Multi-omics islet profiling in type 2 diabetes reveals differential transcriptomics and chromatin accessibility by ancestry

Multi-omics islet profiling in type 2 diabetes reveals differential transcriptomics and chromatin accessibility by ancestry | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Multi-omics islet profiling in type 2 diabetes reveals differential transcriptomics and chromatin accessibility by ancestry Elisabeth F. Heuston, Ayo P. Doumatey, Adebowale A. Adeyemo, Charles N. Rotimi This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5881592/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Type 2 diabetes (T2D) disproportionally affects African Americans (AA) compared to European Americans (EA) in the United States. To investigate how the molecular pathophysiology T2D may vary by ancestry, we analyzed scRNA-Seq and snATAC-Seq data from 85,112 pancreatic islet cells from AA and EA individuals with T2D. AA donors had proportionally fewer beta and alpha cells but more exocrine cells (FDR ≤ 1.5x10 − 3 ) compared to EA. AA donors beta cells displayed upregulated exocrine genes (e.g., PNLIP , PRSS1 ) and mildly downregulated INS compared to EA. However, AA donor alpha cells expressed more INS , and higher HNF4A and HNF4G transcription factor binding accessibility. In contrast, EA exocrine cells expressed higher INS , GCG , SST , and PPY , with a corresponding increase of HNF4A and HNF4G binding site accessibility. These data demonstrate key differences in both islet cell composition and insulin signaling between AA and EA individuals with T2D. Biological sciences/Genetics/Genomics/Epigenomics Biological sciences/Genetics/Genomics/Transcriptomics Health sciences/Diseases/Endocrine system and metabolic diseases/Diabetes/Type 2 diabetes mellitus Ancestry ATAC-Seq Genomics Pancreatic islets RNA-Seq Single cell Transcriptomics Type 2 diabetes Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Individuals of African ancestry in the United States are disproportionally affected by type 2 diabetes mellitus (T2D) and are more likely to experience associated cardiometabolic comorbidities including hypertension, retinopathy, and hyperlipidemia [1]. However, most omics studies have focused on individuals from populations of European and Asian ancestries [2] and African ancestry populations remain understudied despite experiencing a greater burden of the disorder. The pancreas is one of the major organs involved in T2D pathophysiology given its central role in insulin secretion. Pancreatic islets are mostly composed of alpha and beta cells that produce hormones (e.g., glucagon and insulin, respectively) to regulate endocrine signaling [3, 4]. The islets are encircled by acinar and ductal exocrine cells that secrete enzymes to mediate digestion and nutrient release. Historically, the endocrine and exocrine functions of the pancreas were considered distinct and largely independent of each other. However, an increasing body of evidence identifies extensive cross-talk between these populations, suggesting that their relationship is significantly more complicated [5, 6]. For example, it has been suggested that transdifferentiation from ductal progenitors, acinar cells, and alpha cells, into insulin-producing cells does occur, demonstrating cellular plasticity between pancreatic cell populations [7-10]. Thus, studies of pancreatic cell populations are essential to understand the molecular changes associated with T2D in the pancreas and potentially illuminate pathophysiological changes observed in individuals living with T2D. Taking advantage of the Human Pancreas Analysis Program (HPAP) [11], which provides pancreatic islet single cell multi-omic studies of individuals of multiple ancestries, we investigated islet cell molecular profiles between African Americans (AA) and European Americans (EA). We analyzed single cell RNA (scRNA-Seq) and single nucleus ATAC (snATAC-Seq) sequencing profiles of AA and EA pancreas donors from HPAP with T2D. Our findings reveal differences in islet cell composition and insulin-expressing subpopulations between AA and EA donors, and reveal novel mechanisms that could provide explanations for some of the pathophysiologic features of T2D reported in AA individuals. Research Design and Methods Data access Raw sequencing data of hand-picked islet cells for single cell RNA (scRNA-Seq) and single nucleus ATAC (scATAC-Seq) sequencing were accessed through the data portal of The Human Pancreas Analysis Program (version 2.1.2). Donor inclusion criteria included: Ancestry (“Caucasian” OR “African American”); Disease (“T2D”); single cell RNA capture (10X Genomics Chromium 3’ Gene Expression platform with v2, v3, or v3.1 chemistries) or single cell ATAC capture (10X Genomics Chromium Single Cell ATAC platform v1 chemistry). Single cell RNA sequencing data processing FASTQ files were demultiplexed in Cell Ranger_7.1.0 (10x Genomics) and mapped to the GRCh38-2020-A transcriptome assembly. Ambient RNA was excluded with CellBender_0.3.0 [12]. Quality control (QC) and clustering was performed with Seurat_5.0.1 [13], SeuratObject_5.0.0, and DoubletFinder_2.0.3 [14]. scRNA-Seq cell filtering was performed as follows: nFeature_RNA ≥ 200, nFeature_RNA ≤ 2500, percentage reads mapped mitochondrial RNAs (PMMR) ≤ 10. Predicted doublet rate was set at 0.05. Batch correction was performed SCTransform and anchor integration. Clusters were called using a resolution of 0.4 (total cell population) or 0.2 (beta cell subclustering). Down sampling of cells per group was done using Seurat’s subset (downsample) function. Differential abundance analyses were performed with scProportion_0.0.0.9000 [15] and MiloR_2.0.0 [16], and presented as log2 of the fold difference (log2FD). Over representation analyses (ORA) and gene set enrichment analyses (GSEA) were performed in WebGestalt 2024 [17]. Significance was defined as FDR ≤ 1x10 -1 . Single cell ATAC sequencing data processing FASTQ files were demultiplexed in Cell Ranger-ATAC version 2.1.0 (10X Genomics) and mapped to the cellranger-arc-GRCh38-2020-A-2.0.0 genome assembly. QC, clustering, and integration with single cell RNA sequencing data was performed in ArchR_1.0.2 [18]. snATAC-Seq filtering was performed as follows: nFrags > 1000, nFrags 8. Batch correction was performed with Harmony [19]. Data and code availability Source data are available at the Data Portal of the Human Pancreas Analysis Program (https://hpap.pmacs.upenn.edu/). The code necessary to reproduce the primary results of this study are available at the git repository https://github.com/heustonefNIH/PancT2D. Results RNA and ATAC characteristics of islets in T2D Single cell RNA-Seq data of dissected islets from donors with T2D were obtained from HPAP. Characteristics of the sample donors are shown in Table 1 . After QC, a total of 6679 scRNA-Seq cells from six donors of African ancestry (AA) and 16231 cells from six donors of European ancestry (EA) were analyzed ( Online Supplementary Data 1 ). Unbiased clustering with a conservative resolution of 0.4 identified 17 RNA clusters (designated RNA0, RNA1, etc.) ( Figure 1A, Online Supplementary Data 2 ). Differential gene expression analysis (DGE) and canonical marker genes led to identification of four endocrine cell clusters, two exocrine cell clusters, two mesenchyme cell clusters, four immune cell clusters, three mixed clusters, and one cluster each of epithelial and endothelial cells ( Figure 1B, Online Supplement Figure 1, Online Supplementary Data 3 ). We also analyzed snATAC-Seq data from HPAP and performed integrated transcriptional and chromatin analysis. Of thirteen total donors with snATAC-Seq data, ten also had independently assayed (i.e., not multiome-matched) scRNA-Seq data. After QC, a total of 23226 cells from five AA donors and 38976 cells from eight EA donors were analyzed. Nearest-neighbor analysis generated 25 clusters (designated ATAC1, ATAC2, etc., Figure 2A ). We used gene scores based on transcriptional start site (TSS) accessibility to predict transcriptional profiles and to assign each RNA cluster to a chromatin profile ( Figure 2B ). Differential distribution of beta cell clusters and regulation of beta cell function are observed between AA and EA Our conservative clustering parameters identified one beta cell cluster with an absence of other definitive cell markers ( Figure 1 ). Beta1 (RNA2) exhibited upregulated islet amyloid polypeptide ( IAPP ), SIX homeobox 3 ( SIX3 ) and insulin ( INS ) RNA and was associated with four ATAC clusters: ATAC1, ATAC23, ATAC24, and ATAC25. These ATAC clusters were enriched for chromatin remodeling motifs for proteins such as cohesion complex component (RAD21), suppressor of zeste 12 (SUZ12), and enhancer of zeste 2 polycomb repressive complex 2 subunit (EZH2) (ATAC23) ( Online Supplementary Data 4 ). We performed differential abundance testing (DA) to determine if AA and EA cells were distributed differently across RNA clusters ( Figure 3A ). We found that AA cells proportions in the Beta1 cluster were significantly lower than in EA cells (6% of total AA versus 16% of total EA; log2FD (fold difference) = -1.4, FDR ≤ 2x10 -3 ; Figure 3B, Figure 3C ). Given the large discrepancy in numbers between EA and AA cells in this cluster (2634 EA versus 413 AA), the EA Beta1 cluster was randomly downsampled (see Methods) to 413 cells for subsequent beta cell comparisons. Beta1 DGE between AA and EA found regenerating family member 1 alpha ( REG1A ) , pancreatic lipase ( PNLIP ) , serine protease 1 ( PRSS1 ) , and serine protease 2 ( PRSS2) (log2FC ≥ 4.4, FDR ≤ 3x10 -7 ) were upregulated in AA, INS moderately downregulated in AA (log2FC = 0.26, FDR ≤ 3x10 -6 ), and enrichment of genes involved in Glucose Intolerance (ER = 5.6, FDR ≤ 2x10 -2 ), Fat Malabsorption (ER = 24, FDR ≤ 2x10 -3 ), and Chronic Pancreatitis (ER = 223, FDR ≤ 4x10 -10 ) compared to EA ( Online Supplementary Data 5 ). These data show that AA cells within the Beta1 cluster express exocrine cell markers at a higher level than do EA cells. To assess DA of AA and EA cells across beta cell subsets [20], we subclustered Beta1 and identified four subpopulations ( Figure 4A, Online Supplementary Data 6 ). Although the total number of beta cells was significantly lower in AA, we found different distributions of existing beta cells within three of the four subsets. Beta1b contained 38% of all Beta1 AA cells (155 cells) compared to 29% of all EA Beta1 cells (752 cells; FDR = 4x10 -3 ). Beta1c contained 20% of all Beta1 AA (84 cells) compared to 25% of all Beta1 EA (649 cells; FDR = 4x10 -2 ). Beta1d contained 3% of Beta1 AA (11 cells) compared to 5% of Beta1 EA (142 cells; FDR = 8x10 -3 ). Beta1a was not significantly different