Is Hybridization a Source of Adaptive Venom Variation in Rattlesnakes? A Test, Using a Crotalus scutulatus × viridis Hybrid Zone in Southwestern New Mexico

Abstract

Venomous snakes often display extensive variation in venom composition both between and within species. However, the mechanisms underlying the distribution of different toxins and venom types among populations and taxa remain insufficiently known. Rattlesnakes (Crotalus, Sistrurus) display extreme inter- and intraspecific variation in venom composition, centered particularly on the presence or absence of presynaptically neurotoxic phospholipases A2 such as Mojave toxin (MTX). Interspecific hybridization has been invoked as a mechanism to explain the distribution of these toxins across rattlesnakes, with the implicit assumption that they are adaptively advantageous. Here, we test the potential of adaptive hybridization as a mechanism for venom evolution by assessing the distribution of genes encoding the acidic and basic subunits of Mojave toxin across a hybrid zone between MTX-positive Crotalus scutulatus and MTX-negative C. viridis in southwestern New Mexico, USA. Analyses of morphology, mitochondrial and single copy-nuclear genes document extensive admixture within a narrow hybrid zone. The genes encoding the two MTX subunits are strictly linked, and found in most hybrids and backcrossed individuals, but not in C. viridis away from the hybrid zone. Presence of the genes is invariably associated with presence of the corresponding toxin in the venom. We conclude that introgression of highly lethal neurotoxins through hybridization is not necessarily favored by natural selection in rattlesnakes, and that even extensive hybridization may not lead to introgression of these genes into another species.

Extremely Divergent Haplotypes in Two Toxin Gene Complexes Encode Alternative Venom Types within Rattlesnake Species

“…Two plausible explanations that could account for these observations are (1) the Band APLA2 haplotypes were both present in a polymorphic common ancestor of C. atrox and C. scutulatus and sorted differently between the lineages, or (2) the type B haplotype may have been introduced into the C. scutulatus lineage from C. atrox or some other hemorrhagic species at some time in the past. We note that viable hybrids between C. scutulatus with C. atrox have been obtained in captivity [41] and that hybridization between C. scutulatus and C. viridis in the wild has been well documented [25, 42]. However, incomplete lineage sorting and past hybridization events are notoriously difficult to disentangle [43–46] and will require more extensive sampling of both species to assess.…”

“…One key issue is the genetic basis of the A/B polymorphisms and whether differences in venom type are due to differences in toxin gene structure, number, or regulation. The failure to amplify one or more neurotoxin gene subunits in type B C. scutulatus [24, 25]and type B C. horridus [26] have raised the possibility that gene content varies between venom types. On the other hand, structural differences in MP genes have also been proposed to underlie alternative venom types [27].…”

When one phenotype is not enough – divergent evolutionary trajectories govern venom variation in a widespread rattlesnake species

“…To understand whether differences at the genome level, i.e. presence or absence of coding genes, explain differential expression of venom proteins in the phenotypes at the population level, as previously documented for MTX and SVMPs [14-15], we compared genomic and venom proteome profiles of 50 individuals from multiple localities, and PCR-amplified 14 toxin genes belonging to five gene families (Table S1). We discovered that proteomic presence or absence of individual toxins was invariably associated with presence or absence of the coding genes.…”

“…All primer sets were first tested on the representative individuals; finally, 12 toxin genes were selected for further investigation (Table S1). In addition, we used previously designed primers for the acidic (MTXa) and basic (MTXb) subunit genes of Mojave toxin [14]. In total, 98-163 individuals were screened for toxin gene presence, PCR products were checked on 1.5% agarose gel, and a subset were sequenced to verify consistency of primer specificity across the geographic distribution.…”

“…Run quality, peak calling and allele size scoring was performed with the GeneMapper 4.0 software (Applied Biosystems). The Mohave rattlesnake is known to hybridize in the wild with the Prairie rattlesnake, C. viridis , where the ranges of the two species come into contact in south-western New Mexico [14]. To avoid bias or errors in estimates of population genetic structure and toxin gene distribution, all potential hybrids were removed from the analysis.…”