Venoms as drug libraries
Modern drug discovery is increasingly in need of a novel paradigm to replace the classical blockbuster-based economic model of large pharmaceutical companies. Loss of patent protection, the rise in the generics industry, mounting research and development costs and increasing regulatory pressure are multiple factors that motivate efforts at implementing new approaches to drug discovery. The successes registered with therapeutic antibodies have motivated a recent paradigm shift for the industry, with increased recognition that biological molecules, and particularly peptides, are valid leads for innovative therapeutics.
Animal venoms can play a prime role in drug discovery efforts as they constitute a vast and essentially untapped resource of novel pharmacologically active molecules Citation[1]. Evolution has led to the development of a venomous function in a wide variety of invertebrates (e.g., sea anemones, corals, jellyfish, marine mollusks, spiders, scorpions, hymenopteran insects and marine worms), as well as vertebrates, such as snakes. The evolutionary process has refined the biological diversity and the potency of the venoms so that they now essentially constitute vast libraries of pharmacologically active, receptor-targeted molecules that are preoptimized for the medicinal chemist Citation[2].
Animal-venom toxins range from small organic molecules, such as spider polyamines, up to large proteins of more than 100 kDa, such as latrotoxin from black widow spider venom. Invertebrate venoms mostly contain a large array of polypeptides, 1–10 kDa in size, most of them tightly folded and stabilized by several disulfide bridges Citation[3]. Conversely, snake venoms contain proteins belonging to several structural classes, such as peptides, small neurotoxins, phospholipases and various enzymes Citation[4].
The structural and pharmacological diversity of disulfide-rich venom peptides is stunning in its complexity. Of the approximately 500 cone snail species (Conus spp.), 15 structural families have been defined that correspond to more than 11 pharmacological classes. Spiders constitute an even more amazing potential source of novel drug leads: the approximately 41,000 described species represent an ecologically megadiverse group, which may produce an estimated 12 million or more biologically active peptides Citation[5], of which, the current number of reported peptide sequences is only approximately 600 Citation[101].
The limited investigations of the pharmacological potential of venom components have, nevertheless, had a significant impact on drug discovery. The best example is the development of the blockbuster drug captopril, an antihypertensive synthetic molecule that structurally and functionally mimics the bradykinin-potentiating peptides first discovered in the venom of the Brazilian snake Bothrops jararacaCitation[6]. Another venom-derived drug is Prialt® (ziconotide), the synthetic version of ω-conotoxin MVIIA, a peptide from the venom of the marine snail Conus magus, which possesses strong analgesic activity mediated via a blockade of N-type calcium channels Citation[7]. Exenatide (Byetta®), a synthetic version of exendin-4 from the saliva of the Gila monster lizard, is a peptide agonist of the glucagon-like peptide receptor that has been approved as an adjuvant in the treatment of adults with Type 2 diabetes Citation[8] and is forecast to reach blockbuster status. TM-601, currently in Phase II human trials, is a modified form of the scorpion peptide chlorotoxin that selectively targets receptors on glioma cells (a diffuse form of brain cancer) without binding to healthy surrounding neurons Citation[9]. These examples illustrate how the high receptor subtype selectivity of animal peptide toxins can be harnessed for cell- or tissue-selective therapies or diagnostic applications.
In the past decade, significant progress in receptor characterization, and the identification of multiple isoforms of ion channels and G-protein-coupled receptors has resulted in a plethora of ‘druggable’ targets that remain devoid of selective pharmacology. This necessitates a more target-oriented approach to drug discovery whereby future drugs will be tailored to interact selectively with specific receptor isoforms, leading to more-focused treatments and a reduction in undesirable side effects Citation[10]. Cell proliferation in cancer is one example where selected ion channel subtypes are overexpressed, therefore becoming important therapeutic or diagnostic targets Citation[11].
