The A-superfamily of Conotoxins

Abstract

The generation of functional novelty in proteins encoded by a gene superfamily is seldom well documented. In this report, we define the A-conotoxin superfamily, which is widely expressed in venoms of the predatory cone snails (Conus), and show how gene products that diverge considerably in structure and function have arisen within the same superfamily. A cDNA clone encoding α-conotoxin GI, the first conotoxin characterized, provided initial data that identified the A-superfamily. Conotoxin precursors in the A-superfamily were identified from six Conus species: most (11/16) encoded α-conotoxins, but some (5/16) belong to a family of excitatory peptides, the κA-conotoxins that target voltage-gated ion channels. α-Conotoxins are two-disulfide-bridged nicotinic antagonists, 13–19 amino acids in length; κA-conotoxins are larger (31–36 amino acids) with three disulfide bridges. Purification and biochemical characterization of one peptide, κA-conotoxin MIVA is reported; five of the other predicted conotoxins were previously venom-purified. A comparative analysis of conotoxins purified from venom, and their precursors reveal novel post-translational processing, as well as mutational events leading to polymorphism. Patterns of sequence divergence and Cys codon usage define the major superfamily branches and suggest how these separate branches arose. The generation of functional novelty in proteins encoded by a gene superfamily is seldom well documented. In this report, we define the A-conotoxin superfamily, which is widely expressed in venoms of the predatory cone snails (Conus), and show how gene products that diverge considerably in structure and function have arisen within the same superfamily. A cDNA clone encoding α-conotoxin GI, the first conotoxin characterized, provided initial data that identified the A-superfamily. Conotoxin precursors in the A-superfamily were identified from six Conus species: most (11/16) encoded α-conotoxins, but some (5/16) belong to a family of excitatory peptides, the κA-conotoxins that target voltage-gated ion channels. α-Conotoxins are two-disulfide-bridged nicotinic antagonists, 13–19 amino acids in length; κA-conotoxins are larger (31–36 amino acids) with three disulfide bridges. Purification and biochemical characterization of one peptide, κA-conotoxin MIVA is reported; five of the other predicted conotoxins were previously venom-purified. A comparative analysis of conotoxins purified from venom, and their precursors reveal novel post-translational processing, as well as mutational events leading to polymorphism. Patterns of sequence divergence and Cys codon usage define the major superfamily branches and suggest how these separate branches arose. The predatory cone snails use their biochemically and pharmacologically complex venoms to capture prey, defend against predators, and compete with other animals in their environment (1Olivera B.M. Annu. Rev. Ecol. Syst. 2002; 33: 25-42Crossref Scopus (185) Google Scholar). The active components of these venoms are mostly small, disulfide-rich peptides, 10–30 amino acids in length (“conotoxins”) (2Olivera 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. Science. 1990; 249: 257-263Crossref PubMed Scopus (512) Google Scholar). As a group, cone snails (genus Conus) are very successful (>500 species), arguably the largest single genus of marine invertebrates (3Kohn A.J. Perron F.E. Life History and Biogeography Patterns in Conus. Clarendon Press, Oxford1994Google Scholar). During speciation, a remarkable hypermutation of Conus venom peptides occurs (4Woodward S.R. Cruz L.J. Olivera B.M. Hillyard D.R. EMBO J. 1990; 1: 1015-1020Crossref Scopus (172) Google Scholar, 5Olivera B.M. Mol. Biol. Cell. 1997; 8: 2101-2109Crossref PubMed Scopus (329) Google Scholar). The entire spectrum of 50–200 peptides that comprises the conotoxin repertoire of a particular Conus species is distinct from the peptide complement of the venom of any other Conus species. Thus, it is estimated that a total of >50,000 different peptides can potentially be expressed in the venoms of living Conus. A simplifying conceptual framework for addressing the formidable molecular complexity of the conotoxins is that the 50,000 peptides are