The stress of Arctic warming on polar bears.

Arctic ecosystems are changing rapidly in response to climate warming. While Arctic mammals are highly evolved to these extreme environments, particularly with respect to their stress axis, some species may have limited capacity to adapt to this change. We examined changes in key components of the stress axis (cortisol and its carrier protein-corticosteroid binding globulin [CBG]) in polar bears (Ursus maritimus) from western Hudson Bay (N = 300) over a 33 year period (1983-2015) during which time the ice-free period was increasing. Changing sea ice phenology limits spring hunting opportunities and extends the period of onshore fasting. We assessed the response of polar bears to a standardized stressor (helicopter pursuit, darting, and immobilization) during their onshore fasting period (late summer-autumn) and quantified the serum levels of the maximum corticosteroid binding capacity (MCBC) of CBG, the serum protein that binds cortisol strongly, and free cortisol (FC). We quantified bear condition (age, sex, female with cubs or not, fat condition), sea ice (breakup in spring-summer, 1 year lagged freeze-up in autumn), and duration of fasting until sample collection as well as cumulative impacts of the latter environmental traits from the previous year. Data were separated into "good" years (1983-1990) when conditions were thought to be optimal and "poor" years (1991-2015) when sea ice conditions deteriorated and fasting on land was extended. MCBC explained 39.4% of the variation in the good years, but only 28.1% in the poor ones, using both biological and environmental variables. MCBC levels decreased with age. Changes in FC were complex, but more poorly explained. Counterintuitively, MCBC levels increased with increased time onshore, 1 year lag effects, and in poor ice years. We conclude that MCBC is a biomarker of stress in polar bears and that the changes we document are a consequence of climate warming.

The stress of being alone: Removal from the colony, but not social subordination, increases fecal cortisol metabolite levels in eusocial naked mole-rats.

In many social species, hierarchical status within the group is associated with differences in basal adrenocortical activity. We examined this relationship in naked mole-rats (Heterocephalus glaber), eusocial rodents with arguably the most extreme social hierarchies of all mammals. This species lives in colonies where breeding is restricted to one socially dominant 'queen' and her male consorts, and all other individuals are reproductively suppressed 'subordinates'. The relationship between cortisol and social status in naked mole-rats has not fully been elucidated, as prior results on this topic have been contradictory. We used non-invasive feces sampling to measure baseline cortisol levels in eight laboratory colonies of naked mole-rats, to either replicate or reject rank differences. First, we successfully validated an assay to measure fecal cortisol metabolites (FCMs). Removal from the colony for the validation experiment, either alone or with an opposite sex conspecific, induced prolonged elevation of FCM levels on a scale of days to weeks. This increase in cortisol did not prevent the removed animals from sexually maturing. In colony-housed animals, we found no relationship between rank in the social hierarchy and FCM levels. Further, queens, breeding males, and reproductively suppressed subordinates all had equivalent FCM levels. We conclude that this species shows little evidence of the 'stress of dominance' or 'stress of subordination' and that reproductive suppression in naked mole-rats is not driven by elevated cortisol levels.

Assessing Stress in Arctic Lemmings: Fecal Metabolite Levels Reflect Plasma Free Corticosterone Levels.

