UC Davis

Obesity is a complex disease characterized by excessive fat accumulation that leads to decrease in health and increased risk of developing numerous health complications such as metabolic syndrome, type 2 diabetes mellitus, cardiovascular diseases, and other pathological conditions. Fundamentally, obesity is a consequence of long-term energy imbalance where energy intake surpasses energy expenditure, but the mechanisms behind energy imbalance are influenced by numerous biological and environmental factors, such as genetics and diet. Both the complex etiology and heterogeneous nature of obesity present challenges to effective long-term prevention and treatment of obesity at the population level. For example, population-based diet recommendations have had limited success in mitigating obesity because of the variation in other factors that differ at the individual level to impact physiological response to diet and obesity development, such as differences in genetic background. Since genetics and diet are crucial determinants in the regulation of energy balance, it is necessary to broaden our understanding of how genetic background and diet interact relative to the development of obesity for improving recommendations for weight loss. Animal models are indispensable for discerning the effect of genetic factors from environmental factors on the manifestation of the phenotype of interest. The murine model has been especially crucial for the discovery of mechanisms that influence energy balance and obesity development, such as appetite signaling. Of all available mouse models, the Collaborative Cross (CC) mouse panel is a particularly excellent model system for comparing the effects of genetic background to environmental effects. Derived from elaborate intercrosses of 5 classically inbred mouse strains and 3 wild-derived mouse strains, the CC is a large recombinant inbred mouse population with the degree of genetic and phenotypic diversity reflective of the human population. The CC simultaneously provides both tremendous genetic diversity and the ability to use genetic “replicates” which can mimic twin studies. In this work, replicates from 22 CC strains were placed on either a high protein diet or high fat high sucrose diet challenge for eight weeks. Body composition and circulating analyte levels were assessed both at baseline before the diet challenge and post-diet to compare the impact of genetic background (strain) and diet on adiposity and clinical traits associated with metabolism. The second chapter of this work focused on determining how much genetic background and diet contribute to the development of obesity, whether diet alters susceptibility to developing obesity, and whether differences in diet macronutrient composition result in more beneficial phenotypic outcomes. Both at baseline and post-diet, the CC exhibited a wide range of phenotypic variation for adiposity and circulating analytes by strain; after the diet challenge, phenotypic differences were much larger between strains than diet, suggesting that genetics play a much bigger role in the development of obesity than diet. Similar to the observation in humans, the individual CC strains responded differently to diet where certain strains gained weight on one diet or the other, while others stayed consistently lean or consistently fat regardless of diet, indicating that genetics largely determines whether an individual will become obese, but the effect of diet can be larger or smaller depending on specific genetics. When examining the effect of diet by itself, certain traits differed significantly by diet such as body weight and cholesterol levels, while others did not differ by diet such as adiposity and triglyceride levels, demonstrating that whether and how macronutrient composition influences phenotypic change depends on the trait. Surprisingly, when correlations were performed between adiposity and traditional markers of metabolic syndrome (such as circulating triglycerides, glucose, cholesterol, and insulin), only the correlation between insulin and adiposity stayed significant both before and after the diet challenge. The third chapter of this work explored the relationship between genetics, diet, and hepatic gene expression relative to obesity since the liver regulates biological processes that impact adiposity accumulation, such as lipogenesis and metabolism of macronutrients. To relate the phenotypic results and findings from chapter two to hepatic gene expression, correlations were performed using phenotype data and microarray data, revealing 2,552 genes whose expression levels were significantly correlated with adiposity. In general, the effect of strain was much stronger than diet on hepatic gene expression as demonstrated by differential gene expression analysis which found over 9,000 genes differentially expressed by strain compared to 1,344 genes differentially expressed by diet. Interestingly, diet differentially expressed genes (DEGs) were enriched for many biological pathways associated with substrate metabolism, whereas strain DEGs were enriched for pathways less sensitive to environmental perturbations. Because common obesity is caused by multiple genes, weighted gene co-expression network analysis (WGCNA) was performed to identify clusters of related genes grouped into “modules”. Multiple gene modules were found that differed in average expression by both diet and strain, where three of the gene modules were correlated with adiposity and enriched for biological pathways related to obesity development. By combining all the analyses above and searching in the genome-wide association studies (GWAS) catalog, the list of obesity candidate genes found via (GWAS) in humans can be narrowed down to increase the success of future functional validations studies.