encoded by relatively few gene superfamilies, which have greatly diversified within this genus. In general, each conotoxin superfamily has distinctive and conserved sequence features in the conotoxin precursors encoded by the gene superfamily. Several superfamilies have been characterized, including the O-, T-, P-, I-, and M-superfamilies that have between two and five disulfide cross-links (4Woodward S.R. Cruz L.J. Olivera B.M. Hillyard D.R. EMBO J. 1990; 1: 1015-1020Crossref Scopus (172) Google Scholar, 6Walker C. Steel D. Jacobsen R.B. Lirazan M.B. Cruz L.J. Hooper D. Shetty R. DelaCruz R.C. Nielsen J.S. Zhou L. Bandyopadhyay P. Craig A. Olivera B.M. J. Biol. Chem. 1999; 274: 30664-30671Abstract Full Text Full Text PDF PubMed Scopus (125) Google Scholar, 7Lirazan M.B. Hooper D. Corpuz G.P. Ramilo C.A. Bandyopadhyay P. Cruz L.J. Olivera B.M. Biochemistry. 2000; 39: 1583-1588Crossref PubMed Scopus (65) Google Scholar, 8Miles L.A. Dy C.Y. Nielsen J. Barnham K.J. Hinds M.G. Olivera B.M. Bulaj G. Norton R.S. J. Biol. Chem. 2002; 277: 43033-43040Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar, 9Jimenez E.C. Shetty R.P. Lirazan M. Rivier J. Walker C. Abogadie F.C. Yoshikami D. Cruz L.J. Olivera B.M. J. Neurochem. 2003; 85: 610-621Crossref PubMed Scopus (88) Google Scholar). 1G. P. Corpuz, R. B. Jacobsen, E. C. Jimenez, M. Watkins, C. Walker, C. Colledge, J. E. Garrett, D. McDougal, W. Li, W. R. Gray, D. R. Hillyard, J. Rivier, J. M. McIntosh, L. J. Cruz, and B. M. Olivera, submitted for publication. 1G. P. Corpuz, R. B. Jacobsen, E. C. Jimenez, M. Watkins, C. Walker, C. Colledge, J. E. Garrett, D. McDougal, W. Li, W. R. Gray, D. R. Hillyard, J. Rivier, J. M. McIntosh, L. J. Cruz, and B. M. Olivera, submitted for publication. The mature conotoxins in each conotoxin superfamily have a characteristic arrangement of cysteine residues, the oxidative folding of which results in a three-dimensional scaffold with a specific disulfide connectivity (“the Cys framework”). The conserved Cys superfamily framework largely determines the three-dimensional conformation of the polypeptide backbone of all peptides that belong to the superfamily. The very first toxin purified and characterized from a Conus venom was α-conotoxin GI, a competitive nicotinic antagonist from the venom of the geography cone, Conus geographus (10Gray W.R. Luque A. Olivera B.M. Barrett J. Cruz L.J. J. Biol. Chem. 1981; 256: 4734-4740Abstract Full Text PDF PubMed Google Scholar). α-GI was shown to be a 13-amino acid peptide with two disulfide cross-links. The biochemical studies on α-GI established for the first time that the biologically active constituents of Conus venoms were small, multiply cross-linked peptides, generically referred to as conotoxins. Although several conotoxin superfamilies have been described, the gene superfamily to which α-conotoxin GI belongs has not been previously defined in the literature. We show that the A-conotoxin gene superfamily, to which α-conotoxin GI belongs, is found in fish-hunting, snail-hunting, and worm-hunting species of Conus and has undergone both structural and functional diversification. Isolation and Sequencing of cDNA Clones of α-Conotoxin GI—cDNA clones encoding α-conotoxin GI were isolated from a size-fractionated cDNA library previously constructed by J. Hunspurger from Conus geographus duct mRNA using pUC13 as cloning vector and Escherichia coli strain MC1061F′laclq as the host cell. Clones were identified by screening them with a mixed oligonucleotide probe corresponding to the first six amino acids of GI (5′-GCXGGRTTRCARCAYTC-3′, where R = AorG, Y = CorT,and X = AGT or C). The probe was end-labeled with [γ-32P]ATP using T4 polynucleotide kinase (11Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning. Cold Spring Harbor Press, Cold Spring Harbor, NY1989Google Scholar) and hybridized to nitrocellulose filters for 40–60 h in amino and in and two in were identified by clones the sequence of a clones were by screening using probe corresponding to the five amino acids the of clones were in with T. E. M. PubMed Scopus Google Scholar). was purified by in Sequencing of in both was by the A. PubMed Scopus Google Scholar) using from The sequence shown in was by the cDNA with five in the and five in the