"Dose-response" models with mixed parameters by the example of venom toxicity estimation of the common European adder Vipera berus

The toxicity analysis of samples of the common European adder Vipera berus venom was made. The adders were captured in different regions of the European part of Russia from populations with mixed features of nominative subspecies of V. b. berus and Nikolsky''s adder V. b. nikolskii. The values of the average lethal doses (DL50) for white laboratory mice with the intraperitoneal injections were defined by three different methods. The appendix contains functions in the algorithmic language R which assess the values of isoeffective doses for a given set of animal lethal probabilities. "Dose-response" dependences were constructed with the use of GLMEM – general linear mixed-effects models – for various combinations of concomitant factors potentially affecting the result of toxicometric experiment (e.g. sex of experimental animals, subspecies signs of the common adder, capture area, venom color etc.). The optimal model, which corresponded to the minimum of the information AIC-criterion, included both adder subspecies and their habitat. The presence of geographical gradient with an approximate direction from Perm region to Lipetsk region was found in the zone of V. b. berus and V. b. nikolskii intergradation. In this direction in the adder populations with mixed signs the diagnostic features of V. b. nikolskii are intensified and the venom toxicity for mice increases. This identified trend corresponds to the modern notion about the history of the resettlement of the two forms of vipers and their hybridization. The necessity of redescription of "nikolskii" form is noted, because its typical territory is situated in the contact zone with V. b. berus. The toxicity of the venom to mice can be used as a diagnostic sign for the redescription of taxon.