Regardless of the tremendous potential of animal venoms, the unfortunate truth remains that the current state of technologies dedicated to venom characterization does not allow full investigation of the enormous array of venom peptides, as most venomous animals are of very small size. Thus, the feeble amounts of venom available totally preclude the application of ‘classical’ biochemistry techniques, such as bioassay-guided fractionation and Edman sequencing of purified peptides. As will now be discussed, application of the most cutting-edge genomics and proteomics technologies can help solve the current conundrum.
Proteomics-based investigation strategies: venom profiling for identification of pharmacological & structural peptide classes
Venom analysis at the nanogram-to-microgram scale presents challenges similar to those encountered in proteomics studies in which only very limited amounts of biological material of very high complexity may be available to the investigator. Venom research faces a supplementary challenge with the lack of appropriate protein databases for the animals studied. As a consequence, a ‘classical’ proteomics approach based on peptide mass fingerprinting and tandem mass spectrometry (MS) peptide fragmentation will not be sufficient for venom investigation, except in some groups (e.g., snakes) where a large body of venom protein information already exists Citation[4]. For all other groups of venomous animals, de novo peptide sequencing must be implemented.
The first step in venom investigation by MS is usually mass profiling of crude venom or fractions from chromatographic separation. Several recent studies on scorpion, spider, snake and cone snail venoms have demonstrated that venom peptides number in the hundreds Citation[12], with up to 1000 masses detected in spider venom by MALDI-TOF MS Citation[5] or in Conus venom by electrospray-qTOF MS Citation[13,14]. A conservative estimate of the total number of peptides in all described venomous animal species approaches 20 million and will certainly rise in the future with increases in the performance of MS instrumentation.
However, determination of intact large peptide masses does not shed light on the structure and pharmacological profile of components. To solve the issue, an approach known as ‘venom landscapes’ based on liquid chromatography MALDI MS has been recently proposed Citation[5]. 3D profiles generated by plotting mass spectra versus retention time reveal that the various peptides of related pharmacology are separated by mass and hydrophobicity and group into ‘islands’ in the plot. This approach identifies groups of unknown toxins and enables selective isolation of toxins of similar pharmacology from related species. This bypasses the cumbersome process of bioassay-guided fractionation for generating target-specific peptide libraries. It is, therefore, of absolute necessity to obtain amino acid sequence information in order to further separate the various pharmacological classes of toxins; the implementation of de novo MS-sequencing strategies is, thus, key to the in-depth investigation of venoms.
Toxin identification: top-down sequencing of large peptides?
De novo sequencing of large peptide remains a true challenge. In addition to 1–5-kDa peptides, ‘long’ peptides are common in venoms and comprise scorpion sodium channel toxins (6–9 kDa), some spider peptides, such as the MIT-like toxins (7–8 kDa), and the neurotoxic components of snake venom (6–9 kDa). Early de novo sequencing work conducted on wasp venom peptides used post-source decay fragmentation Citation[15]. The development of proteomics has permitted the wide application of collision-induced dissociation to de novo peptide sequencing. However, contrary to proteomics studies in which uninterpreted tandem MS spectra are searched against a database of proteins in a ‘bottom-up’ approach, in organisms with unknown genomes, true de novo peptide sequencing is necessary in a ‘top-down’ approach. Approaches involving collision-induced dissociation were used in the characterization of conotoxins, from Conus monile and Conus virgoCitation[16], and small snake peptides, such as BPPs Citation[17], poly-His and poly-Gly peptides Citation[18] and sarafotoxins Citation[19]. These last results emphasize the increasing role of Fourier-transform ion cyclotron resonance MS, which brings high resolution and mass accuracy to de novo sequencing, allowing the unambiguous assignment of fragment masses and quasi-isobaric residues, such as lysine and glutamine, which differ by only 0.003638 Da. One promising tool for MS sequencing of toxins is the MALDI matrix 1,5-DAN which reduces disulfide linkages in the gas phase and enhances in-source decay, permitting rapid top-down sequencing of Conus and snake toxins via fragmentation in a TOF analyzer Citation[20].