sequence of the gene for α-conotoxin GI and the predicted The sequence is and the toxin is in the are in and the is In two amino acids corresponding to of six are in the by a in the sequence of in a and Sequencing of A-superfamily from Several Conus were from cDNA of Conus and C. and the mRNA from the venom of C. C. C. and C. mRNA was from the venom of several snails in each The cDNA were using the in (11Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning. Cold Spring Harbor Press, Cold Spring Harbor, NY1989Google Scholar) or the were using a of one corresponding to the sequence and the other to the the with was for using as of products were by in in with and the products were purified by in a The vector for cloning was a which we is with by cloning the and a encoding for and was by a that has a the and to with and the products were with of T4 of in the of for the vector and for the in a to of The was for for to the a of vector and was for of and of were and the was for were for the of the by of the purified with and by of the purified by of the T. E. M. PubMed Scopus Google Scholar) was using from and the Purification of κA-conotoxin Conus venom was with acid and as previously by Yoshikami D. W.R. Olivera B.M. J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar). The was on a with a other with a that all A was acid and was in The of was by from of as previously D. Olivera B.M. PubMed Scopus Google Scholar, Olivera B.M. M. Jacobsen R.B. W.R. Cruz L.J. Hillyard D.R. A. Yoshikami D. J. PubMed Google Scholar, G. Cruz L.J. Hillyard D.R. J. Rivier W.R. J. DelaCruz A. Yoshikami D. Olivera B.M. Biochemistry. PubMed Scopus Google Scholar). and Sequencing of peptide was to of of MIVA was to of MIVA and in of mixed with of and for of in was and the was for in the of acid was to on which was with a of in The peptide, to was mixed with of of the and of The was of was of the by as for the Sequencing of the peptide and the was on the the of and of a cDNA α-Conotoxin cDNA sequence of the α-conotoxin GI the established for conotoxin gene superfamilies by Hillyard and (4Woodward S.R. Cruz L.J. Olivera B.M. Hillyard D.R. EMBO J. 1990; 1: 1015-1020Crossref Scopus (172) Google the on conotoxin superfamilies was (1Olivera B.M. Annu. Rev. Ecol. Syst. 2002; 33: 25-42Crossref Scopus (185) Google Scholar). The encoding α-conotoxin GI has the characteristic including a a the by a and a the α-conotoxin GI the sequence is amino the is and the toxin comprises amino acids that are predicted to be to the 13-amino acid α-conotoxin GI, the is The generation of the mature toxin of the of the and the from the sequence to A cDNA sequence was which to be a in the α-conotoxin GI The sequence is for of six as shown in where the α-GI sequence and the cDNA encoding the as α-conotoxin are The of the six to a mature toxin predicted to have two amino acids the the sequence between α-GI and be of cDNA Clones from C. from Conus we characterized cDNA clones from the venom duct of the cone, C. a species venom has been characterized Ramilo C. W.R. R. Olivera B.M. Cruz L.J. Biochemistry. PubMed Scopus Google and several were previously purified from the venom of this species. cDNA clones were from a C. venom duct cDNA library as distinct were which are with α-conotoxin GI in As two of these previously characterized from C. are α-conotoxin and α-conotoxin The is peptide in sequence to other α-conotoxins, but with disulfide A of the amino acid of the three predicted GI, and that the sequence are In general, sequence divergence from the to of the in the is most a single amino acid that from the with divergence in the and the are three of toxin on sequence the first GI, and show sequence to each other to the other two in and The of sequence divergence is in the of the peptides of conotoxin are and the toxin are In the of the cDNA the that the in length of and and the the of to a conotoxin The sequence of α-conotoxin GI from C. geographus is for in a In to the α-conotoxin and cDNA the clone from C. encoded the of Although the predicted mature peptide has the characteristic Cys of the it is very in amino acid sequence from GI, and sequence The predicted mature peptide has not been purified from A from the analysis of the C. cDNA was a of clones that sequence to but