moncuri VI/VII 30 6 6/6/2/ NMR 30405* 6CEG* 223* 6 κ-PVIIA C. VI/VII 27 6 6/6/3/ O1 NMR 1AV3, 46 purpurascens 5 1KCP cyclic-PVIIA* VI/VII 27 6 6/6/3/ NMR 25847* 2N8E* 199* 5 ω-SO3 C. striatus VI/VII 25 6 6/6/3/ O1 NMR 1FYG 61 4 ω-SVIB C. striatus VI/VII 26 6 6/6/3/ O1 NMR 1MVJ 71 5 t7a C. tulipa VI/VII 30 6 6/3/4/ O1 NMR 1EYO 57 4 ω-TxVII C. textile VI/VII 26 6 6/6/3/ O1 NMR 1F3K 58 3 δ-TxVIA C. textile VI/VII 27 6 6/6/3/ O1 NMR 1FU3 60 4 G117 (G11.1) * C. XI 32 8 6/5/3/ I3 NMR 30406* 6CEI* 208* geographus 3/6 ι-RXIA C. radiatus XI 46 8 6/5/2/ I1 NMR 15175 2P4L , 104, 4 2JTU 130 ι-RXIA [BTr33>W] XI 46 8 6/5/2/ NMR 15174 2JRY 105 4 Sub-fold C2 (three residues in first loop) Gm9a C. IX 27 6 3/5/3/ P NMR 1IXT 67 gloriamaris 1/4 cyclic-Gm9a* IX 30 6 3/5/3/ NMR 25128* 2MSO* 195* 1/4 cyclic-Bru9a* IX 27 6 3/3/3/ NMR 25129* 2MSQ* 196* 1/4 Fold D: four cysteines, disulfide bonds with ribbon connectivity [connectivity 1-4, 2-3] Sub-fold D1 (disulfide 2-3 in a staple conformation) 39

Ar1248* C. araneosus X 12 4 4/2 NMR 19103* 2M62* 176* χ-MrIA C. X 13 4 4/2 T NMR 6891 2EW4 102 marmoreus cyclic-MrIA X 15 4 4/2 NMR 2J15 49 χ-MrIB amidated C. X 13 4 4/2 NMR 1IEO 66 marmoreus *α-GI ribbon isoform I 13 4 3/5 NMR 1XGB 23 Sub-fold D2 (disulfide 2-3 in a hook conformation) *α-AuIB ribbon I 15 4 4/6 NMR 1MXP 14 isoform *α-BuIA ribbon I 4 4/4 NMR 2NS3 114 isoform *α-ImI deamidated I 12 4 4/3 NMR 2IGU 106 ribbon isoform *α-ImI [P6A] ribbon I 12 4 4/3 NMR 2IFI 108 isoform *α-ImI [P6K] ribbon I 12 4 4/3 NMR 2IFZ 107 isoform *α-ImI [P6K] ribbon I 12 4 4/3 NMR 2IFJ 109 deamidated isoform *χ-CMrVIA ribbon 11 4 4/2 2B5P 95 isoform *χ-CMrVIA amidated X 11 4 4/2 NMR 2IHA 113 ribbon isoform Qc16a* C. quercinus XVI 11 4 2/4 Q NMR 20128* 227* τ-CnVA* C. consors V 14 4 5 NMR 18972* 3ZKT* 151* Fold E: four cysteines, mirror of fold A [connectivity 1-3, 2-4] χ-CMrVIA C. X 11 4 4/2 NMR 2B5Q 96 marmoreus Fold F: four cysteines, disulfide bonds collinear [connectivity 1-3, 2-4] α-Pu14a C. pulicarius XIV 23 4 10/1/ A NMR 21015 147 3 Fold G: four cysteines, parallel disulfide bonds [connectivity 1-3, 2-4] κ-PlXIVA C. litteratus XIV 25 4 3/10/ J NMR 6951 2FQC 103 1 Fold H: six cysteines [connectivity 1-5, 2-4, 3-6] Ar1446* C. araneosus III 14 6 3/3/2 NMR 19102* 2M61* 175* BtIIIA C. betulinus III 14 6 3/2/2 M NMR 154 Mr3e C. III 16 6 4/3/1 M NMR 15195 2EFZ 101 marmoreus Fold I: six cysteines

α-PIVA [O7P,O13P] IV 25 6 7/2/1/ NMR 1P1P 75 6 α-EIVA C. ermineus IV 30 6 7/2/1/ NMR 5869 1PQR 76 7 Cctx* C. consors IV 30 6 7/2/1/ A NMR 18897* 4B1Q* 155* 3 Fold J: two cysteines, cystine stabilized turn Sub-fold J1* (contryphan in water with P3 cis, P6 trans) contryphan-Lo* C. loroisii 8 2 5 NMR 19132* 2M6G* 166* contryphan-R C. radiatus 8 2 5 NMR 1QFB 47 contryphan-R [Δ1] 7 2 5 NMR 1DG0 45 contryphan-Sm C. 8 2 5 NMR 1DFY, 38, 39 stercusmusca 1DFZ rum contryphan-Vc2* C. victoriae 7 2 5 O2 NMR 30152* 5L34* 218* contryphan-Vn C. 9 2 5 NMR 1NXN 43 ventricosus cyclic-contryphan 8 2 5 NMR 1D7T 37 conopressin-T C. tulipa 9 2 4 NMR 20007 124 Sub-fold J2* (contryphan in methanol with P3 cis, P6 trans; contryphan-In with P6 trans) contryphan-In P6 trans* C. inscriptus 8 2 5 NMR 19129*, 2M6D*, 163*, (water, methanol) 19131* 2M6F* 165* contryphan-Lo* C. loroisii 8 2 5 NMR 19133* 2M6H* 167* (methanol) Sub-fold J3* (contryphan-In with P6 cis) contryphan-In P6 cis* C. inscriptus 8 2 5 NMR 19128*, 2M6C*, 162*, (water, methanol) 19130* 2M6E* 164* Fold K: no cysteine, fully helical conantokin-G C. 17 0 B1 NMR 1AD7, 40, 41, geographus 1AWY, 44, 216* 1ONU, 2MZM* conantokin-G[+10O] * 18 NMR 25491* 2MZL* 215* conantokin-RlB* C. rolani 18 0 B1 NMR 25490* 2MZK* 214* Conantokin-RlB 18 0 NMR 30176* 5TBG* 210* [O10P] * Conantokin-RlB 17 0 NMR 30178* 5TBQ* 211* [delO10] * Conantokin-RlB 18 0 NMR 30179* 5TBR* 212* [O10A] * Conantokin-RlB 17 0 NMR 25465* 2MYZ* 213* [K8N,A9Q,del10O] * 41

conantokin-T C. tulipa 21 0 NMR 1ONT 42 Fold L: no cysteine, 3/10 helix and coil conomarphin C. 15 0 M NMR 7397 2YYF 115 marmoreus conomarphin [d13>D] 15 0 NMR 2JQC 116 Fold M*: helix-loop-helix fold VilXIVA* C. villepinii XIV 27 4 3/11/ R NMR 30508* 6EFE* 221* 3 Im23a* C. imperialis XXIII 42 6 6/3/9/ K NMR 18141* 2LMZ* 193* 14 Fold N*: six cysteines [connectivity 1-6, 2-4, 3-5] reg3b* C. regius III 15 6 3/4/2 NMR 30385* 6BX9* 207* Fold O*: six cysteines, disulfide stabilised β-sandwich [connectivity 1-3,2-5,4-6] δ/κ-Mo3964* C. monile XXVII 37 6 6/8/5/ NMR 25302* 2MW7* 184* 6 Fold P*: two cysteines, single disulfide-directed β-hairpin contryphan-Vc1* C. victoriae 31 2 12 O2 NMR 25585* 2N24* 217* contryphan-Vc1[1-22] * 22 2 12 NMR 30124* 5KKM* 219* Fold Q*: two β-sheet sandwich / disulfide stabilised dimeric α-GeXXA* C. generalis XX 50 10 11/4/ D X-ray 4X9Z* 201* 3/4/1/ 9/1 Fold R*: four helix bundle con-ikot-ikot* C. striatus 86 13 4/6/1 con- X-ray 4U5G*, 187*, 4/7/8/ ikot- 4U5H*, 188*, 5/16/ ikot 4U5Bj*, 189j*, 4/3 4U5Cj*, 190j*, 4U5Dj*, 226j*, 4U5Ej*, 224j*, 4U5Fj* 225j* Insulin fold*

Con-Ins G1* C. 43 6 X-ray 5JYQ* 220* geographus Kunitz fold: large protein with two disulfide bonds [connectivity 1-4,2-3]

Conkunitzin-S1 C. striatus XIV 60 4 24/20 X-ray 1Y62 48 /3 Conkunitzin-S2 C. striatus XIV 65 4 24/20 NMR 2J6D 117 /3

* Peptide name, database (BMRB, PDB or ConoServer) entry, or fold or sub-fold name that was not described in our review in

2014. a A brief description of the folds and sub-folds is provided in Figures 29 and 30. The names of non-natural synthetic variants are indented in the case where the fold is the same as the wild-type conopeptide, whereas the name of the variant is preceded by an asterisk in the case where it adopts a different fold from the wild-type. b Only native conotoxin are provided with a Conus (C.) species. c Cysteine frameworks are defined in Table 1. d The number of cysteine residues (# cysteines) is counted in the sequence of the mature peptide region in the precursor, before modification to cystines. e The “loop size” designates the length of the inter-cysteine segments defined in the cysteine frameworks. f “Method” refers to the experimental method used to determine the 3D structures. If two different experimental methods were used for the same conopeptide, identifiers are provided on two separate lines. g The database identifiers in the Biological Magnetic Resonance dataBank (BMRB), Protein Data Bank (PDB) and ConoServer database are provided. Distinct structural studies are catalogued as different entries in ConoServer, and therefore each entry in

ConoServer can be associated with a BMRB and/or a PDB entry. Some conopeptide 3D structures are only found in ConoServer as they were not deposited by their authors in the PDB or BMRB. h The X-ray structure is in complex with an acetylcholine binding protein (AChBP). i The connectivity of -KIIIA is different from that of other Framework III conotoxins with the same fold but the backbone conformation is similar. j Complex with AMPA receptor GluA2 (GluR2)

Folds are designated with a single upper-case letter, and sub-folds are indicated by appending a number to the letter name of the fold (e.g. “A1” is a sub-fold of “Fold A”). Since the initial publication of the conotoxin fold nomenclature, new structures have helped refine the definition of the folds (Figure 29) and seven entirely new folds have been discovered (Figure 30), one of which is the insulin fold and the five others we named M to R. Pictorial descriptions of the folds for which the definition has not changed are in

Figure 29 of Akondi et al. (Akondi et al., 2014).

4.1 Structures of frameworks with four cysteines 43

The 11 new structures of native conotoxins from Framework I fit all into sub-fold A1, similarly to all other native Framework I conotoxins with four residues in their first loop. An additional, five crystal structures of complexes between Framework I conotoxins and the acetylcholine binding protein (AChBP) have been determined (4/4 BuIA, 4/7 GIC, 4/7 LsIA, 4/7 LvIA, 4/7 PeIA), adding to the four conotoxin/AChBP structures previously determined (4/7 PnIA variant, 4/7 TxIA, and 4/3 ImI twice).

Interestingly, the Framework I conotoxins ImI, BuIA, GIC, LvIA, and PeIA have been determined both in the apo state by NMR spectroscopy (Chi et al., 2004, Chi et al., 2006, Daly et al., 2011, Lamthanh et al.,

1999, Luo et al., 2014) and in complex with AChBP by crystallography (Hansen et al., 2005, Hone et al.,

2018, Lin et al., 2016, Xu et al., 2017).

Figure 29: Structural variation and inclusion of new cysteine frameworks from the folds initially defined in Akondi et al., 2014. (A) Comparison of Framework XI conotoxins adopting sub-fold C1 with the Framework

VI/VII conotoxin MVIIC. (B) Comparison of sub-fold C2 gm9a and cyclic bru9a. (C) Comparison of Framework V conotoxin

CnVA with Framework I conotoxin ribbon AuIB. (D) Comparison of the connectivity of two Framework III conotoxins adopting

Fold H with the Framework III mr3e. Comparison of Framework IV PIVA variant and EIVA with the o-glycosylated Framework

IV CcTx (its N-terminal glycosylated tail is in pink and glycosylation in orange). Structurally conserved regions are shown in red or orange, other regions that are conformationally distinct are in green or white. The disulfide bonds are of utmost importance for defining the folds and are clearly represented as yellow sticks. The α-carbon of hemi-cystines are shown as spheres and numbered along the sequence.

The crystal structure of BuIA/AChBP, reported in the PDB [ID: 4EZ1] has not been published in a peer-reviewed publication but is included in this analysis given the availability of the coordinates. A comparison of the structure in complex and in solution revealed a globally similar conformation, but Loop

2 and the disulfide bonds of BuIA, GIC and PeIA display different conformations in the bound and free states. The C-terminus of these three toxins shifted by 4.4, 2.4 and 1.6 Å, respectively, between the two states and this change seems to be linked to the high-energy conformations of disulfide bonds in the NMR- derived structures (Figure 31), suggesting possible problems in the disulfide bond parameters of the force fields used to derive these structures.

The conformation of disulfide bonds cannot always be determined unambiguously by using NMR spectroscopy hence creating problems for their classification. We suggest using sub-fold designations D1 and D2 to identify single disulfide bond variants. A recently-published machine learning method that can more accurately predict disulfide conformations using NMR chemical shifts (Armstrong et al., 2018) will help improve the description of disulfide bonds in NMR solution structures of conotoxins..

Figure 30: Structures of new conotoxin folds. The conotoxin folds are noted M to N, with the exception of the conoinsulin fold. Disulfide bonds are shown as yellow sticks. The α-carbons of hemi-cystines are shown as spheres and numbered along the sequence. The disulfide bonds are shown in yellow. The α-helices that define Fold M are in red. For Fold P, the structural segments in orange and blue are shared between Contryphan-Vc1 and Fold C (here conotoxin GS). Folds Q, R and

Insulin are dimeric and each monomer is either in grey or red.

Figure 31: Analysis of the conformation and energy of disulfides of the structure of Framework I.

Conotoxins BuIA, ImI, GIC, LvIA and PeIA, which have been studied by both NMR spectroscopy in solution and by

48 crystallography in complex with AChBP. The distribution of disulfide bond energies (A), χ1 and χ1’ dihedral angles (C), χ2 and

χ2’ dihedral angles (D) and χ3 dihedral angles (E) have been computed from a set of 7,183 disulfide bonds from a non-redundant representation set of high-resolution crystal structures created with Culled PDB. The energy was computed according to the formula of Katz et al. 1986 (Katz and Kossiakoff, 1986). Under each histogram, the measurements made in NMR solution structures and in X-ray crystallography are in red and blue, respectively. Panel B shows the superimposition of the NMR-solution structure and X-ray structure of PeIA in red and blue, respectively.

Conotoxins α-RgIA and α-Vc1.1 have attracted substantial attention for their potential as analgesic molecules and several variants aiming at modulating their activity, selectivity or bioavailability have been structurally characterized. The backbone conformation of α-4/7-Vc1.1 is unchanged when its first or second disulfide bond is replaced by a cis or trans dicarba-bridge, respectively (van Lierop et al., 2013). By contrast, a large conformational change of the second loop is seen upon replacing the first cystine with a trans dicarba-bridge. Backbone cyclization of α-Vc1.1 also does not affect its fold and neither nor does the replacement of one of its disulfide bonds by two hydrophobic residues. The structure of the second loop of

α-4/3-RgIA is affected by the replacement of its disulfide bond by a dicarba bridge in either cis or trans

(Chhabra et al., 2014), suggesting that α-4/7-conotoxins are more amenable to the introduction of dicarba bridges.

Framework X conotoxins as well as Framework I conotoxins with a non-natural ribbon disulfide connectivity form small β-hairpin structures classified as Fold D. Framework V conotoxin τ-CnVA, which targets the somatostatin receptor, displays a similar hairpin conformation (Petrel et al., 2013), with highest similarity with sub-fold D2 toxins (Figure 29C). Interestingly, the structure of another Framework V conotoxin, TxVA, iswas very different as it foldsed into a compact structure (Rigby et al., 1999), suggesting that Framework V conotoxins have heterogeneous folds.

Framework XIV also contains four cysteine residues, although its pattern definition is somewhat loose, as it is defined as having at least one residue separating each of the cysteine residues. Because of this loose definition, conotoxins with a Framework XIV adopt very different structures (Akondi et al., 2014).

The structures of Framework XIV conotoxins pu14a, pl14a and LtXIVA were classified into Folds F, G and

A2 (Akondi et al., 2014). In contrast to these three toxins, which display a globular disulfide connectivity

(i.e. 1-3, 2-4), the structure of Framework XIV conotoxin Vil14a displays a ribbon connectivity (1-4, 2-3)

(Moller et al., 2018). The structure of Vil14a was described as a α-helix-loop-α-helix fold (Moller et al.,

2018), which is similar to that of scorpion cystine-stabilized α-helix-loop-helix (Cs α/α) toxins, and we classified it as Fold M (Figure 30). A similar fold is adopted by the Framework XXIII conotoxin Im23a (Ye et al., 2012), which comprises three disulfide bonds with a bead connectivity. These two toxins belong to different gene superfamilies and there is no overlap of disulfide bonds, suggesting convergent evolution of the toxins toward a similar structure.

4.2 Structures of frameworks with six cysteines

Fold C represents the cystine knot fold, which is a common structural motif, and comprises a knotted arrangement of three disulfide bonds (Postic et al., 2018). Framework VI/VII is frequently displayed by conotoxins (Kaas et al., 2010) and corresponds to the inhibitory cystine knot pattern C-C-CC-C-C, which is a sub-category of the cystine knot fold. Framework VI/VII conotoxins are all classified as sub-fold C1

(Akondi et al., 2014), together with the structures of two Framework XI conotoxins, RXIA (Buczek et al.,

2007) and G117 (PDB 6CEI; no peer-reviewed publication yet). Framework XI has eight cysteines, and the structures of both RXIA and G117 display in addition to the cystine knot a disulfide bond linking the Loop

3 of the knot motif to the C-terminus (Figure 29A). The sub-fold C2 is displayed by Framework IX conotoxins, gm9a (Miles et al., 2002) and bru9a (which was engineered to be cyclic in the experimental structure) (Akcan et al., 2015). The main structural difference between sub-folds C1 and C2 is a different conformation of the N-terminus as well as a different spatial location of the first disulfide bond.

Conotoxins adopting Fold B have the same connectivity as cystine knot conotoxins (Fold C), i.e. 1-

4, 2-5, 3-6, but their three disulfide bonds are not knotted (Akondi et al., 2014). In the current status of our knowledge, conotoxins adopting Fold B also have a Framework III. Sub-fold B1 displays a turn of α-helix in the second loop, and these conotoxins target voltage-gated sodium channel. The second loop of the two

50 conotoxins identified to adopt a sub-fold B2, i.e., μ-PIIIE and μ-PIIIF (Van Wagoner et al., 2003, Van

Wagoner and Ireland, 2003), do not have a helical content, and these two conotoxins block nAChRs.

Interestingly, the conformation of Fold B conotoxins is similar to that of Fold A, which mainly comprises toxins acting as antagonists of the nAChR (Akondi et al., 2014). Framework III conotoxins can also display different disulfide connectivity and folds from that of Fold B. Framework III Ar1446 (PDB 2M61; no peer- reviewed publication) has a disulfide connectivity 1-6, 2-4, 3-5 and adopts Fold H (Figure 29D). Fold H was already represented by two Framework III toxins: μ-BtIIIA (Akcan et al., 2013) and mr3e (Du et al.,

2007). μ-BtIIIA has the same connectivity as ar1446 but mr3e has a distinct connectivity, 1-5, 2-4, 3-6.

Framework III conotoxin reg3b has the same connectivity as ar1446 and μ-BtIIIA but it folds differently

(Franco et al., 2018). Its structure was described as a constrained multi-turn scaffold, and we have classified it as Fold N (Figure 30). Conotoxins displaying the same framework and connectivity can therefore fold differently.

The structure of Framework IV conotoxin CcTx is the first of an O-glycosylated conotoxin

(Hocking et al., 2013). Its structure is similar to that of the two Framework IV conotoxins that adopt Fold

I. The definition of this fold is loose as only the segments from Cys3 to Cys5 and the relative position of

Cys6 seem to be really conserved (Figure 29). In contrast to the two other Fold I toxins, the connectivities of the first two disulfide bonds of CcTx are exchanged, i.e. CcTx has the connectivity 1-3, 2-5, 4-6 and the other toxins have the connectivity 1-5, 2-3, 4-6. The same connectivity was discovered for the four cysteine

Framework XXVII Mo3964 (Kancherla et al., 2015), which displays the new Fold O (Figure 30). Fold I does not display any regular secondary structure whereas Fold O comprises two β-sheets, one made of three

β-strands and the other of two β-strands.