between AA and EA populations. Compared to other beta cell subsets, Beta1b expressed genes associated with Amyloid-beta Formation (ER = 25, FDR ≤ 4x10 -2 ) and IL6-mediated signaling (ER = 47, FDR ≤ 5x10 -2 , Online Supplementary Data 5 ), was mildly enriched for AA cells (log2FD = 0.39, FDR ≤ 4x10 -3 ), and had a lower percentage of mitochondrial-mapping RNAs (FDR ≤ 5x10 -6 ; Figure 4B ). Interestingly macrophages, which a can be activated by IL6/IL-6R signaling, were also significantly enriched in AA (RNA14, log2FD = 0.92, FDR ≤ 1.5x10 -3 ). In contrast to Beta1b, Beta1c had upregulated solute carrier family 35 member F1 ( SLC35F1 ) and protein phosphatase 1 regulatory inhibitor subunit 1A ( PPP1R1A ) expression than other Beta1 subpopulations, was mildly depleted for AA (log2FD = -0.28, FDR ≤ 5x10 -2 ), and had a higher percentage of mitochondrial-mapping RNAs (FDR ≤ 7x10 -3 ; Figure 4B ). Together these data show that fewer AA cells exist within the beta cell compartment, and those that do have a distinct transcriptional profile compared to their EA counterparts. Alpha cell clusters show differential abundance and regulation between AA and EA Three clusters expressed canonical alpha cell markers (e.g., glucagon ( GCG ) and transthyretin ( TTR )) in the absence of other definitive markers: Alpha1 (RNA0), Alpha2 (RNA9), and Alpha3 (RNA16) ( Figure 1 ). Alpha1 and Alpha2 were together associated with six ATAC clusters and shared similar DNA binding motifs including GATA binding protein 3 (GATA3) (ATAC2; ATAC5), tripartite motif-containing 28 (TRIM28) (ATAC3), RAD21 (ATAC21), hepatocyte nuclear factor 4 gamma (HNF4G), hepatocyte nuclear factor 4 alpha (HNF4A) (ATAC 5), and for chromatin remodeling-associated proteins including SUZ12, E2F transcription factor 6 (E2F6), and TATA-box binding protein associated factor 1 (TAF1) (ATAC21) ( Online Supplementary Data 4 ). Compared to other RNA clusters, Alpha1 had the highest expression of canonical alpha cell genes including GCG, TTR, and lysyl oxidase like 4 ( LOXL4 ), and was significantly enriched for predicted targets of the NRSF Transcription Factor (ER = 12, FDR ≤ 2x10 -2 ). Alpha2 expressed high ROS proto-oncogene 1 receptor tyrosine kinase ( ROS1 ) and olfactory receptor family 5 subfamily AU member 1 ( OR5AU1 ) and was significantly enriched for Macrophage-derived Foam Cell Differentiation (ER = 60, FDR ≤ 5x10 -3 ). Alpha3 contained fewer than 1% of the total population of cells and was excluded from further analyses. AA cells were significantly depleted (i.e. had lower proportions) in the Alpha1 cell cluster (log2FD ≥ 1.1, FDR ≤ 1x10 -3 ; Figure 3B-C ) and expressed significantly higher IAPP (log2FC = 8.6, FDR ≤ 2x10 -3 ) and INS (log2FC = 2.2, FDR ≤ 2x10 -5 ) compared to EA ( Online Supplementary Data 7 ), although GCG expression was not significantly different between these groups. In contrast, EA cells expressed significantly higher retinol binding protein 1 ( RBP1) (retinol transport), transketolase ( TKT ) (glycolysis), and fatty acid binding protein 5 ( FABP5 ) (lipid synthesis) (log2FC ≥ 1.7, FDR ≤ 8x10 -8 ). These data show that, although the largest alpha cell population is depleted (lower) in AA, these cells express higher levels of insulin than those of EA donors. Exocrine cell clusters display differential expression of insulin RNA between AA and EA Two clusters expressed canonical exocrine cell markers (e.g., REG1A and PNLIP ): Exocrine1 (RNA1) and Exocrine 2 (RNA3). Exocrine1 was associated with 3 ATAC clusters ( Figure 2A ) and was enriched for TFBS including retinoid X receptor alpha (RXRA), histone deacetylase 2 (HDAC2), MYB proto-oncogene like 2 (MYBL2) (ATAC17; ATAC18), TRIM28 (ATAC18), and HNF4G and HNF4A (ATAC19) ( Online Supplementary Data 4 ). Exocrine2-associated chromatin profiles overlapped in part with ATAC17 but predominantly with ATAC24 (which was assigned to Beta1, Figure 2B ). Compared to other RNA clusters, Exocrine1 expressed the highest levels of canonical acinar cell genes including PNLIP, PRSS1, and PRSS2 , and was significantly enriched for genes associated with Protein and Fat Digestion and Absorption (ER ≥ 24, FDR ≤ 0.001) and Peptidase Activities (ER ≥ 7, FDR ≤ 2x10 -10 ). In contrast, Exocrine2 expressed upregulated progastricsin ( PGC ) and lymphotoxin beta ( LTB ) (previously associated with pancreatic inflammatory responses), and was enriched for processes associated with Detoxification (ER ≥ 15.0, FDR ≤ 1x10 -6 ) ( Online Supplementary Data 5 ). Exocrine1 versus Exocrine2 DGE, followed by gene set enrichment analysis (GSEA), showed that Exocrine1 had higher MT-RNR2 like 1 ( MTRNR2L1 ) and (glycoprotein 2) GP2 expression and was enriched for Integrin Signaling (NES = 1.6, FDR ≤ 1x10 -1 ). In contrast, Exocrine2 had higher (Wnt family member 4 ( WNT4 ), ubiquitin C-terminal hydrolase L1 ( UCHL1 ) and NK6 homeobox 1 ( NKX6-1 ) expression and was enriched for targets of NKX6.1 (NES = 1.9, FDR ≤ 1x10 -1 ) and RE1-silencing transcription factor (REST) (NES = 2.3, FDR ≤ 7x10 -3 ). AA cells were significantly enriched in Exocrine1 (log2FD = 0.94, FDR ≤ 2x10 -3 ) and Exocrine2 (log2FD = 1.45, FDR ≤ 2x10 -3 ) compared to EA cells. Despite having a lower abundance of cells, Exocrine1 EA cells expressed significantly higher INS (log2FC = 4.4, FDR ≤ 2x10 -9 ) than AA cells ( Online Supplementary Data 8 ), while Exocrine2 EA cells expressed significantly higher INS (log2FC = 4.4, FDR ≤ 2x10 -9 ), GCG (log2FC = 2.2, FDR ≤ 2x10 -130 ), somatostatin ( SST ) (log2FC = 1.8, FDR ≤ 8x10 -112 ), and pancreatic polypeptide ( PPY ) (log2FC = 1.9, FDR ≤ 1x10 -115 ) ( Figure 4C ) than AA cells. Finally, we tested whether cells expressing exocrine markers co-expressed endocrine markers, or if cells expressing either exocrine or endocrine markers were co-clustering. In Exocrine1, more EA cells expressed both PNLIP (exocrine) and either INS , GCG , or SST , compared to AA cells ( Online Supplement Figure 2 ). In Exocrine2, higher percentages of EA cells co-expressed both exocrine and endocrine markers. Together these data show that signals of insulin-expressing non-beta cells arise from different cell populations in AA and EA T2D pancreatic tissue donors. Discussion African Americans with T2D display a number of pathophysiologic differences when compared with European Americans with T2D, including differences in beta-cell function, peripheral and hepatic insulin sensitivity, hepatic insulin clearance and hepatic glucose dysregulation [21–23]. These differences are already present in individuals without T2D and are found in concert with markers of subclinical inflammation and oxidative stress (among others) [24, 25]. In the present study, we focus on the pancreas, one of the key organs in T2D pathophysiology, hypothesizing that single cell analysis of pancreatic islets cells may illuminate some of these pathophysiologic features. We analyzed over 85,000 single cells for either RNA or ATAC profiles of pancreatic cells in AA and EA individuals with T2D. Our findings show that alpha and beta cells are significantly depleted (i.e. had significantly lower proportions) in AA donor cell populations when contrasted with EA donor cell populations, while exocrine cells are comparatively depleted (lower) in EA donor cell populations. Alongside these distributional differences, alternative sources of insulin transcripts apparently come from alpha cells in AA donors and exocrine cells in EA donors. The findings of markedly smaller numbers of alpha and beta cells in AA as well as different islet transcriptomic profiles (when compared with EA) are noteworthy because early beta cell dysfunction in T2D is a known feature in individuals of African ancestry [1]. Reduction in beta cell numbers and/or beta cell dysfunction result in relative or absolute insulin deficiency that is a feature of the dysglycemia seen in T2D while increasing evidence suggests that alpha cell dysfunction is an important contributor to hyperglycemia in T2D [26]. Loss of beta cell function in T2D can attribute to a number of mechanisms including decrease in beta cell mass/volume due to apoptosis, transdifferentiation and dedifferentiation [27]. In our study, we found enrichment in TF-like (e.g., RAD21) and chromatin remodeling-associated TFBS (e.g. EZH2), both of which are implicated in apoptosis [28, 29]. EZH2 is known to mediate glucolipotoxicity induced apoptosis of beta cells [29]. Interestingly, pharmacological inhibition of EZH2 in humans affects the regenerative of beta like cell ability with insulin release in vitro [30]. In this study, beta cells from AA donors were more strongly associated with amyloid deposition, IL-6 signaling, and insulin expression, but deficient in a population of PPP1R1A -expressing cells. Loss of PPP1R1A has been associated with impaired insulin secretion and glucose metabolism [31, 32] and amyloid accumulation has been known for its detrimental effects on beta cell function. Amyloid deposits are characteristic pathological features of T2D associated with the misfolding of islet amyloid polypeptide (IAPP), the second most abundant protein expressed by beta cells and is secreted along with insulin in response to diet (e.g., glucose, fat, and amino acids). Islet amyloid deposits appear to be population-dependent (for example occurring in over 80% of European ancestry individuals in contrast to only 30–40% of Japanese individuals) and has been suggested to be one of the features of differences in pathological and pathophysiological differences of metabolic diseases between populations [27, 33–35]. Our findings are consistent with this notion and provide a potential explanation for the observed beta cell secretory defects in African Americans [1]. While the role of IL-6 signaling in islet cells is complex [36–38], it should be noted that IL-6 is pleiotropic and has both pro-inflammatory and anti-inflammatory/regenerative effects depending on cellular conditions. IL-6 also plays a role in macrophage-mediated inflammation in T2D [39]. Higher levels of inflammation have been noted in individuals of African ancestry with T2D, and the observed increased abundance of resident macrophages in the pancreas of AA, combined with upregulation of IL-6 related signaling genes and increased amyloid deposition, may offer an explanation for increased beta cell dysfunction in these individuals. Compensation for severe beta cell loss have been previously reported with alpha cells, PP cells, and exocrine cells being converted to insulin producing cells under extenuating circumstances like T2D [40–42]. Consistent with this induction of insulin-producing cells from other cell populations, we found that exocrine cells from EA donors are the predominant source of non-beta