The effect of overfeeding and obesity on canine adipose tissue and skeletal muscle transcriptomes

Overweight dogs have a reduced life expectancy and increased risk of chronic disease. During obesity development, adipose tissue undergoes major expansion and remodeling, but the biological processes involved are not well understood. The objective of study 1 was to analyze global gene expression profiles of adipose tissue in dogs, fed a high-fat (47% kcal/g) diet, during the transition from a lean to obese phenotype. Nine female beagles were randomized to ad libitum (n=5) feeding or body weight maintenance (n=4). Subcutaneous adipose tissue biopsy, skeletal muscle biopsy and blood samples were collected, and dual x-ray absorptiometry measurements were taken at 0, 4, 8, 12, and 24 wk of feeding. Ad libitum feeding increased (P<0.05) body weight, body fat mass, adipocyte size and serum leptin concentrations. Microarrays displayed 1,665 differentially expressed genes in adipose tissue over time in the ad libitum fed dogs. Alterations were observed in many homeostatic processes including metabolism, oxidative stress, mitochondrial homeostasis, and extracellular matrix. Our data implies that during obesity development subcutaneous adipose tissue has a large capacity for expansion, which is accompanied by tissue remodeling and short-term adaptations to the metabolic stresses associated with ad libitum feeding. Gene expression changes between 12 and 24 wk indicate a transition from the initial adaptive response to ad libitum feeding and the onset of a chronic obese state. The objective of study 2 was to identify differentially expressed genes and enriched functional pathways between subcutaneous and gonadal adipose of lean and obese dogs at 24 wk of the animal experiment described in study 1. Subcutaneous adipocytes of obese dogs were larger than obese gonadal and lean subcutaneous and gonadal adipocytes. A total of 946 and 703 transcripts were differentially expressed between gonadal and subcutaneous adipose tissue in obese and lean dogs, respectively. Both lean and obese dog gene lists had enrichment of the complement and coagulation cascade, lysosomal and systemic lupus erythematosus pathways. Obese dogs had enrichment of extracellular matrix-receptor interaction and renin-angiotensin system pathways. Lean dogs had enrichment of glutathione metabolism, focal adhesion, synthesis and degradation of ketone bodies and amino sugar metabolism. This study demonstrates that there are a core set of genes that differentiate adipose tissue depots regardless of obesity status, that may underlie differences in depot metabolism and inflammation. Skeletal muscle, as a large and insulin-sensitive tissue, is an important contributor to metabolic homeostasis and energy expenditure. Only a limited number of studies have compared skeletal muscle gene expression of lean and obese dogs. The objective of study 3 was to identify genes and functional classes differentially expressed between lean and obese dog skeletal muscle at 24 wk of the animal experiment described in study 1. Microarrays displayed 77 differentially expressed skeletal muscle transcripts between lean and obese dogs. Alterations were observed in genes pertaining to the functional classes of signaling, transport, protein catabolism and proteolysis, protein modification, development, transcription and apoptosis, cell cycle and differentiation. The small amount of differentially expressed genes in skeletal muscle indicates that adipose tissue expansion may buffer skeletal muscle from the metabolic stresses associated with ad libitum feeding. In study 1, zinc-alpha2 glycoprotein (ZAG) was highly expressed (up to 60-fold vs. baseline) in subcutaneous adipose tissue during weight gain, with increased expression at 4, 8, and 12 wk. At 24 wk, however, ZAG expression had returned to basal levels. Studies in both humans and rodents suggest that adipose tissue ZAG expression is down-regulated during obesity. Therefore, the objective of study 4 was to determine if ZAG was increased with increasing body weight and body condition score in a clinical population of dogs. The clinical population studied had a narrow body condition score range (2-6), but wide body weight range (3.0-29.5 kg). ZAG expression (ZAG/cyclophilin A) ranged from 0.68-3.13, but there was no influence (P>0.05) of body condition score or body weight on its expression. Gonadal adipose tissue expression of ZAG was not influenced by body condition score in this clinical population, which could be due to the depot measured or lack of obese dogs in this population. Our results indicate that adipose tissue has a large capacity to store and adapt to excess calories during the transition from a lean to obese phenotype. The capacity for adipose tissue to expand may buffer skeletal muscle from the effects of excess calories during this period. Even though large phenotypic differences were noted between lean and obese dogs, transcriptome analysis revealed a core set of genes differentiating subcutaneous and gonadal adipose tissue in both groups