Aims/hypothesis:
An R1420H variation in sulfonylurea receptor 1 (SUR1), a subunit of the K
ATP
channel, was previously identified in an Indigenous community in Arizona where a homozygous carrier (1420HH) had hyperinsulinaemic hypoglycaemia during infancy (HHI), suggestive of a K
ATP
channel loss of function (LoF). Interestingly, heterozygous carriers of this variation (1420RH, occurring in 3% of the community), had a twofold increased risk of type 2 diabetes. We aimed to create an isogenic induced pluripotent stem cell (iPSC)-derived pancreatic islet (SC-islet)-based platform to test whether the R1420H variant results in K
ATP
channel LoF, and to examine the distinct temporal effects of SUR1 1420HH and 1420RH K
ATP
channel variations on insulin secretion from developing and mature SC-islets.
Methods:Using CRISPR-Cas9, isogenic iPSCs with all three genotypes (SUR1 1420RR, 1420RH and 1420HH) were generated from two different parental Indigenous American iPSC lines (IS1, isogenic cell lines derived from parental cell line 1; and IS2, isogenic cell lines derived from parental cell line 2). These isogenic cell lines were used to generate immature SC-islets (resembling fetal islets) and mature SC-islets (resembling adult islets), which were used to assess insulin secretion dynamics during different stages of development and identify differences in gene expression by single-cell RNA-seq. This study was consistent with the CONSIDER statement for research studies among Indigenous American communities.
Results:
Immature SUR1 1420HH SC-islets secreted 3.4-fold (IS1,
p
<0.001) and 4.2-fold (IS2,
p
=0.001) more insulin under basal conditions than normal (SUR1 1420RR) SC-islets. Modest hyperinsulinaemia was also seen in immature SUR1 1420RH SC-islets (2.2-fold [IS1] and 2.3-fold [IS2]) but the results were not statistically significant. After maturation, the 1420HH SC-islets failed to achieve glucose responsiveness whereas the 1420RH SC-islets achieved biphasic insulin secretion but had significantly lower glucose responsiveness than normal SC-islets (AUC for insulin secretion [as a % of total insulin] under high glucose challenge: 1.04 vs 0.56 in normal vs 1420RH SC-islets,
p
<0.001). Diazoxide reduced hyperinsulinaemia in SUR1 1420RH and 1420HH immature SC-islets, while tolbutamide elicited a greatly diminished or undetectable insulin secretory response from mature SUR1 1420RH SC-islets (13.2-fold increase in insulin secretion) and 1420HH SC-islets (1.9-fold increase) compared with normal SC-islets (31.5-fold increase). Results were directionally comparable for both IS1 and IS2 SC-islets. SUR1 1420RH SC-islets also responded to the glucokinase activator dorzagliatin with improvement in first-phase insulin secretory response (first-phase stimulation index: 3.9-fold vs 7.3-fold,
p
=0.01 [IS1, 11 mmol/l glucose ± dorzagliatin]; 5.5-fold vs 9.0-fold,
p
=0.13 [IS1, 20 mmol/l glucose ± dorzagliatin]). Single-cell RNA-seq identified dysregulated genes in SUR1 1420RH SC-beta cells, including lower expression of glycolytic genes and upregulation of
G6PC2
, which could explain the lower insulin secretory response to glucose.
Conclusions/interpretation:
A SUR1 R1420H variant identified in an Indigenous American population causes hyperinsulinaemia in homozygous immature SC-islets during basal conditions and these SC-islets fail to achieve glucose responsiveness after maturation. In the heterozygous state, modest hyperinsulinaemia is observed in immature SC-islets, which after maturation have significantly lower glucose responsiveness. These results demonstrate that SUR1 R1420H is a K
ATP
channel LoF variation and suggest that a lower insulin secretory response during adulthood is the cause of the higher type 2 diabetes risk in individuals heterozygous for this variation. We also show that the isogenic iPSC-based platform can be used to test therapeutic agents to treat HHI in infants homozygous for LoF K
ATP
channels and screen for drugs that can improve glucose-responsive insulin secretion in adult heterozygous carriers.
Data availability:
See the database of Genotypes and Phenotypes (dbGaP; dbgap.ncbi.nlm.nih.gov/home; accession no.: phs002490.v1.p1) for details concerning data requests. All source codes can be found in the GitHub repository under
https://github.com/Koushik-Cheranda/SC-islet-scRNAseq-analysis-R-codes
.