The cascading effects of human food on hibernation and cellular aging in free-ranging black bears

Human foods have become a pervasive subsidy in many landscapes, and can dramatically alter wildlife behavior, physiology, and demography. While such subsidies can enhance wildlife condition, they can also result in unintended negative consequences on individuals and populations. Seasonal hibernators possess a remarkable suite of adaptations that increase survival and longevity in the face of resource and energetic limitations. Recent work has suggested hibernation may also slow the process of senescence, or cellular aging. We investigated how use of human foods influences hibernation, and subsequently cellular aging, in a large-bodied hibernator, black bears (Ursus americanus). We quantified relative telomere length, a molecular marker for cellular age, and compared lengths in adult female bears longitudinally sampled over multiple seasons. We found that bears that foraged more on human foods hibernated for shorter periods of time. Furthermore, bears that hibernated for shorter periods of time experienced accelerated telomere attrition. Together these results suggest that although hibernation may ameliorate cellular aging, foraging on human food subsidies could counteract this process by shortening hibernation. Our findings highlight how human food subsidies can indirectly influence changes in aging at the molecular level.

between individuals (relative telomere length, RTL). Though any reliably amplified non-variable copy gene can be employed for standardization 2 , we previously optimized this method using HNRPF gene 3 and telomere primers telg and telc 4,5 .
DNA concentration was determined with Qubit 2.0 Fluorometer (Life Technologies) and DNA quality assessed using gel-electrophoresis. Telomere and single-copy gene PCR were conducted on separate 96-well plates, with identical preparation except for primers (see Kirby, Alldredge & Pauli 2017 for details). Each sample was analyzed in triplicate within a plate and the average used in subsequent analyses (coefficient of variations for T: within-plate = 14%, between-plate = 17%; S: within-plate = 8%, between-plate = 9%). Standard curves were generated from a mixture of 6 randomly chosen bear samples run in triplicate on each plate and diluted to 0.5, 1, 2.5, 6, and 10 ng/µl. Real-time PCR was conducted using an Eppendorf Mastercycler ep realplex, followed by baseline correction in LinRegPCR, and sample quantification using the standard curve method 6 .
We examined relative telomere lengths at first sampling for each bear and calculated telomere length change (averaged over months, to account for differences in sampling time) within each individual between sampling periods and throughout the entire study (n = 30).

Oxidative stress analyses
We measured oxidative stress in bear serum samples, using the d-ROM and the oxy-adsorbent tests (Diacron International, Italy). The d-ROM test measures oxidative damage via the concentration of hydroperoxide, a reactive oxygen metabolite (ROM) that results from an attack of reactive oxygen species on organic substrates (e.g. nucleotides, proteins). Following the manufacturer's protocol, 1.5 µl of bear serum was mixed with 300 µl of an acidic buffer solution, 3 µl of a chromogenic mixture, and incubated for 90 minutes at 37 °C. In these acidic conditions, iron is released from proteins catalyzing hydroperoxide to generate alkoxyl and peroxyl radicals, which react with the chromogenic mixture to produce a color intensity that is proportional to its concentration and read at 505 nm with a spectrophotometer. The concentration of hydroperoxide (expressed as mg H2O2 dl -1 ) was calculated by comparison with a calibrator solution with an oxidative activity of 0.08 mg dl -1 (equivalent to that of H2O2). The oxyadsorbent test measures the total antioxidant capacity of the sample by measuring the ability of the serum to oppose the massive oxidative action of a hypochlorous acid (HClO) solution.
Briefly, serum was first diluted 1:100 with distilled water, 2 µl of the diluted sample was mixed with 200 µl of the oxidant (HClO-based) solution, and incubated at 37 °C for 10 minutes. After incubation, 2 µl of the chromogenic solution was added and the resulting color read with a microplate spectrophotometer at 505 nm, with the color intensity inversely related to the antioxidant capacity, expressed as µmol HClO ml -1 neutralized. For each assay, all samples were analyzed in triplicate and the averages were compared to standard solutions. The inter-assay coefficients of variation were 0.10 and 0.06 for d-ROM and oxy-adsorbent tests, respectively.

Stable isotope analyses
Hair samples were rinsed three times with 2: