King's College London

Research portal

Cell-type heterogeneity in adipose tissue is associated with complex traits and reveals disease-relevant cell-specific eQTLs

Research output: Contribution to journalArticle

Original languageEnglish
Pages (from-to)1013-1024
Number of pages12
JournalAmerican Journal of Human Genetics
Volume104
Issue number6
Early online date23 May 2019
DOIs
Publication statusPublished - 6 Jun 2019

Documents

King's Authors

Abstract

Adipose tissue is an important endocrine organ with a role in many cardiometabolic diseases. It is comprised of a heterogeneous collection of cell-types which can differentially impact disease phenotypes. Cellular heterogeneity can also confound ‘omic analyses, but is rarely taken into account in analysis of solid-tissue transcriptomes. Here, we investigate cell-type heterogeneity in two population-level subcutaneous adipose tissue RNA-seq datasets (TwinsUK, N =766 and GTEx, N=326) by estimating the relative proportions of four distinct cell types (adipocytes, macrophages, CD4+ t-cells and Micro-Vascular Endothelial Cells). We find significant cellular heterogeneity within and between the TwinsUK and GTEx adipose datasets. We find that adipose cell-type composition is heritable and confirm the positive association between adipose-resident macrophage proportion and obesity (BMI), but find a stronger BMI-independent association with DXA-derived body-fat distribution traits. We benchmark the impact of adipose tissue cell-composition on a range of standard analyses, including phenotype-gene expression association, co-expression networks and cis-eQTL discovery. Our results indicate that it is critical to account for cell-type composition when combining adipose transcriptome datasets, in co-expression analysis and in differential expression analysis with obesity-related traits. We applied Gene expression by Cell Type Proportion interaction models (G × Cell) to identify 26 cell-type specific eQTLs in 20 genes, including 4 autoimmune disease GWAS loci. These results identify cell-specific eQTLs and demonstrate the potential of in-silico deconvolution of bulk tissue to identify cell-type restricted regulatory variants.

Download statistics

No data available

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454