Pancreatic islet cells are critical for maintaining normal blood glucose levels, and their malfunction underlies diabetes development and progression. We used single-cell RNA sequencing to determine the transcriptomes of 1,492 human pancreatic α, β, δ, and PP cells from non-diabetic and type 2 diabetes organ donors. We identified cell-type-specific genes and pathways as well as 245 genes with disturbed expression in type 2 diabetes. Importantly, 92% of the genes have not previously been associated with islet cell function or growth. Comparison of gene profiles in mouse and human α and β cells revealed species-specific expression. All data are available for online browsing and download and will hopefully serve as a resource for the islet research community.

Proinsulin is a misfolding-prone protein, making its biosynthesis in the endoplasmic reticulum (ER) a stressful event. Pancreatic β-cells overcome ER stress by activating the unfolded protein response (UPR) and reducing insulin production. This suggests that β-cells transition between periods of high insulin biosynthesis and UPR-mediated recovery from cellular stress. We now report the pseudotime ordering of single β-cells from humans without diabetes detected by large-scale RNA sequencing. We identified major states with ) low UPR and low insulin gene expression,) low UPR and high insulin gene expression, or ) high UPR and low insulin gene expression. The latter state was enriched for proliferating cells. Stressed human β-cells do not dedifferentiate and show little propensity for apoptosis. These data suggest that human β-cells transition between states with high rates of biosynthesis to fulfill the body's insulin requirements to maintain normal blood glucose levels and UPR-mediated recovery from ER stress due to high insulin production.

Glucagon supports glucose homeostasis by stimulating hepatic gluconeogenesis, in part by promoting the uptake and conversion of amino acids into gluconeogenic precursors. Genetic disruption or pharmacologic inhibition of glucagon signaling results in elevated plasma amino acids and compensatory glucagon hypersecretion involving expansion of pancreatic α cell mass. Recent findings indicate that hyperaminoacidemia triggers pancreatic α cell proliferation via an mTOR-dependent pathway. We confirm and extend these findings by demonstrating that glucagon pathway blockade selectively increases expression of the sodium-coupled neutral amino acid transporter Slc38a5 in a subset of highly proliferative α cells and that Slc38a5 controls the pancreatic response to glucagon pathway blockade; most notably, mice deficient in Slc38a5 exhibit markedly decreased α cell hyperplasia to glucagon pathway blockade-induced hyperaminoacidemia. These results show that Slc38a5 is a key component of the feedback circuit between glucagon receptor signaling in the liver and amino-acid-dependent regulation of pancreatic α cell mass in mice.

This study provides an assessment of the Fluidigm C1 platform for RNA sequencing of single mouse pancreatic islet cells. The system combines microfluidic technology and nanoliter-scale reactions. We sequenced 622 cells, allowing identification of 341 islet cells with high-quality gene expression profiles. The cells clustered into populations of α-cells (5%), β-cells (92%), δ-cells (1%), and pancreatic polypeptide cells (2%). We identified cell-type-specific transcription factors and pathways primarily involved in nutrient sensing and oxidation and cell signaling. Unexpectedly, 281 cells had to be removed from the analysis due to low viability, low sequencing quality, or contamination resulting in the detection of more than one islet hormone. Collectively, we provide a resource for identification of high-quality gene expression datasets to help expand insights into genes and pathways characterizing islet cell types. We reveal limitations in the C1 Fluidigm cell capture process resulting in contaminated cells with altered gene expression patterns. This calls for caution when interpreting single-cell transcriptomics data using the C1 Fluidigm system.

