Elsevier

Toxicon

Volume 53, Issue 6, May 2009, Pages 672-679
Toxicon

Functional basis of a molecular adaptation: Prey-specific toxic effects of venom from Sistrurus rattlesnakes

https://doi.org/10.1016/j.toxicon.2009.01.034Get rights and content

Abstract

Understanding the molecular bases of adaptations requires assessing the functional significance of phenotypic variation at the molecular level. Here we conduct such an assessment for an adaptive trait (snake venom proteins) which shows high levels of interspecific variation at the molecular level. We tested the toxicity of venom from four taxa of Sistrurus rattlesnakes with different diets towards 3 representative prey (mice, lizards and frogs). There were significant differences among prey in their overall susceptibility to Sistrurus venom, with frogs being an order of magnitude more resistant than mice or lizards. However, only in mice was there substantial variation in the toxicity of venom from different Sistrurus taxa, with the variation being roughly correlated with the incidence of mammals in the snake's diet. A comparative analysis using published data of the toxicity of rattlesnake and outgroup (Agkistrodon) venoms to mice confirms that both the gain and loss of toxicity to mammals were major modes of venom evolution in Sistrurus catenatus and Sistrurus miliarius. Our findings identify toxicity to mammals as a major axis along which venom evolution has occurred among Sistrurus rattlesnakes, with little evidence for evolutionary changes in toxicity towards the other prey tested. They also emphasize the need to consider ecological and evolutionary factors other than diet alone as causes of variation in venom toxicity.

Introduction

Identifying the molecular basis of adaptations in natural populations is an important yet largely unrealized goal in evolutionary biology, despite its potential to address fundamental questions about the role of different types of selective and genetic mechanisms as the basis for adaptive variation in phenotype (Golding and Dean, 1998, Orr, 2005). A key step in this research approach is identifying the functional significance of phenotypic variation at the molecular level. This will be most successful in systems where the possible function of the variation can be narrowly defined due to the nature of the adaptation. In this sense, predator–prey systems offer such a clearly defined phenotypic interface because the functional goals of the traits directly involved in killing the prey by the predator or resisting predation by the prey are clear (Brodie and Brodie, 1999).

Venoms produced by snakes in the Colubroidea are an example of a trait in a predator which shows high levels of variation at the molecular level and which also has a clearly defined function, namely the capture and digestion of prey. Venomous snakes such as rattlesnakes produce a complex mixture of up to 40 distinct proteins of several different families (Mackessy, 2008). Specialized venom glands located in the upper jaw synthesize and store venom which is then injected into prey via long, hollow fangs. Detailed and comprehensive characterization of the genes that underlie this variation and of the proteins they encode are becoming increasingly common for snakes (e.g. Sanz et al., 2006, Pahari et al., 2007, Gibbs and Rossiter, 2008), but functional characterization of this variation, in terms of effects on prey, still is poorly characterized. In general, the working hypothesis is that the high level of variation in venom at the inter- or intraspecific level (for a review see Chippaux et al., 1991) allows snakes to specialize on different prey (e.g. Mackessy, 1988, Daltry et al., 1996). Support for this hypothesis has come from studies showing a correlation between diet and venom variation in adult snakes (Daltry et al., 1996), associations between ontogenetic shifts in diet and venom composition (Mackessy, 1988), and more rarely, direct tests which have shown that venom produced by a particular age class of snake is most toxic to its preferred prey (e.g. Mackessy, 1988, Andrade and Abe, 1999, Jorge da Silva and Aird, 2001, Urdaneta et al., 2002, Mackessy et al., 2006).

