Individual variation in ACTH-induced cortisol levels in females of a livebearing fish at different gestational stages

Abstract

Individuals vary in their baseline levels of stress hormones (predictive homeostasis) and in their stress responses (reactive homeostasis). Variation in normal reactive scope, both predictive and reactive homeostasis, may be important for understanding how endocrine traits respond to selection. Reactive homeostasis is the increase in glucocorticoid (GCs) hormones above baseline. Individuals at different life history stages, such as gestation in females, may show variation in normal reactive scope. We performed an adrenocorticotropic hormone (ACTH) challenge and measured changes in circulating GCs to estimate the reactive scope of female sailfin mollies (Poecilia latipinna) at different gestational states. We measured cortisol, primary GC in teleost fishes, to obtain baseline release rates prior to injection with either ACTH or saline control. Using water-borne hormones, we measured cortisol release rates at four time intervals post-injection. Females were then sacrificed to determine the developmental stage of embryos, if present, and the number of developing embryos or mature ova. We found that ACTH-injected females had significant increases in cortisol releases rates, whereas cortisol release rates of control females did not change during the 4 h post-injection period. We found high repeatability in predictive homeostasis of cortisol and moderate repeatability in reactive homeostasis and a phenotypic correlation between predictive and reactive homeostasis. Gestational state did not affect female predictive or reactive homeostasis. We applied the reactive scope model to P. latipinna and gained a further understanding of how among- and within-individual variation in both predictive and reactive homeostasis are partitioned and how these traits vary under certain life-history conditions.