Lack of Correlation Between Adiponectin Multimers and Obesity-Related Changes in Insulin Sensitivity in Dogs
In human beings, adiponectin is implicated in the development of obesity-induced insulin resistance and of type 2 diabetes mellitus. Obese dogs also develop insulin resistance, but type 2 diabetes is not well documented in dogs. Adiponectin circulates as low- and high-molecular weight (HMW) multimers. Changes in HMW adiponectin account for most of the variation in human total adiponectin concentrations. This study aimed to examine whether total and/or HMW adiponectin are associated with obesity-related insulin resistance in dogs.
Twelve client-owned dogs were recruited. Six obese dogs (body condition score (BCS) 8-9/9) and six lean dogs (BCS 4-5/9) were matched for age and gender. Insulin sensitivity was measured using a frequently-sampled glucose tolerance test with Bergman's Minimal Model analysis. We measured the proportion of adiponectin that was HMW multimers (SA) using velocity centrifugation on sucrose gradients followed by Western blotting. Total adiponectin was measured by ELISA. Absolute HMW adiponectin concentration was calculated as SA multiplied by total adiponectin concentration. Linear regression was used to assess effects of obesity on each of insulin sensitivity, total adiponectin, angular-transformed SA, and absolute HMW adiponectin concentration and the effects of total adiponectin, SA, and absolute HMW adiponectin on insulin sensitivity. Results are reported as mean±SEM or regression coefficient±SE.
Obese dogs were half as insulin sensitive as lean dogs (3.0±0.8 versus 6.5±1.5x10-4 L•mU-1•min-1, respectively; p=0.04). SA did not vary with obesity (obese 0.76±0.05 versus lean 0.76±0.06, p=0.98). Total and absolute HMW adiponectin were numerically lower in obese dogs than lean dogs, but the associations were not statistically significant (total adiponectin: obese 15±2 versus lean 25±9 µg/mL, p=0.26; absolute HMW adiponectin: obese 11.9±2.2 versus lean 20.8±8.5 µg/mL, p=0.29). Insulin sensitivity did not vary significantly with total adiponectin (ß-coef. -0.009±0.081; p=0.91), SA (ß-coef. -6.8±8.2; p=0.37), or absolute HMW adiponectin (ß-coef. -0.025±0.085; p=0.78); similar results were observed after accounting for adiposity in the regression analysis. When one glucose intolerant dog was excluded a higher SA was associated with lower insulin sensitivity (ß-coef. -23±7.2, p=0.01) but neither total (ß-coef. -0.3±0.08, p=0.70) nor absolute HMW adiponectin (0.06±0.08, p=0.52) varied with insulin sensitivity.
We conclude that 1) unlike human beings, in dogs the proportion of HMW adiponectin (SA) is not decreased with obesity, and 2) SA does not appear to mediate obesity-induced changes in insulin sensitivity in dogs. The differences between species in adiponectin physiology might help to explain differences in susceptibility to type 2 diabetes mellitus.