Rare variants in HNF1A, a pancreatic beta cell transcription factor, have been shown to be associated with T2D and monogenic diabetes. Here, we examine the effect of functionally validated rare (MAF<1%) nonsynonymous variants (RNVs) in a functional domain of HNF1A on T2D risk, and whether their effect is influenced by an overall genetic susceptibility to T2D.

Using exome sequencing data of 3,443 African ancestry (AA) and 157,182 European ancestry (EA) individuals in the UK Biobank and 5,018 AA, 5,6EA and 7,1Hispanic-Latino Ancestry (HA) individuals in the BioMe Biobank, we identified 132 AA, 287 EA, and 64 HA RNV carriers (NRNVs= 30) of HNF1A variants that impair protein function ≤ 60%. T2D Polygenic Risk Scores (PRSs) were built using either trans-ethnic or EA GWAS summary statistics and standardized per ancestry. We tested the association between RNV carrier status and the PRS with T2D risk, and meta-analyzed the ancestry-specific results.

Carrying any of the HNF1A RNV mutations was associated with a 1.5 increase in the odds of T2D in the EA, albeit non-significantly (P=0.06) , but not in the AA (OR=1.1, P=0.6) or HA (OR=1 , P=1) . We then restricted analyses to RNVs within a functional domain, which results in more pronounced effects in the AA (OR=1.5, P=0.07) and EA (OR=1.9, P=0.002) but not in the HA (OR=0.8, P=0.5) . Per SD increase in the PRS, there is a significant 1.7, 1.6 and 2.1 higher odds of T2D in AA, EA and HA individuals, respectively. Even though we observed no interaction between the PRS and HNF1A RNVs (P >; 0.05) , the association between RNVs and T2D was more pronounced among EA individuals with a high PRS. Functional domain carriers (NEA = 50) in the top PRS quintile had a 3.6 (P=0.0008) higher odds of T2D as compared to non-carriers in the same quintile.

Our study shows the value of large biobanks in studying rare variants, the importance of variant selection, and the need to include non-European ancestry populations. We demonstrate the utility of combining rare and common variant risks to better assess T2D risk.


L.Stalbow: None. M.H.Preuss: None. N.Chami: None. R.A.J.Smit: None. R.J.Loos: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.