Immune variation during pregnancy suggests immune component-specific costs of reproduction in a viviparous snake with disparate life-history strategies.

Growing evidence suggests the existence of trade-offs between immune function and reproduction in diverse taxa. Among vertebrates, however, there is still a taxonomic bias toward studies in endotherms, particularly birds. We tested the hypothesis that reproduction entails immune-related costs in the viviparous garter snake, Thamnophis elegans, from populations that exhibit two life-history strategies, termed ecotypes, with contrasting paces of life. Between the two ecotypes, we predicted lower immune function in gravid than non-gravid females of both strategies, but with relatively larger immunity costs in the ecotype that generally invests more in current reproduction. Across individuals, we predicted greater immune costs for females investing more in the present specific reproductive event (i.e., higher fecundity) irrespective of their ecotype. We assessed leukocyte profiles and measured bactericidal capacity of plasma (innate immunity) and T- and B-lymphocyte proliferation (adaptive immunity) in gravid and non-gravid females in their natural habitats. We also collected data on reproductive output from these same gravid females brought into captivity. Gravid females of both ecotypes showed lower T-lymphocyte proliferation responses to concanavalin A than non-gravid females, but no differential costs were observed between ecotypes. The remaining immune variables did not vary between gravid and non-gravid females. Among gravid females within each ecotype, those with larger reproductive output showed lower total leukocyte counts, suggesting a fecundity-dependent trade-off. Our study contributes to the comparative ecoimmunology of vertebrates by highlighting the immune component-specificity of trade-offs between reproduction and immune function and showing that costs can be fecundity-dependent in some, but not all cases.

Innate immune performance and steroid hormone profiles of pregnant versus nonpregnant cottonmouth snakes (Agkistrodon piscivorus).

Squamates (lizards and snakes) have independently evolved viviparity over 100 times, and exhibit a wide range of maternal investment in developing embryos from the extremes of lecithotrophic oviparity to matrotrophic viviparity. This group therefore provides excellent comparative opportunities for studying endocrine and immune involvement during pregnancy, and their possible interactions. We studied the cottonmouth (Agkistrodon piscivorus), since they exhibit limited placentation (e.g., ovoviviparity), allowing comparison with squamate species hypothesized to require considerable maternal immune modulation due to the presence of a more extensive placental connection. Furthermore, the cottonmouth's biennial reproductive cycle provides an opportunity for simultaneously comparing pregnant and non-pregnant females in the wild. We document significantly elevated concentrations of progesterone (P4) and significantly lower concentrations of estradiol (E2) in pregnant females relative to non-pregnant females. Pregnant females had lower plasma bacteria lysis capacity relative to non-pregnant females. This functional measure of innate immunity is a proxy for complement performance, and we also determined significant correlations between P4 and decreased complement performance in pregnant females. These findings are consistent with studies that have determined P4's role in complement modulation during pregnancy in mammals, and thus this study joins a growing number of studies that have demonstrated convergent and/or conserved physiological mechanisms regulating viviparous reproduction in vertebrates.