However, studies showing associations between diet and venom have been criticized because they rarely test the key assumption that differences in venom composition are in fact correlated with increased toxicity towards more commonly consumed prey (Sasa, 1999, Mebs, 1999). Other research has also found no association between venom composition and diet (Williams et al., 1988). Further, a few studies with front-fanged snakes (viperids and elapids) that have conducted direct tests of venom toxicity on a range of prey have yielded variable results, ranging from positive associations (see above) to negative associations between toxicity and prey preference (Heatwole and Poran, 1995, Heatwole and Powell, 1998, Mebs, 2001) possibly due to coevolutionary interactions between snakes and their prey. The venom of the Brown Treesnake (Boiga irregularis), a rear-fanged colubrid snake, has been shown to have taxon-specific effects, with preferred prey (lizards and birds) being an order of magnitude more sensitive to the venom than mice (Mackessy et al., 2006). Further, a highly specific toxin from the venom of this species, which comprises ∼10% of the total venom, is very potent towards lizards and birds but is non-toxic to mammals (Pawlak et al., 2009). Thus, the functional association between venom composition and diet, particularly for front-fanged snakes, remains unclear, likely because ecological and evolutionary factors other than selection in relation to diet can potentially influence venom composition in different species (Sasa, 1999, Wüster et al., 1999, Mebs, 2001).

Given this uncertainty, we feel that one productive way forward is to conduct comprehensive studies on the causes and functional consequences of venom composition in a small group of phylogenetically similar species that nonetheless show high levels of variation in diet. One such group is rattlesnakes in the genus Sistrurus, which inhabit a range of ecologically diverse habitats across North America (Campbell and Lamar, 2004). Here we report on the toxicity of venom from the four taxa of Sistrurus rattlesnakes (Sistrurus miliarius barbouri [Pygmy Rattlesnake], Sistrurus catenatus catenatus, Sistrurus catenatus tergeminus, and Sistrurus catenatus edwardsii [Eastern, Western, and Desert Massasauga rattlesnakes, respectively]) with different diets towards 3 representative prey (mice, lizards, and frogs). Recent phylogenetic analyses based on mitochondrial and nuclear DNA indicate that S. miliarius is basal to all three S. catenatus subspecies, whereas the named S. catenatus subspecies fall into two distinct clades: one consisting of S. c. catenatus alone and the other consisting of both S. c. tergeminus and S. c. edwardsii (Kubatko and Gibbs, unpublished data). This work complements our recent efforts to accumulate detailed information on the genetic and proteomic basis of venom variation in this group of snakes (Sanz et al., 2006, Pahari et al., 2007, Gibbs and Rossiter, 2008) and place it in the context of venom evolution in all rattlesnakes (Mackessy, 2008).

Diet studies show that different taxa of Sistrurus rattlesnakes vary in the degree to which they specialize on endothermic vs. ectothermic prey (Holycross and Mackessy, 2002; T.M. Farrell and P.G. May, unpublished data). Specifically, there are snakes that largely specialize on mammals (S. c. catenatus) vs. frogs and lizards (S. m. barbouri) as well as snakes that bridge this dietary transition by eating mammals, lizards and frogs (S. c. tergeminus and S. c. edwardsii). Previous researchers (Daltry et al., 1996, Chijiwa et al., 2003) have argued that physiological features of prey related to their thermoregulatory strategy (e.g. body temperature, muscle physiology, or aerobic vs. anaerobic escape locomotion – see Wilmer et al., 2004) may exert a significant selection pressure for distinct venom proteins. Earlier studies have documented toxicity of Sistrurus venoms towards mice, but only for a limited set of taxa and without comparisons of toxicity towards non-mammalian prey (Githens, 1935, Minton, 1956, Kocholaty et al., 1971).

Our goals in this study were as follows: using test animals that were representative of the major classes of prey consumed by different Sistrurus taxa, we conducted LD50 studies to determine (1) if prey-specific effects are present and if so, how they vary across taxa in relation to diet, and (2) if toxicity towards different prey covaries (which would support trade-offs in toxicity towards different prey as a mechanism underlying patterns of toxicity among taxa). Finally, using published data and comparative analyses, we address the question of how toxicity towards mammals evolved in these rattlesnakes.

