Xhibited highest identity to cytotoxin 1 (97 ) [43], cytotoxin 2 (89 ) and cytotoxin 3 (84 ) [44], purified from Hemachatus haemachatus venom. Hemachatoxin differs from cytotoxin 1 [43] in two amino acid positions (Leu27Met28 is replaced by Met27Leu28). This difference was confirmed by ESI-MS (CNBr cleavage site and mass of peptides, Table S1), Edman degradation (Figure S3A, S3B and S3C) and electron density map (see below).Hemachatoxin from Ringhals Cobra VenomFigure 3. Structure of hemachatoxin. (A) Ribbon representation of the hemachatoxin 374913-63-0 site monomer. Cysteine bonds are shown in yellow. b-strands, N- and C- terminals are labeled. (B) Electron density map. A sample final 2Fo-Fc map of hemachatoxin shows the region from Tyr23 to Lys29. The map is I-BRD9 contoured at a level of 1s. (C) The electrostatic surface potential of hemachatoxin is shown in the same orientation as Figure 3A. Blue indicates positive potential and red indicates negative potential in units kT/e. All the structure related figures of this paper were prepared using the program PyMol [77]. doi:10.1371/journal.pone.0048112.gresidues of hemachatoxin as well as its identity to cardiotoxins/ cytotoxins (Figure 4A). Also, hemachatoxin shared the common three-finger fold and molecular shape when compared to its structural homologues (Figure 4B) [46].DiscussionThe three-dimensional structures of snake venom 3FTxs, particularly that of neurotoxins [15,20,47,48] and cardiotoxins/ cytotoxins [49?2] have been extensively studied. Here we report the structural characterization of a new 3FTx, hemachatoxin from the venom of H. haemachatus. The structural analyses indicate that hemachatoxin belongs to cardiotoxin/cytotoxin subgroup of 3FTx family. It exhibited 97 sequence identity to cytotoxin 1 [43], whose crystal structure has not been determined. ESI-MS, Edman degradation and crystal structure data indicates that hemachatoxin differs from cytotoxin 1 in two amino acid positions (Leu27Met28 is replaced by Met27Leu28) and hence are isoforms. Multiple isoforms of 3FTxs are known to be present in single snake venom [53,54]. As mentioned in the introduction section, 3FTxs, including hemachatoxin, share overall structural similarity (Figure 4B), but they differ from each other in their biological activities. Subtle variations in the size and conformation of b-sheet loops dictate the biological specificities in 3FTxs. For example, the well characterized long-chain (e.g. a-cobratoxin, a-bungarotoxin) and shortchain (e.g. erabutoxin a, toxin-a) neurotoxins that differ in loop size and length of C-terminal extension, exhibit distinct specificity for nAChR subtypes. Short-chain neurotoxins has a longer loop I (12?3 amino acid residues (aa) vs. 9?2 aa in long-chain neurotoxins), a shorter loop II (15?6 aa vs. 19?0 aa in longchain neurotoxins) and C-terminal tail (2 aa vs. 7?4 aa in longchain neurotoxins) when compared to long-chain neurotoxins. This longer loop I of short-chain neurotoxins contains key functional residues that are important for recognizing the nicotinicacetylcholine receptor [55,56], while shorter loop I of long-chain neurotoxins lacks these functional residues. The long C-terminal tail appears 1313429 to `substitute’ for the loop I functional residues and contribute to the receptor binding [57,58]. The deletion of this Cterminal tail reduces the binding affinity [59,60]. Similarly, the difference in the conformations of the three loops appears to dictate the biological specificities of these.Xhibited highest identity to cytotoxin 1 (97 ) [43], cytotoxin 2 (89 ) and cytotoxin 3 (84 ) [44], purified from Hemachatus haemachatus venom. Hemachatoxin differs from cytotoxin 1 [43] in two amino acid positions (Leu27Met28 is replaced by Met27Leu28). This difference was confirmed by ESI-MS (CNBr cleavage site and mass of peptides, Table S1), Edman degradation (Figure S3A, S3B and S3C) and electron density map (see below).Hemachatoxin from Ringhals Cobra VenomFigure 3. Structure of hemachatoxin. (A) Ribbon representation of the hemachatoxin monomer. Cysteine bonds are shown in yellow. b-strands, N- and C- terminals are labeled. (B) Electron density map. A sample final 2Fo-Fc map of hemachatoxin shows the region from Tyr23 to Lys29. The map is contoured at a level of 1s. (C) The electrostatic surface potential of hemachatoxin is shown in the same orientation as Figure 3A. Blue indicates positive potential and red indicates negative potential in units kT/e. All the structure related figures of this paper were prepared using the program PyMol [77]. doi:10.1371/journal.pone.0048112.gresidues of hemachatoxin as well as its identity to cardiotoxins/ cytotoxins (Figure 4A). Also, hemachatoxin shared the common three-finger fold and molecular shape when compared to its structural homologues (Figure 4B) [46].DiscussionThe three-dimensional structures of snake venom 3FTxs, particularly that of neurotoxins [15,20,47,48] and cardiotoxins/ cytotoxins [49?2] have been extensively studied. Here we report the structural characterization of a new 3FTx, hemachatoxin from the venom of H. haemachatus. The structural analyses indicate that hemachatoxin belongs to cardiotoxin/cytotoxin subgroup of 3FTx family. It exhibited 97 sequence identity to cytotoxin 1 [43], whose crystal structure has not been determined. ESI-MS, Edman degradation and crystal structure data indicates that hemachatoxin differs from cytotoxin 1 in two amino acid positions (Leu27Met28 is replaced by Met27Leu28) and hence are isoforms. Multiple isoforms of 3FTxs are known to be present in single snake venom [53,54]. As mentioned in the introduction section, 3FTxs, including hemachatoxin, share overall structural similarity (Figure 4B), but they differ from each other in their biological activities. Subtle variations in the size and conformation of b-sheet loops dictate the biological specificities in 3FTxs. For example, the well characterized long-chain (e.g. a-cobratoxin, a-bungarotoxin) and shortchain (e.g. erabutoxin a, toxin-a) neurotoxins that differ in loop size and length of C-terminal extension, exhibit distinct specificity for nAChR subtypes. Short-chain neurotoxins has a longer loop I (12?3 amino acid residues (aa) vs. 9?2 aa in long-chain neurotoxins), a shorter loop II (15?6 aa vs. 19?0 aa in longchain neurotoxins) and C-terminal tail (2 aa vs. 7?4 aa in longchain neurotoxins) when compared to long-chain neurotoxins. This longer loop I of short-chain neurotoxins contains key functional residues that are important for recognizing the nicotinicacetylcholine receptor [55,56], while shorter loop I of long-chain neurotoxins lacks these functional residues. The long C-terminal tail appears 1313429 to `substitute’ for the loop I functional residues and contribute to the receptor binding [57,58]. The deletion of this Cterminal tail reduces the binding affinity [59,60]. Similarly, the difference in the conformations of the three loops appears to dictate the biological specificities of these.