Overall, the fold landscape of conotoxins with six cysteines in their primary sequence is complex, with different disulfide connectivities present for the same framework or similar fold adopted by toxins with different disulfide connectivities or frameworks (Table 5).

Table 5: Connectivities of cysteine frameworks with six cysteines and their fold.

Disulfide connectivity Cysteine framework Fold

1-2, 3-4, 5-6 XXIII M

1-3, 2-5, 4-6 IV, XXVII I, P

1-4, 2-5, 3-6 III, VI/VII C, B

1-6, 2-4, 3-5 III H, N

1-5, 2-3, 4-6 IV I

1-5, 2-4, 3-6 III H

4.3 Large peptides forming dimers

Framework XX has ten cysteines and the structure of one of its members, αD-GeXXA, was determined by X-ray crystallography (Xu et al., 2015). αD-GeXXA forms a heterodimer stabilized by two inter-chain disulfide bonds formed between the two cysteine residues at the N-terminus (Figure 30). Each chain has 50 amino acids and displays a 30-amino acid domain at the C-terminus, which is maintained by three disulfide bonds with the connectivity 1-4, 2-5, 3-6. The N-terminus of each chain contributes by a 20- amino acid segment to half of a central domain, which is stabilized by two inter-chain disulfide bonds, representing a novel fold, which we named Q.

Conoinsulin Con-Ins G1 comprises two different peptide chains, which are covalently linked by two inter-chain disulfide bonds (Menting et al., 2016) (Figure 30). The structure of Cons-Ins-G1 is very similar to that of the mammalian insulins apart from a missing 10-amino acid segment at the N-terminus of the second chain, which forms a β-strand structure for mammalian insulins. Remarkably, Con-Ins G1 activates the human insulin receptor with only 10-fold lower activity than human insulin.

C. striatus Con-ikot-ikot comprises 86 amino acids, with a greater number of cysteine residues (13) than usual in its primary sequence (Walker et al., 2009). The determination of its structure by X-ray

52 crystallography structure revealed that this protein forms a homodimer stabilized by three inter-chain disulfide bonds and five intra-chain disulfide bonds (Chen et al., 2014a). Con-ikot-ikot was also crystallized in complex with its molecular target, the AMPA (α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid) receptor GluA2 (GluR2) receptor (Chen et al., 2014a). AMPA receptors are cation channels activated by glutamate and are the main neuronal receptor in the brain. The crystal structure of the AMPA receptor/con- ikot-ikot complex represents the first experimental structure of a conotoxin in complex with a physiologically relevant molecular target.

4.4 Disulfide-poor conotoxins

Conotoxins with no more than one disulfide bond are classified as disulfide-poor conotoxins (Lebbe and Tytgat, 2016). Amongst them, the conantokins are 20-amino acid helical conotoxins containing γ- carboxyglutamate residues and adopt Fold K (Akondi et al., 2014). They allosterically inhibit NMDA receptors, which are the second major class of glutamate receptors besides the AMPA receptors (Donevan and McCabe, 2000). Some conantokins, such as conantokin G, only adopt a helical conformation in the presence of divalent cations, such as magnesium or calcium (Chen et al., 1998). The chelation of magnesium ions by γ-carboxyglutamate side chains controls the conformation of the conantokins, and their activity at the GluN2B NMDA receptor. The structure of C. rolani conantokin-RlB was determined by

NMR and revealed a kink in the middle of the helix, which was induced by a hydroxyproline (Yuan et al.,

2016). The replacement of this hydroxyproline by a proline or an alanine dramatically reduced its bioactivity.

Contryphans are small conotoxins adopting a five-residue turn structure containing one D-amino acid and stabilized by a single disulfide bond, which we have classified as Fold J (Akondi et al., 2014 ).

The molecular target of contryphans remains unknown, but they induce a depressive or hyperactive phenotype when injected in mice (Drane et al., 2017). The structures of contryphan-In from C. inscriptus and contryphan-Lo from C. loroisii were studied in water and in methanol, leading to the identification of

53 two alternative conformations depending on the cis/trans isomerization of a proline (Sonti et al., 2013). We reclassified the conformation in a water environment observed for six contryphan as sub-fold J1 and the two alternative conformations as sub-folds J2 and J3.

C. victoriae contryphan-Vc1 shares no homology of sequence and structures with other contryphans, but it was named based on the similarity of its signal peptide sequence of its precursor with other members of the contryphan family (Robinson et al., 2014). The structure of contryphan-Vc1 (Robinson et al., 2016) established a new fold that we called Fold P (Figure 30). This fold is characterized by a small β-sheet and a single disulfide bond that links one of the C-terminal β-strands to the N-terminus, which folds back on the β-sheet. This fold was described as single disulfide-directed β-hairpin, and is a stable sub-fold of the

ICK (Robinson et al., 2016). A comparison between Fold P and the structure of Framework VI/VII conotoxin GS, which displays an ICK, is shown in Figure 30. Fold P overlays well with the segment between the third and sixth cysteine residue of the ICK (sub-fold C1), and with the disulfide of Fold P corresponding to the third disulfide of the ICK. It is interesting to speculate that the core of other conotoxin folds could be decomposed into smaller, stable motifs, similarly to protein domains, which have been proposed to have emerged from combination of ancestral sub-domain fragments (Alva et al., 2015).

5. Conotoxin Synthesis

Fmoc and Boc solid phase peptide synthesis (SPPS) remain the methods of choice for the production of conotoxins from the various superfamilies (Merrifield, 1963, Fields and Noble, 1990, Amblard et al.,

2006, Jensen et al., 2013). Fmoc-SPPS is generally preferred due to its simple laboratory setup. Boc-SPPS, which requires a dedicated HF apparatus (Muttenthaler et al., 2015) is a valuable alternative for difficult sequences that cannot be assembled by Fmoc chemistry or for synthetic strategies that are incompatible with the basic Fmoc deprotection chemistry (Schnölzer et al., 1992). Native chemical ligation (Dawson et al., 1994) can also be used to overcome difficult sequences as well as for combinatorial SAR studies

(Hopping et al., 2009) and intramolecular N-to-C-terminal backbone cyclization (Clark and Craik, 2010).

SPPS has the advantage of being rapid and highly-automated, and able to provide mg to kg quantities of peptides for preclinical and clinical studies. It allows the incorporation of unnatural amino acids, post- translational modifications, imaging and other chemical reporter tags, as well as structural modifications such as backbone cyclization, mimetic design and addition of fatty acids or PEG units to improve half-life and bioavailability. Recombinant production of conotoxins in heterologous expression systems such as E. coli and yeast is often used for larger conotoxins with multiple disulfide bonds as well as for the production of conotoxin libraries for screening and SAR purposes (Xia et al., 2006, Pi et al., 2007, Xu et al., 2015). A good guide to venom peptide production in E. coli has been described by King and co-workers (Klint et al.,

2013) and a conotoxin-specific protocol was recently established for α-TxIB (Yu et al., 2018, Wu et al.,

2013). Recombinant expression using auxotroph strains is often employed to produce isotope labelled (e.g.,

15N, 13C) peptides to simplify structure determination by NMR (Mobli and King, 2010).

Fmoc-SPPS and undirected oxidative folding remain the most common strategies for the production of conotoxins since our last review. Directed folding approaches have been used less frequently and usually include the orthogonal thiol-protecting group S-Acm in combination with S-Trt and S-tBu, or the newly introduced S-Allocam. The selenocysteine/cysteine combinations continue to add value in simplifying folding and thereby accelerating SAR studies as well as providing access to more complex molecules, such as the heterodimeric conoinsulins. These as well as other innovations, such as new folding reagents, handy solubility tags, multivalent design, new disulfide bond mimetics and cyclization strategies are described in the following sections.

5.1 Oxidative folding strategies

The bioactive fold is generally induced thermodynamically based on the conotoxin’s sequence including its cysteine frameworks and subsequent disulfide bond connectivities. Even though multiple cysteine residues can potentially form a variety of disulfide bond isomers (Muttenthaler et al., 2010), conotoxins generally fold efficiently into a single predominant bioactive isomer. In addition to the

55 sequence-based folding information, cone snails employ a combination of post-translational processing where N- and C-terminal propeptides act as intramolecular chaperones and folding catalysts to produce the bioactive fold (Gething and Sambrook, 1992, Hartl, 1996, Dobson, 2003, Arolas et al., 2006). In vitro non- directed folding of reduced conotoxins relies mainly on the folding information encoded in the mature sequence in combination with experimentally validated folding conditions (Akondi et al., 2014). 2,2′- dithiopyridine (DTP) (Maruyama et al., 1999) can now be added to these conditions as it accelerates oxidative folding of conotoxins from days to minutes (Giribaldi et al., 2018). Ionic liquids have also been screened to facilitate conotoxin folding, though it remains to be seen if this approach will find broader use in the field (Heimer et al., 2014, Heimer et al., 2018, Sajeevan and Roy, 2018).

Structural implications of different disulfide bond isomers have now been extended to a three- disulfide bond system, where all 15 possible isomers of μ-PIIIA were chemically produced through directed folding (with Cys(Trt), Cys(Acm), Cys(tBu) pairs) and studied by NMR (Heimer et al., 2018). Thatis study highlights the importance of the correct disulfide bond connectivity to obtain bioactivity as well as the limitations of efficiently separating the different isomers by HPLC.

5.2 Directed folding strategies

Directed folding is important for situations where the bioactive fold does not spontaneously form or where non-native modifications have been introduced that disrupt the encoded folding sequence.

Directed folding is generally achieved via orthogonal thiol-protecting groups or strategic cysteine- selenocysteine replacements (Akondi et al., 2014, Muttenthaler et al., 2010). A combination of these strategies was successfully applied for the synthesis of the newly-discovered, three-disulfide-bond- containing heterodimeric conoinsulins that were active in zebrafish hyperglycemia models (Scheme 1)

(Safavi-Hemami et al., 2015).

The cysteine-selenocysteine replacement strategy was also employed in a SAR study on -BuIIIA targeting the Nav1.3 ion channel (Green et al., 2014). Replacing one of the three disulfide bonds (CysI-

CysIV) with a diselenide bond and replacing another one (CysII-CysVI) with alanine, simplified folding substantially, thereby enabling rapid and directed one-pot folding of μ-BuIIIA analogs yielding only single folding isomers that retained activity at Nav1.3. Thatis SAR study provided novel insights regarding the influence of aromatic and basic residues near the C-terminus of μ-BuIIIB in terms of Nav1.3 potency and selectivity, and produced a novel analog ([C5U,C17U,C6A,C23A,E3Dab]-BuIIIB) with 5.5-fold higher potency (Kd = 36 nM) at Nav1.3 compared to native μ-BuIIIB (Kd = 200 nM, U=selenocysteine).

Another orthogonal synthesis was devised to study the antiparallel dimeric N-terminal domain of large αD-GeXXA (Scheme 2) (Hemu and Tam, 2017). The thiol-activating property of DTNB (5,5’- dithiobis-(2-nitrobenzoic acid)) in combination with S-Acm was used to obtain only the antiparallel heterodimer without any of the homodimeric side products.

Scheme 2. Regioselective off-resin folding strategy for the synthesis of the antiparallel dimeric N-terminal domain of αD-GeXXA using DTNB (5,5’-dithiobis-(2-nitrobenzoic acid)) in combination with S-Acm.

DTNB activates the free thiol of the A-chain, thereby facilitating the desired regioselective interchain disulfide bond formation upon mixing it with the A- with the B-chain at pH 7.3. Iodine oxidation removes then the Acm protecting group forming the second intermolecular disulfide bond.

An acid-cleavable solubility tag was developed to facilitate the synthesis, purification and directed disulfide bond formation of the highly-hydrophobic and thus difficult-to-produce class of δ-conotoxins

(Scheme 3) (Peigneur et al., 2014). This strategy enabled the synthesis and fully-directed folding of multiple bioactive δ-conotoxins analogs, a task that was unachievable using strategies without this solubility tag.

Scheme 3. Regioselective off-resin folding strategy for the synthesis of hydrophobic δ-conotoxins using an acid-cleavable solubility tag and Fmoc chemistry. Four lysine residues were coupled to the Rink Amide resin followed by acylation with the phenylacetamido (PAM) linker. Regioselective folding was achieved 59 using pairs of S-Trt, S-Acm, and S-tBu orthogonal protecting groups. The solubility tag was in the end removed by treatment with HF.

RecentlyFurthermore, a new orthogonal thiol protecting group, Allocam

(allyloxycarbonylaminomethyl), was validated with α-LvIA for on-resin regioselective folding (Scheme 4)

(Kondasinghe et al., 2019). This versatile iodine- or palladium-removable protecting group is compatible with the S-Trt and S-Mmt protecting groups, delivers good yields, and will hopefully be further evaluated in future syntheses.

Scheme 4. Regioselective on-resin folding strategy for the synthesis of α-LvIA using a three-glycine spaced

Rink Amide ChemMatrix resin, Fmoc chemistry and (A) a combination of Allocam and Trityl protecting groups, or (B) a combination of Allocam and Mmt/StBu protecting groups. Removal of the Allocam protecting group is either achieved by palladium or iodine. Allocam, allyloxycarbonylaminomethyl; DMSO,

dimethylsulfoxide; DTNP, 2,2’-dithiobis(5-nitropyridine); HS-(CH2)2-OH, 2-mercaptoethanol; Mmt, monomethoxy trityl; NMM, N-methyl morpholine; Npys, 2-(5-nitropyridyl); TFA, trifluoroacetic acid.

5.3 Disulfide bond isosteres / peptide mimetics

Disulfide bonds are inherently unstable under reducing conditions, hence there is a strong interest for isosteric replacement strategies without affecting potency or selectivity. This can be achieved by replacing the disulfide bond by thioether, selenoether, selenylsulfide, diselenide, ditellurium, dicarba or lactam bonds (Akondi et al., 2014). Disulfide bond replacement with 1,2,3-triazole, obtained through copper-catalyzed azide-alkyne cycloaddition (CuAAC) was studied in χ-MrIA and resulted in mimetics with improved plasma stability and stability to reduction (Gori et al., 2015). Full ability to inhibit the norepinephrine transporter was retained with the CysI-CysIV disulfide bond replacement, whereas the

CysII-CysIII replacement yielded reduced activity, highlighting that the 1,2,3-triazole can induce unwanted structural constraints if not strategically placed.

A new synthetic approach to produce dicarba disulfide bond mimetics using a diaminodiacid building block was studied in three-disulfide bond containing μ-SIIIA (Scheme 5) (Guo et al., 2015). Three analogs were produced with each having one disulfide bond replaced with a dicarba bridge that accelerated folding and improved yield compared to native μ-SIIIA but had little impact on their NMR structure.

Unfortunately, no bioactivity experiments were carried out to assess its isosteric properties in terms of potency and selectivity.

Olefin metathesis is an alternative method to produce more stable dicarba disulfide bond mimetics and this was recently employed on α-Vc1.1 (van Lierop et al., 2013) and α-RgIA (Chhabra et al., 2014).

Interestingly, the placement of the dicarba bond influenced receptor selectivity between the α9α10 nAChRs and the gamma-aminobutyric acid (GABA) B (GABAB) receptor. When the dicarba bond was placed in the 1-3 position, activity at GABAB receptor was retained, while when placed in the 2-4 position, inhibition

61 at the nAChR was retained. This trend was observed in both -conotoxins and the structures were also analyzed by NMR.

Scheme 5. Dicarba disulfide bond mimetics. A protected diaminodiacid building block (Fmoc, Alloc and

Ally protecting groups) is used instead of a cysteine residue during the Fmoc-SPPS chain assembly of -

SIIIA. Chain assembly is stopped before the paired cysteine residue. Cyclization is carried out after deprotection of the Alloc, Ally and Fmoc protecting groups. The rest of the peptide sequence is then assembled using standard Fmoc-SPPS. Following cleavage from resin, the remaining two disulfide bonds are formed in 0.1 M Tris buffer at pH 7.5.

The application of a dithiol amino acid that can form two disulfide bonds at a single amino acid site was studied in α-ImI as a surrogate for two adjacent cysteine residues (Scheme 6) (Chen et al., 2014b).

Undirected folding yielded predominantly the native fold with similar structure as α-ImI (determined by

NMR). This analog was 7.6-fold more potent in blocking the nAChR α7 receptor than α-ImI, likely due to the more compact fold induced through the dithiol linker.

Scheme 6. Two disulfide bonds from a single amino acid position. L-4,5-dithiolnorvaline (L-Dtn) is incorporated into α-ImI instead of the first cysteine residue, whereas the adjacent cysteine residue is replaced by alanine. The dithiol analog folds into a bioactive fold that is structurally highly similar to α-

ImI (compared by NMR) and displays a 7.6-fold increased potency at the nAChR α receptor compared to

α-ImI.

5.4 Cyclic conotoxins

N-C backbone cyclization has not been observed in conotoxins naturally, but is an attractive strategy to improve the in vivo half-life (Akondi et al., 2014). Chemically, cyclization is largely carried out via intramolecular native chemical ligation owing to the cysteine-rich nature of the conotoxins. Fitting amino acid linkers are incorporated based on the distances between the N and C termini to ensure little impact onin structure and or bioactivity (Scheme 7). This chemical cyclization approach has been employed and

63 studied extensively with various cyclic α-Vc1.1 analogs in an effort to produce more potent, selective, and stable drug candidates for the treatment of chronic pain targeting the GABAB receptor (Carstens et al.,

2016).

Scheme 7. Intramolecular native chemical ligation to produce N-C backbone cyclized conotoxins. A. The conotoxin sequence is assembled by SPPS including a short linker region (light grey) that bridges the distance between the N- and C-termini (generally determined by NMR). A C-terminal thioester and N- terminal cysteine are required for the intramolecular ligation reaction to occur, which first undergoes transesterification, followed by a S to N acyl shift, resulting in backbone cyclization via a peptide bond. B.

Example of the 3D NMR structure of a cyclized α-conotoxin (cyclic-MII with a 6-residue linker, PDB

2jaw).

Backbone cyclization can also improve folding, as exemplified with α-GeXIVA, where cyclization leads to selective formation of the native ribbon disulfide connectivity without requiring orthogonal protection (Wu et al., 2017). Cyclic α-GeXIVA analogs furthermore have improved stability in human serum and retain biological activity at the human αα nAChR. Cyclization can also be carried out enzymatically using a mutated version of the bacterial transpeptidase, sortase A, as demonstrated with cyclic Vc1.1 (Jia et al., 2014), α-MII (Cheng et al., 2018) and ICK-containing peptide κ-PVIIA (Kwon et al., 2016).

5.5 Multivalent conotoxins

Di- and tetravalent α-ImI dendrimers were synthesized using CuAAC between an azide- polyethylene glycol spacer-modified α-ImI to an alkyne-polylysine dendron (Wan et al., 2015). Structural analysis by NMR and cCircular dichroism (CD) confirmed the α-ImI moieties attached onto the dendrimers had the same 3D structure as native α-ImI, which aligned well with the binding affinity to the AChBP. The

α-ImI dendrimers had 100-fold enhanced potency at the human α7 nAChRs compared to native α-ImI, while no significant potency enhancement was observed at the heteromeric α3β2 and α9α10 nAChRs, indicating that multivalent design can be used to enhance conotoxin potency and selectivity. The same chemical strategy was employed to produce tetra and octavalent χ-MrIA dendrimers. While the in vitro activity of the dendrimers was similar to native χ-MrIA, the dendrimers, in contrast to native χ-MrIA, displayed no anti-allodynic activity when administered intrathecally in a rat model of neuropathic pain, suggesting that the larger dendrimer structures are unable to diffuse through the spinal column tissue and reach the norepinephrine transporter.

Overall, it can be said that conotoxins have become well-recognized model peptides due to their well-defined secondary structures, their diverse disulfide-bond frameworks and their therapeutically relevant, potent and easy-accessible bioactivity (Bingham et al., 2012). The conotoxin field has clearly benefitted from this, as it attracts many researchers from areas often not related to conotoxin research to showcase and validate their innovative new chemistry, which at the same time accelerates conotoxin discovery research often resulting in better therapeutic lead candidates.

6. Pharmacological diversity

6.1 Novel conotoxins acting on ion channels and transporters

6.1.1 Calcium channel modulators

Voltage-gated calcium channels (VGCCs) are central to the transmission, development and maintenance of chronic pain, suggesting that VGCC inhibitors or modulators could be developed into useful drugs to treat neuropathic pain (Zamponi, 2016). Interestingly, many ω-conotoxins are selective N-type or

P/Q type VGCC antagonists and preferentially block nociception in inflammatory pain models, and allodynia and/or hyperalgesia in neuropathic pain models. With the development of the marketed drug Prialt

(ω-MVIIA), arguably ω-conotoxins are the most successful class of conotoxins in terms of therapeutic value and application possibilities. The development, diversity, synthesis, pharmacology, SAR and analysis of ω-conotoxin VGCC interactions have been comprehensively reviewed (Vink and Alewood, 2012, Lewis et al., 2012, Durek and Craik, 2015, Sousa et al., 2013, Catterall and Swanson, 2015, Lee, 2014, Simms and Zamponi, 2014, Zamponi, 2016, Bourinet and Zamponi, 2017).

