Human subcutaneous adipose tissue variability is driven by VEGFA, ACTA2, adipocyte density, and ancestral history of the patient

Abstract

Adipose tissue is a dynamic regulatory organ that has profound effects on the overall health of patients. Unfortunately, inconsistencies in human adipose tissues are extensive and multifactorial including large variability in cellular sizes, lipid content, inflammation, extracellular matrix components, mechanics, and cytokines secreted. Given the high human variability, and since much of what is known about adipose tissue is from animal models, we sought to establish correlations and patterns between biological, mechanical, and epidemiological properties of human adipose tissues. To do this, twenty-six independent variables were cataloged for twenty patients that included patient demographics and factors that drive health, obesity, and fibrosis. A factorial analysis for mixed data (FAMD) was used to analyze patterns in the dataset (with BMI > 25) and a correlation matrix was used to identify interactions between quantitative variables. Vascular endothelial growth factor A (VEGFA) and actin alpha 2, smooth muscle (ACTA2) gene expression were the highest loading in the first two dimensions of the FAMD. The number of adipocytes was also a key driver of patient-related differences, where a decrease in the density of adipocytes was associated with aging. Aging was also correlated with a decrease in overall lipid percentage of subcutaneous tissue (with lipid deposition being favored extracellularly), an increase in transforming growth factor-{beta}1 (TGF{beta}1), and an increase in M1 macrophage polarization. An important finding was that self-identified race contributed to variance between patients in this study, where Black patients had significantly lower gene expression levels of TGF{beta}1 and ACTA2. This finding supports the urgent need to account for patient ancestry in biomedical research to develop better therapeutic strategies for all patients. Another important finding was that TGF{beta} induced factor homeobox 1 (TGIF1), an understudied signaling molecule, is highly correlated with leptin signaling and was correlated with metabolic inflammation. Finally, this study revealed an interesting gene expression pattern where M1 and M2 macrophage markers were correlated with each other, and leptin, in patients with a BMI > 25. This finding supports growing evidence that macrophage polarization in obesity involves a complex, interconnecting network system rather than a full switch in activation patterns from M2 to M1 with increasing body mass. Overall, this study reinforces key findings in animal studies and identifies important areas for future research, where human and animal studies are divergent. Understanding key drivers of human patient variability is required to unravel the complex metabolic health of unique patients.