Novel fragmentation techniques, such as electron-transfer dissociation, probably offer the highest potential for the full sequencing of large venom peptides Citation[21]. In an impressive study, electron-transfer dissociation, when coupled with a targeted chemical derivatization, was shown to greatly facilitate long peptide sequencing and it permitted determination of the full sequence and post-translational modifications of 31 Conus textile peptides, although only a small fraction of the crude venom extract (7%) from a single snail was used Citation[22]. Indeed, the sensitivity of MS instrumentation has now become sufficient to enable peptide sequences to be generated from the venom taken from a single individual. Recently, we reported the analysis of individual small spiders with a body size of less than 5 mm. In this ‘nanovenomics’ approach, nano-liquid chromatography MS, as well as tandem MS analysis, resulted in the generation of more than 380 parent ion masses and several dozen of de novo sequence tags from a single 4.5-mm jumping spider (Salticidae) Citation[23].
Venomics as a new paradigm for venom exploration
The presence of an estimated 20 million peptides in all animal venoms combined means that high-throughput sequence characterization is necessary at the microscale level. One possible solution is an approach known as ‘venomics’, which is based on a combination of MS and molecular biology methods. In short, MS-based de novo peptide sequencing can be used to generate large numbers of peptide sequence tags from crude venoms. These tags can then be used to design specific primers for the large-scale cloning and sequencing of venom components from venom-gland cDNA libraries. Individual studies have demonstrated the feasibility of the two techniques separately Citation[5,24,25], and it is now an opportune time for a combination of these methods to fully explore the molecular diversity of venoms.
Venom gland genomics has been used in cones for the elucidation of gene superfamilies and has permitted the discovery of varied structural and pharmacological groups of conotoxins. More recently, an approach known as ‘exogenomics’ has been used to find receptor-subtype selective conotoxins from related species. In this approach, the direct search for members of a given supergene family allows identification of whole groups of related toxins targeting a given receptor, but having varying pharmacological profiles. With drug discovery in mind, the investigator is taking full advantage of nature’s ability to perform random mutagenesis and generate libraries of target-oriented ligands Citation[26]. In spiders, large-scale random sequencing of cDNA libraries has also permitted the identification of novel structural families Citation[24,25].
Future directions
In the past decade, the ability to mass-profile venoms and to determine the amino acid sequence of venom peptides lagged behind the methods available for studying the pharmacology and 3D structure of these molecules; however, robust MS-based methods have now outpaced developments in high-throughput functional and structural analyses.
The primary technique for determining the structure of venom peptides is NMR spectroscopy. The Protein Data Bank currently contains 170 structures of venom peptides less than 9 kDa, and 141 (82%) of these were determined using NMR. This highlights two points: first, the number of available venom-peptide structures is exceedingly small compared with the number of available sequences; second, significant increases in the number of peptide-toxin structures will require advances in the speed of NMR protein structure determination. It was recently demonstrated that the combination of rapid data acquisition via nonuniform sampling combined with automated spectral analysis and structure calculation Citation[27] enables high-resolution 3D structures of venom peptides less than 10 kDa in size to be routinely determined in a time frame of less than 1 week Citation[28].
The greatest remaining bottleneck with regard to unleashing the full potential of animal venoms is probably the production of sufficient quantities of venom peptides for complete structural/functional characterization. Disulfide-rich toxins lend themselves to native chemical ligation, whereby the peptide can be divided into fragments that can be separately synthesized and then ligated to yield the full-length peptide Citation[29]. Venom peptides have also been produced by overexpression in bacteria Citation[30], yeast Citation[31] or insect cells Citation[32]. Rapid synthesis and peptide engineering techniques are also being developed with special application to disulfide-rich peptides Citation[33]. However, the problem of producing the native disulfide-bond isomer must be addressed, as a toxin with three, four or five disulfide bonds can form 15, 105 or 945 different disulfide-bond isomers, respectively.