not α-conotoxin these two of the predicted mature κA-conotoxin was previously characterized from venom G. Cruz L.J. Hillyard D.R. J. Rivier W.R. J. DelaCruz A. Yoshikami D. Olivera B.M. Biochemistry. PubMed Scopus Google Scholar). The two from two different the clones encoding were from a cDNA library from snails in the A of encoding a peptide as was from cDNA from of C. in the As shown in the of these clones have sequence to the clones that the the GI, and and κA-conotoxins and were found to be amino acid in in the is divergence between the clones that that κA-conotoxins in the encoded mature is between the and The two clones encoding κA-conotoxins are largely in the predicted amino acid sequence for the first amino acids of the the mature toxin of comprises for it is Thus, the is a total of the which is The amino acid of is predicted to be a by a As shown in is codon the in in amino acids the of the sequence in this shown in that the from to is not by a or is a of a of in the of the of the The of the of sequence between the α-conotoxin and κA-conotoxin precursors is The arrangement of Cys in the mature are different for the two the κA-conotoxins are excitatory peptides to voltage-gated ion and not nicotinic antagonists, the the of Olivera B.M. Annu. Rev. 1999; PubMed Scopus Google Scholar). Thus, the two of peptides are in and the of their and belong to the same gene superfamily, have to two of conotoxins that are both and of cDNA Clones from Conus which of the A-conotoxin superfamily are expressed in other we C. the cone, and Conus the cone The results of the analysis are shown in C. one encoding a sequence for α-conotoxin was all of these encoded α-conotoxin α-Conotoxin has been characterized Yoshikami D. W.R. Olivera B.M. J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar, Rivier Olivera B.M. Biochemistry. 1997; PubMed Scopus Google Scholar) and shown not to with the nicotinic the two nicotinic and Yoshikami D. W.R. Olivera B.M. J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar, M. M. A. C. J. 2002; PubMed Google Scholar, A. M. L. J. 2002; PubMed Google Scholar). The A-superfamily sequence from C. encoded a peptide that sequence to κA-conotoxin which we are κA-conotoxin As be we have characterized κA-conotoxin MIVA from C. The C. cDNA library not any α-conotoxin all of the clones that encoded precursors of peptides to κA-conotoxins and MIVA κA-conotoxins and sequence and are to be The predicted are shown in of the A-superfamily in Conus the A-superfamily in Conus species that not on to the A-superfamily conotoxins expressed in the Conus species we identified A-superfamily from cDNA from a Conus and from a Conus the of M. Cruz L.J. Olivera B.M. 39: PubMed Scopus Google all of the species belong to or C. belongs to the and C. belongs to the worm-hunting The analysis of cDNA clones three different of for C. and two for C. is that all five from these species are predicted to of the clones encoded a Conus peptide as κA-conotoxin Thus, in to species as C. C. and C. the species we have are in encoding of the C. or C. peptides have been purified from one of the C. peptides, encoded by clone was as and on six different nicotinic expressed in by previously Yoshikami D. W.R. Olivera B.M. J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar) and The peptide not any of the nicotinic for all The of be to of the peptide for a the that a post-translational in the peptide is for is Isolation and of κA-conotoxin excitatory peptide that and the same as κA-conotoxin was purified from C. venom As shown in initial of the peptide a acid peptide with a in and in and The sequence is to that predicted from a cDNA clone from C. sequence in that and are and of predicted from cDNA predicted sequence that is to of that is in and and are that is in and and are of κA-conotoxin are is is The predicted sequence that is to that is in and and are The are is is in a of the peptide three and are the same for where a for was found in and a of was found in which is is is with the data that a of between and In was the in to the the with that the as of this was by The of the peptide by = was the predicted for the peptide in and the was to where the two predicted from cDNA clones as is with the two and The of can be between two on and