Linking the plasma proteome to genetics in individuals from continental Africa provides insights into type 2 diabetes pathogenesis

Abstract

Individuals of African ancestry remain largely underrepresented in genetic and proteomic studies. Here, we measure the levels of 2,873 proteins using the Olink proximity extension assay in plasma samples from 163 individuals with type 2 diabetes (T2D) or prediabetes and 362 normoglycemic controls from the Ugandan population for the first time. We identify 88 differentially expressed proteins between the two groups and 208 proteins associated with cardiometabolic traits. We link genome-wide data to protein expression levels and construct the first protein quantitative trait locus (pQTL) map in this population. We identify 399 independent associations with 346 (86.7%) cis-pQTLs and 53 (13.3%) trans-pQTLs. 16.7% of the cis-pQTLs and all of the trans-pQTLs have not been previously reported in African-ancestry individuals. Of these, 37 pQTLs have not been previously reported in any population. We find evidence for colocalization between a pQTL for SIRPA and T2D genetic risk. Mendelian randomization analysis identified 20 proteins causally associated with T2D. Our findings reveal proteins causally implicated in the pathogenesis of T2D, which may be leveraged for personalized medicine tailored to African-ancestry individuals.