A number of hurdles still need to be overcome to realize the goal of producing ‘man-made venoms’, in which each individual peptide from a single venom can be produced efficiently and aliquoted into multiwell plates for high-throughput functional analysis. Nevertheless, the enormous potential of venoms as drug sources of the future becomes more of a reality each day as the investigation of venom peptidomes appears to offer almost endless possibilities for drug discovery.
Acknowledgements
We are grateful to Graham M Nicholson (University of Technology Sydney, Australia), for a critical reading of the manuscript, and useful insights on toxin applications in drug discovery. Glenn King acknowledges support from ARC grant DP0774245.
Financial & competing interests disclosure
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
No writing assistance was utilized in the production of this manuscript.
References
- Lewis RJ, Garcia ML. Therapeutic potential of venom peptides. Nat. Rev. Drug Discov.2(10), 790–802 (2003).
- Terlau H, Olivera BM. Conus venoms: a rich source of novel ion channel-targeted peptides. Physiologic. Rev.84(1), 41–68 (2004).
- Escoubas P, Rash L. Tarantulas: eight-legged pharmacists and combinatorial chemists. Toxicon43(5), 555–574 (2004).
- Calvete JJ, Juarez P, Sanz L. Snake venomics. Strategy and applications. J. Mass Spectrom.42(11), 1405–1414 (2007).
- Escoubas P, Sollod B, King GF. Venom landscapes: mining the complexity of spider venoms via a combined cDNA and mass spectrometric approach. Toxicon47(6), 650–663 (2006).
- Hayashi MA, Camargo AC. The bradykinin-potentiating peptides from venom gland and brain of Bothrops jararaca contain highly site specific inhibitors of the somatic angiotensin-converting enzyme. Toxicon45(8), 1163–1170 (2005).
- Miljanich GP. Ziconotide: neuronal calcium channel blocker for treating severe chronic pain. Curr. Med. Chem.11(23), 3029–3040 (2004).
- Bray GM. Exenatide. Am. J. Health. Syst. Pharm.63(5), 411–418 (2006).
- Mamelak AN, Jacoby DB. Targeted delivery of antitumoral therapy to glioma and other malignancies with synthetic chlorotoxin (TM-601). Expert Opin. Drug Deliv.4(2), 175–186 (2007).
- Li RA, Ennis IL, Xue T et al. Molecular basis of isoform-specific micro-conotoxin block of cardiac, skeletal muscle, and brain Na+ channels. J. Bio. Chem.278(10), 8717–8724 (2003).
- Arcangeli A, Crociani O, Lastraioli E, Masi A, Pillozzi S, Becchetti A. Targeting ion channels in cancer: a novel frontier in antineoplastic therapy. Curr. Med. Chem.16(1), 66–93 (2009).
- Nascimento DG, Rates B, Santos DM et al. Moving pieces in a taxonomic puzzle: venom 2D-LC/MS and data clustering analyses to infer phylogenetic relationships in some scorpions from the Buthidae family (scorpiones). Toxicon, 47(6), 628–639 (2006).
- Biass D, Dutertre S, Gerbault A et al. Comparative proteomic study of the veno of the piscivorous cone snail Conus consors.J. Proteomics72(2), 210–218 (2009).
- Davis J, Jones A, Lewis RJ. Remarkable inter- and intra-species complexity of conotoxins revealed by LC/MS. Peptides DOI: 10.1016/j.peptides.2009.1003.1019 (2009) (Epub ahead of print).
- Hisada M, Konno K, Itagaki Y, Naoki H, Nakajima T. Advantages of using nested collision induced dissociation/post-source decay with matrix-assisted laser desorption/ionization time-of-flight mass spectrometry: sequencing of novel peptides from wasp venom. Rapid Commun. Mass Spectrom.14(19), 1828–1834 (2000).
- Sudarslal S, Singaravadivelan G, Ramasamy P et al. A novel 13 residue acyclic peptide from the marine snail, Conus monile, targets potassium channels. Biochem. Biophys. Res. Comm.317(3), 682–688 (2004).
- Soares MR, Oliveira-Carvalho AL, Wermelinger LS et al. Identification of novel bradykinin-potentiating peptides and C-type natriuretic peptide from Lachesismuta venom. Toxicon46(1), 31–38 (2005).
- Favreau P, Cheneval O, Menin L et al. The venom of the snake genus Atheris contains a new class of peptides with clusters of histidine and glycine residues. Rapid Commun. Mass Spectrom.21(3), 406–412 (2007).
- Quinton L, Le Caer JP, Phan G et al. Characterization of toxins within crude venoms by combined use of Fourier transform mass spectrometry and cloning. Anal. Chem.77(20), 6630–6639 (2005).
- Quinton L, Demeure K, Dobson R, Gilles N, Gabelica V, Pauw ED. New method for characterizing highly disulfide-bridged peptides in complex mixtures: application to toxin identification from crude venoms. J Proteome Res.6(8), 3216–3223 (2007).
- Coon JJ. Collisions or electrons? Protein sequence analysis in the 21st Century. Anal. Chem.81(9), 3208–3215(2009).
- Ueberheide BM, Fenyö D, Alewood PF, Chait BT. Rapid sensitive analysis of cysteine rich peptide venom components. Proc. Natl Acad. Sci. USA106(17), 6910–6915 (2009).
- Escoubas P, Dauly C, Nicholson GM, King GF. Spider venoms for novel therapeutics: from venomics to drug candidates. Presented at: XVI International Congress of the International Society on Toxinology. Recife, Brazil, 15–20 March 2009.
- Kozlov S, Malyavka A, McCutchen B et al. A novel strategy for the identification of toxinlike structures in spider venom. Proteins59(1), 131–140 (2005).
- Sollod BL, Wilson D, Zhaxybayeva O, Gogarten JP, Drinkwater R, King GF. Were arachnids the first to use combinatorial peptide libraries? Peptides26(1), 131–139 (2005).
- Holford M, Zhang MM, Gowd KH et al. Pruning nature: biodiversity-derived discovery of novel sodium channel blocking conotoxins from Conus bullatus. Toxicon53(1), 90–98 (2009).
- Güntert P. Automated structure determination from NMR spectra. Eur. Biophys. J.38(2), 129–143 (2009).
- Mobli M, Bermel W, Miljenovic T, Pierens GK, King GF. ASAP-NMR: automated processing and analysis of non-uniformly sampled NMR data. Presented at: Proceedings of the Australian and New Zealand Society for Magnetic Resonance (ANZMAG). Couran Cove, Queensland, Australia, 7–11 December 2008.
- Jensen JE, Durek T, Alewood PF, Adams DJ, King GF, Rash LD. Chemical synthesis and folding of APETx2, a potent and selective inhibitor of acid sensing ion channel 3. Toxicon54(1), 56–61 (2009).
- Tedford HW, Fletcher JI, King GF. Functional significance of the β-hairpin in the insecticidal neurotoxin ω-atracotoxin-Hv1a. J. Biol. Chem.276, 26568–26576 (2001).
- Anangi R, Chen C-Y, Cheng C-H et al. Expression of snake venom toxins in Pichia pastoris. Toxin Rev.26, 169–187 (2007).
- Escoubas P, Bernard C, Lambeau G, Lazdunski M, Darbon H. Recombinant production and solution structure of PcTx1, the specific peptide inhibitor of ASIC1a proton-gated cation channels. Protein Sci.12(7), 1332–1343 (2003).
- Bulaj G. Integrating the discovery pipeline for novel compounds targeting ion channels. Curr. Opin. Chem. Biol.12(4), 441–447 (2008).
Website
- ArachnoServer. Spider toxin database www.arachnoserver.org