Interest in the ecology of stress in wild populations has triggered the development of noninvasive methods for quantifying stress hormones. Measurement of fecal corticosteroid metabolites (FCMs) is one such method, but it is still unclear whether FCMs can be a reliable proxy of free plasma glucocorticoids. To assess the validity of this assumption, we carried out a robust assessment on brown lemmings (Lemmus trimucronatus) from Bylot Island, Nunavut, Canada, that were hand captured and anesthetized and related plasma glucocorticoid levels to fecal metabolite glucocorticoid levels. We examined endogenous factors that could explain interindividual variability. Blood corticosterone was measured from samples obtained on capture and 30 min later, and FCM levels were measured from animals kept in captivity for 72 h. Plasma free corticosterone increased 135-fold over baseline values 30 min after capture, which confirmed that initial handling was perceived as a stressor. We found that FCM levels were highly related with free (marginal [Formula: see text] = 0.53) but not with total ([Formula: see text] = 0.02) corticosterone levels, regardless of age, sex, and reproductive condition. FCM levels started increasing 2 h after capture and reached maximum levels 4 h after capture. No circadian rhythm in FCMs was found. Plasma total corticosterone levels were much higher in adult females compared with adult males, but this difference was much smaller when measuring free corticosterone levels and FCM levels. Our results suggest that FCM levels are good measures of stress by being closely related to plasma free corticosterone levels in brown lemmings.

Assessment of the stress response in Columbian ground squirrels: laboratory and field validation of an enzyme immunoassay for fecal cortisol metabolites.

Stress responses play a critical role in the ecology and demography of wild animals, and the analysis of fecal hormone metabolites is a powerful noninvasive method to assess the role of stress. We characterized the metabolites of injected radiolabeled cortisol in the urine and feces of Columbian ground squirrels and validated an enzyme immunoassay for measuring fecal cortisol metabolites (FCM) with a 5 alpha-3beta,11 beta-diol structure by stimulation and suppression of adrenocortical activity and by evaluation of the circadian pattern of FCM excretion. In addition, we also evaluated the impact of capture, handling, and acclimation to the laboratory on FCM. Cortisol is highly metabolized, with virtually none being excreted, and of the radiolabeled cortisol injected, 31% was recovered in urine and 6.5% in feces. The lag time between cortisol injection and its appearance in urine and feces was 4.5 +/- 0.82 (SE) h and 7.0 +/- 0.53 (SE) h, respectively. FCM levels varied over the day, reflecting circadian variation in endogenous cortisol. Dexamethasone decreased FCM levels by 33%, and ACTH increased them by 255%. Trapping and housing initially increased FCM levels and decreased body mass, but these reversed within 3-7 d, indicating acclimation. Finally, FCM levels were modestly repeatable over time (r=0.57) in wild, live trapped, nonbreeding animals, indicating that FCMs provide a measure of the squirrel's stress-axis state. This assay provides a robust noninvasive assessment of the stress response of the Columbian ground squirrel and will facilitate an integration of its life history and physiology.

A non-invasive technique for analyzing fecal cortisol metabolites in snowshoe hares (Lepus americanus).

To develop non-invasive techniques for monitoring steroid stress hormones in the feces of free-living animals, extensive knowledge of their metabolism and excretion is essential. Here, we conducted four studies to validate the use of an enzyme immunoassay for monitoring fecal cortisol metabolites in snowshoe hares (Lepus americanus). First, we injected 11 hares with radioactive cortisol and collected all voided urine and feces for 4 days. Radioactive metabolites were recovered predominantly in the urine (59%), with only 8% recovered in the feces. Peak radioactivity was detected an average of 3.5 and 5.7 h after injection in the urine and feces, respectively. Second, we investigated diurnal rhythms in fecal cortisol metabolites by measuring recovered radioactivity 2 days after the radioactive cortisol injection. The total amount of radioactivity recovered showed a strong diurnal rhythm, but the amount of radioactivity excreted per gram of feces did not, remaining constant. Third, we injected hares with dexamethasone to suppress fecal cortisol metabolites and 2 days later with adrenocorticotropic hormone to increase fecal cortisol metabolites. Dexamethasone decreased fecal cortisol metabolites concentrations by 61% and adrenocorticotropic hormone increased them by 1,000%, 8-12 h after injection. Fourth, we exposed hares to a simulated predator (dog). This increased the fecal cortisol metabolites concentrations by 175% compared with baseline concentrations 8-12 h after exposure. Thus, this enzyme immunoassay provides a robust foundation for non-invasive field studies of stress in hares.