Authors Anna Morros-Nuevo, Jodie Wainwright, Jessica Pavey , Natalie Wallis, Eleanor Raffan
Affiliations Department of Physiology, Development and Neuroscience, University of Cambridge, Cambridge, Cambridgeshire, CB2 3DY, England, United Kingdom.
Presentation Type Talk
Abstract
A canine obesity epidemic mirrors that observed in the human population and is affected by many of the same risk factors – modern pet dogs have increasingly less active lifestyles and ready access to calorie dense food. Obesity is commonly dismissed as the consequence of lax diet and exercise management despite being a highly heritable trait in many species. We used the electronic health records of 1.1 million purebred dogs to calculate the probability of dogs within a breed being overweight or obese (Breed Obesity/Overweight Probability). In >14,000 dogs we quantified eating behaviour and owner management of diet and exercise, using the Dog Obesity Risk Assessment (DORA) questionnaire. Breed Average Food Motivation Score was calculated accounting differences in the distribution of sex, neuter status and age between breeds.
There was wide variation in both breed average Food Motivation Score and Breed Obesity/Overweight Probability across breeds and there was a strong positive correlation between the two variables (r=0.711, p=4.48×10-8). Indeed over half the variability in Breed Obesity/Overweight Probability was explained by differences in Breed Average Food Motivation Score (R2 =0.5053, p=4.48×10-8). These data suggest genetics are a strong driver for obesity risk and that this is largely mediated by alterations in eating behaviour.
At an individual level, Food Motivation Score had the greatest magnitude and most significant effect on Body Condition Score (BCS) with up to 15% of the variation in adiposity (BCS=1.64/9, p<2×10-16) attributable to this factor alone. The population was stratified by Food Motivation Score tertiles, using this as a surrogate indicator of dogs’ genetic risk for obesity. We show known obesity risk factors such as sex, age and neuter status had a small but significant effect on BCS, that differed according to underlying risk with the effect size of risk factors being greater in dogs with higher food motivation.
We further tested how environmental exposure to food and exercise affected obesity outcome depending on food motivation tertile. Exercise and Food Restriction tended to reduce BCS and did so to a greater extent in highly food motivated dogs compared to low food motivated dogs. Owner Intervention to control weight (e.g. measuring food, tailoring food to exercise) was found to have a contrasting effect in different food motivation tertiles with a significant negative (protective) effect in highly food motivated dogs but a positive and significant effect in dogs with a low food drive.
These data shed light on how genetically determined eating behaviours (food motivation) are a powerful driver of weight gain in pet dogs and are highly heritable. Owner control of environmental access to food and activity are of particular importance in high risk individuals (i.e. highly food motivated dogs) which has important implications for targeting clinical interventions to prevent and treat obesity. Our approach has broader relevance to the study of complex traits with large environmental components.