cell insulin, supported by higher chromatin accessibility of transcription factor binding proteins like HNF4A and HNF4G, which promote beta cell differentiation [43, 44]. In contrast, alpha cells are the primary source of non-beta cell insulin expression in AA donors in this study, supported by both higher expression of IAPP and increased chromatin accessibility of HNF4A and HNF4G transcription factor binding sites. Interestingly, IL-6 is associated with alpha cell mass expansion in murine models, a process which is thought to be necessary for functional beta cell compensation [45, 46]. Another known compensatory mechanism for beta cell loss is increased vascular function [46], which may in part explain our observed increase in endothelial cells from AA donors. Together, these findings show that differences in cellular composition, gene expression and transcriptional regulation play an important role in islet pathophysiology in T2D in AA and EA individuals. In summary, we analyzed single cell transcriptional and chromatin accessibility profiles of pancreatic islets from AA and EA donors with T2D. We found important cell distribution differences between the two groups (notably reduced proportions of beta and alpha cells in AA individuals) as well as differences in gene expression and chromatin accessibility. There was also evidence of differences in sources of non-beta cell insulin suggesting that insulin compensatory mechanisms in T2D differs between the two groups. These findings add to our knowledge of islet function in T2D across diverse populations. Declarations Acknowledgements This manuscript used data acquired from the Human Pancreas Analysis Program (HPAP-RRID:SCR_016202) Database (https://hpap.pmacs.upenn.edu/), a Human Islet Research Network (RRID:SCR_014393) consortium (UC4-DK-112217, U01-DK-123594, UC4-DK-112232, and U01-DK-123716). The contents of this publication are solely the responsibility of the authors and do not necessarily represent the official view of the National Institutes of Health. This research was supported by the Intramural Research Program of the Center for Research on Genomics and Global Health (CRGGH). The CRGGH is supported by the National Human Genome Research Institute, the National Institute of Diabetes and Digestive and Kidney Diseases, and the Office of the Director at the National Institutes of Health (1ZIAHG200362). This work utilized the computational resources of the NIH HPC Biowulf cluster (https://hpc.nih.gov). Contributions E.F.H., A.A.A. and C.N.R conceptualized and supervised the project. E.F.H., A.A.A and A.P.D. analyzed the data. E.F.H, A.P.D, and C.N.R interpreted the findings. E.F.H. drafted the manuscript. All authors reviewed, edited and approved the manuscript. Code availability The code necessary to reproduce the primary results of this study are available at the git repository https://github.com/heustonefNIH/PancT2D. Authors’ relationships and activities The authors declare no competing interests. References Gaillard, T.R. and K. Osei, Racial Disparities in the Pathogenesis of Type 2 Diabetes and its Subtypes in the African Diaspora: A New Paradigm. J Racial Ethn Health Disparities, 2016. 3 (1): p. 117-28. Kučera, J., et al., Human White Adipose Tissue Metabolome: Current Perspective. Obesity (Silver Spring), 2018. 26 (12): p. 1870-1878. Capozzi, M.E., et al., Glucagon lowers glycemia when β cells are active. JCI Insight, 2019. 4 (16). Al-Hasani, K., et al., EZH2 inhibitors promote beta-like cell regeneration in young and adult type 1 diabetes donors. Signal Transduct Target Ther, 2024. 9 (1): p. 2. Bartolome, A., The Pancreatic Beta Cell: Editorial. Biomolecules, 2023. 13 (3). Ciochina, M., et al., The Impact of Pancreatic Exocrine Diseases on the beta-Cell and Glucose Metabolism-A Review with Currently Available Evidence. Biomolecules, 2022. 12 (5). Qadir, M.M.F., et al., Single-cell resolution analysis of the human pancreatic ductal progenitor cell niche. Proc Natl Acad Sci U S A, 2020. 117 (20): p. 10876-10887. Gribben, C., et al., Ductal Ngn3-expressing progenitors contribute to adult beta cell neogenesis in the pancreas. Cell Stem Cell, 2021. 28 (11): p. 2000-2008 e4. Liu, G., et al., Charting a high-resolution roadmap for regeneration of pancreatic beta cells by in vivo transdifferentiation from adult acinar cells. Sci Adv, 2023. 9 (21): p. eadg2183. Overton, D.L. and T.L. Mastracci, Exocrine-Endocrine Crosstalk: The Influence of Pancreatic Cellular Communications on Organ Growth, Function and Disease. Front Endocrinol (Lausanne), 2022. 13 : p. 904004. Shapira, S.N., et al., Understanding islet dysfunction in type 2 diabetes through multidimensional pancreatic phenotyping: The Human Pancreas Analysis Program. Cell Metab, 2022. 34 (12): p. 1906-1913. Fleming, S.J., et al., Unsupervised removal of systematic background noise from droplet-based single-cell experiments using CellBender. Nat Methods, 2023. 20 (9): p. 1323-1335. Hao, Y., et al., Dictionary learning for integrative, multimodal and scalable single-cell analysis. Nat Biotechnol, 2024. 42 (2): p. 293-304. McGinnis, C.S., L.M. Murrow, and Z.J. Gartner, DoubletFinder: Doublet Detection in Single-Cell RNA Sequencing Data Using Artificial Nearest Neighbors. Cell Syst, 2019. 8 (4): p. 329-337 e4. Miller, S.A., et al., LSD1 and Aberrant DNA Methylation Mediate Persistence of Enteroendocrine Progenitors That Support BRAF-Mutant Colorectal Cancer. Cancer Res, 2021. 81 (14): p. 3791-3805. Dann, E., et al., Differential abundance testing on single-cell data using k-nearest neighbor graphs. Nat Biotechnol, 2022. 40 (2): p. 245-253. Elizarraras, J.M., et al., WebGestalt 2024: faster gene set analysis and new support for metabolomics and multi-omics. Nucleic Acids Res, 2024. 52 (W1): p. W415-W421. Granja, J.M., et al., ArchR is a scalable software package for integrative single-cell chromatin accessibility analysis. Nat Genet, 2021. 53 (3): p. 403-411. Korsunsky, I., et al., Fast, sensitive and accurate integration of single-cell data with Harmony. Nat Methods, 2019. 16 (12): p. 1289-1296. Liu, J.S. and M. Hebrok, All mixed up: defining roles for beta-cell subtypes in mature islets. Genes Dev, 2017. 31 (3): p. 228-240. Goedecke, J.H. and T. Olsson, Pathogenesis of type 2 diabetes risk in black Africans: a South African perspective. J Intern Med, 2020. 288 (3): p. 284-294. Ladwa, M., et al., Ethnic differences in beta cell function occur independently of insulin sensitivity and pancreatic fat in black and white men. BMJ Open Diabetes Res Care, 2021. 9 (1). Lorenzo, C., et al., Insulin resistance, beta-cell dysfunction, and conversion to type 2 diabetes in a multiethnic population: the Insulin Resistance Atherosclerosis Study. Diabetes Care, 2010. 33 (1): p. 67-72. Feairheller, D.L., et al., Racial differences in oxidative stress and inflammation: in vitro and in vivo. Clin Transl Sci, 2011. 4 (1): p. 32-7. Morris, A.A., et al., Differences in systemic oxidative stress based on race and the metabolic syndrome: the Morehouse and Emory Team up to Eliminate Health Disparities (META-Health) study. Metab Syndr Relat Disord, 2012. 10 (4): p. 252-9. Moon, J.S. and K.C. Won, Pancreatic alpha-Cell Dysfunction in Type 2 Diabetes: Old Kids on the Block. Diabetes Metab J, 2015. 39 (1): p. 1-9. Mizukami, H. and K. Kudoh, Diversity of pathophysiology in type 2 diabetes shown by islet pathology. J Diabetes Investig, 2022. 13 (1): p. 6-13. Pati, D., N. Zhang, and S.E. Plon, Linking sister chromatid cohesion and apoptosis: role of Rad21. Mol Cell Biol, 2002. 22 (23): p. 8267-77. Dahlby, T., et al., Enhancer of Zeste Homolog 2 (EZH2) Mediates Glucolipotoxicity-Induced Apoptosis in beta-Cells. Int J Mol Sci, 2020. 21 (21). Naina Marikar, S., et al., Pharmacological inhibition of human EZH2 can influence a regenerative β-like cell capacity with in vitro insulin release in pancreatic ductal cells. Clinical Epigenetics, 2023. 15 (1): p. 101. Cataldo, L.R., et al., The MafA-target gene PPP1R1A regulates GLP1R-mediated amplification of glucose-stimulated insulin secretion in beta-cells. Metabolism, 2021. 118 : p. 154734. Taneera, J., et al., Unraveling the significance of PPP1R1A gene in pancreatic beta-cell function: A study in INS-1 cells and human pancreatic islets. Life Sci, 2024. 345 : p. 122608. Abedini, A. and A.M. Schmidt, Mechanisms of islet amyloidosis toxicity in type 2 diabetes. FEBS Lett, 2013. 587 (8): p. 1119-27. Van Hulle, F., et al., Formation of amyloid in encapsulated human pancreatic and human stem cell-generated beta cell implants. Am J Transplant, 2021. 21 (6): p. 2090-2099. Wirth, F., et al., A human antibody against pathologic IAPP aggregates protects beta cells in type 2 diabetes models. Nat Commun, 2023. 14 (1): p. 6294. Oliveira, V.R., et al., Pre-treatment with IL-6 potentiates beta-cell death induced by pro-inflammatory cytokines. BMC Mol Cell Biol, 2023. 24 (1): p. 11. Linnemann, A.K., et al., Interleukin 6 protects pancreatic beta cells from apoptosis by stimulation of autophagy. FASEB J, 2017. 31 (9): p. 4140-4152. Akbari, M. and V. Hassan-Zadeh, IL-6 signalling pathways and the development of type 2 diabetes. Inflammopharmacology, 2018. 26 (3): p. 685-698. Li, H., et al., Macrophages, Chronic Inflammation, and Insulin Resistance. Cells, 2022. 11 (19). Zhou, Q., et al., In vivo reprogramming of adult pancreatic exocrine cells to beta-cells. Nature, 2008. 455 (7213): p. 627-32. Thorel, F., et al., Conversion of adult pancreatic alpha-cells to beta-cells after extreme beta-cell loss. Nature, 2010. 464 (7292): p. 1149-54. Furuyama, K., et al., Diabetes relief in mice by glucose-sensing insulin-secreting human alpha-cells. Nature, 2019. 567 (7746): p. 43-48. Wang, G., et al., Integrating genetics with single-cell multiomic measurements across disease states identifies mechanisms of beta cell dysfunction in type 2 diabetes. Nat Genet, 2023. 55 (6): p. 984-994. Ng, N.H.J., et al., HNF4A and HNF1A exhibit tissue specific target gene regulation in pancreatic beta cells and hepatocytes. Nat Commun, 2024. 15 (1): p. 4288. Ellingsgaard, H., et al., Interleukin-6 regulates pancreatic alpha-cell mass expansion. Proc Natl Acad Sci U S A, 2008. 105 (35): p. 13163-8. Almaca, J., et al., Young capillary vessels rejuvenate aged pancreatic islets. Proc Natl Acad Sci U S A, 2014. 