The of amino acid and data the sequence of κA-conotoxin MIVA as The pharmacologically active peptides of Conus venoms have shown remarkable The different peptides found in Conus venoms can be Conus peptide in several different Conus belong to a Conus peptide gene superfamily. Conus peptide gene superfamily has a of the most of which is the conserved In to of gene in which the the are the conserved in conotoxin gene superfamilies the is are the most conserved Thus, the sequence is of which superfamily a particular Conus peptide belongs The A-conotoxin superfamily is defined by from species of Conus C. C. and C. as well as from two species a and C. a worm-hunting species cDNA clones that A-conotoxin superfamily these precursors have the and a mature toxin The acid sequence for the sequence is found any in a of the different conserved sequence as a for of the A-conotoxin gene superfamily. In is a divergence of mature conotoxin a specific α-conotoxin and κA-conotoxin MIVA both from C. have but is in the mature toxin of and amino acid of sequence and mature toxin In amino acids are post-translational are to be to isolated are in A. in a shown in is a of the of the sequence with the of mature toxin is that not are the amino acids of the but codon usage is some of the are from of Conus species to have in the or (1Olivera B.M. Annu. Rev. Ecol. Syst. 2002; 33: 25-42Crossref Scopus (185) Google Scholar, M. Cruz L.J. Olivera B.M. 39: PubMed Scopus Google the of is The of the superfamily have a conserved the show to the A-superfamily (4Woodward S.R. Cruz L.J. Olivera B.M. Hillyard D.R. EMBO J. 1990; 1: 1015-1020Crossref Scopus (172) Google Scholar). Thus, each Conus peptide superfamily has characteristic A of of a particular superfamily is the arrangement of cysteine residues, which in the framework of the peptides to that superfamily. Although most in the mature toxin are the cysteine are in general, each Cys one of the two is a previously for other conotoxins E. M. Mol. Biol. PubMed Scopus Google Scholar). are α-GI and the first two Cys have in codon usage from all of the other A-superfamily conotoxins in this in GI and in all The of this codon are be that and in the conotoxins use for the first Cys all other Cys are encoded by in most other A-superfamily between toxin and predicted amino acids are α-Conotoxins and and κA-conotoxin were previously isolated by Ramilo C. W.R. R. Olivera B.M. Cruz L.J. Biochemistry. PubMed Scopus Google Ramilo C.A. L. Yoshikami D. W.R. R. J. G. Olivera B.M. Cruz L.J. Biochemistry. PubMed Scopus Google and Craig G. Cruz L.J. Hillyard D.R. J. Rivier W.R. J. DelaCruz A. Yoshikami D. Olivera B.M. Biochemistry. PubMed Scopus Google mature mature mature mature in a conotoxins from their precursors by a characteristic sequence two amino or to the of the mature the as are to by a well characterized R.C. J. Biol. Chem. Full Text PDF PubMed Google Scholar, D. J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar). The data suggest that both can be complex for some of the A-conotoxin superfamily. where one not have predicted a from the cDNA clone mature conotoxin sequence be found in venom are shown in Thus, for α-conotoxin is the α-conotoxin the by a of the The most complex of post-translational is in κA-conotoxin as in The the is but the initial by the of by a and of the to In in to in which amino acids are κA-conotoxin MIVA has amino that have been a that has been to and two of and that be In two peptides of this family purified from venom, and a from Conus P. D. J. A. Biochemistry. 1999; PubMed Scopus Google a single is not this is to a or a in post-translational is that in and all peptides MIVA and the Conus is a or that is and are not G. Cruz L.J. Hillyard D.R. J. Rivier W.R. J. DelaCruz A. Yoshikami D. Olivera B.M. Biochemistry. PubMed Scopus Google Scholar, P. D. J. A. Biochemistry. 