Specific fat distributions are risk factors for complex diseases, including coronary heart disease and obstructive sleep apnea. To demonstrate the utility of high-diversity mouse models for elucidating genetic associations, we describe the phenotyping and heritability of fat distributions within the five classical inbred and three wild-derived founder mouse strains of the Collaborative Cross and Diversity Outbred mice. Measurements of subcutaneous and internal fat volumes in the abdomen, thorax and neck, and fat volumes in the tongue and pericardium were obtained using magnetic resonance imaging in male mice from the A/J (n = 12), C57BL/6J (n = 17), 129S1/SvlmJ (n = 12), NOD/LtJ (n = 14), NZO/HILtJ (n = 12), CAST/EiJ (n = 14), PWK/PhJ (n = 12), and WSB/EiJ (n = 15) strains. Phenotypes were compared across strains using analysis of variance and heritability estimated as the proportion of phenotypic variability attributable to strain. Heritability ranged from 44% to 91% across traits, including >70% heritability of tongue fat. A majority of heritability estimates remained significant controlling for body weight, suggesting genetic influences independent of general obesity. Principal components analysis supports genetic influences on overall obesity and specific to increased pericardial and intra-neck fat. Thus, among the founder strains of the Collaborative Cross and Diversity Outbred mice we observed significant heritability of subcutaneous and internal fat volumes in the neck, thorax and abdomen, pericardial fat volume and tongue fat volume, consistent with genetic architecture playing an important role in explaining trait variability. Findings pave the way for studies utilizing high-diversity mouse models to identify genes affecting fat distributions and, in turn, influencing risk for associated complex disorders.

Introduction Research suggests greater obstructive sleep apnea (OSA) severity in African-Americans than Caucasians. However, the underlying mechanisms causing this ethnic disparity are unknown. To evaluate possible mechanisms, we compared the size of the tongue between African American and Caucasian OSA patients using magnetic resonance imaging (MRI), controlling for age, body mass index (BMI) and apnea-hypoxia index (AHI). Given prior evidence of more severe OSA in African Americans, we hypothesized these patients would have larger soft tissue volumes compared to Caucasians. Methods Upper airway soft tissue volumes, (total tongue, tongue fat, lateral walls, pterygoids, total soft tissue) were quantified using MRI and compared between Caucasian (n=133) and African American (n=175) patients with moderate OSA. Analyses were conducted using regression models controlling for age, sex, BMI and AHI. Results Among all OSA patients, African Americans had higher BMI than Caucasians (40.0±8.6 vs. 37.1±8.1 kg/m2, p=0.0024) and a higher proportion of females (66.3% vs. 36.1%; p<0.0001). There were no significant differences in age (p=0.143) or AHI (p=0.314). Controlling for these covariates, there were no differences between African American and Caucasian OSA patients in tongue fat volume (mean [95% confidence interval] difference = 479 [-3156, 4115] mm3; p=0.794). However, African Americans had a 13,286 (6,439, 20,132) mm3 larger total tongue volume compared to Caucasians (p=0.0002). Larger volumes in African Americans were also observed for the soft palate (p<0.0001), retropalatal lateral walls (p=0.003), pterygoid (p=0.034) and total soft tissue volume (p=0.0003). Conclusion African Americans were observed to have larger volumes of the tongue, soft palate, retropalatal lateral walls, pterygoids and total soft tissue volume compared to Caucasians, although there were no differences in AHI and tongue fat volume. The study contributes to the overall understanding of ethnic-specific pathology of OSA and can potentially influence how African Americans and Caucasians are diagnosed and treated specifically for the disease. Support