111 (49): p. 17612-7. Additional Declarations No competing interests reported. Supplementary Files HeustonSingleCellIsletsandAncestrySupplementaryTables.xlsx Supplementary Data Supplementary Data 1 Table including donor identifier and corresponding ethnicity, assays, and technical replicates used in this study Supplementary Data 2 Number of cells per cluster based on either donor identifier or replicate identifier Supplementary Data 3 Differential gene expression analysis were genes within each cluster are compared to genes across all other clusters Supplementary Data 4 ENCODE motif enrichment per ATAC cluster. Values indicate -log10(P-adj) of normalized enrichment score Supplementary Data 5 ORA analysis for comparisons performed in this study. Note following descriptors for comparisons: · FC1.5 up: All genes at FDR ≤ 0.05 and FC|1.5| were used in ORA · Top100: The 100 genes with highest FC and with FDR ≤ 0.05 were used in ORA · Bottom100: No upregulated genes were identified in this dataset; the 100 genes with the lowest FC and with FDR ≤ 0.05 were used in ORA Supplementary Data 6 Comparison of beta cell populations between African and European ancestries across Beta1 subclusters. Also included are differential gene expression analyses between each beta subcluster when compared to all other beta subclusters Supplementary Data 7 Differential gene analyses between each alpha subcluster when compared to all other alpha subclusters. Also included is differential gene expression analysis between cells from donors of African ancestry and those of European ancestry in Alpha1 Supplementary Data 8 Differential gene analyses between Exocrine1 and Exocrine2. Also included is differential gene expression analysis between cells from donors of African ancestry and those of European ancestry in Exocrine1 Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5881592","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":474200125,"identity":"667b8d33-14b3-471b-9c4f-4f6915ddc0d2","order_by":0,"name":"Elisabeth F. Heuston","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAx0lEQVRIiWNgGAWjYBACAxiDvb2B4TDxWg4AMc+ZAyRruZHAwEyUFnP2HuPPHyruyPFIvj14uIChVs7gAAEtlj1nzCQOnHlmzCOdl3B4BsNxY4JaDG7kmDEcbDucuF86x+AwD8OxxJkNhLTcf2P8AailvkfyDLFabvAYSAC1JPBI8IC01CT2E9DBYHAmrUzizJnDhj08IIcZHDDmJ6jl+OHNHyoqDsvzsJ8x/sxTUSfHRkgLugnExSYKqCNdyygYBaNgFAx7AAA440Pz+NQshQAAAABJRU5ErkJggg==","orcid":"","institution":"National Human Genome Research Institute, National Institutes of Health","correspondingAuthor":true,"prefix":"","firstName":"Elisabeth","middleName":"F.","lastName":"Heuston","suffix":""},{"id":474200126,"identity":"5c1630c6-12e5-45de-903a-aaa0f6ca2854","order_by":1,"name":"Ayo P. Doumatey","email":"","orcid":"","institution":"National Human Genome Research Institute, National Institutes of Health","correspondingAuthor":false,"prefix":"","firstName":"Ayo","middleName":"P.","lastName":"Doumatey","suffix":""},{"id":474200127,"identity":"e579bc04-0f03-449d-9324-9baa97f544b3","order_by":2,"name":"Adebowale A. Adeyemo","email":"","orcid":"","institution":"National Human Genome Research Institute, National Institutes of Health","correspondingAuthor":false,"prefix":"","firstName":"Adebowale","middleName":"A.","lastName":"Adeyemo","suffix":""},{"id":474200128,"identity":"430a2f60-c516-4a8b-88ac-e7688f165d71","order_by":3,"name":"Charles N. Rotimi","email":"","orcid":"","institution":"National Human Genome Research Institute, National Institutes of Health","correspondingAuthor":false,"prefix":"","firstName":"Charles","middleName":"N.","lastName":"Rotimi","suffix":""}],"badges":[],"createdAt":"2025-01-22 14:23:04","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5881592/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5881592/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":86843842,"identity":"d16b2bfb-30d7-4c5a-bdfd-70887a6870ba","added_by":"auto","created_at":"2025-07-16 08:26:09","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":245712,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eA\u003c/strong\u003e. Cell types identified in dissected islet cells. UMAP shows single cell distribution, colored according to assigned identity. A total of 22910 single cell RNA profiles remained after quality control filtering.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eB\u003c/strong\u003e. scRNA cluster cell type identities defined by marker gene expression. Transcript levels for each cell in the cluster is averaged. Circle size represents the percentage of cells expressing the gene. Color represents average gene expression within the cluster (pale green = low, black = high).\u003c/p\u003e","description":"","filename":"HeustonSingleCellIsletsandAncestryFigure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5881592/v1/24f31f5404cd5a77ccfdb226.jpg"},{"id":86844547,"identity":"4f7cfce1-483c-4736-a485-59ffb9b21955","added_by":"auto","created_at":"2025-07-16 08:34:10","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":719437,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eA\u003c/strong\u003e. Clustering assignments of snATAC-Seq profiles generates 25 clusters. UMAP shows distribution of cells based on similarities of accessible chromatin profiles. A total of 62199 snATAC-Seq profiles remained after quality control filtering.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eB\u003c/strong\u003e. Integrated scRNA-Seq and snATAC-Seq profiles define unique chromatin regulatory regions among scRNA clusters. (Top) Average gene score across snATAC-Seq cells in the cluster. Selected marker genes are highlighted on the right. Blue = low TSS accessibility. Yellow = high TSS accessibility. Rows = hierarchically clustered genes. Columns = hierarchically clustered snATAC-Seq clusters. (Bottom) snATAC-Seq and scRNA-Seq integration performed in ArchR by associating gene scores with transcription level profiles. Rows = scRNA-Seq cluster cell type. Columns = snATAC-Seq clusters. Values indicate number of snATAC-Seq cells assigned to the scRNA-Seq cluster. Colors indicate increasing percentage of snATAC-Seq cells in the cluster assigned to scRNA-Seq clusters (White = lowest percentage, Green = highest percentage). Bottom row indicates assignment of snATAC-Seq cluster to scRNA-Seq cluster.\u003c/p\u003e","description":"","filename":"HeustonSingleCellIsletsandAncestryFigure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5881592/v1/af4904ee34761a9027695952.jpg"},{"id":86843844,"identity":"dad5c0c1-c39e-4361-8e16-73dcd833197a","added_by":"auto","created_at":"2025-07-16 08:26:10","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":405258,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eA\u003c/strong\u003e. Distribution of AA and EA cells in pancreatic cell populations. UMAP of all cells, colored according to donor ethnicity (AA = red; EA = blue).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eB\u003c/strong\u003e. Breakdown of cell types within AA (top) or EA (bottom) cell populations. Colors correspond to broad identity categories, and patterns designate different subclusters: Exocrine (blue), Alpha (red), Beta (green), Mixed (brown), Epithelial (pink), Endothelial (orange), Immune (black).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eC\u003c/strong\u003e. Differential abundance (DA) of AA and EA cells across transcriptional clusters. (Left) Putative cell identity and RNA cluster ID. Clusters significantly enriched for EA or AA cells (FDR ≤\u0026nbsp;0.05, log2FD ≥\u0026nbsp;0.58) are indicated in blue or red, respectively. (Middle) Neighborhood analysis to assess differential abundance, as calculated by Milo. (Right) Log2(Fold Difference) [FDR] as calculated by scProportionTest. Significance is defined as Log2FD ≥ 5.8x10\u003csup\u003e-1\u003c/sup\u003e and FDR ≤ 5x10\u003csup\u003e-2\u003c/sup\u003e. Non-significant values are shows in gray italics.\u003c/p\u003e","description":"","filename":"HeustonSingleCellIsletsandAncestryFigure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5881592/v1/f472f23ea5a6adae4bcee07e.jpg"},{"id":86843845,"identity":"ba10aed1-e6cc-4414-95e3-09616c2078c2","added_by":"auto","created_at":"2025-07-16 08:26:10","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":534373,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eA\u003c/strong\u003e. AA and EA cells have different profiles of marker genes across beta cell subclusters. Subclustering of Beta1 cells generated four populations: Beta1a (1254 cells), Beta1b (907 cells), Beta1c (733 cells), Beta1d (153 cells).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eB\u003c/strong\u003e. AA and EA cells have different levels of mitochondrial mapping RNAs within beta subsets. AA had significantly fewer mitochondrial mapping RNAs in Beta1b (FDR ≤\u0026nbsp;4x10\u003csup\u003e-3\u003c/sup\u003e), and significantly more in Beta1c (FDR ≤\u0026nbsp;4x10\u003csup\u003e-2\u003c/sup\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eC\u003c/strong\u003e. Exocrine cells express endocrine markers in EA cells, but not AA cells. Violin plots show expression levels of exocrine cell markers (A, B), beta cells (C), delta cells (D), alpha cells (E), and PP cells (F). Expression is shown as violin, per-cell expression is shown as points. White = AA. Gray = EA. Left violin = Exocrine1 cluster cells. Right violin = Exocrine2 cluster cells.\u003c/p\u003e","description":"","filename":"HeustonSingleCellIsletsandAncestryFigure4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5881592/v1/ebaaa5bed33d5e752bfac876.jpg"},{"id":92271613,"identity":"cd360c6b-3898-4876-8569-a368dcdb783c","added_by":"auto","created_at":"2025-09-26 14:47:16","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2828744,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5881592/v1/e98a3bd5-ef8e-4fec-b712-2cea0dcadcf0.pdf"},{"id":86844539,"identity":"5ac4559a-9fd2-4bed-adeb-cc51a5304995","added_by":"auto","created_at":"2025-07-16 08:34:10","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":4365964,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Data\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplementary Data 1\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTable including donor identifier and corresponding ethnicity, assays, and technical replicates used in this study\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplementary Data 2\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNumber of cells per cluster based on either donor identifier or replicate identifier\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplementary Data 3\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDifferential gene expression analysis were genes within each cluster are compared to genes across all other clusters\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplementary Data 4\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eENCODE motif enrichment per ATAC cluster. Values indicate -log10(P-adj) of normalized enrichment score\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplementary Data 5\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eORA analysis for comparisons performed in this study. Note following descriptors for comparisons:\u003c/p\u003e\n\u003cp\u003e· FC1.5 up: All genes at FDR ≤ 0.05 and FC|1.5| were used in ORA\u003c/p\u003e\n\u003cp\u003e· Top100: The 100 genes with highest FC and with FDR ≤ 0.05 were used in ORA\u003c/p\u003e\n\u003cp\u003e· Bottom100: No upregulated genes were identified in this dataset; the 100 genes with the lowest FC and with FDR ≤ 0.05 were used in ORA\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplementary Data 6\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eComparison of beta cell populations between African and European ancestries across Beta1 subclusters. Also included are differential gene expression analyses between each beta subcluster when compared to all other beta subclusters\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplementary Data 7\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDifferential gene analyses between each alpha subcluster when compared to all other alpha subclusters. Also included is differential gene expression analysis between cells from donors of African ancestry and those of European ancestry in Alpha1\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplementary Data 8\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDifferential gene analyses between Exocrine1 and Exocrine2. Also included is differential gene expression analysis between cells from donors of African ancestry and those of European ancestry in Exocrine1\u003c/p\u003e","description":"","filename":"HeustonSingleCellIsletsandAncestrySupplementaryTables.