Although C. magus derived ω-MVIIA (Prialt, Elan) is approved by the FDA to treat severe chronic pain in patients unresponsive to opioid therapy, it produces dose limiting neurological and psychiatric side- effects that narrow its therapeutic index (Sanford, 2013, Bourinet and Zamponi, 2017). Therefore, second generation ω-conotoxin discovery was redirected to ω-conotoxins with better efficacy and fewer side effects.

Although C. catus derived CVID (AM336, AMRAD) had better selectivity towards Cav2.2 and fewer side effects compared to ω-MVIIA, clinical trials were stalled after Phase I/IIa due to lack of commercial interest in developing intrathecal therapies (Kolosov et al., 2011) (Sousa et al., 2018a). A series of ω-MVIIA/SO-

3 hybrids also showed reduced ω-MVIIA toxicity associated with methionine 12, providing a new avenue for the design of ω-conotoxins with faster off rate and lower toxicity that can reduce toxicity (Wang et al.,

2016). Currently no new ω-conotoxins are in preclinical development.

Despite the therapeutic promise, the rate of discovery for novel ω-conotoxins has slowed in the past

5 years. New discoveries seem to be limited by the availability of novel fish-hunting species where the

66 majority of characterized ω-conotoxins (CVIA-F, GVIA, MVIIA-D, SO3, SVIA and SVIB) have been discovered. From the literature, it is evident that fish hunters belonging to clades Pionoconus and

Gastridium produce ω-conotoxins as major components of their venom, but so far, no mammalian active

ω-conotoxins were reported from other fish hunters of clades such as Chelyconus, Phasmoconus and Textila.

Initially, ω-conotoxins were believed to be a part of the “motor cabal” of fish hunters that allows rapid immobilization of prey. However, recent realization of separate predation and defense mechanisms demonstrated that ω-conotoxins are an essential part of the defensive strategy in C. geographus (Dutertre et al., 2014). ω-Conotoxins are occasionally found in the venoms of worm and mollusk hunters (TxVII,

PnVIA, PnVIB, PuIA, PuIIA) and the sequence properties of these peptides diverge from fish-hunting ω- conotoxins (Figure 32). Indeed, the first mammalian active analgesic ω-conotoxins (MoVIA and MoVIB) from a worm-hunting cone snail was only isolated and characterized recently from the crude venom of C. moncuri (Sousa et al., 2018a).

A B ω-Conotoxins from fish hunting species MVIIA CKGKGAK CSR---LMYD CCTGSC--RS GKC* CVID CKSKGAK CSK---LMYD CCSGSCSGTV GRC* CVIE CKGKGAS CRR---TSYD CCTGSC--RS GRC* CVIF CKGKGAS CRR---TSYD CCTGSC--RL GRC* SO3 CKAAGKP CSR---IAYN CCTGSC--RS GKC* SVIA CRSSGSO CGV---TSI- -CCGRC--YR GKCT* SVIB CKLKGQS CRK---TSYD CCSGSC-GRS GKC* GVIA CKSOGSS CSO---TSYN CCR-SCNOYT KRCY*

ω-Conotoxins from mollusk hunting species C TxVII CKQADEP CDV---FSLD CCTGIC---L GVCMW IC 50 (µ M ) IC 50 (µ M ) PnVIB DDDCEPPGNF CGM-IKIGPP CCSGWCFFAC A ω -C o n o to xin h C a v 2.2 in S H-S Y5Y rD R G n eu ro n N -ty p e PnVIA GCLEVDYF CGIPFANNGL CCSGNCVFVC TPQ flu o rim etric a ssa y s c u rren ts MoVIA 0.33 0.08 ω-Conotoxins from worm hunting species MoVIB 0.60 0.18 MoVIA CKPOGSK CSO---SMRD CCT-TCISYT KRCRKYYN MoVIB-[R13Y] 3.47 0.90 MoVIB CKPOGSK CSO---SMRD CCT-TCISYT KRCRKYY PuIA RDCRPVGQY CGIPYEHNWR CCSQLCAIIC VS MVIIA 0.024 0.052 PuIIA TCNTPTQY CTL----HRH CCSLYCHKTI HACA MVIIA-[Y13R] Inactive at 30µM Not detected

Figure 32: Diversity of ω-conotoxins. (A) Sequence comparisons of ω-conotoxins derived from fish- hunting, mollusk-hunting and worm-hunting cone snails. (B) Concentration dependent inhibition of high voltage-activated (HVA) calcium channel currents in isolated rat DRG neurons, obtained for first mammalian active, worm-hunting venom derived ω-MoVIA and ω-MoVIB and synthetic mutant MoVIB-

[R13Y] (Sousa et al., 2018b). (C) Comparison of the Cav2.2 inhibiting effects of worm-hunting venom derived ω-MoVIA and ω-MoVIB with fish-hunting venom derived Cav2.2 inhibitor ω-MVIIA and their 67 single point mutants (Sousa et al., 2018b).

ω-MoVIA and ω-MoVIB potently inhibited human Cav2.2 in fluorometric assays and rat Cav2.2 in patch clamp studies (Figure 32B), and both potently displaced radiolabeled ω-GVIA (125I-GVIA) from human SH-SY5Y cells and fish brain membranes (IC50 2-9 pM) (Sousa et al., 2018a). Interestingly, in both of these peptides, an arginine has replaced the tyrosine at position 13, which was found to be critical for the biological activity of piscivorous ω-conotoxins (Lewis et al., 2012). However, the SAR studies investigating the presence of arginine and tyrosine in position 13 of worm-hunting and fish-hunting venom derived ω-conotoxins, respectively, have shown a significant loss in activity in synthesized MoVIB[R13Y], and MVIIA[Y13R]. These results indicate that arginine at position 13 is preferred over tyrosine in ω- conotoxins from worm hunters, contrary to ω-conotoxins from fish hunters, providing new insight into the structural features required for high-affinity interactions of ω-conotoxins at Cav2.2 channels.

6.1.2 Calcium channel modulation via GABAB receptor

Whereas ω-conotoxins are highly selective pore blockers which are receptor independent modulators of Cav2.2 channels, Cav2.2 channels are under the powerful control of several GPCRs, including

GABAB receptors and various members of the opioid receptor family, presenting multiple ways to target

Cav2.2 channels for therapeutic purposes (Zamponi, 2016, Zamponi and Currie, 2013). Interestingly, a subset of α-conotoxins (Vc1.1, RgIA, PeIA, AuIB and Eu1.6) have potent analgesic properties. Among these selective blockers of α9α10 nAChRs, α-Vc1.1 and α-RgIA potently inhibited high voltage-activated

(HVA) calcium channels via a GABAB receptor mediated mechanism (Adams and Berecki, 2013, Cai et al., 2018, Lewis et al., 2012). Cyclic cVc1.1[D11A,E14A], engineered to specifically target HVA calcium channels, is >8000-fold more selective for GABAB receptor mediated inhibition of HVA calcium channels compared to α9α10 nAChRs, and is still a potent analgesic in a mouse model of chronic visceral hypersensitivity further underpinning this mechanism of action (Figure 33A) (Sadeghi et al., 2018).

Vc1.1 GCCSDPRCNY DHPEIC* Vc1.1[D11E,E14A] GCCSDPRCNY EHPAIC CyclicVc1.1[D11E,E14A] GCCSDPRCNY EHPAICGGAAGG A B Eu1.6 GCCSNPACML KNPNKC* (i) Vc1.1 GCCSDPRCNY DHPEI C * (i ) RgIA GCCSDPRCRY RCR Vc1.1[D11E,E14A] GCCSDPRCNY EHP A I C RgIA4 GCCTDPRCXY QCY CyclicVc1.1[D11E,E14A] GCCSDPRCNY EHP A ICGGAAGG Eu1.6 GCCSNPACML KNPNKC* (ii) (ii) % In h ib itiRgIAo n o f th e GICCC50 (SDPRn M ) H VCARY C a RC%R In h ib itio n o f th e h-nAChR Subtype 1C 50 P ep ti d e H V A C aRgIA4 c u rren ts GcCCu rreTnDPRts in mCiXc eY D R QG ChYα 9α 10 c u rre n ts α9α10 1.5 nM (100 n M ) n eu ro n s (1 µ M p ep tid e) α2β2 >10 µM Vc1.1 30.4 ± 3.2 1.7 64.0 ± 3.5 α2β4 >10 µM Vc1.1[D11E,E14A] 38.4 ± 2.5 2.5 ± 1.1 3 ± 2.6 α3β2 >10 µM CyclicVc1.1[D11E,E14A] 48.9 ± 4.7 3.3 ± 1.1 6.4 ± 2.9 α3β4 >10 µM (iii) α4β2 >10 µM α4β4 >10 µM α7 >1.8 µM Α1β1δ! >10 µM

CyclicVc1.1[D11E,E14A]

Figure 33: Chemically engineered selective analogs of Vc1.1 (A) and RgIA (B). (A) (i) Sequence alignment of the analogs of Vc1.1. Blue color is used to mark the residues that are different to the native peptide. Green color residues show the linker used for cyclization. (ii) Biological activity of Vc1.1 analogs in comparison to the native peptide showing, the inhibitory effect of the peptides on HVA Ca currents in mice DRG neurons (percentage inhibition and IC50) and the percentage inhibition of human  currents expressed in Xenopus oocytes. (iii) Three dimensional solution structure of the most active analog cVc1.1[D11A,E14A (PDB 6CGX). (B) (i) Sequence alignment of the α9α10 selective synthetic analog

RgIA4 and native RgIA. (ii) Selectivity of RgIA4 across different human nAChR subtypes expressed in

Xenopus oocytes.

Conversely, McIntosh et al., have proposed an alternative hypothesis that α9α10 nAChRs is the primary analgesic target (Grau et al., 2018, Romero et al., 2017). A recently developed RgIA analog, RgIA4 has low nM potency and high selectivity for human, mouse and rat α9α10 nAChRs (Figure 33B) and elicits long lasting protection against oxaliplatin-induced cold allodynia (Christensen et al., 2017) and prevents chemotherapy induced neuropathic pain (Romero et al., 2017) in mice and rats. It is interesting to see that both proposed mechanisms are supported by several lines of evidence. However, possible cross-talk

between the α9α10 nAChRs and GABAB receptor mediated inhibition of calcium channels to elicit an analgesic effect remains to be addressed.

α-Eu1.6 is another atypical α-conotoxin discovered from the venom of C. eburneus. It was reported to be a potent Cav2.2 blocker with weak block at the α7 and α3β4 nAChRs. It has analgesic activity in rat partial sciatic nerve injury and chronic constriction injury pain models (Liu et al., 2018).

6.2 Novel conotoxins acting on nicotinic acetylcholine receptors

6.2.1 nAChR modulators

nAChR blockers remain by far the largest group of characterized conotoxins with the majority being

α-conotoxins from the A superfamily. Given their vast distribution it is no surprise that nAChR blockers top the list of known conotoxins. nAChRs are pentameric ligand-gated cation channels that play key roles in synaptic transmission in the central and peripheral nervous system (Vetter and Lewis, 2012, Giribaldi and Dutertre, 2018, Abraham and Lewis, 2018). α7 nAChR subtypes are thought to be an ancestral form evolved in lower organisms that do not rely on fast excitatory mechanisms and are capable of inducing downstream signaling mechanisms in non-neuronal cells (Papke, 2014). Given their diverse functionality, nAChRs are a major target of prey capture and defense strategies in cone snails, which explains the abundance and diversity of nAChR modulators in their venom. α-Conotoxins antagonize both muscle and neuronal homomeric or heteromeric nAChR subtypes with exquisite selectivity and potency. Most α- conotoxins are competitive antagonists that bind to the acetylcholine ligand-binding site located at the interface between subunits. Their high selectivity has been instrumental in decoding molecular, cellular and physiological functions of these specific nAChR isoforms and to understand orthosteric ligand recognition at the nAChR and pair-wise interactions defining subtype selectivity (Dineley et al., 2015, Hartmut et al.,

2018, Abraham and Lewis, 2018, Jin et al., 2017).

α-Conotoxins have been extensively reviewed in recent years due to their abundancey and pharmacological importance (Durek and Craik, 2015, Abraham and Lewis, 2018, Lewis et al., 2012, Jin et

70 al., 2017, Giribaldi and Dutertre, 2018). They have demonstrated a range of potential applications, including, diagnostic agents, skeletal muscle relaxants and therapeutic agents for the treatment of cancer, mood, psychotic, cognitive or cardiovascular disorders, drug addiction, urinary incontinence and pain. NHowever, no α-conotoxin based drug has been developed to date, although they have helped to reveal how binding at the orthosteric site is mediated (Harvey, 2014). Extensive structure activity relationship (SAR) studies have been conducted with the use of AChBP co-crystal structures, systematic alanine scans, combinatorial chemistry, and structure-based mutagenesis methods to elucidate the interactions between α-conotoxins and nAChR isoforms and to rationally guide the synthesis and development of more potent and selective antagonists of therapeutic relevance (Abraham et al., 2017, Romero et al., 2017, Sadeghi et al., 2018,

Carstens et al., 2016, Daniel and Clark, 2017, Kompella et al., 2015).

Most α-conotoxins discovered in the last five years fall in the category of classical nAChR antagonists. Although many α-conotoxins have been discovered acting selectively on different subunits of nAChRs, VnIB discovered in C. ventricosus is the first native α-conotoxin to potently and selectively antagonize the neuronal α6β4 nAChR (van Hout et al., 2019). In contrast to most of α-conotoxins, recently identified α-MrIC from the transcriptome of C. marmoreus (Dutertre et al., 2013) is atypical despite similar

3D structure and disulfide bond connectivity, as it is a biased agonist at the α7 nAChR in the presence of type-II nAChR positive allosteric modulator (Jin et al., 2014, Mueller et al., 2015). This activity suggests that α-MrIC has a unique receptor state dependence probably due to the differences in primary sequence, including an extended hydrophobic N-terminal. However, the precise molecular factors contributing to this unusual pharmacology remain to be identified.

6.2.2 Atypical α-conotoxins targeting nAChRs

While α-conotoxins of A superfamily are the most common nAChR modulators found in cone snail venoms, other conotoxins targeting nAChR receptors are distributed across eight superfamilies (B, D, J, L,

M, O1, S, T). Among these nAChR inhibitory αD-conotoxins are relatively more characterized. αD-

71 conotoxins exist as covalently linked homomeric dimers, hence the difficulty in synthesis of these relatively large molecules. However, a recent study characterized α9α10 subtype selective αD-GeXXA (Xu et al.,

2015). The high-resolution crystal structure of αD-GeXXA (Figure 30) helped to reveal that binding is mediated through the C-terminal domain of αD-GeXXA (Xu et al., 2015). It was suggested that αD-GeXXA cooperatively binds to two inter-subunit interfaces on the top surface of the nAChR, and thereby allosterically disturbs the opening of the receptor. The internal dimeric N-terminal domain (NTD) of αD-

GeXXA (Figure 34A) preferentially binds to the β subunits of the nAChR and the NTD part of αD-GeXXA acts as “lid-covering” nAChR inhibitor (Figure 34B), displaying a novel inhibitory mechanism distinct from other allosteric ligands of nAChRs (Yang et al., 2017). The novel antagonistic mechanism of αD-

GeXXA via a new binding site on nAChRs provides a new avenue for the design of new nAChR-targeting compounds. Also, the advances in preparing an active dimeric NTD made it possible to generate more selective peptides (Figure 34A) and to highlight the importance of the all four arginine residues in the short

NTD dimer since replacing this arginine residue with glutamine renders the analog inactive. The inhibitory activity of CTD (C-terminal domain) and the NTD of αD-GeXXA on hα9α10 nAChR are similar, while the activity of monomeric CTD is too low to be detected at rα1β1δ nAChR, indicating the selectivity of

NTD peptide to rodent muscle type nAChR (Figure 34C). Interestingly, a study on defensive strategies of worm-hunting cone snails revealed that αD-conotoxins are almost exclusively injected during defensive stings by C. vexillum (Prashanth et al., 2016). Further investigation of defensive venom from worm-hunting cone snail species might reveal novel αD conotoxins with interesting pharmacology.

A αD-GeXXA DVHRPC QSVRPGRVWGKCCLTRLC STMCCARADC TC VYHTWRGHGC SC VM MVC S C GHGRWTHYVCT C DARACCMTSCLRTLCCKGWVRGPRVSQ C PRHVD

B 6 18 Short-NTD C QSVRPGRVWGKC

C KGWVRGPRVSQC 18 6

1 13 cNTD C QSVRPGRVWGKP G C

C KGWVRGPRVSQP hα9α10 (IC 50) rα1β1δ! (IC 50) 27 15 Peptide

αD-GeXXA 28 nM 743 nM 1 13 cNTD-RQ C QSVQ PGQ VWGKP CTD 2.02 µM ND G NTD 2.33 µM 5.88 µM C KGWVQ GPQ VSQP 27 15 cNTD 2.66 µM 3.91 µM

Figure 34: N terminal mutants derived from αD-GeXXA and their functionality. (A). Active N terminal domain (green) of αD-GeXXA and its mutants, dimeric NTD, cyclic NTD (cNTD) and RQ mutants of cNTD. (B) Top view of the αD-GeXXA NTD (cyan) bound to the top surface of the human

α4(pink)β2(baige) nAChR subtype. Model based on the crystal structure of dimeric αD-GeXXA (PDB

4X9Z). (C) Inhibition of human α9α10 and rat α1β1δε nAChRs by αD-GeXXA, CTD of αD-GeXXA and

NTD mutants (Yang et al., 2017).

αS-conotoxins of the S superfamily with five disulfide bonds are another unusual sub-class of nAChR inhibitors. To date only two peptides of this sub-class have been identified from fish-hunting cone snails, namely αS-RVIIIA from C. radiates (Teichert et al., 2005) and αS-GVIIIB from C. geographus

(Christensen et al., 2015). αS-GVIIIB is an exclusive competitive inhibitor of α9α10, which binds to the orthosteric binding site at the interface of the α9/α10 subunits (Christensen et al., 2015). Also the B3 superfamily peptide α-VxXXIVA, which inhibits the α9α10 nAChR receptor with low micromolar affinity was found through a cDNA library search of C. vexillium (Luo et al., 2013). Another interesting α9α10 nAChR subtype selective allosteric inhibitor, GeXIVA was isolated from C. generalis (Luo et al., 2015).

GeXIVA belongs to the O1 superfamily, although the mature conotoxin is quite distinct from other

73 members of the O1 superfamily due to its high density of arginine residues (9 arginine residues in the 28 amino acid peptide).

6.3 Sodium channel modulators

There are nine different mammalian sodium channel α-subunits, NaV1.1 to 1.9, which form the ion conducting pore, and co-assemble with the auxiliary β‐subunits (β1 to β4) modulating channel gating and trafficking (Bennett et al., 2019, Deuis et al., 2017). Nav isoforms 1.1, 1.2, and 1.6 are present in both the

CNS and PNS, while Nav1.3 is only present in the CNS and 1.7-1.9 are only present in the PNS (Kwong and Carr, 2015). Isoforms 1.4 and 1.5 are mostly expressed in skeletal muscle and the heart, respectively

(Kwong and Carr, 2015). Given the critical role of Nav channels in the CNS and PNS, it may not come as a surprise that cone snails have evolved a number of different ways to selectively target these channels.

To date, five classes of conotoxins are known to targeting Nav channels (µ-, µO-, µO-S-, δ-, and - conotoxins). Despite them acting on the same target, they are structurally distinct and elicit different pharmacology at Nav channels. µ, µO and µO-S conotoxins are Nav channel inhibitors while δ- and - conotoxins are activators. µ-Conotoxins bind to site 1 located near the ion-conducting pore where they block the pore. Several conotoxins acting on Nav1.4, Nav1.2 and Nav1.8 have been characterized and reviewed before (Lewis et al., 2012). Since that review very few new discoveries of Nav channel blockers have been reported (Chen et al., 2018). The adverse side effects of undesirable rapid paralysis and death of inflammatory and neuropathic pain models upon intravenous applications of µ-conotoxins, presumably due to off-target activity at skeletal subtype NaV1.4, may have discouraged the search for novel Nav channels blockers from cone snails as pain therapeutics. Efforts have been made to chemically engineer µ-conotoxins to selectively target therapeutically important Nav isoforms (Chen et al., 2018). µ-Conotoxins have also been explored beyond their therapeutic potential, e.g., as cosmetics. µ-CnIIIC, which is commercialized as

XEPTM-018, is a potent antagonist of Nav1.4 with an IC50 of 1.3 nM (Violette et al., 2012). A cream formulation of µ-CnIIIC is applied topically (Lirikos Marine Botoxin1; Amorepacific) to smooth facial lines in aesthetic dermatology (Del Rio-Sancho et al., 2017).

µO-conotoxins inhibit Nav channels via modulating voltage sensors in domain II to restrict channel opening, therefore they are not pore blockers. Due to their hydrophobic nature and therefore challenging synthesis, not many µO-conotoxins have been characterized and no novel µO-conotoxin has been discovered since 2014.

