Genetics of adipose cell-type expression and cardiometabolic traits

点击翻译按钮获取中文摘要

Abstract A global obesity epidemic is driving the concomitant rapid increase in the prevalence of cardiometabolic disorders (CMDs), including hypertriglyceridemia, type 2 diabetes (T2D), hypertension, and non-alcoholic fatty liver disease (NAFLD). Population- and sex-specific differences in CMD predisposition exist; however, the biological mechanisms underlying these differences are not well understood. Previous large-scale genome-wide association studies (GWAS) have reliably identified CMD-associated variants in multiple populations; however, functional understanding of the biological mechanisms of the GWAS variants remains challenging. One major obstacle is the limited knowledge of the relevant cell types in which GWAS variants affect gene expression. Bulk tissue gene expression data exist for CMD-relevant tissues, such as subcutaneous adipose, but these data exhibit considerable heterogeneity, including both cell type and cell state within each cell type. Subcutaneous adipose is an important human endocrine tissue for CMDs, and it is possible to collect high-quality adipose tissue samples from healthy individuals. However, the contributions of many adipose genes to CMDs and CMD traits are still poorly understood. The current lack of cell-type expression reference data sets limits fine-scale regional transcriptional assessment of GWAS variant effects. In addition, local expression quantitative trait locus (cis- eQTL) analyses are confounded by cell-type-specific expression differences, which hamper replication efforts across independent bulk RNA-sequenced (RNA-seq) cohorts. To address these knowledge gaps and identify genetic effects on adipose cell-type gene expression, we will perform single nucleus RNA-sequencing (snRNA- seq) in frozen subcutaneous adipose tissue biopsy samples from 300 well-characterized individuals, generate fine-scale estimates of cell-type proportions, identify study-wide and personalized cell-type-specific differences corresponding to cardiometabolic trait levels, and experimentally test GWAS variants for allelic effects on cell- type-specific expression. We hypothesize that by elucidating adipose tissue cell-type expression from 300 existing frozen adipose biopsies, we can leverage available GWAS and adipose bulk RNA-seq data (n=3,230; 45% female) from diverse populations to identify the relevant cell types for hundreds of CMD genes. In our preliminary studies, we have successfully performed snRNA-seq in frozen human subcutaneous adipose tissue biopsies and performed multi-omic studies integrating GWAS results with bulk adipose RNA-seq data, identifying hundreds of colocalized loci for CMD traits. Our approach will leverage the existing wealth of information available in GWAS and bulk adipose RNA-seq cohorts to elucidate the largely unknown cell types of biological mechanisms in the adipose tissue that drive CMDs. Success of the proposed study will substantially improve understanding of cell-type-specific transcriptional mechanisms of CMD diseases and traits in a key human metabolic tissue.