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5881592/v1/2e79f21f00288f57b281c89c.xlsx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Multi-omics islet profiling in type 2 diabetes reveals differential transcriptomics and chromatin accessibility by ancestry","fulltext":[{"header":"Introduction","content":"\u003cp\u003eIndividuals of African ancestry in the United States are disproportionally affected by type 2 diabetes mellitus (T2D) and are more likely to experience associated cardiometabolic comorbidities including hypertension, retinopathy, and hyperlipidemia [1]. However, most omics studies have focused on individuals from populations of European and Asian ancestries [2] and African ancestry populations remain understudied despite experiencing a greater burden of the disorder. The pancreas is one of the major organs involved in T2D pathophysiology given its central role in insulin secretion. Pancreatic islets are mostly composed of alpha and beta cells that produce hormones (e.g., glucagon and insulin, respectively) to regulate endocrine signaling [3, 4]. The islets are encircled by acinar and ductal exocrine cells that secrete enzymes to mediate digestion and nutrient release. Historically, the endocrine and exocrine functions of the pancreas were considered distinct and largely independent of each other. However, an increasing body of evidence identifies extensive cross-talk between these populations, suggesting that their relationship is significantly more complicated [5, 6]. For example, it has been suggested that transdifferentiation from ductal progenitors, acinar cells, and alpha cells, into insulin-producing cells does occur, demonstrating cellular plasticity between pancreatic cell populations [7-10]. Thus, studies of pancreatic cell populations are essential to understand the molecular changes associated with T2D in the pancreas and potentially illuminate pathophysiological changes observed in individuals living with T2D.\u003c/p\u003e\n\u003cp\u003eTaking advantage of the Human Pancreas Analysis Program (HPAP) [11], which provides pancreatic islet single cell multi-omic studies of individuals of multiple ancestries, we investigated islet cell molecular profiles between African Americans (AA) and European Americans (EA). We analyzed single cell RNA (scRNA-Seq) and single nucleus ATAC (snATAC-Seq) sequencing profiles of AA and EA pancreas donors from HPAP with T2D. Our findings reveal differences in islet cell composition and insulin-expressing subpopulations between AA and EA donors, and reveal novel mechanisms that could provide explanations for some of the pathophysiologic features of T2D reported in AA individuals.\u003c/p\u003e"},{"header":"Research Design and Methods","content":"\u003cp\u003e\u003cem\u003eData access\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eRaw sequencing data of hand-picked islet cells for single cell RNA (scRNA-Seq) and single nucleus ATAC (scATAC-Seq) sequencing were accessed through the data portal of The Human Pancreas Analysis Program (version 2.1.2). Donor inclusion criteria included: Ancestry (\u0026ldquo;Caucasian\u0026rdquo; OR \u0026ldquo;African American\u0026rdquo;); Disease (\u0026ldquo;T2D\u0026rdquo;); single cell RNA capture (10X Genomics Chromium 3\u0026rsquo; Gene Expression platform with v2, v3, or v3.1 chemistries) or single cell ATAC capture (10X Genomics Chromium Single Cell ATAC platform v1 chemistry).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSingle cell RNA sequencing data processing\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eFASTQ files were demultiplexed in Cell Ranger_7.1.0 (10x Genomics) and mapped to the GRCh38-2020-A transcriptome assembly.\u003cstrong\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003e\u003c/strong\u003eAmbient RNA was excluded with CellBender_0.3.0 [12]. Quality control (QC) and clustering was performed with Seurat_5.0.1 [13], SeuratObject_5.0.0, and DoubletFinder_2.0.3 [14]. scRNA-Seq cell filtering was performed as follows: nFeature_RNA \u0026ge; 200, nFeature_RNA \u0026le; 2500, percentage reads mapped mitochondrial RNAs (PMMR) \u0026le; 10. Predicted doublet rate was set at 0.05. Batch correction was performed SCTransform and anchor integration. Clusters were called using a resolution of 0.4 (total cell population) or 0.2 (beta cell subclustering). Down sampling of cells per group was done using Seurat\u0026rsquo;s \u003cem\u003esubset (downsample)\u003c/em\u003e function. Differential abundance analyses were performed with scProportion_0.0.0.9000 [15] and MiloR_2.0.0 [16], and presented as log2 of the fold difference (log2FD). Over representation analyses (ORA) and gene set enrichment analyses (GSEA) were performed in WebGestalt 2024 [17]. Significance was defined as FDR \u0026le;\u0026nbsp;1x10\u003csup\u003e-1\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSingle cell ATAC sequencing data processing\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eFASTQ files were demultiplexed in Cell Ranger-ATAC version 2.1.0 (10X Genomics) and mapped to the cellranger-arc-GRCh38-2020-A-2.0.0\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003egenome assembly. QC, clustering, and integration with single cell RNA sequencing data was performed in ArchR_1.0.2 [18]. snATAC-Seq filtering was performed as follows: nFrags \u0026gt; 1000, nFrags \u0026lt; 40000, BlacklistRatio \u0026le; 0.03, TSSEnrichment \u0026gt; 8. Batch correction was performed with Harmony [19].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eData and code availability\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSource data are available at the Data Portal of the Human Pancreas Analysis Program (https://hpap.pmacs.upenn.edu/). The code necessary to reproduce the primary results of this study are available at the git repository https://github.com/heustonefNIH/PancT2D.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eRNA and ATAC characteristics of islets in T2D\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSingle cell RNA-Seq data of dissected islets from donors with T2D were obtained from HPAP. Characteristics of the sample donors are shown in \u003cstrong\u003eTable 1\u003c/strong\u003e. After QC, a total of 6679 scRNA-Seq cells from six donors of African ancestry (AA) and 16231 cells from six donors of European ancestry (EA) were analyzed (\u003cstrong\u003eOnline Supplementary Data 1\u003c/strong\u003e). Unbiased clustering with a conservative resolution of 0.4 identified 17 RNA clusters (designated RNA0, RNA1, etc.) (\u003cstrong\u003eFigure 1A, Online Supplementary Data 2\u003c/strong\u003e). Differential gene expression analysis (DGE) and canonical marker genes led to identification of four endocrine cell clusters, two exocrine cell clusters, two mesenchyme cell clusters, four immune cell clusters, three mixed clusters, and one cluster each of epithelial and endothelial cells (\u003cstrong\u003eFigure 1B, Online Supplement Figure 1, Online Supplementary Data 3\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eWe also analyzed snATAC-Seq data from HPAP and performed integrated transcriptional and chromatin analysis. Of thirteen total donors with snATAC-Seq data, ten also had independently assayed (i.e., not multiome-matched) scRNA-Seq data. After QC, a total of 23226 cells from five AA donors and 38976 cells from eight EA donors were analyzed. Nearest-neighbor analysis generated 25 clusters (designated ATAC1, ATAC2, etc., \u003cstrong\u003eFigure 2A\u003c/strong\u003e). We used gene scores based on transcriptional start site (TSS) accessibility to predict transcriptional profiles and to assign each RNA cluster to a chromatin profile (\u003cstrong\u003eFigure 2B\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eDifferential distribution of beta cell clusters and regulation of beta cell function are observed between AA and EA\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOur conservative clustering parameters identified one beta cell cluster with an absence of other definitive cell markers (\u003cstrong\u003eFigure 1\u003c/strong\u003e). Beta1 (RNA2) exhibited upregulated islet amyloid polypeptide (\u003cem\u003eIAPP\u003c/em\u003e), SIX homeobox 3 (\u003cem\u003eSIX3\u003c/em\u003e) and insulin (\u003cem\u003eINS\u003c/em\u003e) RNA and was associated with four ATAC clusters: ATAC1, ATAC23, ATAC24, and ATAC25. These ATAC clusters were enriched for chromatin remodeling motifs for proteins such as cohesion complex component (RAD21), suppressor of zeste 12 (SUZ12), and enhancer of zeste 2 polycomb repressive complex 2 subunit (EZH2) (ATAC23) (\u003cstrong\u003eOnline Supplementary Data 4\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eWe performed differential abundance testing (DA) to determine if AA and EA cells were distributed differently across RNA clusters (\u003cstrong\u003eFigure 3A\u003c/strong\u003e). We found that AA cells proportions in the Beta1 cluster were significantly lower than in EA cells (6% of total AA versus 16% of total EA; log2FD (fold difference) = -1.4, FDR \u0026le; 2x10\u003csup\u003e-3\u003c/sup\u003e; \u003cstrong\u003eFigure 3B, Figure 3C\u003c/strong\u003e). Given the large discrepancy in numbers between EA and AA cells in this cluster (2634 EA versus 413 AA), the EA Beta1 cluster was randomly downsampled (see Methods) to 413 cells for subsequent beta cell comparisons. Beta1 DGE between AA and EA found regenerating family member 1 alpha (\u003cem\u003eREG1A\u003c/em\u003e)\u003cem\u003e,\u003c/em\u003e pancreatic lipase\u003cem\u003e\u0026nbsp;\u003c/em\u003e(\u003cem\u003ePNLIP\u003c/em\u003e)\u003cem\u003e,\u003c/em\u003e serine protease 1 (\u003cem\u003ePRSS1\u003c/em\u003e)\u003cem\u003e,\u003c/em\u003e and serine protease 2 (\u003cem\u003ePRSS2)\u003c/em\u003e (log2FC \u0026ge; 4.4, FDR \u0026le; 3x10\u003csup\u003e-7\u003c/sup\u003e) were upregulated in AA, \u003cem\u003eINS\u003c/em\u003e moderately downregulated in AA (log2FC = 0.26, FDR \u0026le; 3x10\u003csup\u003e-6\u003c/sup\u003e), and enrichment of genes involved in \u003cem\u003eGlucose Intolerance\u003c/em\u003e (ER = 5.6, FDR \u0026le; 2x10\u003csup\u003e-2\u003c/sup\u003e), \u003cem\u003eFat Malabsorption\u003c/em\u003e (ER = 24, FDR \u0026le; 2x10\u003csup\u003e-3\u003c/sup\u003e), and \u003cem\u003eChronic Pancreatitis\u003c/em\u003e (ER = 223, FDR \u0026le; 4x10\u003csup\u003e-10\u003c/sup\u003e) compared to EA (\u003cstrong\u003eOnline Supplementary Data 5\u003c/strong\u003e). These data show that AA cells within the Beta1 cluster express exocrine cell markers at a higher level than do EA cells.