Antagonizing glucagon action represents an attractive therapeutic option for reducing hepatic glucose production in settings of hyperglycemia where glucagon excess plays a key pathophysiological role. We therefore generated REGN1193, a fully human monoclonal antibody that binds and inhibits glucagon receptor (GCGR) signaling in vitro. REGN1193 administration to diabetic ob/ob and diet-induced obese mice lowered blood glucose to levels observed in GCGR-deficient mice. In diet-induced obese mice, REGN1193 reduced food intake, adipose tissue mass, and body weight. REGN1193 increased circulating levels of glucagon and glucagon-like peptide 1 and was associated with reversible expansion of pancreatic α-cell area. Hyperglucagonemia and α-cell hyperplasia was observed in fibroblast growth factor 21-deficient mice treated with REGN1193. Single administration of REGN1193 to diabetic cynomolgus monkeys normalized fasting blood glucose and glucose tolerance and increased circulating levels of glucagon and amino acids. Finally, administration of REGN1193 for 8 weeks to normoglycemic cynomolgus monkeys did not cause hypoglycemia or increase pancreatic α-cell area. In summary, the GCGR-blocking antibody REGN1193 normalizes blood glucose in diabetic mice and monkeys but does not produce hypoglycemia in normoglycemic monkeys. Thus, REGN1193 provides a potential therapeutic modality for diabetes mellitus and acute hyperglycemic conditions.

Bulk RNA sequencing provides the opportunity to understand biology at the whole transcriptome level without the prohibitive cost of single cell profiling. Advances in spatial transcriptomics enable to dissect tissue organization and function by genome-wide gene expressions. However, the readout of both technologies is the overall gene expression across potentially many cell types without directly providing the information of cell type constitution. Although several in-silico approaches have been proposed to deconvolute RNA-Seq data composed of multiple cell types, many suffer a deterioration of performance in complex tissues. Here we present AdRoit, an accurate and robust method to infer the cell composition from transcriptome data of mixed cell types. AdRoit uses gene expression profiles obtained from single cell RNA sequencing as a reference. It employs an adaptive learning approach to alleviate the sequencing technique difference between the single cell and the bulk (or spatial) transcriptome data, enhancing cross-platform readout comparability. Our systematic benchmarking and applications, which include deconvoluting complex mixtures that encompass 30 cell types, demonstrate its preferable sensitivity and specificity compared to many existing methods as well as its utilities. In addition, AdRoit is computationally efficient and runs orders of magnitude faster than most methods.

Limitations in cell proliferation are important for normal function of differentiated tissues and essential for the safety of cell replacement products made from pluripotent stem cells, which have unlimited proliferative potential. To evaluate whether these limitations can be established pharmacologically, we exposed pancreatic progenitors differentiating from human pluripotent stem cells to small molecules that interfere with cell cycle progression either by inducing G 1 arrest or by impairing S phase entry or S phase completion and determined growth potential, differentiation, and function of insulin-producing endocrine cells. We found that the combination of G 1 arrest with a compromised ability to complete DNA replication promoted the differentiation of pancreatic progenitor cells toward insulin-producing cells and could substitute for endocrine differentiation factors. Reduced replication fork speed during differentiation improved the stability of insulin expression, and the resulting cells protected mice from diabetes without the formation of cystic growths. The proliferative potential of grafts was proportional to the reduction of replication fork speed during pancreatic differentiation. Therefore, a compromised ability to enter and complete S phase is a functionally important property of pancreatic endocrine differentiation, can be achieved by reducing replication fork speed, and is an important determinant of cell-intrinsic limitations of growth.

Aging improves pancreatic β-cell function in mice. This is a surprising finding because aging is typically associated with functional decline. We performed single-cell RNA sequencing of β-cells from 3- and 26-month-old mice to explore how changes in gene expression contribute to improved function with age. The old mice were healthy and had reduced blood glucose levels and increased β-cell mass, which correlated to their body weight. β-Cells from young and old mice had similar transcriptome profiles. In fact, only 193 genes (0.89% of all detected genes) were significantly regulated (≥2-fold; false discovery rate < 0.01; normalized counts > 5). Of these, 183 were down-regulated and mainly associated with pathways regulating gene expression, cell cycle, cell death, and survival as well as cellular movement, function, and maintenance. Collectively our data show that β-cells from very old mice have transcriptome profiles similar to those of young mice. These data support previous findings that aging is not associated with reduced β-cell mass or functional β-cell decline in mice.