Adiponectin Concentration Does Not Correlate with Obesity-Associated Changes in Fasting Insulin Sensitivity in Dogs
Decreased circulating adiponectin concentrations are associated with obesity-related changes in insulin sensitivity in all species tested to date, including laboratory rodents and human beings. A recent study in Japan reported significantly lower adiponectin concentrations in overweight and obese dogs. The aim of the current study was to assess whether adiponectin is associated with obesity-associated changes in fasting insulin sensitivity in spontaneously overweight and obese dogs.
We recruited 123 client-owned lean, overweight, and obese dogs in Australia. Age, gender, breed, and body weight were recorded. Dogs were fasted for 24 hours prior to blood collection. Body condition score was assessed on a 9-point scale by one veterinarian (KRV). Plasma total adiponectin was measured by commercial ELISA, validated in the previous Japanese study, and by Western blotting. Fasting insulin sensitivity was estimated as the product of fasting plasma insulin and glucose concentrations (HOMA). Adiponectin measured by ELISA and Western blotting were compared using Pearson's correlation coefficient. Linear regression models were used to assess effects of adiposity on total plasma adiponectin concentration and HOMA, adiponectin on HOMA, and to check for confounding of the association between adiposity and adiponectin due to age, sex, neuter status, gender, breed genetic background, body weight, and estimated lean weight. Raw data from a previous study of 71 dogs from 4 Japanese veterinary clinics was also obtained and the observed association between obesity and adiponectin concentrations was examined for confounding to assess potential reasons for the disparity between the results of the two studies.
There was close correlation between plasma adiponectin concentrations measured by ELISA and estimated by Western blot (correlation coef. 0.92, p<0.01). There was a strong association between adiposity and HOMA (ß-coef. 0.78; p<0.01) which did not change when adiponectin was accounted for. There was no significant association between adiposity and plasma total adiponectin concentration (ß-coef. -1.2, p=0.38). There was no association between total adiponectin concentration and fasting insulin sensitivity (ß-coef. -0.006, p=0.57). Accounting for age, sex, neuter status, gender, breed genetics, and body weight did not substantially alter the observed weak association between adiposity and adiponectin from the Australian study or the observed strong negative association from the Japanese study.
We conclude that 1) total adiponectin concentration was not strongly associated with adiposity in Australian dogs with naturally occurring obesity, and 2) total adiponectin concentration does not appear to mediate obesity-associated fasting insulin resistance in dogs.