\u003c/p\u003e\n\u003cp\u003eTo assess DA of AA and EA cells across beta cell subsets [20], we subclustered Beta1 and identified four subpopulations (\u003cstrong\u003eFigure 4A, Online Supplementary Data 6\u003c/strong\u003e). Although the total number of beta cells was significantly lower in AA, we found different distributions of existing beta cells within three of the four subsets. Beta1b contained 38% of all Beta1 AA cells (155 cells) compared to 29% of all EA Beta1 cells (752 cells; FDR = 4x10\u003csup\u003e-3\u003c/sup\u003e). Beta1c contained 20% of all Beta1 AA (84 cells) compared to 25% of all Beta1 EA (649 cells; FDR = 4x10\u003csup\u003e-2\u003c/sup\u003e). Beta1d contained 3% of Beta1 AA (11 cells) compared to 5% of Beta1 EA (142 cells; FDR = 8x10\u003csup\u003e-3\u003c/sup\u003e). Beta1a was not significantly different between AA and EA populations. Compared to other beta cell subsets, Beta1b expressed genes associated with \u003cem\u003eAmyloid-beta Formation\u003c/em\u003e (ER = 25, FDR \u0026le; 4x10\u003csup\u003e-2\u003c/sup\u003e) and \u003cem\u003eIL6-mediated signaling\u003c/em\u003e (ER = 47, FDR \u0026le; 5x10\u003csup\u003e-2\u003c/sup\u003e, \u003cstrong\u003eOnline Supplementary Data 5\u003c/strong\u003e), was mildly enriched for AA cells (log2FD = 0.39, FDR \u0026le;\u0026nbsp;4x10\u003csup\u003e-3\u003c/sup\u003e), and had a lower percentage of mitochondrial-mapping RNAs (FDR \u0026le;\u0026nbsp;5x10\u003csup\u003e-6\u003c/sup\u003e; \u003cstrong\u003eFigure 4B\u003c/strong\u003e). Interestingly macrophages, which a can be activated by IL6/IL-6R signaling, were also significantly enriched in AA (RNA14, log2FD = 0.92, FDR \u0026le;\u0026nbsp;1.5x10\u003csup\u003e-3\u003c/sup\u003e). In contrast to Beta1b, Beta1c had upregulated solute carrier family 35 member F1 (\u003cem\u003eSLC35F1\u003c/em\u003e) and protein phosphatase 1 regulatory inhibitor subunit 1A (\u003cem\u003ePPP1R1A\u003c/em\u003e) expression than other Beta1 subpopulations, was mildly depleted for AA (log2FD = -0.28, FDR \u0026le; 5x10\u003csup\u003e-2\u003c/sup\u003e), and had a higher percentage of mitochondrial-mapping RNAs (FDR \u0026le;\u0026nbsp;7x10\u003csup\u003e-3\u003c/sup\u003e; \u003cstrong\u003eFigure 4B\u003c/strong\u003e). Together these data show that fewer AA cells exist within the beta cell compartment, and those that do have a distinct transcriptional profile compared to their EA counterparts.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAlpha cell clusters show differential abundance and regulation between AA and EA\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThree clusters expressed canonical alpha cell markers (e.g., glucagon (\u003cem\u003eGCG\u003c/em\u003e) and transthyretin (\u003cem\u003eTTR\u003c/em\u003e)) in the absence of other definitive markers: Alpha1 (RNA0), Alpha2 (RNA9), and Alpha3 (RNA16) (\u003cstrong\u003eFigure 1\u003c/strong\u003e). Alpha1 and Alpha2 were together associated with six ATAC clusters and shared similar DNA binding motifs including GATA binding protein 3 (GATA3) (ATAC2; ATAC5), tripartite motif-containing 28 (TRIM28) (ATAC3), RAD21 (ATAC21), hepatocyte nuclear factor 4 gamma (HNF4G), hepatocyte nuclear factor 4 alpha (HNF4A) (ATAC 5), and for chromatin remodeling-associated proteins including SUZ12, E2F transcription factor 6 (E2F6), and TATA-box binding protein associated factor 1 (TAF1) (ATAC21) (\u003cstrong\u003eOnline Supplementary Data 4\u003c/strong\u003e). Compared to other RNA clusters, Alpha1 had the highest expression of canonical alpha cell genes including \u003cem\u003eGCG, TTR,\u0026nbsp;\u003c/em\u003eand\u003cem\u003e\u0026nbsp;\u003c/em\u003elysyl oxidase like 4 (\u003cem\u003eLOXL4\u003c/em\u003e), and was significantly enriched for predicted targets of the \u003cem\u003eNRSF Transcription Factor\u0026nbsp;\u003c/em\u003e(ER = 12, FDR \u0026le; 2x10\u003csup\u003e-2\u003c/sup\u003e). Alpha2 expressed high ROS proto-oncogene 1 receptor tyrosine kinase (\u003cem\u003eROS1\u003c/em\u003e)\u003cem\u003e\u0026nbsp;\u003c/em\u003eand olfactory receptor family 5 subfamily AU member 1 (\u003cem\u003eOR5AU1\u003c/em\u003e) and was significantly enriched for \u003cem\u003eMacrophage-derived Foam Cell Differentiation\u003c/em\u003e (ER = 60, FDR \u0026le; 5x10\u003csup\u003e-3\u003c/sup\u003e). Alpha3 contained fewer than 1% of the total population of cells and was excluded from further analyses.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAA cells were significantly depleted (i.e. had lower proportions) in the Alpha1 cell cluster (log2FD \u0026ge; 1.1, FDR \u0026le; 1x10\u003csup\u003e-3\u003c/sup\u003e; \u003cstrong\u003eFigure 3B-C\u0026nbsp;\u003c/strong\u003e) and expressed significantly higher \u003cem\u003eIAPP\u003c/em\u003e (log2FC = 8.6, FDR \u0026le; 2x10\u003csup\u003e-3\u003c/sup\u003e) and \u003cem\u003eINS\u003c/em\u003e (log2FC = 2.2, FDR \u0026le; 2x10\u003csup\u003e-5\u003c/sup\u003e) compared to EA (\u003cstrong\u003eOnline Supplementary Data 7\u003c/strong\u003e), although \u003cem\u003eGCG\u0026nbsp;\u003c/em\u003eexpression was not significantly different between these groups. In contrast, EA cells expressed significantly higher retinol binding protein 1 (\u003cem\u003eRBP1)\u003c/em\u003e (retinol transport), transketolase (\u003cem\u003eTKT\u003c/em\u003e) (glycolysis), and fatty acid binding protein 5 (\u003cem\u003eFABP5\u003c/em\u003e) (lipid synthesis) (log2FC \u0026ge; 1.7, FDR \u0026le; 8x10\u003csup\u003e-8\u003c/sup\u003e). These data show that, although the largest alpha cell population is depleted (lower) in AA, these cells express higher levels of insulin than those of EA donors.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eExocrine cell clusters display differential expression of insulin RNA between AA and EA\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTwo clusters expressed canonical exocrine cell markers (e.g., \u003cem\u003eREG1A\u003c/em\u003e and \u003cem\u003ePNLIP\u003c/em\u003e): Exocrine1 (RNA1) and Exocrine 2 (RNA3). Exocrine1 was associated with 3 ATAC clusters (\u003cstrong\u003eFigure 2A\u003c/strong\u003e) and was enriched for TFBS including retinoid X receptor alpha (RXRA), histone deacetylase 2 (HDAC2), MYB proto-oncogene like 2 (MYBL2) (ATAC17; ATAC18), TRIM28 (ATAC18), and HNF4G and HNF4A (ATAC19) (\u003cstrong\u003eOnline Supplementary Data 4\u003c/strong\u003e). Exocrine2-associated chromatin profiles overlapped in part with ATAC17 but predominantly with ATAC24 (which was assigned to Beta1, \u003cstrong\u003eFigure 2B\u003c/strong\u003e). Compared to other RNA clusters, Exocrine1 expressed the highest levels of canonical acinar cell genes including\u003cem\u003e\u0026nbsp;PNLIP, PRSS1,\u0026nbsp;\u003c/em\u003eand\u003cem\u003e\u0026nbsp;PRSS2\u003c/em\u003e, and was significantly enriched for genes associated with \u003cem\u003eProtein and Fat Digestion and Absorption\u003c/em\u003e (ER \u0026ge; 24, FDR \u0026le; 0.001) and \u003cem\u003ePeptidase Activities\u003c/em\u003e (ER \u0026ge; 7, FDR \u0026le; 2x10\u003csup\u003e-10\u003c/sup\u003e). In contrast, Exocrine2 expressed upregulated progastricsin (\u003cem\u003ePGC\u003c/em\u003e)\u003cem\u003e\u0026nbsp;\u003c/em\u003eand lymphotoxin beta (\u003cem\u003eLTB\u003c/em\u003e) (previously associated with pancreatic inflammatory responses), and was enriched for processes associated with \u003cem\u003eDetoxification\u0026nbsp;\u003c/em\u003e(ER \u0026ge; 15.0, FDR \u0026le;\u0026nbsp;1x10\u003csup\u003e-6\u003c/sup\u003e) (\u003cstrong\u003eOnline Supplementary Data 5\u003c/strong\u003e). Exocrine1 versus Exocrine2 DGE, followed by gene set enrichment analysis (GSEA), showed that Exocrine1 had higher MT-RNR2 like 1 (\u003cem\u003eMTRNR2L1\u003c/em\u003e) and (glycoprotein 2) \u003cem\u003eGP2\u0026nbsp;\u003c/em\u003eexpression\u003cem\u003e\u0026nbsp;\u003c/em\u003eand was enriched for \u003cem\u003eIntegrin Signaling\u003c/em\u003e (NES = 1.6, FDR \u0026le; 1x10\u003csup\u003e-1\u003c/sup\u003e). In contrast, Exocrine2 had higher (Wnt family member 4 (\u003cem\u003eWNT4\u003c/em\u003e), ubiquitin C-terminal hydrolase L1 (\u003cem\u003eUCHL1\u003c/em\u003e) and NK6 homeobox 1 (\u003cem\u003eNKX6-1\u003c/em\u003e) expression and was enriched for targets of NKX6.1 (NES = 1.9, FDR \u0026le; 1x10\u003csup\u003e-1\u003c/sup\u003e) and RE1-silencing transcription factor (REST) (NES = 2.3, FDR \u0026le; 7x10\u003csup\u003e-3\u003c/sup\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAA cells were significantly enriched in Exocrine1 (log2FD = 0.94, FDR \u0026le;\u0026nbsp;2x10\u003csup\u003e-3\u003c/sup\u003e) and Exocrine2 (log2FD = 1.45, FDR \u0026le;\u0026nbsp;2x10\u003csup\u003e-3\u003c/sup\u003e) compared to EA cells. Despite having a lower abundance of cells, Exocrine1 EA cells expressed significantly higher \u003cem\u003eINS\u003c/em\u003e (log2FC = 4.4, FDR \u0026le; 2x10\u003csup\u003e-9\u003c/sup\u003e) than AA cells (\u003cstrong\u003eOnline Supplementary Data 8\u003c/strong\u003e), while Exocrine2 EA cells expressed significantly higher \u003cem\u003eINS\u003c/em\u003e (log2FC = 4.4, FDR \u0026le; 2x10\u003csup\u003e-9\u003c/sup\u003e), \u003cem\u003eGCG\u003c/em\u003e (log2FC = 2.2, FDR \u0026le; 2x10\u003csup\u003e-130\u003c/sup\u003e), somatostatin (\u003cem\u003eSST\u003c/em\u003e) (log2FC = 1.8, FDR \u0026le; 8x10\u003csup\u003e-112\u003c/sup\u003e), and pancreatic polypeptide (\u003cem\u003ePPY\u003c/em\u003e) (log2FC = 1.9, FDR \u0026le; 1x10\u003csup\u003e-115\u003c/sup\u003e) (\u003cstrong\u003eFigure 4C\u003c/strong\u003e) than AA cells. Finally, we tested whether cells expressing exocrine markers co-expressed endocrine markers, or if cells expressing either exocrine or endocrine markers were co-clustering. In Exocrine1, more EA cells expressed both \u003cem\u003ePNLIP\u003c/em\u003e (exocrine) and either \u003cem\u003eINS\u003c/em\u003e, \u003cem\u003eGCG\u003c/em\u003e, or \u003cem\u003eSST\u003c/em\u003e, compared to AA cells (\u003cstrong\u003eOnline Supplement Figure 2\u003c/strong\u003e). In Exocrine2, higher percentages of EA cells co-expressed both exocrine and endocrine markers. Together these data show that signals of insulin-expressing non-beta cells arise from different cell populations in AA and EA T2D pancreatic tissue donors.