Abstract 139 – Non-invasive age estimation based on DNA methylation using hair in brown bears

Shiori Nakamura, Jumpei Yamazaki, Naoya Matsumoto, Kyougo Hagino, Hideyuki
Sakamoto, Masami Yamanaka, Mina Jimbo, Yojiro Yanagawa, Hideyuki Ito, Toshio Tsubota,
Michito Shimozuru

Age is a crucial factor in elucidating the ecology and management of bears. Recently, it has
been discovered that DNA methylation levels of certain cytosine-phosphate-guanine (CpG) sites
change with age, and they have begun to be used as indicators for age estimation in various
animal species. Previously, we have established an age estimation method based on blood DNA
methylation levels in brown bears. This method has the advantage of being less invasive and
simpler than the conventional tooth method, but it requires the capture of bears. Therefore, in
this study, we aimed to establish a novel age estimation method using non-invasively obtained
hair samples by identifying DNA regions where methylation levels change with age. Hair
samples were collected from captive and wild brown bears of known ages. The captive
individuals, kept at the Noboribetsu Bear Park, consisted of 27 bears (16 males and 11 females)
aged 2 to 31 years old, while the wild individuals included 8 female bears aged 8 to 25 years old
captured alive on the Shiretoko Peninsula, Hokkaido, Japan. Through bisulfite pyrosequencing,
we measured methylation levels of DNA extracted from hair roots in regions adjacent to 12
genes. Positive correlations were observed between age and DNA methylation levels in three
regions, and the best age estimation model was based on DNA methylation levels at two CpG
sites. This model demonstrated high accuracy, with a mean absolute error of 2.7 years and a
median absolute error of 2.4 years after leave-one-out cross-validation. These results suggest
that the age of brown bears can be estimated using the methylation level of hair root-derived
DNA as an indicator. Combining this novel tool with hair trapping survey, widely used for
population estimation, could allow us to non-invasively determine the age structure of wild
bear populations.