1999; PubMed Scopus Google Scholar). κA-conotoxin MIVA is the first of this family characterized that to have two The the first Cys and the of post-translational in κA-conotoxins this of the A-superfamily from the α-conotoxin Thus, the A-conotoxin superfamily comprises two well biochemically of peptides, the α-conotoxin family and the κA-conotoxin The are to nicotinic which belong to the superfamily, the of the voltage-gated ion superfamily. of Conotoxin results of this conotoxin several The of the encoding the different conotoxins are different A-superfamily or from different Conus species. are a few in which different be including κA-conotoxins κA-conotoxins and several of these as κA-conotoxins are a few amino acid the two mature are these mostly from single the α-conotoxin are two amino acids in with but the two are this have by the of six to the α-GI or the of six from the gene or The most in mature toxin is between κA-conotoxins and in this amino acids are in with As is in a of in one gene has been by a in the other gene In or to the the are in and is that κA-conotoxin was found in C. from the these have a different from to this species. is a in and was with to the κA-conotoxins were found for the α-conotoxin and cDNA clones were from both of C. and sequence were between the two for these of the most from this was the that peptides to the A-superfamily not have a single arrangement of cysteine in the mature conotoxin a of other conotoxin superfamilies (1Olivera B.M. Annu. Rev. Ecol. Syst. 2002; 33: 25-42Crossref Scopus (185) Google Scholar, Olivera B.M. Rev. PubMed Scopus Google Scholar). Thus, the and the have the Cys and α-Conotoxin GI and all other have the cysteine in the sequence as Cys to the for In all peptides with this the disulfide connectivity is which we to as the scaffold (1Olivera B.M. Annu. Rev. Ecol. Syst. 2002; 33: 25-42Crossref Scopus (185) Google Scholar). results that a of peptides, with a different Cys Cys belongs to the this is by κA-conotoxin these in Cys and the κA-conotoxins be to have different and polypeptide backbone the peptides are and not nicotinic antagonists, the two are both and a is a between two branches of the α-conotoxin the sequence and the sequence is C. and C. belong to the same of cone the α-conotoxin is not the sequence of the to and from C. to any of the in the first amino α-conotoxin sequence divergence from two predicted from C. and C. cDNA clones and in and the first of the of these are three can be the conotoxin the conotoxin and the κA-conotoxin peptides that belong within the same not in sequence in the in by three amino in peptides different are diverge by of amino acids in the in the peptides and not belong to any of the and are from each other and define within the in The the of amino acids for a of conotoxins. The and between within a and between of in a the is widely in fish-hunting, and it has been that this be the Cys of the α-conotoxin family (1Olivera B.M. Annu. Rev. Ecol. Syst. 2002; 33: 25-42Crossref Scopus (185) Google Scholar). The is it is found in two of Conus and is as a that have arisen in the of the genus. the from the arrangement of shown in for α-conotoxin how the from in the first to in the first have peptide was from by to a Cys codon the first two Cys in peptide, this the from to three this be by the of cysteine as is found in α-conotoxin of the first cysteine codon in α-conotoxin and all in the codon in GI and the where is a codon that not The be a in the of the first of from to with in the of the Cys is shown in Thus, the codon can be by the first Cys in conotoxins from the Cys of the both of which use the In the κA-conotoxins have the conserved for the first two Cys as is found in the α-conotoxin and all are encoded by in all κA-conotoxins are for Cys for the first two in any with this codon usage is a to that to the divergence between κA-conotoxins and κA-conotoxin have been a Cys of conotoxin were by a three Cys all for a specific of how this have In this we have defined the A-superfamily and several major one of which is and the analysis of has events as that suggest how superfamily branches have Molecular that events as that toxin divergence are not in this is that A-superfamily have a in the of the B. Olivera, and D. Hillyard, it was previously that the found in conotoxin superfamily a in the hypermutation in these peptides as occurs B.M. Walker C. Hooper D. R. Shetty R. M. Bandyopadhyay P. Hillyard D.R. 1999; PubMed Scopus Google Scholar). a of the from A-superfamily from different Conus species has not been and this has not been events from or the analysis that as structural and functional novelty within a single conotoxin gene superfamily.