Does chronic stress mediate predator-prey interactions in wild fish? An experimental approach using exogenous cortisol implants.

The hypothalamic-pituitary-interrenal (HPI) axis serves as one of the central neuroendocrine systems in mediating the stress response in teleosts. However, HPI axis stimulation, particularly over a prolonged activation, can result in fitness-related consequences. Previous work has shown that stressed teleosts exhibit higher rates of predation when compared to unstressed conspecifics. However, the mechanism(s) underlying this effect, at the physiological and behavioural levels, are currently unknown. My overarching hypothesis is that because of higher metabolic expenditures under chronic cortisol treatment, an individual fish should exhibit riskier behavioural phenotypes leading to higher predator vulnerability. I tested this hypothesis using sunfish (Leopmis sp.) that were implanted with cocoa butter containing either cortisol or as a sham control. Treated animals were subject to a thorough metabolic and physiological profiling. Fish were then assessed for a suite of behavioural traits related to anti-predator and risk-taking behaviours and were accompanied by a field assessment of predator mortality. Cortisol-treatment did result in a significant alteration to the animal's metabolic physiology which included a higher standard metabolic rate, ammonia excretion rates, hepatic [glycogen], and blood [glucose], against respective controls. Despite an altered metabolic state, I observed no alteration to behavioural phenotypes associated with anti-predator or risk-taking behaviours under cortisol-treatment. Similarly, predation rates were comparable across all of my treatment groups. These data indicate that the HPI axis, in this context, is not the driving mechanism underlying higher predation rates in stressed teleosts. It could be that another physiological system associated with the stress response or in conjunction with the HPI axis may be driving this effect. Additionally, metabolism-behavioural interactions are often highly context specific and may not have been manifested under the current environmental context. Lastly, the timescale used in my study may have been too acute with fish being behaviourally resilient. Together, this work provides a foundational assessment looking at the mechanistic role of the stress response in dictating predator-prey interactions. While further work investigating metabolic and behavioural interactions is required, particularly at the individual level, these data could prove useful in better understanding anthropogenic impacts on aquatic ecosystems; a highly relevant consideration in the contemporary Anthropocene.