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eAfrican Americans with T2D display a number of pathophysiologic differences when compared with European Americans with T2D, including differences in beta-cell function, peripheral and hepatic insulin sensitivity, hepatic insulin clearance and hepatic glucose dysregulation [21\u0026ndash;23]. These differences are already present in individuals \u003cem\u003ewithout\u003c/em\u003e T2D and are found in concert with markers of subclinical inflammation and oxidative stress (among others) [24, 25]. In the present study, we focus on the pancreas, one of the key organs in T2D pathophysiology, hypothesizing that single cell analysis of pancreatic islets cells may illuminate some of these pathophysiologic features. We analyzed over 85,000 single cells for either RNA or ATAC profiles of pancreatic cells in AA and EA individuals with T2D. Our findings show that alpha and beta cells are significantly depleted (i.e. had significantly lower proportions) in AA donor cell populations when contrasted with EA donor cell populations, while exocrine cells are comparatively depleted (lower) in EA donor cell populations. Alongside these distributional differences, alternative sources of insulin transcripts apparently come from alpha cells in AA donors and exocrine cells in EA donors. The findings of markedly smaller numbers of alpha and beta cells in AA as well as different islet transcriptomic profiles (when compared with EA) are noteworthy because early beta cell dysfunction in T2D is a known feature in individuals of African ancestry [1]. Reduction in beta cell numbers and/or beta cell dysfunction result in relative or absolute insulin deficiency that is a feature of the dysglycemia seen in T2D while increasing evidence suggests that alpha cell dysfunction is an important contributor to hyperglycemia in T2D [26].\u003c/p\u003e \u003cp\u003eLoss of beta cell function in T2D can attribute to a number of mechanisms including decrease in beta cell mass/volume due to apoptosis, transdifferentiation and dedifferentiation [27]. In our study, we found enrichment in TF-like (e.g., RAD21) and chromatin remodeling-associated TFBS (e.g. EZH2), both of which are implicated in apoptosis [28, 29]. EZH2 is known to mediate glucolipotoxicity induced apoptosis of beta cells [29]. Interestingly, pharmacological inhibition of EZH2 in humans affects the regenerative of beta like cell ability with insulin release in vitro [30]. In this study, beta cells from AA donors were more strongly associated with amyloid deposition, IL-6 signaling, and insulin expression, but deficient in a population of \u003cem\u003ePPP1R1A\u003c/em\u003e-expressing cells. Loss of \u003cem\u003ePPP1R1A\u003c/em\u003e has been associated with impaired insulin secretion and glucose metabolism [31, 32] and amyloid accumulation has been known for its detrimental effects on beta cell function. Amyloid deposits are characteristic pathological features of T2D associated with the misfolding of islet amyloid polypeptide (IAPP), the second most abundant protein expressed by beta cells and is secreted along with insulin in response to diet (e.g., glucose, fat, and amino acids). Islet amyloid deposits appear to be population-dependent (for example occurring in over 80% of European ancestry individuals in contrast to only 30\u0026ndash;40% of Japanese individuals) and has been suggested to be one of the features of differences in pathological and pathophysiological differences of metabolic diseases between populations [27, 33\u0026ndash;35]. Our findings are consistent with this notion and provide a potential explanation for the observed beta cell secretory defects in African Americans [1]. While the role of IL-6 signaling in islet cells is complex [36\u0026ndash;38], it should be noted that IL-6 is pleiotropic and has both pro-inflammatory and anti-inflammatory/regenerative effects depending on cellular conditions. IL-6 also plays a role in macrophage-mediated inflammation in T2D [39]. Higher levels of inflammation have been noted in individuals of African ancestry with T2D, and the observed increased abundance of resident macrophages in the pancreas of AA, combined with upregulation of IL-6 related signaling genes and increased amyloid deposition, may offer an explanation for increased beta cell dysfunction in these individuals.\u003c/p\u003e \u003cp\u003eCompensation for severe beta cell loss have been previously reported with alpha cells, PP cells, and exocrine cells being converted to insulin producing cells under extenuating circumstances like T2D [40\u0026ndash;42]. Consistent with this induction of insulin-producing cells from other cell populations, we found that exocrine cells from EA donors are the predominant source of non-beta cell insulin, supported by higher chromatin accessibility of transcription factor binding proteins like HNF4A and HNF4G, which promote beta cell differentiation [43, 44]. In contrast, alpha cells are the primary source of non-beta cell insulin expression in AA donors in this study, supported by both higher expression of IAPP and increased chromatin accessibility of HNF4A and HNF4G transcription factor binding sites. Interestingly, IL-6 is associated with alpha cell mass expansion in murine models, a process which is thought to be necessary for functional beta cell compensation [45, 46]. Another known compensatory mechanism for beta cell loss is increased vascular function [46], which may in part explain our observed increase in endothelial cells from AA donors. Together, these findings show that differences in cellular composition, gene expression and transcriptional regulation play an important role in islet pathophysiology in T2D in AA and EA individuals.\u003c/p\u003e \u003cp\u003eIn summary, we analyzed single cell transcriptional and chromatin accessibility profiles of pancreatic islets from AA and EA donors with T2D. We found important cell distribution differences between the two groups (notably reduced proportions of beta and alpha cells in AA individuals) as well as differences in gene expression and chromatin accessibility. There was also evidence of differences in sources of non-beta cell insulin suggesting that insulin compensatory mechanisms in T2D differs between the two groups. These findings add to our knowledge of islet function in T2D across diverse populations.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003e Acknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis manuscript used data acquired from the Human Pancreas Analysis Program (HPAP-RRID:SCR_016202) Database (https://hpap.pmacs.upenn.edu/), a Human Islet Research Network (RRID:SCR_014393) consortium (UC4-DK-112217, U01-DK-123594, UC4-DK-112232, and U01-DK-123716). The contents of this publication are solely the responsibility of the authors and do not necessarily represent the official view of the National Institutes of Health. This research was supported by the Intramural Research Program of the Center for Research on Genomics and Global Health (CRGGH). The CRGGH is supported by the National Human Genome Research Institute, the National Institute of Diabetes and Digestive and Kidney Diseases, and the Office of the Director at the National Institutes of Health (1ZIAHG200362). This work utilized the computational resources of the NIH HPC Biowulf cluster (https://hpc.nih.gov).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eContributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eE.F.H., A.A.A. and C.N.R conceptualized and supervised the project. E.F.H., A.A.A and A.P.D. analyzed \u0026nbsp;the data. E.F.H, A.P.D, and C.N.R interpreted the findings. \u0026nbsp;E.F.H. drafted the manuscript. All authors reviewed, edited and approved the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCode availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe code necessary to reproduce the primary results of this study are available at the git repository https://github.com/heustonefNIH/PancT2D.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; relationships and activities\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eGaillard, T.R. and K. Osei, \u003cem\u003eRacial Disparities in the Pathogenesis of Type 2 Diabetes and its Subtypes in the African Diaspora: A New Paradigm.\u003c/em\u003e J Racial Ethn Health Disparities, 2016. \u003cstrong\u003e3\u003c/strong\u003e(1): p. 117-28.\u003c/li\u003e\n\u003cli\u003eKučera, J., et al., \u003cem\u003eHuman White Adipose Tissue Metabolome: Current Perspective.\u003c/em\u003e Obesity (Silver Spring), 2018. \u003cstrong\u003e26\u003c/strong\u003e(12): p. 1870-1878.\u003c/li\u003e\n\u003cli\u003eCapozzi, M.E., et al., \u003cem\u003eGlucagon lowers glycemia when \u0026beta; cells are active.\u003c/em\u003e JCI Insight, 2019. \u003cstrong\u003e4\u003c/strong\u003e(16).\u003c/li\u003e\n\u003cli\u003eAl-Hasani, K., et al., \u003cem\u003eEZH2 inhibitors promote beta-like cell regeneration in young and adult type 1 diabetes donors.\u003c/em\u003e Signal Transduct Target Ther, 2024. \u003cstrong\u003e9\u003c/strong\u003e(1): p. 2.\u003c/li\u003e\n\u003cli\u003eBartolome, A., \u003cem\u003eThe Pancreatic Beta Cell: Editorial.\u003c/em\u003e Biomolecules, 2023. \u003cstrong\u003e13\u003c/strong\u003e(3).\u003c/li\u003e\n\u003cli\u003eCiochina, M., et al., \u003cem\u003eThe Impact of Pancreatic Exocrine Diseases on the beta-Cell and Glucose Metabolism-A Review with Currently Available Evidence.\u003c/em\u003e Biomolecules, 2022. \u003cstrong\u003e12\u003c/strong\u003e(5).\u003c/li\u003e\n\u003cli\u003eQadir, M.M.F., et al., \u003cem\u003eSingle-cell resolution analysis of the human pancreatic ductal progenitor cell niche.\u003c/em\u003e Proc Natl Acad Sci U S A, 2020. \u003cstrong\u003e117\u003c/strong\u003e(20): p. 10876-10887.