In an effort to identify disease-causing mechanisms, numerous genome-wide studies across different ancestries have discovered genetic variants influencing type 2 diabetes (T2D) risk. Amongst identified diabetes-associated loci, a PAX4 coding variant (p.Arg192His) is uniquely and reproducibly associated with altered T2D-risk in East Asian populations. PAX4 encodes a transcription factor, and in mice Pax4 is essential for beta cell development. Whether PAX4 plays a similar role in humans and how diabetes associated PAX4 variants impact beta cell development and/or function is unknown. We combined in vivo studies in human variant carriers with disease modelling in human stem cell models to address these questions. Carriers of the PAX4 p.Arg192His allele have reduced islet cell function compared to non-carriers (AIRg: p.Arg192 756.9 vs. p.His192 530.8 p<0.002) . Deletion of PAX4 by CRISPR-Cas9 genome-editing in isogenic human induced pluripotent stem cell (hiPSC) -derived beta-like cells caused de-repression of alpha cell gene expression, suggesting decreased function in carriers of PAX4 variants may be due to altered beta cell development. hiPSCs derived from carriers of either the PAX4 p.Arg192His or a novel p.Tyr186X allele identified in a proband with early onset diabetes, had increased polyhormonal endocrine cell formation and reduced insulin content following in vitro differentiation towards beta-like cells; correction of the diabetes-associated PAX4 alleles reversed these phenotypic changes. Our in vitro and in silico molecular studies of diabetes associated alleles support reduced PAX4 expression and/or function. PAX4 knockdown in the EndoC-βH1 human beta cell model altered hormone gene expression, reduced total insulin content, and impaired insulin secretion.

In conclusion, we have demonstrated the importance of PAX4 for human endocrine cell development and provide a potential mechanism for diabetes-associated alleles through de-repression of alpha cell gene expression.


N. Krentz: None. S. Hoon: None. D. Gardner: None. S. Kao: None. E. Tai: Advisory Panel; Amgen Inc., Merck Sharp & Dohme Corp., Novartis AG, Novo Nordisk, Stock/Shareholder; Abbott, AbbVie Inc. A. L. Gloyn: Other Relationship; Genentech, Inc., Roche Pharmaceuticals. A. Teo: Stock/Shareholder; BetaLife Pte Ltd. H. Lau: None. F. Abaitua: Employee; Vertex Pharmaceuticals Incorporated. M. Perez-alcantara: None. J. Chan: None. J. N. Ajeian: None. B. Champon: None. H. Sun: None. A. Jha: None.


Wellcome (200837) ; National Institutes of Health (U01-DK105535, U01-DK085545, UM1DK126185, U01DK123743, U24DK098085, P30DK116074) ; Lee Foundation Grant (SHTX/LFG/002/2018) Paris-NUS 2021-06-R/UP-NUS (ANR-18-IDEX-0001)

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