| Izvleček: | <p>Most variants associated with type 2 diabetes (T2D) predisposition in genome-wide association studies (GWAS) act through defects in insulin secretion. In principle, these could result from deficiencies in islet development and/or mature islet function: most functional studies have focused on the latter. To explore the contribution of disturbed islet development to T2D pathogenesis, I characterised the transcriptome (using RNA-seq), accessible chromatin (ATAC-seq), H3K27ac histone marks (ChIP-seq) and DNA methylation patterns (using the 850k Illumina array) of three human iPSC lines (from different donors) differentiated in vitro across 7 stages (n=24) of pancreatic beta cell development.</p>
<p>Principal component analysis, enriched pathways and transcription factors regulating the transcriptome and DNA regulatory marks analysed indicate that the hiPSC-derived model from which the data originated recapitulates key aspects of beta cell development, and achieves more adult beta cells (though still immature) than previously-published differentiation models. Analysis of differentially expressed genes and regulatory marks within each stage indicated that both the most adult stage (beta-like cells) and less differentiated stages were enriched in T2D-associated signals, pointing to the contribution of both developing and adult islets to T2D risk. Novel potential regulators of islet development were identified (e.g. REST or CUX2), including genes within loci associated with T2D by GWAS (e.g. PROX1, ONECUT1/HNF6) and TFs whose binding was disrupted by T2D-associated variants (e.g. CTCF, ZBTB33). The DNA regulatory information was used to lever the posterior probabilities of association of variants within each T2D-associated locus, refining fine-mapping at these developmentally active regions. Integration of DNA regulatory marks with transcriptomic information facilitated the identification of active enhancers in islet development and their possible target genes, including candidate gene effectors of T2D-associated variants. </p>
<p>One of these candidate genes, PROX1, was functionally tested in islet development in the laboratory, aiming to modulate its expression using CRISPR activation, interference, and knock-out. PROX1 was successfully upregulated in the undifferentiated hiPSC system using CRISPRa. Finally, it was possible to replicate the differentiation model in different cell lines (KOLFC2) than those that originated the omics data (SB). These new cell lines had the CRISPRa/i machinery integrated in the genome, providing more tools to study the effect of candidate genes in islet development and T2D.</p>
|
|---|