Three-Finger Toxin Diversification in the Venoms of Cat-Eye Snakes (Colubridae: Boiga)

The Asian genus Boiga (Colubridae) is among the better studied non-front-fanged snake lineages, because their bites have minor, but noticeable, effects on humans. Furthermore, B. irregularis has gained worldwide notoriety for successfully invading Guam and other nearby islands with drastic impacts on the local bird populations. One of the factors thought to allow B. irregularis to become such a noxious pest is irditoxin, a dimeric neurotoxin composed of two three-finger toxins (3FTx) joined by a covalent bond between two newly evolved cysteines. Irditoxin is highly toxic to diapsid (birds and reptiles) prey, but roughly 1000 × less potent to synapsids (mammals). Venom plays an important role in the ecology of all species of Boiga, but it remains unknown if any species besides B. irregularis produce irditoxin-like dimeric toxins. In this study, we use transcriptomic analyses of venom glands from five species [B. cynodon, B. dendrophila dendrophila, B. d. gemmicincta, B. irregularis (Brisbane population), B. irregularis (Sulawesi population), B. nigriceps, B. trigonata] and proteomic analyses of B. d. dendrophila and a representative of the sister genus Toxicodryas blandingii to investigate the evolutionary history of 3FTx within Boiga and its close relative. We found that 92.5% of Boiga 3FTx belong to a single clade which we refer to as denmotoxin-like because of the close relation between these toxins and the monomeric denmotoxin according to phylogenetic, sequence clustering, and protein similarity network analyses. We show for the first time that species beyond B. irregularis secrete 3FTx with additional cysteines in the same position as both the A and B subunits of irditoxin. Transcripts with the characteristic mutations are found in B. d. dendrophila, B. d. gemmicincta, B. irregularis (Brisbane population), B. irregularis (Sulawesi population), and B. nigriceps. These results are confirmed by proteomic analyses that show direct evidence of dimerization within the venom of B. d. dendrophila, but not T. blandingii. Our results also suggest the possibility of novel dimeric toxins in other genera such as Telescopus and Trimorphodon. All together, this suggests that the origin of these peculiar 3FTx is far earlier than was appreciated and their evolutionary history has been complex.