ABSTRACT
OBJECTIVES: Methylmercury metabolism was investigated in Pacific walruses (Odobenus rosmarus divergens) from St. Lawrence Island, Alaska, United States. METHODS: Total mercury and methylmercury concentrations were measured in fecal samples and paired colon samples (n = 16 walruses). Gut microbiota composition and diversity were determined using 16S rRNA gene sequencing. Associations between fecal and colon mercury and the 24 most prevalent gut microbiota taxa were investigated using linear models. RESULTS: In fecal samples, the median values for total mercury, methylmercury, and %methylmercury (of total mercury) were 200 ng/g, 4.7 ng/g, and 2.5%, respectively, while in colon samples, the median values for the same parameters were 28 ng/g, 7.8 ng/g, and 26%, respectively. In fecal samples, methylmercury was negatively correlated with one Bacteroides genus, while members of the Oscillospirales order were positively correlated with both methylmercury and %methylmercury (of total mercury). In colon samples, %methylmercury (of total mercury) was negatively correlated with members of two genera, Romboutsia and Paeniclostridium. CONCLUSIONS: Median %methylmercury (of total mercury) was 10 times higher in the colon compared to the fecal samples, suggesting that methylmercury was able to pass through the colon into systemic circulation. Fecal total mercury and/or methylmercury concentrations in walruses were comparable to some human studies despite differences in seafood consumption rates, suggesting that walruses excreted less mercury. There are no members (at this time) of the Oscillospirales order which are known to contain the genes to methylate mercury, suggesting the source of methylmercury in the gut was from diet and not in vivo methylation.
ABSTRACT
The International Species Information System has set forth an extensive database of reference intervals for zoologic species, allowing veterinarians and game park officials to distinguish normal health parameters from underlying disease processes in captive wildlife. However, several recent studies comparing reference values from captive and free-ranging animals have found significant variation between populations, necessitating the development of separate reference intervals in free-ranging wildlife to aid in the interpretation of health data. Thus, this study characterizes reference intervals for six biochemical analytes, eleven hematologic or immune parameters, and three hormones using samples from 219 free-ranging African lions ( Panthera leo ) captured in Kruger National Park, South Africa. Using the original sample population, exclusion criteria based on physical examination were applied to yield a final reference population of 52 clinically normal lions. Reference intervals were then generated via 90% confidence intervals on log-transformed data using parametric bootstrapping techniques. In addition to the generation of reference intervals, linear mixed-effect models and generalized linear mixed-effect models were used to model associations of each focal parameter with the following independent variables: age, sex, and body condition score. Age and sex were statistically significant drivers for changes in hepatic enzymes, renal values, hematologic parameters, and leptin, a hormone related to body fat stores. Body condition was positively correlated with changes in monocyte counts. Given the large variation in reference values taken from captive versus free-ranging lions, it is our hope that this study will serve as a baseline for future clinical evaluations and biomedical research targeting free-ranging African lions.