μO§-Conotoxins are the most recently discovered group of Na channel blockers and currently μO§-

Conotoxin GVIIJ isolated from C. geographus remains the only member of this group (Gajewiak et al.,

2014). Compared to typical Nav channel blocking conotoxins, μO§-GVIIJ has seven7 cysteine residues with six of them6 cross-linked to form three disulfide bonds and Cyscysteine 24 being S-cysteinylated.

GVIIJ defines site 8 on sodium channels between S5 and SS5 of DII where the peptide-channel complex is stabilized by a disulfide tether between Cys24 of the peptide and Cys910 of rat (r)Nav1.2 (Gajewiak et al.,

2014). Electrophysiological studies have shown that S-glutathionylated GVIIJ inhibited all TTX sensitive rNaV1 subtypes. Although the mechanism of S-glutathionylated GVIIJ mediated channel inhibition remains to be determined, it was confirmed that GVIIJ is not a classical pore blocker (Gajewiak et al., 2014) and thiol-oxidizing or disulfide-reducing agents could protect rNav1.2 from S-glutathionylated GVIIJ.

Therefore, in a follow-up study, a series of GVIIJ analogues were synthesized with Cys24 disulfide-bonded to various thiols to probe the redox state of extracellular cysteines in the Nav channels (Zhang et al., 2015).

The results of this study indicated that Cys910 in wild-type Nav1.2 has a free thiol and Cys910 is disulfide- bonded to Cys918 and Cys912, respectively, shedding light on the redox states of extracellular cysteines of the Nav channels. Furthermore, given GVIIJ block is β-subunit dependent, GVIIJ was used to establish that action potentials in A-fibers of the rodent sciatic nerves are mediated primarily by NaV1.6 associated with

NaVβ2 or NaVβ4 (Wilson et al., 2015).

δ-Conotoxins activate Nav channels via delaying the channel inactivation mechanism, which results in persistent neuronal firing. It is suggested that δ-conotoxins interact with the hydrophobic residues located in the S3/S4 linker of domain IV of the Nav channel (Lewis et al., 2012). Until recently, only δ-conotoxins isolated from fish hunters activated mammalian Nav channels (PVIA, SVIA, EVIA, NgVIA) (Figure 35)

(Aman et al., 2015). However, δ-conotoxins isolated from worm-hunting species of the Tesseliconus clade,

TsVIA (C. tessulatus) (Aman et al., 2015), ErVIA (C. eburneus) (Aman et al., 2015) and SuVIA (C. suturatus) (Jin et al., 2015a) are also potent activators of vertebrate Nav channels. In particular, SuVIA activates human Nav1.3, Nav1.4 and Nav1.6 channels with low nanomolar EC50 (Figure 35). These were the first reports of vertebrate active δ-conotoxins found from non-piscivorous species. The striking sequence similarity between δ-conotoxins from fish and worm hunters led to the hypothesis that δ-conotoxins from worm hunters primarily evolved for defence against larger predators, including fishes, and were ultimately repurposed to facilitate the dietary shift to fish-hunting lineages from ancestral vermivore diet (Jin et al.,

2015a). Despite these new discoveries, the therapeutic potential of δ-conotoxins is not well explored. The main constraint of studying δ-conotoxin pharmacology is the challenging synthesis and purification due to their hydrophobic nature. This challenge has recently been overcome by employing a solubility tag (Scheme

3) and we hope that this will soon provide more insights into their pharmacology and provide novel tools to decipher Nav channel activity mechanisms further.

A C. gloriomaris (M) GmVIA VKPCRKEGQL CD--PIFQNCCRGWNCVLFC V C. purpurascens(F) PVIA EACYAOGTF CGIKOGL--CCSEFCLPGVC FG* C. tessulatus (W) TsVIA CAAFGSF CGL-PGLVDCCSGRCFI-VC LL C. sutaratus (W) SuVIA CAGIGSF CGL-PGLVDCCSDRCFI-VC LP C. eburneus (W) ErVIA CAGIGSF CGL-PGLVDCCSGRCFI-VC LP

B C

p s

e Target EC50 (nM) R

d Nav1.4 4.99 ± 0.92

e s

i Nav1.3 3.98 ± 0.97 l

a Nav1.6 1.27 ± 0.56

m r

o Nav1.7 1.27 ± 0.56 N

Log [SuVIA] (M)

Figure 35: Recently discovered δ-conotoxins and their activities. (A) Alignment of recently discovered

δ-conotoxins from worm- and mollusk-hunting cone snails. UPGMA tree with 100 bootstrap replicates shows that molluscivorous δ-conotoxins are closely related to piscivorous δ-PVIA, while molluscivorous

δ-GmVIA clusters separately. (B) Effect of δ-SuVIA on different subtypes of Nav channels analyzed using 76 fluorimetric FLIPR assays and SH-SY5Y cells. Figure adapted from Jin et al., 2015. (C) Given the sequence and functional similarity of molluscivorous δ-conotoxins to piscivorous δ-conotoxins, it can be proposed that δ-conotoxins initially evolved for defense in worm-hunting cone snails and were then repurposed for predation in fish hunters.

6.4 Potassium channel modulators

Potassium channels (Kv) are one of the most abundant and diverse family of ion channels, with ~70 different isoforms. To match this diversity of targets, multiple families of conotoxins have evolved to modulate voltage gated potassium channels and there exist eight classes of K-channel blockers (κ- conotoxins), namely, κA, κO, κM, κJ, κI, contryphans, conkunitzins and conorfamides. The pharmacology of the κ-conotoxins was described in 2012 (Lewis et al., 2012) and no significant discoveries have been made since then. However, some µ-conotoxins can also selectively antagonize voltage-gated potassium channels of the Kv1 family. For instance, µ-PIIIA and µ-SIIIA inhibit Kv1.1 and Kv1.6 channels in the nanomolar range, while inactive on Kv1.2-1.5 and Kv2.1 (Leipold et al., 2017). The activity of these toxins on Kv channels is largely determined by the pore regions of the channels, including the “turret” domain

(Leipold et al., 2017). These results have not only revealed multiple targets of some µ-conotoxins but also raise an important question on considering µ-conotoxins as analgesics since off-target effects can potentially lead to severe side effects.

Cono-RFamide-Sr3 remains the first K-channel blocker identified from the relatively small class of conopeptides, Cono-RFamide (Figure 37) (Campos-Lira et al., 2017). Cono-RFamide-Sr3 is a 15-amino acid long disulfide-free peptide, and the third conorfamide identified from C. spurius. Cono-RFamide-Sr3 blocks the activity of shaker voltage dependent K channels. Although Cono-RFamide-Sr3 has only moderate affinity (2.7 ± 0.35 µM), it has high specificity to shaker subtype of Kv channels over Shab, Shaw,

Shal and hEag channels (Campos-Lira et al., 2017) presenting a new class of K-channel blockers that remains to be explored.

6.5 Activities beyond voltage- and ligand-gated ion channels

The last three decades of Conus research has highlighted that the main mechanisms underlying the biological activity of Conus venom are predominantly potent modulation of either voltage-gated or ligand- gated ion channels (Lewis et al., 2012). Given the diversity, abundance and high therapeutic promise, discovery has been directed towards identifying ligands for these known targets, mainly pain modulators.

The majority of venom peptides that has been pharmacologically characterized comes from the venoms of piscivorous Conus, mostly from the Indo-Pacific. Integrated venomics together with high-throughput peptide production (Turchetto et al., 2017) and screening technologies (Vetter, 2012, Inserra et al., 2013,

Himaya and Lewis, 2018) have however enabled comprehensive profiling and characterization of venoms from rare species without having to use large quantities of venom. These advances catapulted Conus venom research way beyond the initial focus of fish hunter venoms and ion channel modulators.

With integrated venomics approaches have revealeding another layer of conotoxin diversity, it is clear that the majority of conotoxins remains uncharacterized, including interesting and newly discovered atypical conotoxins. Characterization of conotoxins with novel structural features has revealed new activities; among these many resemble endogenous neuropeptides or hormones, suggesting their evolutionary origin.

In addition to the previously characterized neuropeptide like sequence families (conopressins, contulakins and conantokins) found in cone snail venoms (Lewis et al., 2012, Akondi et al., 2014), several novel classes of hormone/neuropeptide-like molecules targeting neuroendocrine processes have been discovered in the recent past revealing the fascinating strategies being employed by the cone snails for envenomation, suggesting the potential significance of nontoxic components of cone snail venoms. This section discusses the pharmacology of novel conotoxin families with previously unknown activities for conotoxins. 78

6.5.1 Conoinsulins

Conoinsulins induce hypoglycemic shock in fish (Safavi-Hemami et al., 2015) and make up a novel class of conotoxins that is widely distributed across fish-hunting cone snails. Con-Ins G1 is the smallest naturally occurring insulin analog ever described and it was suggested that a smaller size may be associated with better bioavailability (Robinson and Safavi-Hemami, 2016). Sequence similarity to vertebrate insulins and smaller size presented a unique opportunity to investigate the pharmacological potential of these special molecules evolved to affect glucose homeostasis. In a detailed functional and SAR study, it was found that

Con-Ins G1 is a naturally occurring B-chain-minimized mimetic of human insulin (Menting et al., 2016).

A B Human Insulin Blood glucose level in B-Chain receptor (isoform B) Zebra fish A-Chain Con-Ins activation (mg /dL) EC50 (nM) Con-Ins G1 GVVyHCCHRPCSNAEFKKYC* TFDTOKHRCGSYITNSYMDLCYR Con-Ins G1 16.3 145 ± 58.8 Con-Ins G3 GIVyVCCDNPCTVATLRTFCH NSDTPKHRCGSELADQYVQLCH* C o n -I ns G3 242.0 186.2 ± 65.9 Con-Ins TIA GVVyHCCHRPCSNAEFKKFC* NSDTOKYRCGSyIPNSYIDLCF Con-Ins TIB GVVyHCCHRPCSNAEFKKFC* NSDTOKYRCGSDIPNSYMDLCF Con-Ins T1A 1.5 77.8 ± 38.2 Con-Ins T2 GVVyHCCKRACSNAYFMQFC* NSDTPWNRCGSQITDSYRyLCPH Con-Ins T1B 12.0 - Con-Ins K1 GIVyDCCYNDCTDEKLKEYCHTLQ SDSGTTLVRRRLCGSyLVTYLGELCLGN Con-Ins T2 15.5 199.2 ± 89.1 Con-Ins K2 VIVGDCCDNYCTDERLKGYCASLLGL DSGTTPDRDHSCGGyLVDRLVKLCPSN Con-Ins K1 30.5 166.2 ± 94.2 Con-Ins K2 373.2 196.8 ± 62.5 Human Insulin 1.5 92 ± 45.5

Figure 36: Cone snail venom derived conoinsulins and their activities. (A) Alignment of the A and B chains of conoinsulins found in fish-hunting cone snails C. geographus (Con-InsG1 and G3), C. tulipa (Con

Ins-T1A, T1B and T2) and C. kinoshitai (Con-Ins K1 and K2). (B) Effect of conoinsulins on activation of the isoform B of human insulin receptor and the blood glucose level of streptozotocin (STZ)- induced diabetes model of zebrafish. Dose of 65 ng peptide/body weigh was injected (i.p) into the zebrafish following injection of STZ (1.5 g/kg).

Although Con-Ins G1 lacks the human insulin C-terminal part of the B chain that mediates engagement into the insulin receptor and assembly of the hormone’s hexameric storage form, it retains the all other structural features of vertebrate insulin including the A chain and canonical disulfide-bonding 79 patterns (Robinson and Safavi-Hemami, 2016). Despite lacking the equivalent B chain residues, monomeric conoinsulins (Con-Ins G1, Con-Ins G3, Con-Ins T1A, Con-Ins TIB, Con-Ins T2, Con-Ins K1, Con-Ins K2)

(Figure 36A) strongly bind to the B isoform of the human insulin receptor and activate receptor signaling

(Ahorukomeye et al., 2019). Moreover, when tested in zebrafish and mice, conoinsulins lower blood glucose in the streptozotocin-induced model of diabetes (Figure 36B) (Ahorukomeye et al., 2019).

Conoinsulins contain a structural element that mitigates the lack of an equivalent to residue PheB24 in the B chain of human insulins that is known to be critical for binding to the receptor. Crystal structure of

Con-Ins G1 and complementary modelling suggest that the absence of PheB24 is mitigated by residues

TyrB15 and TyrB20 (Menting et al., 2016). Moreover, Con-Ins G1 contains four post-translationally modified residues (A4 and B10 are γ-carboxyglutamates, A-chain C-terminal residue CysA20 is amidated

B3 is a hydroxyproline) and omission of these PTMs reduces the extent of receptor activation approximately by eight-fold, demonstrating the importance of these PTMs to functionality (Menting et al.,

2016). Molecular modelling studies revealed that in highly active Con Ins T1A, structurally important

TyrB20 observed in Con Ins G1 is replaced with PheB20, and still acts as a surrogate for the receptor- engaging residue PheB24 of the human insulins. In the model of Con-Ins K1 bound to the hIR, positions

B15 and B20 appear to play a key role in receptor binding, although the aromatic residues present in Con-

Ins G1 and T1A at these positions are replaced by leucine (Ahorukomeye et al., 2019).

Historically, shortening of the B-chain of human insulin to create rapid-acting insulins have failed since the C-terminal region of the B-chain is critical for its activity (Menting et al., 2013). However, these novel conoinsulins provide a novel scaffold for the design of an improved class of therapeutic human insulin mimetic analogs that are intrinsically monomeric and rapid-acting.

6.5.2 RF-amide peptides

Cono-RFamide-Sr1 (Maillo et al., 2002) and Cono-RFamide-Sr2 (Aguilar et al., 2008) from C. spurius were the first conopeptides discovered from an Atlantic worm-hunting Conus species. Cono-

RFamides have high sequence similarity with the molluskan cardioexcitatory RFamide tetrapeptide Phe-

Met-Arg-Phe-NH2 (FMRFamide) and to other FMRFa-related peptides that are found in other invertebrate and vertebrate phyla (Figure 37) (Campos-Lira et al., 2017).

Cono-RFamide-Sr1 elicits hyperactivity in mice similar to what is observed with the neuropeptide

FMRF–NH2. Structurally different cCono-RFamide-Sr2 has paralytic activity towards limpet Patella opea and elicits hyperactivity in the freshwater snail Pomacea paludosa and in mice when injected. Cono-

RFamide-Sr2 has sequence similarity to RFamide neuropeptides of marine and fresh water mollusks explaining its biological activities against mollusks. Cono-RFamide-Sr2 sequence is also highly similar to

RFamide neuropeptides of polychetes, the prey of C. spurius, providing a plausible ecological role in predation. Recently, a new RFamide like conotoxin, cono-RFamide-Vc1, was discovered from C. victoriae

(Robinson et al., 2015). Similar to cCono-RFamide-SrI, hyperactivity was observed upon injection of low doses of conorfamide-Vc1 into mice. At higher doses (2.5–20 nmol), complete incapacitation of mice lasting up to 20 min was observed. In further evaluation of its bioactivity, it was found that conorfamide-

Vc1 causes fluctuations in calcium ion levels in a range of neuronal and non-neuronal DRG cells (Robinson et al., 2015).

Human Neuropeptide FF FLFQPQRF* Lymnaea stagnalis Neuropeptide FMRF* ConoRFamide SR1 GPMGWVP VFYRF* ConoRFamide SR3 ATSGPMGWLP VFYRF* ConoRFamide SR2 GPMγDPL γIIRI* ConoRFamide Vc1.1 HSGFLLAWSG PRNRFVR* ConoRFamide Tx1.1 RPRF* ConoRFamide As1a RIKKP IFIAFPRF* ConoRFamide As2a RIRKP IFIAFPRF*

Figure 37: Sequence alignment of all characterized conoRFamides with structurally similar human neuropeptide FF and Lymnaea stagnalis derived neuropeptide FMRFamide.

The pharmacological target of cono-RFamides was identified with the discovery of cono-

RPRFamide-Tx1.1 (Reimers et al., 2017). Cono-RPRFamide-Tx1.1 specifically enhances nociceptor-

81 specific ion channel ASIC3 currents via increasing the excitability of sensory neurons. Injection of cono-

RPRFamide-Tx1.1 into the gastrocnemius muscle enhanced acid-induced muscle pain in mice that was abolished by genetic inactivation of ASIC3. Furthermore, it was confirmed that cono-RPRFamide-Tx1.1 does not activate or bind to FMRFa-gated Na+ channels, despite belonging to same RFamide family confirming its specificity for ACIC3 channels. The affinity to ASIC3 is ~10-fold higher than FMRFamide, making cono-RPRFamide-Tx1.1 the most potent RFamide modulating ASICs so far (Reimers et al., 2017).

Binding assays revealed that RPRFa interacts with the closed state of the ASIC3 (Reimers et al., 2017), as previously proposed for FMRFa interactions with ASICs (Chen et al., 2006). Interestingly, only four venom-derived toxins have been found to target ASICs to date, the ASIC1a inhibitor tarantula toxin PcTx1

(Escoubas et al., 2000), the ASIC3 inhibitor sea anemone toxin APETx2 (Diochot et al., 2004), the pan-

ASIC activator coral snake toxin MitTx (Bohlen et al., 2011), and the ASIC1a and 1b inhibitors snake toxin mambalgins (Diochot et al., 2012). These toxins have been instrumental in unravelling the physiological functions and the structure of ASICs. The ability of RPRF-amide to enhance acid induced muscle pain in mice via activating ASIC3 subtypes suggests that this peptide can be a useful tool to further study the role of ASIC3 in pain.

Recently two novel cono-RFamides, conorfamide-As1a and conorfamide-As2a from the Mexican cone snail C. austini (Figure 37) (Jin et al., 2019a), were identified. Pharmacological characterization of these peptides revealed that As1a and As2a altered desensitization of ASIC1a and to a lesser degree ASIC3 currents. As2a potentiated ASIC1a currents with an EC50 of 10.9 μM while As1a had little effect on peak currents up to 200 μM. Interestingly, conorfamides have poly-pharmacology, with nAChR activity being the most potent conorfamide target identified to date (Jin et al., 2019a). With these insights into the molecular target of cono-RFamides, it appears that their potential ecological role could be to enhance pain or cause discomfort upon injection in predators by targeting ASICs and/or muscle nAChRs.

6.5.3 Granulin-like conotoxins

C. miles venomics revealed a novel 33-residue C-terminal carboxylate, φ-MiXXVIIA belonging to the novel G2 superfamily (Jin et al., 2017). φ-MiXXVIIA is defined by eight cysteine residues (C-C-C-

CCC-C-C) including the rare sequential cysteine residue triplet (CCC). φ-MiXXVIIA displays a unique topology containing two β-hairpins that resemble the N-terminal domain of the cell proliferation regulator human granulin A (Figure 38).

Figure 38: Comparison of the secondary structure between conotoxins and recombinant human granulin A. φ‐MiXXVIIA (PDB 6PPC), truncated GeXXA (PDB 4X9Z) and Vc7.2 (PDB 6Q5Z) are in red. Granulin A (PDB 2JYE) is in grey. The disulfide bonds are represented by yellow sticks.

Granulin A is an ancestral growth factor involved in development and wound healing with homologous domains spanning most phyla. φ-MiXXVIIA also promotes cell proliferation (EC50 17.85 µM) 83 in the human liver cholangiocyte cell line H69. However, the cell proliferation levels were considerably lower compared to the most potent granulin analog from Opisthorchis viverinni liver fluke parasite (11.5 nM) (Smout et al., 2011). Although not a potent inducer of cell proliferation, φ-MiXXVIIA protects from apoptotic cell death (EC50 2.2 µM), when proliferation was restricted with nutrient limitations (Jin et al.,

2017). Conotoxins with anti-apoptotic activity are very rare; s-cal14.1a from C. califonicus has micromolar apoptotic activity in human lung cancer cells (Oroz-Parra et al., 2016). Given the diversity of conotoxins and the number of non-characterized conotoxins it is however not surprising to find unexpected biological activities. The most interesting evolutionary feature is that cone snails have weaponized ancestral molecules like neuropeptides and hormones for prey capture and defense.

AInterestingly, another recent study has reported an H superfamily peptide H-Vc7.2 (Figure 38) from C. victoriae with a structure resembling the N-terminal region of the human granulin protein (Nielsen et al., 2019). It was unusual that despite having typical ICK connectivities (CysI-CysIV/CysI-CysV/CysIII-

CysVI) of cysteine framework VI/VII, H-Vc7.2 has acquired a granulin-like fold. Further studies on the effects of varying loop lengths in cysteine framework VI/VII proteins will help determine how the mini- granulin vs. ICK folds evolved. Interestingly, the C terminal domain of the D-GeXXA (Figure 38) also has a granulin fold (Nielsen et al., 2019), suggesting that granulin-like structures might be more common in toxins than previously recognised.

