New conotoxins define the novel I3-superfamily.

We purified two novel conotoxins, designated as ca11a and ca11b, from the venom of Conus caracteristicus. Based on the amino acid sequence of mature ca11a, we cloned its full-length cDNA. Based on the signal peptide of ca11a, several ca11a-like conotoxins were cloned from C. caracteristicus and C. pulicarius. These novel conotoxins have an I-superfamily cysteine pattern but with a novel signal peptide sequence, suggesting they belong to a new branch of I-superfamily, designated as I(3)-superfamily. Additionally, two O-superfamily conotoxins were also cloned based on the signal peptide of ca11a, suggesting a possible evolutionary relationship between O- and I-superfamilies.

Isolation and cloning of a conotoxin with a novel cysteine pattern from Conus caracteristicus.

A new conotoxin, ca16a, containing 8 cysteine residues was purified, sequenced, and cloned from a worm-hunting snail, Conus caracteristicus. This conotoxin is an extremely hydrophilic peptide comprising 34 residues, with 4 acidic and 4 basic residues. It is rich in polar Gly, Ser, and Thr residues and includes a hydroxylated Pro residue. The cysteine arrangement pattern of ca16a (-C-C-CC-C-CC-C-, designated as framework #16) is distinct from that of other known conotoxins. Furthermore, the signal peptide sequence of this conotoxin does not share any homology with those of other conotoxins. Leu residues account for almost 50% of its 20-residue signal peptide. The unique cysteine framework and signal peptide sequence of ca16a suggest that it belongs to a new conotoxin superfamily.

Two different groups of signal sequence in M-superfamily conotoxins.

M-superfamily conotoxins can be divided into four branches (M-1, M-2, M-3 and M-4) according to the number of amino acid residues in the third Cys loop. In general, it is widely accepted that the conotoxin signal peptides of each superfamily are strictly conserved. Recently, we cloned six cDNAs of novel M-superfamily conotoxins from Conus leopardus, Conus marmoreus and Conus quercinus, belonging to either M-1 or M-3 branch. These conotoxins, judging from the putative peptide sequences deducted from cDNAs, are rich in acidic residues and share highly conserved signal and pro-peptide region. However, they are quite different from the reported conotoxins of M-2 and M-4 branches even in their signal peptides, which in general are considered highly conserved for each superfamily of conotoxins. The signal sequences of M-1 and M-3 conotoxins composed of 24 residues start with MLKMGVVL-, while those of M-2 and M-4 conotoxins composed of 25 residues start with MMSKLGVL-. It is another example that different types of signal peptides can exist within a superfamily besides the I-conotoxin superfamily. In addition to the different disulfide connectivity of M-1 conotoxins from that of M-4 or M-2 conotoxins, the sequence alignment, preferential Cys codon usage and phylogenetic tree analysis suggest that M-1 and M-3 conotoxins have much closer relationship, being different from the conotoxins of other two branches (M-4 and M-2) of M-superfamily.

From the identification of gene organization of alpha conotoxins to the cloning of novel toxins.

In the venoms of cone snails, alpha conotoxins are competitive antagonists of nicotinic acetylcholine receptors. Eleven novel cDNA and eight partial gene sequences (including two pseudogenes) of alpha conotoxins were identified from five species of cone snail. As expected, every cDNA encodes a precursor of prepropeptide. In all the partial genes of alpha conotoxins identified, there is a long intron inserted at a fixed position in the pro-region, dividing the encoding region into two exons. The mutation rate in exon I (encoding the signal peptide and a part of pro-region) is much lower than that in exon II (encoding the other part of pro-region, the mature peptide and 3' untranslational region). Interestingly, the sequences at the 5' and 3' end of introns are highly conserved. In addition, in the identified introns exist long dinucleotide (e.g. "GT", "CA") or trinucleotide ("CAT") repeats. In the special case of Pu 1.1, there are five almost identical repeats of a 150 bp sequence in the long intron. Taking advantage of the conserved 3' end sequence of intron, 16 alpha conotoxins, as well as a pseudogene and three kappa A conotoxins, were identified from their genomic DNAs. Based on the comparison of these cDNA and gene sequences, a hypothesis of the alpha conotoxin evolution was proposed.

Characterization of novel M-superfamily conotoxins with new disulfide linkage.

The M-superfamily with the typical Cys framework (-CC-C-C-CC-) is one of the seven major superfamilies of conotoxins found in the venom of cone snails. Based on the number of residues in the last Cys loop (between C4 and C5), M-superfamily conotoxins can be provisionally categorized into four branches (M-1, M-2, M-3, M-4) [Corpuz GP, Jacobsen RB, Jimenez EC, Watkins M, Walker C, Colledge C, Garrett JE, McDougal O, Li W, Gray WR, et al. (2005) Biochemistry44, 8176-8186]. Here we report the purification of seven M-superfamily conotoxins from Conus marmoreus (five are novel and two are known as mr3a and mr3b) and one known M-1 toxin tx3a from Conus textile. In addition, six novel cDNA sequences of M-superfamily conotoxins have been identified from C. marmoreus, Conus leopardus and Conus quercinus. Most of the above novel conotoxins belong to M-1 and M-2 and only one to M-3. The disulfide analyses of two M-1 conotoxins, mr3e and tx3a, revealed that they possess a new disulfide bond arrangement (C1-C5, C2-C4, C3-C6) which is different from those of the M-4 branch (C1-C4, C2-C5, C3-C6) and M-2 branch (C1-C6, C2-C4, C3-C5). This newly characterized disulfide connectivity was confirmed by comparing the HPLC profiles of native mr3e and its two regioselectively folded isoforms. This is the first report of three different patterns of disulfide connectivity in conotoxins with the same cysteine framework.