Identification of the genes and processes mediating genetic association signals for complex diseases represents a major challenge. As many of the genetic signals for type 2 diabetes (T2D) exert their effects through pancreatic islet-cell dysfunction, we performed a genome-wide pooled CRISPR loss-of-function screen in a human pancreatic beta cell line. We assessed the regulation of insulin content as a disease-relevant readout of beta cell function and identified 580 genes influencing this phenotype. Integration with genetic and genomic data provided experimental support for 20 candidate T2D effector transcripts including the autophagy receptor CALCOCO2. Loss of CALCOCO2 was associated with distorted mitochondria, less proinsulin-containing immature granules and accumulation of autophagosomes upon inhibition of late-stage autophagy. Carriers of T2D-associated variants at the CALCOCO2 locus further displayed altered insulin secretion. Our study highlights how cellular screens can augment existing multi-omic efforts to support mechanistic understanding and provide evidence for causal effects at genome-wide association studies loci.

Pelvic floor muscle (PFM) injury during childbirth is a key risk factor for pelvic floor disorders that affect millions of women worldwide. Muscle stem cells (MuSCs), supported by the fibro-adipogenic progenitors (FAPs) and immune cells, are indispensable for the regeneration of injured appendicular skeletal muscles. However, almost nothing is known about their role in PFM regeneration following birth injury. To elucidate the role of MuSCs, FAPs, and immune infiltrate in this context, we used radiation to perturb cell function and followed PFM recovery in a validated simulated birth injury (SBI) rat model. Non-irradiated and irradiated rats were euthanized at 3,7,10, and 28 days post-SBI (dpi). Twenty-eight dpi, PFM fiber cross-sectional area (CSA) was significantly lower and the extracellular space occupied by immune infiltrate was larger in irradiated relative to nonirradiated injured animals. Following SBI in non-irradiated animals, MuSCs and FAPs expanded significantly at 7 and 3 dpi, respectively; this expansion did not occur in irradiated animals at the same time points. At 7 and 10 dpi, we observed persistent immune response in PFMs subjected to irradiation compared to non-irradiated injured PFMs. CSA of newly regenerated fibers was also significantly smaller following SBI in irradiated compared to non-irradiated injured PFMs. Our results demonstrate that the loss of function and decreased expansion of MuSCs and FAPs after birth injury lead to impaired PFM recovery. These findings form the basis for further studies focused on the identification of novel therapeutic targets to counteract postpartum PFM dysfunction and the associated pelvic floor disorders.

Population level variation and molecular mechanisms behind insulin secretion in response to carbohydrate, protein, and fat remain uncharacterized despite ramifications for personalized nutrition.We now define prototypical insulin secretion dynamics in response to the three macronutrients in islets from 140 cadaveric donors, including those diagnosed with type 2 diabetes. We leverage the insulin response heterogeneity and use transcriptomics and proteomics to identify molecular pathways of specific nutrient responsiveness. Surprisingly, we find robust insulin secretion to fatty acid stimulus in ~8% of donors, challenging the idea that fat has negligible effects on insulin release. By comparing human islets to human embryonic stem cell-derived islet clusters, we show that, unlike glucose-responsiveness, fat hyper-responsiveness is equivalent and may be a hallmark of functionally immature cells. Our study represents the first comparison of dynamic responses to nutrients and multi-omics analysis in human insulin secreting cells. Responses of different people's islets to carbohydrate, protein, and fat lay the groundwork for personalized nutrition.

Genome-wide association studies have identified hundreds of signals for type 2 diabetes (T2D), most of which confer risk through effects on gene expression. We previously identified the transcription factor ZMIZ1 as a probable effector transcript in human islets, but how altered ZMIZ1 expression impacts T2D risk is unknown. We now show that islets from carriers of the T2D-risk alleles have reduced islet insulin content and glucose-stimulated insulin secretion. To elucidate the mechanism for islet-cell dysfunction, we generated β-cell-specific Zmiz1 knockout (Zmiz1βKO) mice. Male and female Zmiz1βKO mice were glucose intolerant with impaired insulin secretion, compared with control littermates. Transcriptomic profiling of Zmiz1βKO islets identified over 500 differentially expressed genes including those involved in β-cell function and maturity which we confirmed at the protein level. After high fat feeding, Zmiz1βKO mice fail to expand β-cell mass and become severely diabetic. Thus, Zmiz1 is required for normal glucose homeostasis and may contribute to T2D risk by maintaining a mature β-cell state and allowing islet mass expansion upon metabolic stress.

No abstract