7. Concluding remarks and outlook

With the relative decline in small molecule drug discovery and anits ensuing slowed translation into drugs many researchers are returning to natural product discovery once again. Of particular interest are conotoxins that are highly potent and selective on a wide range of receptors. CMoreover, constrained peptides of this size are seen as emerging therapeutic agents that address different chemical space to small molecules and biologics (Gongora-Benitez et al., 2014). Potentially, this peptide drug class has lower cost and enhanced pharmacokinetics with better tissue penetration and delivery.

Since our 2014 review in this journal, we have seen major advances in the application of integrated venomics approaches that have dramatically increased the rate of discovery of expressed conotoxin sequences and helped decipher where along the venom duct those associated with prey capture or defense are produced. Although only two new frameworks have been structurally and pharmacologically defined since 2015, many species are currently being studied. With about 750 known species this remains a rapidly growing field encouraging many new scientists to get involved. This advance in discovery research is in stark contrast to the limited progress in the synthesis and structure determination of most classes of conotoxins. In part this has occurred through the lack of robust folding strategies for these cysteine rich peptides, making the production of large conotoxin libraries an ongoing challenge. This lack of readily accessible chemical diversity is compounded by the focus on a handful of readily accessible conotoxin structural classes, with the structures of 20 of the 28 frameworks being poorly characterized or neglected.

Determining the pharmacology of these peptides also remains a major challenge that has not kept pace with discovery. In part this reflects the limited availability of screening strategies and an over-reliance on cell-based assays targeting mammalian membrane proteins, especially ion channels, GPCRs and transporters. Phenotypic screens using the cells, tissues and animal behaviors directly related to the predatory and defensive species targeted by cone snails are expected to have greater potential to address this pharmacological discovery gap. With the pharmaceutical industry beginning to embrace this approach, it is anticipated that such screens will help drive future discovery efforts and provide exciting new peptide drug leads. For example, zebrafish behavioral assays have the potential to identify new members of the nirvana cabal and C. elegans behavioral assays have the potential to identify conotoxins that specifically target worm receptors. With a greater emphasis being placed on the translation of conotoxins into new membrane protein probes, we predict that in the next 5 years this balance will again shift back to structure- function studies and the hope that new conotoxins will be identified with potential to help better understand and/or treat diseases where small molecule approaches have disappointed.

8 References

ABRAHAM, N., HEALY, M., RAGNARSSON, L., BRUST, A., ALEWOOD, P. F. & LEWIS, R. J. 2017. Structural mechanisms for α-conotoxin activity at the human α3β4 nicotinic acetylcholine receptor. Sci Rep, 7, 45466. ABRAHAM, N. & LEWIS, R. J. 2018. Neuronal Nicotinic Acetylcholine Receptor Modulators from Cone Snails. Mar Drugs, 16. ADAMS, D. J. & BERECKI, G. 2013. Mechanisms of conotoxin inhibition of N-type Cav2.2 calcium channels. Biochim Biophys Acta, 1828, 1619-28. AGUILAR, M. B., LOPEZ-VERA, E., ORTIZ, E., BECERRIL, B., POSSANI, L. D., OLIVERA, B. M. & HEIMER DE LA COTERA, E. P. 2005. A novel conotoxin from Conus delessertii with posttranslationally modified lysine residues. Biochemistry, 44, 11130-6. AGUILAR, M. B., LUNA-RAMÍREZ, K. S., ECHEVERRÍA, D., FALCÓN, A., OLIVERA, B. M., HEIMER DE LA COTERA, E. P. & MAILLO, M. 2008. Conorfamide-Sr2, a gamma- carboxyglutamate-containing FMRFamide-related peptide from the venom of Conus spurius with activity in mice and mollusks. Peptides, 29, 186-195. AGUILAR, M. B., ZUGASTI-CRUZ, A., FALCON, A., BATISTA, C. V., OLIVERA, B. M. & DE LA COTERA, E. P. 2013. A novel arrangement of Cys residues in a paralytic peptide of Conus cancellatus (jr. syn.: Conus austini), a worm-hunting snail from the Gulf of Mexico. Peptides, 41, 38-44. AHORUKOMEYE, P., DISOTUAR, M. M., GAJEWIAK, J., KARANTH, S., WATKINS, M., ROBINSON, S. D., FLOREZ SALCEDO, P., SMITH, N. A., SMITH, B. J., SCHLEGEL, A., FORBES, B. E., OLIVERA, B., HUNG-CHIEH CHOU, D. & SAFAVI-HEMAMI, H. 2019. Fish- hunting cone snail venoms are a rich source of minimized ligands of the vertebrate insulin receptor. Elife, 8. AKCAN, M., CAO, Y., CHONGXU, F. & CRAIK, D. J. 2013. The three-dimensional solution structure of mini-M conotoxin BtIIIA reveals a disconnection between disulfide connectivity and peptide fold. Bioorg Med Chem, 21, 3590-6. AKCAN, M., CLARK, R. J., DALY, N. L., CONIBEAR, A. C., DE FAOITE, A., HEGHINIAN, M. D., SAHIL, T., ADAMS, D. J., MARI, F. & CRAIK, D. J. 2015. Transforming conotoxins into cyclotides: Backbone cyclization of P-superfamily conotoxins. Biopolymers, 104, 682-92. AKONDI, K. B., MUTTENTHALER, M., DUTERTRE, S., KAAS, Q., CRAIK, D. J., LEWIS, R. J. & ALEWOOD, P. F. 2014. Discovery, synthesis, and structure-activity relationships of conotoxins. Chem Rev, 114, 5815-47. ALLAM, A., KALNIS, P. & SOLOVYEV, V. 2015. Karect: accurate correction of substitution, insertion and deletion errors for next-generation sequencing data. Bioinformatics, 31, 3421-8. ALVA, V., SODING, J. & LUPAS, A. N. 2015. A vocabulary of ancient peptides at the origin of folded proteins. Elife, 4, e09410. AMAN, J. W., IMPERIAL, J. S., UEBERHEIDE, B., ZHANG, M.-M., AGUILAR, M., TAYLOR, D., WATKINS, M., YOSHIKAMI, D., SHOWERS-CORNELI, P., SAFAVI-HEMAMI, H., BIGGS, J., TEICHERT, R. W. & OLIVERA, B. M. 2015. Insights into the origins of fish hunting in venomous cone snails from studies of Conus tessulatus. PNAS, 112, 5087-5092. AMBLARD, M., FEHRENTZ, J.-A., MARTINEZ, J. & SUBRA, G. 2006. Methods and protocols of modern solid phase peptide synthesis. Molecular Biotechnology, 33, 239-254. ARMSTRONG, D. A., KAAS, Q. & ROSENGREN, K. J. 2018. Prediction of disulfide dihedral angles using chemical shifts. Chem Sci, 9, 6548-6556. AROLAS, J. L., AVILES, F. X., CHANG, J.-Y. & VENTURA, S. 2006. Folding of small disulfide-rich proteins: clarifying the puzzle. Trends in Biochemical Sciences, 31, 292-301.

BARGHI, N., CONCEPCION, G. P., OLIVERA, B. M. & LLUISMA, A. O. 2015a. Comparison of the Venom Peptides and Their Expression in Closely Related Conus Species: Insights into Adaptive Post-speciation Evolution of Conus Exogenomes. Genome Biol Evol, 7, 1797-814. BARGHI, N., CONCEPCION, G. P., OLIVERA, B. M. & LLUISMA, A. O. 2015b. High conopeptide diversity in Conus tribblei revealed through analysis of venom duct transcriptome using two high- throughput sequencing platforms. Mar Biotechnol (NY), 17, 81-98. BARTON, M. E., WHITE, H. S. & WILCOX, K. S. 2004. The effect of CGX-1007 and CI-1041, novel NMDA receptor antagonists, on NMDA receptor-mediated EPSCs. Epilepsy Res, 59, 13-24. BENNETT, D. L., CLARK, A. J., HUANG, J., WAXMAN, S. G. & DIB-HAJJ, S. D. 2019. The Role of Voltage-Gated Sodium Channels in Pain Signaling. Physiol Rev, 99, 1079-1151. BERNALDEZ, J., ROMAN-GONZALEZ, S. A., MARTINEZ, O., JIMENEZ, S., VIVAS, O., ARENAS, I., CORZO, G., ARREGUIN, R., GARCIA, D. E., POSSANI, L. D. & LICEA, A. 2013. A Conus regularis conotoxin with a novel eight-cysteine framework inhibits CaV2.2 channels and displays an anti-nociceptive activity. Mar Drugs, 11, 1188-202. BIGGS, J. S., WATKINS, M., PUILLANDRE, N., OWNBY, J. P., LOPEZ-VERA, E., CHRISTENSEN, S., MORENO, K. J., BERNALDEZ, J., LICEA-NAVARRO, A., CORNELI, P. S. & OLIVERA, B. M. 2010. Evolution of Conus peptide toxins: analysis of Conus californicus Reeve, 1844. Mol Phylogenet Evol, 56, 1-12. BINGHAM, J. P., ANDREWS, E. A., KIYABU, S. M. & CABALTEJA, C. C. 2012. Drugs from slugs. Part II--conopeptide bioengineering. Chem Biol Interact, 200, 92-113. BOHLEN, C. J., CHESLER, A. T., SHARIF-NAEINI, R., MEDZIHRADSZKY, K. F., ZHOU, S., KING, D., SANCHEZ, E. E., BURLINGAME, A. L., BASBAUM, A. I. & JULIUS, D. 2011. A heteromeric Texas coral snake toxin targets acid-sensing ion channels to produce pain. Nature, 479, 410-4. BOURINET, E. & ZAMPONI, G. W. 2017. Block of voltage-gated calcium channels by peptide toxins. Neuropharmacology, 127, 109-115. BRUST, A., PALANT, E., CROKER, D. E., COLLESS, B., DRINKWATER, R., PATTERSON, B., SCHROEDER, C. I., WILSON, D., NIELSEN, C. K., SMITH, M. T., ALEWOOD, D., ALEWOOD, P. F. & LEWIS, R. J. 2009. -Conopeptide pharmacophore development: toward a novel class of norepinephrine transporter inhibitor (Xen2174) for pain. J med chem, 52, 6991-7002. BUCZEK, O., WEI, D., BABON, J. J., YANG, X., FIEDLER, B., CHEN, P., YOSHIKAMI, D., OLIVERA, B. M., BULAJ, G. & NORTON, R. S. 2007. Structure and sodium channel activity of an excitatory I1-superfamily conotoxin. Biochemistry, 46, 9929-40. CAI, X. J., WANG, L. & HU, C. M. 2018. Effects of GABAB receptor activation on spatial cognitive function and hippocampal neurones in rat models of type 2 diabetes mellitus. Biosci Rep, 38. CALDERON-CELIS, F., CID-BARRIO, L., ENCINAR, J. R., SANZ-MEDEL, A. & CALVETE, J. J. 2017. Absolute venomics: Absolute quantification of intact venom proteins through elemental mass spectrometry. J Proteomics, 164, 33-42. CALVETE, J. J. 2017. Venomics: integrative venom proteomics and beyond. Biochem J, 474, 611-634. CAMPOS-LIRA, E., CARRILLO, E., AGUILAR, M. B., GAJEWIAK, J., GÓMEZ-LAGUNAS, F. & LÓPEZ-VERA, E. 2017. Conorfamide-Sr3, a structurally novel specific inhibitor of the Shaker K+ channel. Toxicon, 138, 53-58. CARSTENS, B. B., SWEDBERG, J., BERECKI, G., ADAMS, D. J., CRAIK, D. J. & CLARK, R. J. 2016. Effects of linker sequence modifications on the structure, stability, and biological activity of a cyclic α-conotoxin. Biopolymers, 106, 864-875. CATTERALL, W. A. & SWANSON, T. M. 2015. Structural Basis for Pharmacology of Voltage-Gated Sodium and Calcium Channels. Mol Pharmacol, 88, 141-50. CHEN, F., HUANG, W., JIANG, T. & YU, R. 2018. Determination of the μ-Conotoxin PIIIA Specificity Against Voltage-Gated Sodium Channels from Binding Energy Calculations. Marine drugs, 16, 153.

CHEN, J. S., FAN, C. X., HU, K. P., WEI, K. H. & ZHONG, M. N. 1999. Studies on conotoxins of Conus betulinus. J Nat Toxins, 8, 341-9. CHEN, L., DURR, K. L. & GOUAUX, E. 2014a. X-ray structures of AMPA receptor-cone snail toxin complexes illuminate activation mechanism. Science, 345, 1021-6. CHEN, P., GARRETT, J. E., WATKINS, M. & OLIVERA, B. M. 2008. Purification and characterization of a novel excitatory peptide from Conus distans venom that defines a novel gene superfamily of conotoxins. Toxicon, 52, 139-45. CHEN, S., GOPALAKRISHNAN, R., SCHAER, T., MARGER, F., HOVIUS, R., BERTRAND, D., POJER, F. & HEINIS, C. 2014b. Dithiol amino acids can structurally shape and enhance the ligand- binding properties of polypeptides. Nat. Chem., 6, 1009-1016. CHEN, X., PAUKERT, M., KADURIN, I., PUSCH, M. & GRUNDER, S. 2006. Strong modulation by RFamide neuropeptides of the ASIC1b/3 heteromer in competition with extracellular calcium. Neuropharmacology, 50, 964-74. CHEN, Z., BLANDL, T., PROROK, M., WARDER, S. E., LI, L., ZHU, Y., PEDERSEN, L. G., NI, F. & CASTELLINO, F. J. 1998. Conformational changes in conantokin-G induced upon binding of calcium and magnesium as revealed by NMR structural analysis. J Biol Chem, 273, 16248-58. CHENG, X., HONG, H., ZHOU, Z. & WU, Z. 2018. Enzymatic On-Resin Peptide Cleavage and in Situ Cyclization One-Pot Strategy for the Synthesis of Cyclopeptide and Cyclotide. J Org Chem, 83, 14078-14083. CHHABRA, S., BELGI, A., BARTELS, P., VAN LIEROP, B. J., ROBINSON, S. D., KOMPELLA, S. N., HUNG, A., CALLAGHAN, B. P., ADAMS, D. J., ROBINSON, A. J. & NORTON, R. S. 2014. Dicarba analogues of α-conotoxin RgIA. Structure, stability, and activity at potential pain targets. J Med Chem, 57, 9933-44. CHI, S. W., KIM, D. H., OLIVERA, B. M., MCINTOSH, J. M. & HAN, K. H. 2004. Solution conformation of α-conotoxin GIC, a novel potent antagonist of α3β2 nicotinic acetylcholine receptors. Biochem J, 380, 347-52. CHI, S. W., KIM, D. H., OLIVERA, B. M., MCINTOSH, J. M. & HAN, K. H. 2006. NMR structure determination of α-conotoxin BuIA, a novel neuronal nicotinic acetylcholine receptor antagonist with an unusual 4/4 disulfide scaffold. Biochem Biophys Res Commun, 349, 1228-34. CHRISTENSEN, S. B., BANDYOPADHYAY, P. K., OLIVERA, B. M. & MCINTOSH, J. M. 2015. αS- Conotoxin GVIIIB Potently and Selectively Blocks α9α10 Nicotinic Acetylcholine Receptors. Biochem Pharmacol, 96, 349-56. CHRISTENSEN, S. B., HONE, A. J., ROUX, I., KNIAZEFF, J., PIN, J. P., UPERT, G., SERVENT, D., GLOWATZKI, E. & MCINTOSH, J. M. 2017. RgIA4 Potently Blocks Mouse α9α10 nAChRs and Provides Long Lasting Protection against Oxaliplatin-Induced Cold Allodynia. Front Cell Neurosci, 11, 219. CLARK, R. J. & CRAIK, D. J. 2010. Native chemical ligation applied to the synthesis and bioengineering of circular peptides and proteins. Biopolymers, 94, 414-22. CORPUZ, G. P., JACOBSEN, R. B., JIMENEZ, E. C., WATKINS, M., WALKER, C., COLLEDGE, C., GARRETT, J. E., MCDOUGAL, O., LI, W., GRAY, W. R., HILLYARD, D. R., RIVIER, J., MCINTOSH, J. M., CRUZ, L. J. & OLIVERA, B. M. 2005. Definition of the M-conotoxin superfamily: characterization of novel peptides from molluscivorous Conus venoms. Biochemistry, 44, 8176-86. CUNY, H., KOMPELLA, S. N., TAE, H. S., YU, R. & ADAMS, D. J. 2016. Key Structural Determinants in the Agonist Binding Loops of Human β2 and β4 Nicotinic Acetylcholine Receptor Subunits Contribute to α3β4 Subtype Selectivity of α-Conotoxins. J Biol Chem, 291, 23779-23792. DALY, N. L., CALLAGHAN, B., CLARK, R. J., NEVIN, S. T., ADAMS, D. J. & CRAIK, D. J. 2011. Structure and activity of α-conotoxin PeIA at nicotinic acetylcholine receptor subtypes and GABAB receptor-coupled N-type calcium channels. J Biol Chem, 286, 10233-7.

DANIEL, J. T. & CLARK, R. J. 2017. Molecular Engineering of Conus Peptides as Therapeutic Leads. Adv Exp Med Biol, 1030, 229-254. DAO, F. Y., YANG, H., SU, Z. D., YANG, W., WU, Y., HUI, D., CHEN, W., TANG, H. & LIN, H. 2017. Recent Advances in Conotoxin Classification by Using Machine Learning Methods. Molecules, 22. DAWSON, P. E., MUIR, T. W., CLARK-LEWIS, I. & KENT, S. B. 1994. Synthesis of proteins by native chemical ligation. Science, 266, 776-9. DEL RIO-SANCHO, S., CROS, C., COUTAZ, B., CUENDET, M. & KALIA, Y. N. 2017. Cutaneous iontophoresis of μ-conotoxin CnIIIC-A potent NaV1.4 antagonist with analgesic, anaesthetic and myorelaxant properties. Int J Pharm, 518, 59-65. DEUIS, J. R., MUELLER, A., ISRAEL, M. R. & VETTER, I. 2017. The pharmacology of voltage-gated sodium channel activators. Neuropharmacology, 127, 87-108. DINELEY, K. T., PANDYA, A. A. & YAKEL, J. L. 2015. Nicotinic ACh receptors as therapeutic targets in CNS disorders. Trends Pharmacol Sci, 36, 96-108. DING, H., DENG, E. Z., YUAN, L. F., LIU, L., LIN, H., CHEN, W. & CHOU, K. C. 2014. iCTX-type: a sequence-based predictor for identifying the types of conotoxins in targeting ion channels. Biomed Res Int, 2014, 286419. DIOCHOT, S., BARON, A., RASH, L. D., DEVAL, E., ESCOUBAS, P., SCARZELLO, S., SALINAS, M. & LAZDUNSKI, M. 2004. A new sea anemone peptide, APETx2, inhibits ASIC3, a major acid‐ sensitive channel in sensory neurons. The EMBO journal, 23, 1516-1525. DIOCHOT, S., BARON, A., SALINAS, M., DOUGUET, D., SCARZELLO, S., DABERT-GAY, A.-S., DEBAYLE, D., FRIEND, V., ALLOUI, A. & LAZDUNSKI, M. 2012. Black mamba venom peptides target acid-sensing ion channels to abolish pain. Nature, 490, 552. DOBSON, C. M. 2003. Protein folding and misfolding. Nature (London, United Kingdom), 426, 884-890. DONEVAN, S. D. & MCCABE, R. T. 2000. Conantokin G is an NR2B-selective competitive antagonist of N-methyl-D-aspartate receptors. Mol Pharmacol, 58, 614-23. DRANE, S. B., ROBINSON, S. D., MACRAILD, C. A., CHHABRA, S., CHITTOOR, B., MORALES, R. A., LEUNG, E. W., BELGI, A., ESPINO, S. S., OLIVERA, B. M., ROBINSON, A. J., CHALMERS, D. K. & NORTON, R. S. 2017. Structure and activity of contryphan-Vc2: Importance of the d-amino acid residue. Toxicon, 129, 113-122. DU, W. H., HAN, Y. H., HUANG, F. J., LI, J., CHI, C. W. & FANG, W. H. 2007. Solution structure of an M-1 conotoxin with a novel disulfide linkage. The FEBS journal, 274, 2596-602. DUREK, T. & CRAIK, D. J. 2015. Therapeutic conotoxins: a US patent literature survey. Expert Opin Ther Pat, 25, 1159-73. DUTERTRE, S., JIN, A. H., KAAS, Q., JONES, A., ALEWOOD, P. F. & LEWIS, R. J. 2013. Deep venomics reveals the mechanism for expanded peptide diversity in cone snail venom. Mol Cell Proteomics, 12, 312-29. DUTERTRE, S., JIN, A. H., VETTER, I., HAMILTON, B., SUNAGAR, K., LAVERGNE, V., DUTERTRE, V., FRY, B. G., ANTUNES, A., VENTER, D. J., ALEWOOD, P. F. & LEWIS, R. J. 2014. Evolution of separate predation- and defence-evoked venoms in carnivorous cone snails. Nat Commun, 5, 3521. DUTERTRE, S. & LEWIS, R. J. 2012. Cone Snail Biology, Bioprospecting and Conservation. In: GOTSIRIDZE-COLUMBUS, N. (ed.) Snails: Biology, Ecology and Conservation. New York: Nova Science Publishers, Inc. DUTT, M., DUTERTRE, S., JIN, A. H., LAVERGNE, V., ALEWOOD, P. F. & LEWIS, R. J. 2019. Venomics Reveals Venom Complexity of the Piscivorous Cone Snail, Conus tulipa. Mar Drugs, 17. ELLIGER, C. A., RICHMOND, T. A., LEBARIC, Z. N., PIERCE, N. T., SWEEDLER, J. V. & GILLY, W. F. 2011. Diversity of conotoxin types from Conus californicus reflects a diversity of prey types and a novel evolutionary history. Toxicon, 57, 311-22.

