Generation of glucose-responsive insulin producing cells from human induced pluripotent stem cells and the role of rs7903146, TCF7L2 and TGFbeta signalling.

Abstract

Type 2 diabetes (T2D) is characterised by increased insulin resistance in tissues such as skeletal muscle, adipocytes, and liver, and a progressive loss of pancreatic beta cells throughout the development of the disease. This results in high blood glucose levels that if left untreated lead to serious health complications. To date, many single nucleotide polymorphism (SNP) variants linked to the risk of T2D have been identified via multiple genome wide association studies. However, T2D is challenging to study due to difficulties in getting fresh pancreatic tissue and the absence of disease models that reflect human genetics. Thus, the underlying mechanisms for most T2D-associated variants are unclear and remain to be investigated. The SNP rs7903146 that is located within the TCF7L2 gene confers the strongest risk to T2D to date, with its T-allele implicated in increased TCF7L2 expression. This has led to the notion that rs7903146 is located within an enhancer that regulates TCF7L2 transcript levels in beta cells. In this study, we aimed to investigate the role of rs7903146 in beta cell development and function using induced pluripotent stem cell (iPSC)-derived pancreatic beta cells. In order to accomplish this aim, we first derived and characterised two iPSC lines, called MANZ2-2 and MANZ4-37. We confirmed that both MANZ lines were successfully reprogrammed back to a pluripotent state, while retaining their genomic integrity. Next, we utilised gene editing approaches to generate various genetically modified iPSC lines that: (1) lacked the rs7903146 locus (rsKO), (2) displayed doxycycline-inducible TCF7L2 expression (TR) and (3) carried a heterozygous deletion in TCF7L2 (TCFKO). In parallel, we developed a novel in vitro protocol that converts iPSCs into glucose responsive insulin producing cells that express key beta cell markers. The genetically modified iPSC lines were differentiated using our optimised protocol revealing that rsKO-derived beta cells displayed altered expression of ACSL5 and INS compared to isogenic controls. However, contrary to prior reports, TCF7L2 expression was not affected by loss of the rs7903146 region, arguing against this locus containing an essential enhancer element. Unfortunately, the TR lines failed to show inducible expression of TCF7L2 and were abandoned. The TCFKO lines gave rise to pancreatic progenitor cells with weak marker expression that subsequently displayed poor beta cell differentiation. This result adds further evidence that TCF7L2 acts during the development of the pancreas to promote beta cells maturation. Over the course of developing our protocol, it became apparent that TGFβ signalling also plays an important role in beta cell maturation. To investigate this, single-cell RNA sequencing was performed on the cells generated from our protocol in the presence and absence of an inhibitor (ALK5ill) of TGFβ signalling. This analysis identified several differentially expressed genes including ribosome and proliferation genes, members of the ID family, encoding proteins that competitively inhibit basic helix-loop-helix (bHLH) transcription factors and the non-coding RNA NEAT1, that were downregulated in ALK5ill-treated cells. Based on these findings, we propose a model in which TGFβ signalling downregulates ID gene expression thereby promoting the activity of key beta cell bHLH factors, such as NEUROD1 and E proteins, enhancing beta cell maturation and reducing proliferation. Overall, this work has pioneered iPSC technologies in New Zealand and combined state-of-the-art beta cell differentiation, gene editing and scRNA-Seq techniques to further our understanding of T2D and TGFβ signalling in beta cells. More broadly, our platform adds new research capabilities to New Zealand researchers, ensuring we remain internationally competitive, and provides new tools to explore the genetic basis of T2D risk.