Section snippets

Sistrurus venom samples

Venoms were extracted manually from single adult snakes from three subspecies of S. catenatus and from S. m. barbouri using standard methods (Mackessy, 1988). Snakes were from the following locations: S. c. catenatus, Killdeer Plains Wildlife Area, Wyandot County, Ohio; S. c. tergeminus, Cheyenne Bottoms Wildlife Area, Barton County, Kansas: S. c. edwardsii, Lincoln County, Colorado and S. m. barbouri, central Florida. Protein concentration of each venom sample was assayed in triplicate

Prey preference

Differential utilization of prey is apparent for the four taxa of Sistrurus (Fig. 1). Mammals are the main prey type taken by S. c. catenatus, whereas lizards and frogs (anurans) are the primary prey of S. m. barbouri. For S. c. tergeminus, mammals followed by lizards are the preferred prey, whereas the opposite is seen for S. c. edwardsii. Overall, as a species, S. catenatus includes a greater proportion of mammals in its diet than does S. miliarius (63% vs. 15%, respectively).

Prey-specific effects

Table 1 shows IP

Experimental issues

Our demonstration of strong prey-specific effects of Sistrurus venoms is dependent on a number of important assumptions related to the design of our study. In particular, because our goal was an initial survey of toxicity across broad categories of prey that are representative of animals consumed by Sistrurus, we limited our tests to easily-obtained animals that may or may not show sensitivities to venom that are representative of other species of that type. This assumption needs to be tested

Acknowledgements

We thank Doug Wynn and the Gibbs Lab for field assistance, Todd Castoe, Liam Kean, Laura Kubatko and Roman Lanno for help with analyses, and Terry Farrell and Peter May for allowing access to extensive unpublished data on the diet of S. m. barbouri in Florida. Scientific collecting permits were issued by Colorado Division of Wildlife (0456) and Kansas Wildlife and Parks (SC-147-96) to S.P.M. This research was supported by funds from the Ohio State University and the University of Northern

References (48)

  • V. Williams et al.

    Variation in venom proteins from isolated populations of tiger snakes (Notechis ater niger, N. scutatus) in south Australia

    Toxicon

    (1988)
  • D.V. Andrade et al.

    Relationship of venom ontogeny and diet in Bothrops

    Herpetologica

    (1999)
  • J.L. Behler et al.

    The Audobon Society Guide to North American Reptiles and Amphibians

    (1979)
  • A.L. Bieber et al.

    Rattlesnake neurotoxins: biochemical and biological aspects

    J. Toxicol.-Toxin Rev.

    (1990)
  • M.M. Bradford

    A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding

    Anal. Biochem.

    (1976)
  • E.D. Brodie et al.

    Predator–prey arms

    Bioscience

    (1999)
  • J.A. Campbell et al.

    The Venomous Reptiles of the Western Hemisphere

    (2004)
  • T.A. Castoe et al.

    Bayesian mixed models and the phylogeny of pitvipers (Viperidae: Serpentes)

    Mol. Phylogenet. Evol.

    (2007)
  • Y.-H. Chen et al.

    Molecular evolution and structure–function relationships of crotoxin-like and asparagine-6-containing phospholipases A2 in pit viper venoms

    Biochem. J.

    (2004)
  • T. Chijiwa et al.

    Interisland evolution of Trimeresurus flavoviridis venom phospholipase A2 isozymes

    J. Mol. Evol.

    (2003)
  • R. Conant et al.

    A Field Guide to Reptiles and Amphibians: Eastern and Central North America

    (1991)
  • J.C. Daltry et al.

    Diet and snake venom evolution

    Nature

    (1996)
  • C.H. Ernst

    Venomous Reptiles of North America

    (1979)
  • H.L. Gibbs et al.

    Rapid evolution by positive selection and gene gain and loss: PLA2 venom genes in closely related Sistrurus rattlesnakes with divergent diets

    J. Mol. Evol.

    (2008)
  • Cited by (132)

    View all citing articles on Scopus
    View full text