ENGLAND, L. J., IMPERIAL, J., JACOBSEN, R., CRAIG, A. G., GULYAS, J., AKHTAR, M., RIVIER, J., JULIUS, D. & OLIVERA, B. M. 1998. Inactivation of a serotonin-gated ion channel by a polypeptide toxin from marine snails. Science, 281, 575-8. ESCOUBAS, P., DE WEILLE, J. R., LECOQ, A., DIOCHOT, S., WALDMANN, R., CHAMPIGNY, G., MOINIER, D., MÉNEZ, A. & LAZDUNSKI, M. 2000. Isolation of a tarantula toxin specific for a class of proton-gated Na+ channels. J Biol Chem, 275, 25116-25121. ESCOUBAS, P. & KING, G. F. 2009. Venomics as a drug discovery platform. Expert Rev Proteomics, 6, 221-4. FAN, Y. X., SONG, J., SHEN, H. B. & KONG, X. 2011. PredCSF: an integrated feature-based approach for predicting conotoxin superfamily. Protein Pept Lett, 18, 261-7. FIELDS, G. B. & NOBLE, R. L. 1990. Solid-phase peptide synthesis utilizing 9-fluorenylmethoxycarbonyl amino acids. Int. J. Pept. Protein Res., 35, 161-214. FIGUEROA-MONTIEL, A., BERNALDEZ, J., JIMENEZ, S., UEBERHIDE, B., GONZALEZ, L. J. & LICEA-NAVARRO, A. 2018. Antimycobacterial Activity: A New Pharmacological Target for Conotoxins Found in the First Reported Conotoxin from Conasprella ximenes. Toxins (Basel), 10. FRANCO, A., DOVELL, S., MOLLER, C., GRANDAL, M., CLARK, E. & MARI, F. 2018. Structural plasticity of mini-M conotoxins - expression of all mini-M subtypes by Conus regius. Febs j, 285, 887-902. GAJEWIAK, J., AZAM, L., IMPERIAL, J., WALEWSKA, A., GREEN, B. R., BANDYOPADHYAY, P. K., RAGHURAMAN, S., UEBERHEIDE, B., BERN, M., ZHOU, H. M., MINASSIAN, N. A., HAGAN, R. H., FLINSPACH, M., LIU, Y., BULAJ, G., WICKENDEN, A. D., OLIVERA, B. M., YOSHIKAMI, D. & ZHANG, M. M. 2014. A disulfide tether stabilizes the block of sodium channels by the conotoxin μO section §-GVIIJ. PNAS, 111, 2758-63. GAO, B., PENG, C., YANG, J., YI, Y., ZHANG, J. & SHI, Q. 2017. Cone Snails: A Big Store of Conotoxins for Novel Drug Discovery. Toxins (Basel), 9. GENOMIC RESOURCES DEVELOPMENT, C., ARTHOFER, W., BANBURY, B. L., CARNEIRO, M., CICCONARDI, F., DUDA, T. F., HARRIS, R. B., KANG, D. S., LEACHE, A. D., NOLTE, V., NOURISSON, C., PALMIERI, N., SCHLICK-STEINER, B. C., SCHLOTTERER, C., SEQUEIRA, F., SIM, C., STEINER, F. M., VALLINOTO, M. & WEESE, D. A. 2015. Genomic resources notes accepted 1 August 2014-30 September 2014. Mol Ecol Resour, 15, 228-9. GETHING, M. J. & SAMBROOK, J. 1992. Protein folding in the cell. Nature, 355, 33-45. GIRIBALDI, J. & DUTERTRE, S. 2018. α-Conotoxins to explore the molecular, physiological and pathophysiological functions of neuronal nicotinic acetylcholine receptors. Neurosci Lett, 679, 24- 34. GIRIBALDI, J., LACONDE, G., ENJALBAL, C., DUTERTRE, S., WILSON, D., DALY NORELLE, L., NICKE, A., EL HAMDAOUI, Y., FAUCHERRE, A. & MOHA OU MAATI, H. 2018. Synthesis, Structure and Biological Activity of CIA and CIB, Two α-Conotoxins from the Predation-Evoked Venom of Conus catus. Toxins, 10. GONGORA-BENITEZ, M., TULLA-PUCHE, J. & ALBERICIO, F. 2014. Multifaceted roles of disulfide bonds. Peptides as therapeutics. Chem Rev, 114, 901-26. GORI, A., WANG, C.-I. A., HARVEY, P. J., ROSENGREN, K. J., BHOLA, R. F., GELMI, M. L., LONGHI, R., CHRISTIE, M. J., LEWIS, R. J., ALEWOOD, P. F. & BRUST, A. 2015. Stabilization of the cysteine-rich conotoxin MrIA by using a 1,2,3-triazole as a disulfide bond mimetic. Angew. Chem., Int. Ed., 54, 1361-1364. GRAU, V., RICHTER, K., HONE, A. J. & MCINTOSH, J. M. 2018. Conopeptides [V11L;V16D]ArIB and RgIA4: Powerful Tools for the Identification of Novel Nicotinic Acetylcholine Receptors in Monocytes. Front Pharmacol, 9, 1499. GREEN, B. R., BULAJ, G. & NORTON, R. S. 2014. Structure and function of μ-conotoxins, peptide-based sodium channel blockers with analgesic activity. Future Med Chem, 6, 1677-98.

GUO, Y., SUN, D.-M., WANG, F.-L., HE, Y., LIU, L. & TIAN, C.-L. 2015. Diaminodiacid bridges to improve folding and tune the bioactivity of disulfide-rich peptides. Angew. Chem., Int. Ed., 54, 14276-14281. HANSEN, S. B., SULZENBACHER, G., HUXFORD, T., MARCHOT, P., TAYLOR, P. & BOURNE, Y. 2005. Structures of Aplysia AChBP complexes with nicotinic agonists and antagonists reveal distinctive binding interfaces and conformations. Embo j, 24, 3635-46. HANSSON, K., FURIE, B., FURIE, B. C. & STENFLO, J. 2004. Isolation and characterization of three novel Gla-containing Conus marmoreus venom peptides, one with a novel cysteine pattern. Biochem Biophys Res Commun, 319, 1081-7. HARTL, F. U. 1996. Molecular chaperones in cellular protein folding. Nature (London), 381, 571-580. HARTMUT, C., RILEI, Y., HAN-SHEN, T., N, K. S. & J, A. D. 2018. α-Conotoxins active at α3- containing nicotinic acetylcholine receptors and their molecular determinants for selective inhibition. Br J Pharmacol, 175, 1855-1868. HARVEY, A. L. 2014. Toxins and drug discovery. Toxicon, 92, 193-200. HEIMER, P., TIETZE, A. A., BAEUML, C. A., RESEMANN, A., MAYER, F. J., SUCKAU, D., OHLENSCHLAEGER, O., TIETZE, D. & IMHOF, D. 2018. Conformational μ-conotoxin PIIIA isomers revisited: Impact of cysteine pairing on disulfide-bond assignment and structure elucidation. Anal. Chem., 90, 3321-3327. HEIMER, P., TIETZE, A. A., BOEHM, M., GIERNOTH, R., KUCHENBUCH, A., STARK, A., LEIPOLD, E., HEINEMANN, S. H., KANDT, C. & IMHOF, D. 2014. Application of Room- Temperature Aprotic and Protic Ionic Liquids for Oxidative Folding of Cysteine-Rich Peptides. ChemBioChem, 15, 2754-2765. HEMU, X. & TAM, J. P. 2017. Macrocyclic Antimicrobial Peptides Engineered from ω-Conotoxin. Curr. Pharm. Des., 23, 2131-2138. HIMAYA, S. W., JIN, A. H., DUTERTRE, S., GIACOMOTTO, J., MOHIALDEEN, H., VETTER, I., ALEWOOD, P. F. & LEWIS, R. J. 2015. Comparative Venomics Reveals the Complex Prey Capture Strategy of the Piscivorous Cone Snail Conus catus. J Proteome Res, 14, 4372-81. HIMAYA, S. W. A. & LEWIS, R. J. 2018. Venomics-Accelerated Cone Snail Venom Peptide Discovery. Int J Mol Sci, 19. HIMAYA, S. W. A., MARI, F. & LEWIS, R. J. 2018. Accelerated proteomic visualization of individual predatory venoms of Conus purpurascens reveals separately evolved predation-evoked venom cabals. Sci Rep, 8, 330. HOCKING, H. G., GERWIG, G. J., DUTERTRE, S., VIOLETTE, A., FAVREAU, P., STOCKLIN, R., KAMERLING, J. P. & BOELENS, R. 2013. Structure of the O-glycosylated conopeptide CcTx from Conus consors venom. Chemistry, 19, 870-9. HONE, A. J., TALLEY, T. T., BOBANGO, J., HUIDOBRO MELO, C., HARARAH, F., GAJEWIAK, J., CHRISTENSEN, S., HARVEY, P. J., CRAIK, D. J. & MCINTOSH, J. M. 2018. Molecular determinants of α-conotoxin potency for inhibition of human and rat α6β4 nicotinic acetylcholine receptors. J Biol Chem, 293, 17838-17852. HOPPING, G., LEWIS, R. J. & ALEWOOD, P. F. 2009. Rapid Access to ω-Conotoxin Chimeras using Native Chemical Ligation. Aust. J. Chem., 62, 1333-1338. HU, H., BANDYOPADHYAY, P. K., OLIVERA, B. M. & YANDELL, M. 2011. Characterization of the Conus bullatus genome and its venom-duct transcriptome. BMC Genomics, 12, 60. HU, H., BANDYOPADHYAY, P. K., OLIVERA, B. M. & YANDELL, M. 2012. Elucidation of the molecular envenomation strategy of the cone snail Conus geographus through transcriptome sequencing of its venom duct. BMC Genomics, 13, 284. INSERRA, M. C., KOMPELLA, S. N., VETTER, I., BRUST, A., DALY, N. L., CUNY, H., CRAIK, D. J., ALEWOOD, P. F., ADAMS, D. J. & LEWIS, R. J. 2013. Isolation and characterization of α- conotoxin LsIA with potent activity at nicotinic acetylcholine receptors. Biochemical pharmacology, 86, 791-799.

ISRAEL, M. R., TAY, B., DEUIS, J. R. & VETTER, I. 2017. Sodium Channels and Venom Peptide Pharmacology. Adv Pharmacol, 79, 67-116. JENSEN, K. J., TOFTENG SHELTON, P., PEDERSEN, S. L., ARMISHAW, C. J., STRØMGAARD, K., SHABANPOOR, F., HOSSAIN, M. A., LIN, F., WADE, J. D., AKCAN, M., CRAIK, D. J., BAUMANN, L., STEINHAGEN, M., BECK-SICKINGER, A. G., ROODBEEN, R., HANSEN, P. R., MUNK, J. K., ROSI, F., TRIOLA, G., HØJLYS-LARSEN, K. B., BLIXT, O., CLO, E., MALIK, L. & VANIER, G. S. 2013. Peptide synthesis and applications, New York City, Humana Press. JIA, X., KWON, S., WANG, C.-I. A., HUANG, Y.-H., CHAN, L. Y., TAN, C. C., ROSENGREN, K. J., MULVENNA, J. P., SCHROEDER, C. I. & CRAIK, D. J. 2014. Semienzymatic cyclization of disulfide-rich peptides using sortase A. J Biol Chem, 289, 6627-6638. JIANG, H., WANG, C. Z., XU, C. Q., FAN, C. X., DAI, X. D., CHEN, J. S. & CHI, C. W. 2006. A novel M-superfamily conotoxin with a unique motif from Conus vexillum. Peptides, 27, 682-9. JIMENEZ, E. C., SHETTY, R. P., LIRAZAN, M., RIVIER, J., WALKER, C., ABOGADIE, F. C., YOSHIKAMI, D., CRUZ, L. J. & OLIVERA, B. M. 2003. Novel excitatory Conus peptides define a new conotoxin superfamily. J Neurochem, 85, 610-21. JIN, A. H., CRISTOFORI-ARMSTRONG, B., RASH, L. D., ROMAN-GONZALEZ, S. A., ESPINOSA, R. A., LEWIS, R. J., ALEWOOD, P. F. & VETTER, I. 2019a. Novel conorfamides from Conus austini venom modulate both nicotinic acetylcholine receptors and acid-sensing ion channels. Biochem Pharmacol, 164, 342-348. JIN, A. H., DEKAN, Z., SMOUT, M. J., WILSON, D., DUTERTRE, S., VETTER, I., LEWIS, R. J., LOUKAS, A., DALY, N. L. & ALEWOOD, P. F. 2017. Conotoxin -MiXXVIIA from the Superfamily G2 Employs a Novel Cysteine Framework that Mimics Granulin and Displays Anti- Apoptotic Activity. Angew Chem Int Ed Engl, 56, 14973-14976. JIN, A. H., DUTERTRE, S., DUTT, M., LAVERGNE, V., JONES, A., LEWIS, R. J. & ALEWOOD, P. F. 2019b. Transcriptomic-Proteomic Correlation in the Predation-Evoked Venom of the Cone Snail, Conus imperialis. Mar Drugs, 17. JIN, A. H., DUTERTRE, S., KAAS, Q., LAVERGNE, V., KUBALA, P., LEWIS, R. J. & ALEWOOD, P. F. 2013. Transcriptomic messiness in the venom duct of Conus miles contributes to conotoxin diversity. Mol Cell Proteomics, 12, 3824-33. JIN, A. H., ISRAEL, M. R., INSERRA, M. C., SMITH, J. J., LEWIS, R. J., ALEWOOD, P. F., VETTER, I. & DUTERTRE, S. 2015a. -Conotoxin SuVIA suggests an evolutionary link between ancestral predator defence and the origin of fish-hunting behaviour in carnivorous cone snails. Proc Biol Sci, 282. JIN, A. H., VETTER, I., DUTERTRE, S., ABRAHAM, N., EMIDIO, N. B., INSERRA, M., MURALI, S. S., CHRISTIE, M. J., ALEWOOD, P. F. & LEWIS, R. J. 2014. MrIC, a novel α-conotoxin agonist in the presence of PNU at endogenous α7 nicotinic acetylcholine receptors. Biochemistry, 53, 1-3. JIN, A. H., VETTER, I., HIMAYA, S. W., ALEWOOD, P. F., LEWIS, R. J. & DUTERTRE, S. 2015b. Transcriptome and proteome of Conus planorbis identify the nicotinic receptors as primary target for the defensive venom. Proteomics, 15, 4030-40. KAAS, Q., WESTERMANN, J. C. & CRAIK, D. J. 2010. Conopeptide characterization and classifications: an analysis using ConoServer. Toxicon, 55, 1491-509. KAAS, Q., WESTERMANN, J. C., HALAI, R., WANG, C. K. & CRAIK, D. J. 2008. ConoServer, a database for conopeptide sequences and structures. Bioinformatics, 24, 445-6. KAAS, Q., YU, R., JIN, A. H., DUTERTRE, S. & CRAIK, D. J. 2012. ConoServer: updated content, knowledge, and discovery tools in the conopeptide database. Nucleic Acids Res, 40, D325-30. KANCHERLA, A. K., MEESALA, S., JORWAL, P., PALANISAMY, R., SIKDAR, S. K. & SARMA, S. P. 2015. A Disulfide Stabilized β-Sandwich Defines the Structure of a New Cysteine Framework M-Superfamily Conotoxin. ACS Chem Biol, 10, 1847-60. KATZ, B. A. & KOSSIAKOFF, A. 1986. The crystallographically determined structures of atypical strained disulfides engineered into subtilisin. J Biol Chem, 261, 15480-5. 92

KING, G. F. 2011. Venoms as a platform for human drugs: translating toxins into therapeutics. Expert Opin Biol Ther, 11, 1469-84. KING, G. F., GENTZ, M. C., ESCOUBAS, P. & NICHOLSON, G. M. 2008. A rational nomenclature for naming peptide toxins from spiders and other venomous animals. Toxicon, 52, 264-76. KLINT, J. K., SENFF, S., SAEZ, N. J., SESHADRI, R., LAU, H. Y., BENDE, N. S., UNDHEIM, E. A. B., RASH, L. D., MOBLI, M. & KING, G. F. 2013. Production of recombinant disulfide-rich venom peptides for structural and functional analysis via expression in the periplasm of E. coli. PLoS One, 8, e63865. KOLOSOV, A., AURINI, L., WILLIAMS, E. D., COOKE, I. & GOODCHILD, C. S. 2011. Intravenous injection of leconotide, an -conotoxin: synergistic antihyperalgesic effects with morphine in a rat model of bone cancer pain. Pain Med, 12, 923-41. KOMPELLA, S. N., HUNG, A., CLARK, R. J., MARI, F. & ADAMS, D. J. 2015. Alanine scan of α- conotoxin RegIIA reveals a selective α3β4 nicotinic acetylcholine receptor antagonist. J Biol Chem, 290, 1039-48. KONDASINGHE, T. D., SARAHA, H. Y., JACKOWSKI, S. T. & STOCKDILL, J. L. 2019. Raising the bar on-bead: Efficient on-resin synthesis of α-conotoxin LvIA. Tetrahedron Lett., 60, 23-28. KOUA, D., BRAUER, A., LAHT, S., KAPLINSKI, L., FAVREAU, P., REMM, M., LISACEK, F. & STOCKLIN, R. 2012. ConoDictor: a tool for prediction of conopeptide superfamilies. Nucleic Acids Res, 40, W238-41. KWON, S., BOSMANS, F., KAAS, Q., CHENEVAL, O., CONIBEAR, A. C., ROSENGREN, K. J., WANG, C. K., SCHROEDER, C. I. & CRAIK, D. J. 2016. Efficient enzymatic cyclization of an inhibitory cystine knot-containing peptide. Biotechnol. Bioeng., 113, 2202-2212. KWONG, K. & CARR, M. J. 2015. Voltage-gated sodium channels. Curr Opin Pharmacol, 22, 131-9. LAMTHANH, H., JEGOU-MATHERON, C., SERVENT, D., MENEZ, A. & LANCELIN, J. M. 1999. Minimal conformation of the α-conotoxin ImI for the α7 neuronal nicotinic acetylcholine receptor recognition: correlated CD, NMR and binding studies. FEBS Lett, 454, 293-8. LAVERGNE, V., DUTERTRE, S., JIN, A. H., LEWIS, R. J., TAFT, R. J. & ALEWOOD, P. F. 2013. Systematic interrogation of the Conus marmoreus venom duct transcriptome with ConoSorter reveals 158 novel conotoxins and 13 new gene superfamilies. BMC Genomics, 14, 708. LAVERGNE, V., HARLIWONG, I., JONES, A., MILLER, D., TAFT, R. J. & ALEWOOD, P. F. 2015. Optimized deep-targeted proteotranscriptomic profiling reveals unexplored Conus toxin diversity and novel cysteine frameworks. PNAS, 112, E3782-91. LEBBE, E. K. & TYTGAT, J. 2016. In the picture: disulfide-poor conopeptides, a class of pharmacologically interesting compounds. J Venom Anim Toxins Incl Trop Dis, 22, 30. LEE, M. S. 2014. Recent progress in the discovery and development of N-type calcium channel modulators for the treatment of pain. Prog Med Chem, 53, 147-86. LEIPOLD, E., ULLRICH, F., THIELE, M., TIETZE, A. A., TERLAU, H., IMHOF, D. & HEINEMANN, S. H. 2017. Subtype-specific block of voltage-gated K+ channels by μ-conopeptides. Biochem Biophys Res Commun, 482, 1135-1140. LEWIS, R. J., DUTERTRE, S., VETTER, I. & CHRISTIE, M. J. 2012. Conus venom peptide pharmacology. Pharmacol Rev, 64, 259-98. LEWIS, R. J., NIELSEN, K. J., CRAIK, D. J., LOUGHNAN, M. L., ADAMS, D. A., SHARPE, I. A., LUCHIAN, T., ADAMS, D. J., BOND, T., THOMAS, L., JONES, A., MATHESON, J. L., DRINKWATER, R., ANDREWS, P. R. & ALEWOOD, P. F. 2000. Novel  -conotoxins from Conus catus discriminate among neuronal calcium channel subtypes. J Biol Chem, 275, 35335-44. LI, Q., BARGHI, N., LU, A., FEDOSOV, A. E., BANDYOPADHYAY, P. K., LLUISMA, A. O., CONCEPCION, G. P., YANDELL, M., OLIVERA, B. M. & SAFAVI-HEMAMI, H. 2017. Divergence of the Venom Exogene Repertoire in Two Sister Species of Turriconus. Genome Biol Evol, 9, 2211-2225.

