| Summary: | <p>Genome-wide association studies have identified >400 signals which are robustly associated with type 2 diabetes (T2D) risk, a large proportion of which are involved in pancreatic beta cell dysfunction. A major barrier to translational insight however is uncertainty over the genes through which they exert their effect. The aim of this thesis was to perform the first genome-wide pooled CRISPR screen in human beta cells and use a high-throughput and unbiased approach to identify candidate causal genes and novel regulators of human beta cell function.</p>
<p>The cell line EndoC-βH1 is the most authentic cellular model of human beta cells but it is a slow growing and technically challenging cell line which has not been established for genome editing yet. Before approaching the genome-wide screen, I therefore developed the first CRISPR/Cas9 editing pipeline in EndoC-βH1 using a dual sgRNA and lentiviral based approach. As demonstrated for six genes, this strategy induced highly efficient genome editing resulting in complete protein depletion and functional gene knockout. Distinct phenotypes upon comparison with an siRNA-based knockdown strategy highlighted the need for complementary validation of individual results to avoid method specific functional interpretations.</p>
<p>After extensive optimisation of all screening parameters, I performed two independent genome-wide pooled loss-of-function CRISPR screens for intracellular insulin in EndoC-βH1 using a FACS based readout. The screen demonstrated an excellent technical performance and was able to identify functionally relevant hits as shown through identification of known regulators of human beta cell function such as INS, NKX2.2, SLC2A2, G6PC2 or GCK, a strong enrichment of T2D, MODY and beta cell pathways and protein-protein networks known to play a crucial role in the beta cell. In total, 580 genes were classified as having a highly reproducible effect on insulin content. Integration with predicted T2D effector genes whose mechanistic role has not been resolved yet, prioritised C2CD4B, FADS1 and CALCOCO2 for functional follow-up studies.</p>
<p>Finally, individual functional validation studies confirmed the phenotypic effects from the screen for several genes including for CALCOCO2, where silencing of the gene reduced insulin content and secretion in EndoC-βH1, therefore establishing it as a novel regulator of beta cell function and potential causal gene at the TTLL6 GWAS locus. Cellular localisation studies within primary human islets further confirmed that CALCOCO2 is highly expressed across endocrine and exocrine pancreatic cell types.</p>
<p>Overall, the work presented in this thesis has addressed the outstanding challenge of functionally investigating all potential T2D effector genes in human beta cells and has generated a valuable resource and numerous hypotheses for future studies. The biological insights gained from this work have contributed to a better functional understanding of T2D pathogenesis and predisposition and will inform future therapeutic strategies.</p>
|
|---|