LI, Q., WATKINS, M., ROBINSON, S. D., SAFAVI-HEMAMI, H. & YANDELL, M. 2018. Discovery of Novel Conotoxin Candidates Using Machine Learning. Toxins (Basel), 10. LIN, B., XU, M., ZHU, X., WU, Y., LIU, X., ZHANGSUN, D., HU, Y., XIANG, S. H., KASHEVEROV, I. E., TSETLIN, V. I., WANG, X. & LUO, S. 2016. From crystal structure of α-conotoxin GIC in complex with Ac-AChBP to molecular determinants of its high selectivity for α3β2 nAChR. Sci Rep, 6, 22349. LIRAZAN, M. B., HOOPER, D., CORPUZ, G. P., RAMILO, C. A., BANDYOPADHYAY, P., CRUZ, L. J. & OLIVERA, B. M. 2000. The spasmodic peptide defines a new conotoxin superfamily. Biochemistry, 39, 1583-8. LIU, Z., BARTELS, P., SADEGHI, M., DU, T., DAI, Q., ZHU, C., YU, S., WANG, S., DONG, M., SUN, T., GUO, J., PENG, S., JIANG, L., ADAMS, D. J. & DAI, Q. 2018. A novel α-conopeptide Eu1.6 inhibits N-type CaV2.2 calcium channels and exhibits potent analgesic activity. Sci Rep, 8, 1004. LLUISMA, A. O., MILASH, B. A., MOORE, B., OLIVERA, B. M. & BANDYOPADHYAY, P. K. 2012. Novel venom peptides from the cone snail Conus pulicarius discovered through next-generation sequencing of its venom duct transcriptome. Mar Genomics, 5, 43-51. LOUGHNAN, M., NICKE, A., JONES, A., SCHROEDER, C. I., NEVIN, S. T., ADAMS, D. J., ALEWOOD, P. F. & LEWIS, R. J. 2006. Identification of a novel class of nicotinic receptor antagonists: dimeric conotoxins VxXIIA, VxXIIB, and VxXIIC from Conus vexillum. J Biol Chem, 281, 24745-55. LUO, S., CHRISTENSEN, S., ZHANGSUN, D., WU, Y., HU, Y., ZHU, X., CHHABRA, S., NORTON, R. S. & MCINTOSH, J. M. 2013. A novel inhibitor of α9α10 nicotinic acetylcholine receptors from Conus vexillum delineates a new conotoxin superfamily. PLoS One, 8, e54648. LUO, S., ZHANGSUN, D., HARVEY, P. J., KAAS, Q., WU, Y., ZHU, X., HU, Y., LI, X., TSETLIN, V. I., CHRISTENSEN, S., ROMERO, H. K., MCINTYRE, M., DOWELL, C., BAXTER, J. C., ELMSLIE, K. S., CRAIK, D. J. & MCINTOSH, J. M. 2015. Cloning, synthesis, and characterization of αO-conotoxin GeXIVA, a potent α9α10 nicotinic acetylcholine receptor antagonist. PNAS, 112, E4026-35. LUO, S., ZHANGSUN, D., SCHROEDER, C. I., ZHU, X., HU, Y., WU, Y., WELTZIN, M. M., EBERHARD, S., KAAS, Q., CRAIK, D. J., MCINTOSH, J. M. & WHITEAKER, P. 2014. A novel α4/7-conotoxin LvIA from Conus lividus that selectively blocks α3β2 vs. α6/α3β2β3 nicotinic acetylcholine receptors. Faseb j, 28, 1842-53. MAILLO, M., AGUILAR, M. B., LOPEZ-VERA, E., CRAIG, A. G., BULAJ, G., OLIVERA, B. M. & HEIMER DE LA COTERA, E. P. 2002. Conorfamide, a Conus venom peptide belonging to the RFamide family of neuropeptides. Toxicon, 40, 401-7. MANSBACH, R. A., TRAVERS, T., MCMAHON, B. H., FAIR, J. M. & GNANAKARAN, S. 2019. Snails In Silico: A Review of Computational Studies on the Conopeptides. Mar Drugs, 17. MARDIS, E. R. 2017. DNA sequencing technologies: 2006-2016. Nat Protoc, 12, 213-218. MARUYAMA, K., NAGASAWA, H. & SUZUKI, A. 1999. 2,2'-Bispyridyl disulfide rapidly induces intramolecular disulfide bonds in peptides. Peptides, 20, 881-884. MCINTOSH, J. M., SANTOS, A. D. & OLIVERA, B. M. 1999. Conus peptides targeted to specific nicotinic acetylcholine receptor subtypes. Annu Rev Biochem, 68, 59-88. MENTING, J. G., GAJEWIAK, J., MACRAILD, C. A., CHOU, D. H., DISOTUAR, M. M., SMITH, N. A., MILLER, C., ERCHEGYI, J., RIVIER, J. E., OLIVERA, B. M., FORBES, B. E., SMITH, B. J., NORTON, R. S., SAFAVI-HEMAMI, H. & LAWRENCE, M. C. 2016. A minimized human insulin-receptor-binding motif revealed in a Conus geographus venom insulin. Nat Struct Mol Biol, 23, 916-920. MENTING, J. G., WHITTAKER, J., MARGETTS, M. B., WHITTAKER, L. J., KONG, G. K., SMITH, B. J., WATSON, C. J., ZAKOVA, L., KLETVIKOVA, E., JIRACEK, J., CHAN, S. J., STEINER, D. F., DODSON, G. G., BRZOZOWSKI, A. M., WEISS, M. A., WARD, C. W. & LAWRENCE,

M. C. 2013. How insulin engages its primary binding site on the insulin receptor. Nature, 493, 241- 5. MERRIFIELD, R. B. 1963. Solid phase peptide synthesis. I. The synthesis of a tetrapeptide. J. Am. Chem. Soc., 85, 2149-54. MILES, L. A., DY, C. Y., NIELSEN, J., BARNHAM, K. J., HINDS, M. G., OLIVERA, B. M., BULAJ, G. & NORTON, R. S. 2002. Structure of a novel P-superfamily spasmodic conotoxin reveals an inhibitory cystine knot motif. J Biol Chem, 277, 43033-40. MOBLI, M. & KING, G. F. 2010. NMR methods for determining disulfide-bond connectivities. Toxicon, 56, 849-54. MOLLER, C., DOVELL, S., MELAUN, C. & MARI, F. 2018. Definition of the R-superfamily of conotoxins: Structural convergence of helix-loop-helix peptidic scaffolds. Peptides, 107, 75-82. MOLLER, C. & MARI, F. 2011. 9.3 KDa components of the injected venom of Conus purpurascens define a new five-disulfide conotoxin framework. Biopolymers, 96, 158-65. MOLLER, C., RAHMANKHAH, S., LAUER-FIELDS, J., BUBIS, J., FIELDS, G. B. & MARI, F. 2005. A novel conotoxin framework with a helix-loop-helix (Cs α/α) fold. Biochemistry, 44, 15986-96. MONDAL, S., BHAVNA, R., MOHAN BABU, R. & RAMAKUMAR, S. 2006. Pseudo amino acid composition and multi-class support vector machines approach for conotoxin superfamily classification. J Theor Biol, 243, 252-60. MUELLER, A., STAROBOVA, H., INSERRA, M. C., JIN, A. H., DEUIS, J. R., DUTERTRE, S., LEWIS, R. J., ALEWOOD, P. F., DALY, N. L. & VETTER, I. 2015. α-Conotoxin MrIC is a biased agonist at α7 nicotinic acetylcholine receptors. Biochem Pharmacol, 94, 155-63. MUTTENTHALER, M., ALBERICIO, F. & DAWSON, P. E. 2015. Methods, setup and safe handling for anhydrous hydrogen fluoride cleavage in Boc solid-phase peptide synthesis. Nat. Protoc., 10, 1067- 1083. MUTTENTHALER, M., NEVIN, S. T., GRISHIN, A. A., NGO, S. T., CHOY, P. T., DALY, N. L., HU, S.-H., ARMISHAW, C. J., WANG, C.-I. A., LEWIS, R. J., MARTIN, J. L., NOAKES, P. G., CRAIK, D. J., ADAMS, D. J. & ALEWOOD, P. F. 2010. Solving the α-conotoxin folding problem: efficient selenium-directed on-resin generation of more potent and stable nicotinic acetylcholine receptor antagonists. J. Am. Chem. Soc., 132, 3514-3522. NIELSEN, C. K., LEWIS, R. J., ALEWOOD, D., DRINKWATER, R., PALANT, E., PATTERSON, M., YAKSH, T. L., MCCUMBER, D. & SMITH, M. T. 2005. Anti-allodynic efficacy of the chi- conopeptide, Xen2174, in rats with neuropathic pain. Pain, 118, 112-24. NIELSEN, L. D., FOGED, M. M., ALBERT, A., BERTELSEN, A. B., SOLTOFT, C. L., ROBINSON, S. D., PETERSEN, S. V., PURCELL, A. W., OLIVERA, B. M., NORTON, R. S., VASSKOG, T., SAFAVI-HEMAMI, H., TEILUM, K. & ELLGAARD, L. 2019. The three-dimensional structure of an H-superfamily conotoxin reveals a granulin fold arising from a common ICK cysteine framework. J Biol Chem, 294, 8745-8759. NORTON, R. S. 2010. -Conotoxins as leads in the development of new analgesics. Molecules, 15, 2825- 44. NORTON, R. S. & PALLAGHY, P. K. 1998. The cystine knot structure of ion channel toxins and related polypeptides. Toxicon, 36, 1573-83. OLDRATI, V., ARRELL, M., VIOLETTE, A., PERRET, F., SPRUNGLI, X., WOLFENDER, J. L. & STOCKLIN, R. 2016. Advances in venomics. Mol Biosyst, 12, 3530-3543. OLIVERA, B. M., RIVIER, J., CLARK, C., RAMILO, C. A., CORPUZ, G. P., ABOGADIE, F. C., MENA, E. E., WOODWARD, S. R., HILLYARD, D. R. & CRUZ, L. J. 1990. Diversity of Conus neuropeptides. Science, 249, 257-63. OROZ-PARRA, I., NAVARRO, M., CERVANTES-LUEVANO, K. E., ALVAREZ-DELGADO, C., SALVESEN, G., SANCHEZ-CAMPOS, L. N. & LICEA-NAVARRO, A. F. 2016. Apoptosis Activation in Human Lung Cancer Cell Lines by a Novel Synthetic Peptide Derived from Conus californicus Venom. Toxins (Basel), 8, 38. 95

PAPKE, R. L. 2014. Merging old and new perspectives on nicotinic acetylcholine receptors. Biochem Pharmacol, 89, 1-11. PEIGNEUR, S., PAOLINI-BERTRAND, M., GAERTNER, H., BIASS, D., VIOLETTE, A., STOCKLIN, R., FAVREAU, P., TYTGAT, J. & HARTLEY, O. 2014. δ-Conotoxins synthesized using an acid- cleavable solubility tag approach reveal key structural determinants for NaV subtype selectivity. J Biol Chem, 289, 35341-35350. PENG, C., LIU, L., SHAO, X., CHI, C. & WANG, C. 2008. Identification of a novel class of conotoxins defined as V-conotoxins with a unique cysteine pattern and signal peptide sequence. Peptides, 29, 985-91. PENG, C., YAO, G., GAO, B. M., FAN, C. X., BIAN, C., WANG, J., CAO, Y., WEN, B., ZHU, Y., RUAN, Z., ZHAO, X., YOU, X., BAI, J., LI, J., LIN, Z., ZOU, S., ZHANG, X., QIU, Y., CHEN, J., COON, S. L., YANG, J., CHEN, J. S. & SHI, Q. 2016. High-throughput identification of novel conotoxins from the Chinese tubular cone snail (Conus betulinus) by multi-transcriptome sequencing. Gigascience, 5, 17. PENNINGTON, M. W., CZERWINSKI, A. & NORTON, R. S. 2018. Peptide therapeutics from venom: Current status and potential. Bioorg Med Chem, 26, 2738-2758. PETREL, C., HOCKING, H. G., REYNAUD, M., UPERT, G., FAVREAU, P., BIASS, D., PAOLINI- BERTRAND, M., PEIGNEUR, S., TYTGAT, J., GILLES, N., HARTLEY, O., BOELENS, R., STOCKLIN, R. & SERVENT, D. 2013. Identification, structural and pharmacological characterization of -CnVA, a conopeptide that selectively interacts with somatostatin sst3 receptor. Biochem Pharmacol, 85, 1663-71. PHUONG, M. A., MAHARDIKA, G. N. & ALFARO, M. E. 2016. Dietary breadth is positively correlated with venom complexity in cone snails. BMC Genomics, 17, 401. PI, C., LIU, J., WANG, L., JIANG, X., LIU, Y., PENG, C., CHEN, S. & XU, A. 2007. Soluble expression, purification and functional identification of a disulfide-rich conotoxin derived from Conus litteratus. J Biotechnol, 128, 184-93. PI, C., LIU, Y., PENG, C., JIANG, X., LIU, J., XU, B., YU, X., YU, Y., JIANG, X., WANG, L., DONG, M., CHEN, S. & XU, A. L. 2006. Analysis of expressed sequence tags from the venom ducts of Conus striatus: focusing on the expression profile of conotoxins. Biochimie, 88, 131-40. POPE, J. E. & DEER, T. R. 2013. Ziconotide: a clinical update and pharmacologic review. Expert Opin Pharmacother, 14, 957-66. POSTIC, G., GRACY, J., PERIN, C., CHICHE, L. & GELLY, J. C. 2018. KNOTTIN: the database of inhibitor cystine knot scaffold after 10 years, toward a systematic structure modeling. Nucleic Acids Res, 46, D454-d458. PRASHANTH, J. R., BRUST, A., JIN, A. H., ALEWOOD, P. F., DUTERTRE, S. & LEWIS, R. J. 2014. Cone snail venomics: from novel biology to novel therapeutics. Future Med Chem, 6, 1659-1675. PRASHANTH, J. R., DUTERTRE, S., JIN, A. H., LAVERGNE, V., HAMILTON, B., CARDOSO, F. C., GRIFFIN, J., VENTER, D. J., ALEWOOD, P. F. & LEWIS, R. J. 2016. The role of defensive ecological interactions in the evolution of conotoxins. Mol Ecol, 25, 598-615. PRASHANTH, J. R. & LEWIS, R. J. 2015. An efficient transcriptome analysis pipeline to accelerate venom peptide discovery and characterisation. Toxicon, 107, 282-9. PRENTIS, P. J., PAVASOVIC, A. & NORTON, R. S. 2018. Sea Anemones: Quiet Achievers in the Field of Peptide Toxins. Toxins (Basel), 10. REIMERS, C., LEE, C. H., KALBACHER, H., TIAN, Y., HUNG, C. H., SCHMIDT, A., PROKOP, L., KAUFERSTEIN, S., MEBS, D., CHEN, C. C. & GRUNDER, S. 2017. Identification of a cono- RFamide from the venom of Conus textile that targets ASIC3 and enhances muscle pain. PNAS, 114, E3507-E3515. REMIGIO, E. A. & DUDA, T. F., JR. 2008. Evolution of ecological specialization and venom of a predatory marine gastropod. Mol Ecol, 17, 1156-62.

RHOADS, A. & AU, K. F. 2015. PacBio Sequencing and Its Applications. Genomics Proteomics Bioinformatics, 13, 278-89. ROBINSON, S. D., CHHABRA, S., BELGI, A., CHITTOOR, B., SAFAVI-HEMAMI, H., ROBINSON, A. J., PAPENFUSS, A. T., PURCELL, A. W. & NORTON, R. S. 2016. A Naturally Occurring Peptide with an Elementary Single Disulfide-Directed β-Hairpin Fold. Structure, 24, 293-9. ROBINSON, S. D., LI, Q., BANDYOPADHYAY, P. K., GAJEWIAK, J., YANDELL, M., PAPENFUSS, A. T., PURCELL, A. W., NORTON, R. S. & SAFAVI-HEMAMI, H. 2017a. Hormone-like peptides in the venoms of marine cone snails. Gen Comp Endocrinol, 244, 11-18. ROBINSON, S. D., LI, Q., LU, A., BANDYOPADHYAY, P. K., YANDELL, M., OLIVERA, B. M. & SAFAVI-HEMAMI, H. 2017b. The Venom Repertoire of Conus gloriamaris (Chemnitz, 1777), the Glory of the Sea. Mar Drugs, 15. ROBINSON, S. D. & NORTON, R. S. 2014. Conotoxin gene superfamilies. Mar Drugs, 12, 6058-101. ROBINSON, S. D. & SAFAVI-HEMAMI, H. 2016. Insulin as a weapon. Toxicon, 123, 56-61. ROBINSON, S. D., SAFAVI-HEMAMI, H., MCINTOSH, L. D., PURCELL, A. W., NORTON, R. S. & PAPENFUSS, A. T. 2014. Diversity of conotoxin gene superfamilies in the venomous snail, Conus victoriae. PLoS One, 9, e87648. ROBINSON, S. D., SAFAVI-HEMAMI, H., RAGHURAMAN, S., IMPERIAL, J. S., PAPENFUSS, A. T., TEICHERT, R. W., PURCELL, A. W., OLIVERA, B. M. & NORTON, R. S. 2015. Discovery by proteogenomics and characterization of an RF-amide neuropeptide from cone snail venom. J Proteomics, 114, 38-47. ROBINSON, S. D., UNDHEIM, E. A. B., UEBERHEIDE, B. & KING, G. F. 2017c. Venom peptides as therapeutics: advances, challenges and the future of venom-peptide discovery. Expert Rev Proteomics, 14, 931-939. RODRIGUEZ, A. M., DUTERTRE, S., LEWIS, R. J. & MARI, F. 2015. Intraspecific variations in Conus purpurascens injected venom using LC/MALDI-TOF-MS and LC-ESI-TripleTOF-MS. Anal Bioanal Chem, 407, 6105-16. ROMERO, H. K., CHRISTENSEN, S. B., DI CESARE MANNELLI, L., GAJEWIAK, J., RAMACHANDRA, R., ELMSLIE, K. S., VETTER, D. E., GHELARDINI, C., IADONATO, S. P., MERCADO, J. L., OLIVERA, B. M. & MCINTOSH, J. M. 2017. Inhibition of α9α10 nicotinic acetylcholine receptors prevents chemotherapy-induced neuropathic pain. PNAS, 114, E1825-e1832. SADEGHI, M., CARSTENS, B. B., CALLAGHAN, B. P., DANIEL, J. T., TAE, H.-S., O'DONNELL, T., CASTRO, J., BRIERLEY, S. M., ADAMS, D. J., CRAIK, D. J. & CLARK, R. J. 2018. Structure- Activity Studies Reveal the Molecular Basis for GABAB-Receptor Mediated Inhibition of High Voltage-Activated Calcium Channels by α-Conotoxin Vc1.1. ACS Chem. Biol., 13, 1577-1587. SAFAVI-HEMAMI, H., GAJEWIAK, J., KARANTH, S., ROBINSON, S. D., UEBERHEIDE, B., DOUGLASS, A. D., SCHLEGEL, A., IMPERIAL, J. S., WATKINS, M., BANDYOPADHYAY, P. K., YANDELL, M., LI, Q., PURCELL, A. W., NORTON, R. S., ELLGAARD, L. & OLIVERA, B. M. 2015. Specialized insulin is used for chemical warfare by fish-hunting cone snails. PNAS, 112, 1743-8. SAFAVI-HEMAMI, H., LU, A., LI, Q., FEDOSOV, A. E., BIGGS, J., SHOWERS CORNELI, P., SEGER, J., YANDELL, M. & OLIVERA, B. M. 2016. Venom Insulins of Cone Snails Diversify Rapidly and Track Prey Taxa. Mol Biol Evol, 33, 2924-2934. SAJEEVAN, K. A. & ROY, D. 2018. Peptide Sequence and Solvent as Levers to Control Disulfide Connectivity in Multiple Cysteine Containing Venom Toxins. J. Phys. Chem. B, 122, 5776-5789. SANDALL, D. W., SATKUNANATHAN, N., KEAYS, D. A., POLIDANO, M. A., LIPING, X., PHAM, V., DOWN, J. G., KHALIL, Z., LIVETT, B. G. & GAYLER, K. R. 2003. A novel α-conotoxin identified by gene sequencing is active in suppressing the vascular response to selective stimulation of sensory nerves in vivo. Biochemistry, 42, 6904-11. SANFORD, M. 2013. Intrathecal ziconotide: a review of its use in patients with chronic pain refractory to other systemic or intrathecal analgesics. CNS Drugs, 27, 989-1002.

SCHNÖLZER, M., ALEWOOD, P. F., JONES, A., ALEWOOD, D. & KENT, S. B. H. 1992. In situ neutralization in Boc-chemistry solid phase peptide synthesis. Rapid, high yield assembly of difficult sequences. Int J Pept Protein Res, 40, 180-93. SHARPE, I. A., GEHRMANN, J., LOUGHNAN, M. L., THOMAS, L., ADAMS, D. A., ATKINS, A., PALANT, E., CRAIK, D. J., ADAMS, D. J., ALEWOOD, P. F. & LEWIS, R. J. 2001. Two new classes of conopeptides inhibit the α1-adrenoceptor and noradrenaline transporter. Nat Neurosci, 4, 902-7. SIMMS, B. A. & ZAMPONI, G. W. 2014. Neuronal voltage-gated calcium channels: structure, function, and dysfunction. Neuron, 82, 24-45. SMOUT, M. J., MULVENNA, J. P., JONES, M. K. & LOUKAS, A. 2011. Expression, refolding and purification of Ov-GRN-1, a granulin-like growth factor from the carcinogenic liver fluke, that causes proliferation of mammalian host cells. Protein Expr Purif, 79, 263-70. SONTI, R., GOWD, K. H., RAO, K. N., RAGOTHAMA, S., RODRIGUEZ, A., PEREZ, J. J. & BALARAM, P. 2013. Conformational diversity in contryphans from Conus venom: cis-trans isomerisation and aromatic/proline interactions in the 23-membered ring of a 7-residue peptide disulfide loop. Chemistry, 19, 15175-89. SOUSA, S. R., MCARTHUR, J. R., BRUST, A., BHOLA, R. F., ROSENGREN, K. J., RAGNARSSON, L., DUTERTRE, S., ALEWOOD, P. F., CHRISTIE, M. J. & ADAMS, D. J. 2018a. Novel analgesic ω-conotoxins from the vermivorous cone snail Conus moncuri provide new insights into the evolution of conopeptides. Scientific reports, 8, 13397. SOUSA, S. R., MCARTHUR, J. R., BRUST, A., BHOLA, R. F., ROSENGREN, K. J., RAGNARSSON, L., DUTERTRE, S., ALEWOOD, P. F., CHRISTIE, M. J., ADAMS, D. J., VETTER, I. & LEWIS, R. J. 2018b. Novel analgesic -conotoxins from the vermivorous cone snail Conus moncuri provide new insights into the evolution of conopeptides. Sci Rep, 8, 13397. SOUSA, S. R., VETTER, I. & LEWIS, R. J. 2013. Venom peptides as a rich source of cav2.2 channel blockers. Toxins (Basel), 5, 286-314. STAROBOVA, H., S, W. A. H., LEWIS, R. J. & VETTER, I. 2018. Transcriptomics in pain research: insights from new and old technologies. Mol Omics, 14, 389-404. TEICHERT, R. W., JIMENEZ, E. C. & OLIVERA, B. M. 2005. Α S-conotoxin RVIIIA: a structurally unique conotoxin that broadly targets nicotinic acetylcholine receptors. Biochemistry, 44, 7897-902. TERLAU, H. & OLIVERA, B. M. 2004. Conus venoms: a rich source of novel ion channel-targeted peptides. Physiol Rev, 84, 41-68. TERRAT, Y., BIASS, D., DUTERTRE, S., FAVREAU, P., REMM, M., STOCKLIN, R., PIQUEMAL, D. & DUCANCEL, F. 2012. High-resolution picture of a venom gland transcriptome: Case study with the marine snail Conus consors. Toxicon, 59, 34-46. TIETZE, A. A., TIETZE, D., OHLENSCHLAGER, O., LEIPOLD, E., ULLRICH, F., KUHL, T., MISCHO, A., BUNTKOWSKY, G., GORLACH, M., HEINEMANN, S. H. & IMHOF, D. 2012. Structurally diverse -conotoxin PIIIA isomers block sodium channel NaV 1.4. Angew Chem Int Ed Engl, 51, 4058-61. TOSTI, E., BONI, R. & GALLO, A. 2017. micro-Conotoxins Modulating Sodium Currents in Pain Perception and Transmission: A Therapeutic Potential. Mar Drugs, 15. TURCHETTO, J., SEQUEIRA, A. F., RAMOND, L., PEYSSON, F., BRÁS, J. L., SAEZ, N. J., DUHOO, Y., BLÉMONT, M., GUERREIRO, C. I. & QUINTON, L. 2017. High-throughput expression of animal venom toxins in Escherichia coli to generate a large library of oxidized disulphide- reticulated peptides for drug discovery. Microbial cell factories, 16, 6. VAN LIEROP, B. J., ROBINSON, S. D., KOMPELLA, S. N., BELGI, A., MCARTHUR, J. R., HUNG, A., MACRAILD, C. A., ADAMS, D. J., NORTON, R. S. & ROBINSON, A. J. 2013. Dicarba α- conotoxin Vc1.1 analogues with differential selectivity for nicotinic acetylcholine and GABAB receptors. ACS Chem Biol, 8, 1815-21.

VAN WAGONER, R. M. & IRELAND, C. M. 2003. An improved solution structure for -conotoxin PiiiE. Biochemistry, 42, 6347-52. VAN WAGONER, R. M., JACOBSEN, R. B., OLIVERA, B. M. & IRELAND, C. M. 2003. Characterization and three-dimensional structure determination of -conotoxin Piiif, a novel noncompetitive antagonist of nicotinic acetylcholine receptors. Biochemistry, 42, 6353-62. VETTER, I., DAVIS, J. L., RASH, L. D., ANANGI, R., MOBLI, M., ALEWOOD, P. F., LEWIS, R. J. & KING, G. F. 2011. Venomics: a new paradigm for natural products-based drug discovery. Amino Acids, 40, 15-28. VETTER, I. 2012. Development and optimization of FLIPR high throughput calcium assays for ion channels and GPCRs. Adv Exp Med Biol, 740, 45-82. VETTER, I. & LEWIS, R. J. 2012. Therapeutic potential of cone snail venom peptides (conopeptides). Curr Top Med Chem, 12, 1546-52. VINK, S. & ALEWOOD, P. F. 2012. Targeting voltage-gated calcium channels: developments in peptide and small-molecule inhibitors for the treatment of neuropathic pain. Br J Pharmacol, 167, 970-89. VIOLETTE, A., BIASS, D., DUTERTRE, S., KOUA, D., PIQUEMAL, D., PIERRAT, F., STOCKLIN, R. & FAVREAU, P. 2012. Large-scale discovery of conopeptides and conoproteins in the injectable venom of a fish-hunting cone snail using a combined proteomic and transcriptomic approach. J Proteomics, 75, 5215-25. WALKER, C. S., JENSEN, S., ELLISON, M., MATTA, J. A., LEE, W. Y., IMPERIAL, J. S., DUCLOS, N., BROCKIE, P. J., MADSEN, D. M., ISAAC, J. T., OLIVERA, B. & MARICQ, A. V. 2009. A novel Conus snail polypeptide causes excitotoxicity by blocking desensitization of AMPA receptors. Curr Biol, 19, 900-8. WALKER, C. S., STEEL, D., JACOBSEN, R. B., LIRAZAN, M. B., CRUZ, L. J., HOOPER, D., SHETTY, R., DELACRUZ, R. C., NIELSEN, J. S., ZHOU, L. M., BANDYOPADHYAY, P., CRAIG, A. G. & OLIVERA, B. M. 1999. The T-superfamily of conotoxins. J Biol Chem, 274, 30664-71. WAN, J., HUANG, J. X., VETTER, I., MOBLI, M., LAWSON, J., TAE, H. S., ABRAHAM, N., PAUL, B., COOPER, M. A., ADAMS, D. J., LEWIS, R. J. & ALEWOOD, P. F. 2015. α-Conotoxin dendrimers have enhanced potency and selectivity for homomeric nicotinic acetylcholine receptors. J Am Chem Soc, 137, 3209-12. WANG, F., YAN, Z., LIU, Z., WANG, S., WU, Q., YU, S., DING, J. & DAI, Q. 2016. Molecular basis of toxicity of N-type calcium channel inhibitor MVIIA. Neuropharmacology, 101, 137-45. WILSON, D. & DALY, N. L. 2018. Nuclear Magnetic Resonance seq (NMRseq): A New Approach to Peptide Sequence Tags. Toxins (Basel), 10. WILSON, M. J., ZHANG, M. M., GAJEWIAK, J., AZAM, L., RIVIER, J. E., OLIVERA, B. M. & YOSHIKAMI, D. 2015. Α- and β-subunit composition of voltage-gated sodium channels investigated with -conotoxins and the recently discovered O section §-conotoxin GVIIJ. J Neurophysiol, 113, 2289-301. WU, X., HUANG, Y.-H., KAAS, Q., HARVEY, P. J., WANG, C. K., TAE, H.-S., ADAMS, D. J. & CRAIK, D. J. 2017. Backbone cyclization of analgesic conotoxin GeXIVA facilitates direct folding of the ribbon isomer. J. Biol. Chem., 292, 17101-17112. WU, X., WU, Y., ZHU, F., YANG, Q., WU, Q., ZHANGSUN, D. & LUO, S. 2013. Optimal cleavage and oxidative folding of α-conotoxin TxIB as a therapeutic candidate peptide. Mar. Drugs, 11, 3537- 3553. WU, Y., ZHENG, Y. & TANG, H. 2016. Identifying the Types of Ion Channel-Targeted Conotoxins by Incorporating New Properties of Residues into Pseudo Amino Acid Composition. Biomed Res Int, 2016, 3981478. XIA, Z., CHEN, Y., ZHU, Y., WANG, F., XU, X. & ZHAN, J. 2006. Recombinant -conotoxin MVIIA possesses strong analgesic activity. BioDrugs, 20, 275-81. XIANFANG, W., JUNMEI, W., XIAOLEI, W. & YUE, Z. 2017. Predicting the Types of Ion Channel- Targeted Conotoxins Based on AVC-SVM Model. Biomed Res Int, 2017, 2929807. 99

XU, M., ZHU, X., YU, J., YU, J., LUO, S. & WANG, X. 2017. The crystal structure of Ac-AChBP in complex with α-conotoxin LvIA reveals the mechanism of its selectivity towards different nAChR subtypes. Protein Cell, 8, 675-685. XU, S., ZHANG, T., KOMPELLA, S. N., YAN, M., LU, A., WANG, Y., SHAO, X., CHI, C., ADAMS, D. J., DING, J. & WANG, C. 2015. Conotoxin αD-GeXXA utilizes a novel strategy to antagonize nicotinic acetylcholine receptors. Sci Rep, 5, 14261. YANG, L., TAE, H. S., FAN, Z., SHAO, X., XU, S., ZHAO, S., ADAMS, D. J. & WANG, C. 2017. A Novel Lid-Covering Peptide Inhibitor of Nicotinic Acetylcholine Receptors Derived from αD- Conotoxin GeXXA. Mar Drugs, 15. YE, M., HONG, J., ZHOU, M., HUANG, L., SHAO, X., YANG, Y., SIGWORTH, F. J., CHI, C., LIN, D. & WANG, C. 2011. A novel conotoxin, qc16a, with a unique cysteine framework and folding. Peptides, 32, 1159-65. YE, M., KHOO, K. K., XU, S., ZHOU, M., BOONYALAI, N., PERUGINI, M. A., SHAO, X., CHI, C., GALEA, C. A., WANG, C. & NORTON, R. S. 2012. A helical conotoxin from Conus imperialis has a novel cysteine framework and defines a new superfamily. J Biol Chem, 287, 14973-83. YIN, J. B., FAN, Y. X. & SHEN, H. B. 2011. Conotoxin superfamily prediction using diffusion maps dimensionality reduction and subspace classifier. Curr Protein Pept Sci, 12, 580-8. YU, J., ZHU, X., YANG, Y., LUO, S. & ZHANGSUN, D. 2018. Expression in Escherichia coli of fusion protein comprising α-conotoxin TxIB and preservation of selectivity to nicotinic acetylcholine receptors in the purified product. Chem. Biol. Drug Des., 91, 349-358. YUAN, D. D., LIU, L., SHAO, X. X., PENG, C., CHI, C. W. & GUO, Z. Y. 2008. Isolation and cloning of a conotoxin with a novel cysteine pattern from Conus caracteristicus. Peptides, 29, 1521-5. YUAN, L. F., DING, C., GUO, S. H., DING, H., CHEN, W. & LIN, H. 2013. Prediction of the types of ion channel-targeted conotoxins based on radial basis function network. Toxicol In Vitro, 27, 852- 6. YUAN, Y., BALSARA, R. D., ZAJICEK, J., KUNDA, S. & CASTELLINO, F. J. 2016. Discerning the Role of the Hydroxyproline Residue in the Structure of Conantokin Rl-B and Its Role in GluN2B Subunit-Selective Antagonistic Activity toward N-Methyl-d-Aspartate Receptors. Biochemistry, 55, 7112-7122. ZAMPONI, G. W. 2016. Targeting voltage-gated calcium channels in neurological and psychiatric diseases. Nat Rev Drug Discov, 15, 19-34. ZAMPONI, G. W. & CURRIE, K. P. 2013. Regulation of CaV2 calcium channels by G protein coupled receptors. Biochim Biophys Acta, 1828, 1629-43. ZHANG, L., ZHANG, C., GAO, R., YANG, R. & SONG, Q. 2016. Using the SMOTE technique and hybrid features to predict the types of ion channel-targeted conotoxins. J Theor Biol, 403, 75-84. ZHANG, M. M., GAJEWIAK, J., AZAM, L., BULAJ, G., OLIVERA, B. M. & YOSHIKAMI, D. 2015. Probing the Redox States of Sodium Channel Cysteines at the Binding Site of O section §- Conotoxin GVIIJ. Biochemistry, 54, 3911-20. ZHOU, M., WANG, L., WU, Y., ZHU, X., FENG, Y., CHEN, Z., LI, Y., SUN, D., REN, Z. & XU, A. 2013. Characterizing the evolution and functions of the M-superfamily conotoxins. Toxicon, 76, 150-9. ZIEGMAN, R., BRUST, A., JHA, P., CARDOSO, F. C., LEWIS, R. J. & ALEWOOD, P. F. 2019. 'Messy' Processing of -conotoxin MrIA Generates Homologues with Reduced hNET Potency. Mar Drugs, 17.

100

9 Acknowledgements

This study was supported by Fellowships from the National Health and Medical Research Council (PFA,

RJL and DC). MM is supported by the European Research Council under the European Union’s Horizon

2020 research and innovation program (grant agreement no 714366) and by the Australian Research

Council Discovery Early Career Researcher Awards (DE150100784). PFA and RJL were supported by a

NHMRC Program Grant (APP1072113).

Ai-Hua JIN was born in China, where she obtained her B. Pharm degree from Shenyang Pharmaceutical

University, China. She completed her PhD degree at the Institute for Molecular Bioscience, The University of Queensland, Australia in 2006. She is currently working in the laboratory of Professor Paul Alewood and her research interests are transcriptomic proteomic and venomic discovery, chemical synthesis and engineering of novel toxins from Australia’s venomous creatures.

101

Markus Muttenthaler is head of the Neuropeptide Research lab at the Institute of Biological Chemistry at the University of Vienna, Austria, and holds a second appointment at the Institute for Molecular

Bioscience at the University of Queensland, Brisbane, Australia. He graduated as a synthetic chemist in

2004 at the Vienna University of Technology, obtained his Ph.D. in Biological and Medicinal Chemistry in 2009 at the University of Queensland, and carried out postdoctoral research in the Alewood Lab at the

Institute of Molecular Bioscience, in the Dawson lab at the Scripps Research Institute in California, and in the Albericio lab at the Institute for Research in Biomedicine Barcelona. In 2015, he set up his first research group at the Institute for Molecular Bioscience, and in 2017, he received the prestigious ERC Starting Grant and was appointed Associate Professor at the University of Vienna, Austria, where he set up his second research group. His research focuses on bioactive peptides to study human health and disease, with applications in gastrointestinal disorders, chronic pain, breast cancer, and autism.

102

Quentin Kaas is a Senior Research Fellow at the Institute for Molecular Bioscience, The University of

Queensland. He obtained his Chemical Engineering degree in 2001 from the Ecole Normale Supérieure de

Chimie de Montpellier (ENSCM) and his Ph.D. degree in Bioinformatics in 2005 from the University of

Montpellier II. He was then awarded an Australian Postdoctoral fellowship by the Australian Research

Council to undertake three years of postdoctoral research on plant cyclic peptide structure−activity relationships in the laboratory of Professor David Craik at the Institute for Molecular Bioscience, The

University of Queensland, Australia. He is currently working in the laboratory of Professor David Craik, and the focuses of his research are structural bioinformatics and computational modelling studies of toxins extracted from plants and animals. He has developed and currently maintains the only database specialized on sequences and structures of cone snail toxins, ConoServer (http://www.conoserver.org).

103

Sébastien Dutertre was born in France, where he obtained his M.Sc. from the University of Paris XII,

School of Industrial Chemistry of Paris and National Museum of Nature History of Paris in 2001. He received his Ph.D. in Molecular Bioscience in 2006 from the University of Queensland, Australia. He was then awarded an EMBO postdoctoral fellowship to join the laboratory of Prof. Heinrich Betz at the Max-

Planck-Institute for Brain Research, Frankfurt, Germany. Next, he joined Atheris laboratories (Switzerland) in 2008 to work on an innovative and ambitious post-genomic project dedicated to the discovery and development of novel biopharmaceuticals generated by the broad biodiversity of animal venoms. In 2010, he was awarded a University of Queensland postdoctoral fellowship to join the laboratory of Prof. Richard

J. Lewis and develop an integrated approach to accelerate the discovery of novel peptides from cone snail venoms, using second generation sequencing technologies and proteomic methods. Since 2013, he is a permanent researcher at the CNRS in Montpellier (Institut des Biomolécules Max Mousseron, France), where he leads the “Chemical Ecology of Toxins and Venoms” lab. His research interests encompass the discovery of venom peptides, their precise mode of interaction with targeted receptors and their associated therapeutic potential.

104

Himaya S.W.A. was born in Sri Lanka, where she obtained her B.Sc. (Hons) from the University of

Peradeniya in 2008. Then she moved to South Korea where she completed her combined M.Sc. (2010) and

Ph.D. (2013) at the Marine Bioprocess Research Center, Pukyong National University, Busan. She conducted a one-year postdoc position at the same institute under Prof. Se-Kwon Kim, characterizing anti- inflammatory peptides from marine invertebrates. In 2015, she was awarded the University of Queensland

Postdoctoral Fellowship to commence three years of postdoctoral research in the group of Prof. Richard

Lewis at Institute for Molecular Biosciences, The University of Queensland, Australia. Since then she is working in the laboratory of Professor Richard Lewis and her research focuses on use of omics technologies for high-throughput profiling of cone snail venoms to unravel therapeutically valuable venom peptides while understanding the evolution and origin of venom.

105

Richard J Lewis is an NHMRC Principal Research Fellow at the Institute for Molecular Bioscience, The

University of Queensland, Australia. He obtained his Ph.D. degree in Zoology and Chemistry from James

Cook University in Townsville, Australia (1985). He was then employed as a Fisheries Scientist by the

Queensland Department of Primary Industries to continue his Ph.D. research on ciguatera before being appointed as a Principal Investigator at the Centre for Drug Design and Development in 1996 to research conotoxins and their therapeutic potential which led to the establishment of the biotechnology company

Xenome Ltd. that he co-founded in 2000. His research focuses on the application of proteomics, transcriptomic, pharmacological and structural studies to the discovery and development venom peptide, especially in relation to pain. He has trained more than 20 Ph.D. students and is the author of over 300 scientific publications.

106

David J. Craik is an Australian Research Council Laureate Fellow at the Institute for Molecular Bioscience,

The University of Queensland, Australia. He obtained his Ph.D. degree in Organic Chemistry from La

Trobe University in Melbourne, Australia (1981), and undertook postdoctoral studies at Florida State and

Syracuse Universities before taking up a lectureship at the Victorian College of Pharmacy in 1983. He was appointed Professor of Medicinal Chemistry and Head of School in 1988. He moved to the University of

Queensland in 1995 to set up a new biomolecular NMR laboratory. His research focuses on application of

NMR in drug design and on toxins, including conotoxins. His group has a particular focus on structural studies of disulfide-rich proteins and on the discovery and applications of circular proteins and novel protein topologies. He has trained more than 70 Ph.D. students and is the author of 650 scientific publications.

107

Paul F Alewood received his B.Sc. (Hons) from the University of New South Wales, Australia, in 1969.

He received his Ph.D. at the University of Calgary, Canada, in 1974 before studying in Geneva and London.

He moved to the University of Melbourne in 1977 before joining the Victorian College of Pharmacy in

1983 as a lecturer where he began a lifelong interest in peptide drug development. In 1998 he was a founding member of Bond University, Australia, before moving to the University of Queensland in 1991 as Deputy Director and Professorial Fellow of the Centre for Drug Design and Development. He was also a founding member of the Institute of Molecular Bioscience at the University of Queensland in 2000, where his group has been focused on the discovery and synthesis of disulfide rich toxins from Australia’s venomous creatures and peptide drug development since then.

108

TOC Graphic

109