\u003c/li\u003e\n\u003cli\u003eGribben, C., et al., \u003cem\u003eDuctal Ngn3-expressing progenitors contribute to adult beta cell neogenesis in the pancreas.\u003c/em\u003e Cell Stem Cell, 2021. \u003cstrong\u003e28\u003c/strong\u003e(11): p. 2000-2008 e4.\u003c/li\u003e\n\u003cli\u003eLiu, G., et al., \u003cem\u003eCharting a high-resolution roadmap for regeneration of pancreatic beta cells by in vivo transdifferentiation from adult acinar cells.\u003c/em\u003e Sci Adv, 2023. \u003cstrong\u003e9\u003c/strong\u003e(21): p. eadg2183.\u003c/li\u003e\n\u003cli\u003eOverton, D.L. and T.L. Mastracci, \u003cem\u003eExocrine-Endocrine Crosstalk: The Influence of Pancreatic Cellular Communications on Organ Growth, Function and Disease.\u003c/em\u003e Front Endocrinol (Lausanne), 2022. \u003cstrong\u003e13\u003c/strong\u003e: p. 904004.\u003c/li\u003e\n\u003cli\u003eShapira, S.N., et al., \u003cem\u003eUnderstanding islet dysfunction in type 2 diabetes through multidimensional pancreatic phenotyping: The Human Pancreas Analysis Program.\u003c/em\u003e Cell Metab, 2022. \u003cstrong\u003e34\u003c/strong\u003e(12): p. 1906-1913.\u003c/li\u003e\n\u003cli\u003eFleming, S.J., et al., \u003cem\u003eUnsupervised removal of systematic background noise from droplet-based single-cell experiments using CellBender.\u003c/em\u003e Nat Methods, 2023. \u003cstrong\u003e20\u003c/strong\u003e(9): p. 1323-1335.\u003c/li\u003e\n\u003cli\u003eHao, Y., et al., \u003cem\u003eDictionary learning for integrative, multimodal and scalable single-cell analysis.\u003c/em\u003e Nat Biotechnol, 2024. \u003cstrong\u003e42\u003c/strong\u003e(2): p. 293-304.\u003c/li\u003e\n\u003cli\u003eMcGinnis, C.S., L.M. Murrow, and Z.J. Gartner, \u003cem\u003eDoubletFinder: Doublet Detection in Single-Cell RNA Sequencing Data Using Artificial Nearest Neighbors.\u003c/em\u003e Cell Syst, 2019. \u003cstrong\u003e8\u003c/strong\u003e(4): p. 329-337 e4.\u003c/li\u003e\n\u003cli\u003eMiller, S.A., et al., \u003cem\u003eLSD1 and Aberrant DNA Methylation Mediate Persistence of Enteroendocrine Progenitors That Support BRAF-Mutant Colorectal Cancer.\u003c/em\u003e Cancer Res, 2021. \u003cstrong\u003e81\u003c/strong\u003e(14): p. 3791-3805.\u003c/li\u003e\n\u003cli\u003eDann, E., et al., \u003cem\u003eDifferential abundance testing on single-cell data using k-nearest neighbor graphs.\u003c/em\u003e Nat Biotechnol, 2022. \u003cstrong\u003e40\u003c/strong\u003e(2): p. 245-253.\u003c/li\u003e\n\u003cli\u003eElizarraras, J.M., et al., \u003cem\u003eWebGestalt 2024: faster gene set analysis and new support for metabolomics and multi-omics.\u003c/em\u003e Nucleic Acids Res, 2024. \u003cstrong\u003e52\u003c/strong\u003e(W1): p. W415-W421.\u003c/li\u003e\n\u003cli\u003eGranja, J.M., et al., \u003cem\u003eArchR is a scalable software package for integrative single-cell chromatin accessibility analysis.\u003c/em\u003e Nat Genet, 2021. \u003cstrong\u003e53\u003c/strong\u003e(3): p. 403-411.\u003c/li\u003e\n\u003cli\u003eKorsunsky, I., et al., \u003cem\u003eFast, sensitive and accurate integration of single-cell data with Harmony.\u003c/em\u003e Nat Methods, 2019. \u003cstrong\u003e16\u003c/strong\u003e(12): p. 1289-1296.\u003c/li\u003e\n\u003cli\u003eLiu, J.S. and M. Hebrok, \u003cem\u003eAll mixed up: defining roles for beta-cell subtypes in mature islets.\u003c/em\u003e Genes Dev, 2017. \u003cstrong\u003e31\u003c/strong\u003e(3): p. 228-240.\u003c/li\u003e\n\u003cli\u003eGoedecke, J.H. and T. Olsson, \u003cem\u003ePathogenesis of type 2 diabetes risk in black Africans: a South African perspective.\u003c/em\u003e J Intern Med, 2020. \u003cstrong\u003e288\u003c/strong\u003e(3): p. 284-294.\u003c/li\u003e\n\u003cli\u003eLadwa, M., et al., \u003cem\u003eEthnic differences in beta cell function occur independently of insulin sensitivity and pancreatic fat in black and white men.\u003c/em\u003e BMJ Open Diabetes Res Care, 2021. \u003cstrong\u003e9\u003c/strong\u003e(1).\u003c/li\u003e\n\u003cli\u003eLorenzo, C., et al., \u003cem\u003eInsulin resistance, beta-cell dysfunction, and conversion to type 2 diabetes in a multiethnic population: the Insulin Resistance Atherosclerosis Study.\u003c/em\u003e Diabetes Care, 2010. \u003cstrong\u003e33\u003c/strong\u003e(1): p. 67-72.\u003c/li\u003e\n\u003cli\u003eFeairheller, D.L., et al., \u003cem\u003eRacial differences in oxidative stress and inflammation: in vitro and in vivo.\u003c/em\u003e Clin Transl Sci, 2011. \u003cstrong\u003e4\u003c/strong\u003e(1): p. 32-7.\u003c/li\u003e\n\u003cli\u003eMorris, A.A., et al., \u003cem\u003eDifferences in systemic oxidative stress based on race and the metabolic syndrome: the Morehouse and Emory Team up to Eliminate Health Disparities (META-Health) study.\u003c/em\u003e Metab Syndr Relat Disord, 2012. \u003cstrong\u003e10\u003c/strong\u003e(4): p. 252-9.\u003c/li\u003e\n\u003cli\u003eMoon, J.S. and K.C. Won, \u003cem\u003ePancreatic alpha-Cell Dysfunction in Type 2 Diabetes: Old Kids on the Block.\u003c/em\u003e Diabetes Metab J, 2015. \u003cstrong\u003e39\u003c/strong\u003e(1): p. 1-9.\u003c/li\u003e\n\u003cli\u003eMizukami, H. and K. Kudoh, \u003cem\u003eDiversity of pathophysiology in type 2 diabetes shown by islet pathology.\u003c/em\u003e J Diabetes Investig, 2022. \u003cstrong\u003e13\u003c/strong\u003e(1): p. 6-13.\u003c/li\u003e\n\u003cli\u003ePati, D., N. Zhang, and S.E. Plon, \u003cem\u003eLinking sister chromatid cohesion and apoptosis: role of Rad21.\u003c/em\u003e Mol Cell Biol, 2002. \u003cstrong\u003e22\u003c/strong\u003e(23): p. 8267-77.\u003c/li\u003e\n\u003cli\u003eDahlby, T., et al., \u003cem\u003eEnhancer of Zeste Homolog 2 (EZH2) Mediates Glucolipotoxicity-Induced Apoptosis in beta-Cells.\u003c/em\u003e Int J Mol Sci, 2020. \u003cstrong\u003e21\u003c/strong\u003e(21).\u003c/li\u003e\n\u003cli\u003eNaina Marikar, S., et al., \u003cem\u003ePharmacological inhibition of human EZH2 can influence a regenerative \u0026beta;-like cell capacity with in vitro insulin release in pancreatic ductal cells.\u003c/em\u003e Clinical Epigenetics, 2023. \u003cstrong\u003e15\u003c/strong\u003e(1): p. 101.\u003c/li\u003e\n\u003cli\u003eCataldo, L.R., et al., \u003cem\u003eThe MafA-target gene PPP1R1A regulates GLP1R-mediated amplification of glucose-stimulated insulin secretion in beta-cells.\u003c/em\u003e Metabolism, 2021. \u003cstrong\u003e118\u003c/strong\u003e: p. 154734.\u003c/li\u003e\n\u003cli\u003eTaneera, J., et al., \u003cem\u003eUnraveling the significance of PPP1R1A gene in pancreatic beta-cell function: A study in INS-1 cells and human pancreatic islets.\u003c/em\u003e Life Sci, 2024. \u003cstrong\u003e345\u003c/strong\u003e: p. 122608.\u003c/li\u003e\n\u003cli\u003eAbedini, A. and A.M. Schmidt, \u003cem\u003eMechanisms of islet amyloidosis toxicity in type 2 diabetes.\u003c/em\u003e FEBS Lett, 2013. \u003cstrong\u003e587\u003c/strong\u003e(8): p. 1119-27.\u003c/li\u003e\n\u003cli\u003eVan Hulle, F., et al., \u003cem\u003eFormation of amyloid in encapsulated human pancreatic and human stem cell-generated beta cell implants.\u003c/em\u003e Am J Transplant, 2021. \u003cstrong\u003e21\u003c/strong\u003e(6): p. 2090-2099.\u003c/li\u003e\n\u003cli\u003eWirth, F., et al., \u003cem\u003eA human antibody against pathologic IAPP aggregates protects beta cells in type 2 diabetes models.\u003c/em\u003e Nat Commun, 2023. \u003cstrong\u003e14\u003c/strong\u003e(1): p. 6294.\u003c/li\u003e\n\u003cli\u003eOliveira, V.R., et al., \u003cem\u003ePre-treatment with IL-6 potentiates beta-cell death induced by pro-inflammatory cytokines.\u003c/em\u003e BMC Mol Cell Biol, 2023. \u003cstrong\u003e24\u003c/strong\u003e(1): p. 11.\u003c/li\u003e\n\u003cli\u003eLinnemann, A.K., et al., \u003cem\u003eInterleukin 6 protects pancreatic beta cells from apoptosis by stimulation of autophagy.\u003c/em\u003e FASEB J, 2017. \u003cstrong\u003e31\u003c/strong\u003e(9): p. 4140-4152.\u003c/li\u003e\n\u003cli\u003eAkbari, M. and V. Hassan-Zadeh, \u003cem\u003eIL-6 signalling pathways and the development of type 2 diabetes.\u003c/em\u003e Inflammopharmacology, 2018. \u003cstrong\u003e26\u003c/strong\u003e(3): p. 685-698.\u003c/li\u003e\n\u003cli\u003eLi, H., et al., \u003cem\u003eMacrophages, Chronic Inflammation, and Insulin Resistance.\u003c/em\u003e Cells, 2022. \u003cstrong\u003e11\u003c/strong\u003e(19).\u003c/li\u003e\n\u003cli\u003eZhou, Q., et al., \u003cem\u003eIn vivo reprogramming of adult pancreatic exocrine cells to beta-cells.\u003c/em\u003e Nature, 2008. \u003cstrong\u003e455\u003c/strong\u003e(7213): p. 627-32.\u003c/li\u003e\n\u003cli\u003eThorel, F., et al., \u003cem\u003eConversion of adult pancreatic alpha-cells to beta-cells after extreme beta-cell loss.\u003c/em\u003e Nature, 2010. \u003cstrong\u003e464\u003c/strong\u003e(7292): p. 1149-54.\u003c/li\u003e\n\u003cli\u003eFuruyama, K., et al., \u003cem\u003eDiabetes relief in mice by glucose-sensing insulin-secreting human alpha-cells.\u003c/em\u003e Nature, 2019. \u003cstrong\u003e567\u003c/strong\u003e(7746): p. 43-48.\u003c/li\u003e\n\u003cli\u003eWang, G., et al., \u003cem\u003eIntegrating genetics with single-cell multiomic measurements across disease states identifies mechanisms of beta cell dysfunction in type 2 diabetes.\u003c/em\u003e Nat Genet, 2023. \u003cstrong\u003e55\u003c/strong\u003e(6): p. 984-994.\u003c/li\u003e\n\u003cli\u003eNg, N.H.J., et al., \u003cem\u003eHNF4A and HNF1A exhibit tissue specific target gene regulation in pancreatic beta cells and hepatocytes.\u003c/em\u003e Nat Commun, 2024. \u003cstrong\u003e15\u003c/strong\u003e(1): p. 4288.\u003c/li\u003e\n\u003cli\u003eEllingsgaard, H., et al., \u003cem\u003eInterleukin-6 regulates pancreatic alpha-cell mass expansion.\u003c/em\u003e Proc Natl Acad Sci U S A, 2008. \u003cstrong\u003e105\u003c/strong\u003e(35): p. 13163-8.\u003c/li\u003e\n\u003cli\u003eAlmaca, J., et al., \u003cem\u003eYoung capillary vessels rejuvenate aged pancreatic islets.\u003c/em\u003e Proc Natl Acad Sci U S A, 2014. \u003cstrong\u003e111\u003c/strong\u003e(49): p. 17612-7.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":true,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Ancestry, ATAC-Seq, Genomics, Pancreatic islets, RNA-Seq, Single cell, Transcriptomics, Type 2 diabetes","lastPublishedDoi":"10.21203/rs.3.rs-5881592/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5881592/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eType 2 diabetes (T2D) disproportionally affects African Americans (AA) compared to European Americans (EA) in the United States. To investigate how the molecular pathophysiology T2D may vary by ancestry, we analyzed scRNA-Seq and snATAC-Seq data from 85,112 pancreatic islet cells from AA and EA individuals with T2D. AA donors had proportionally fewer beta and alpha cells but more exocrine cells (FDR\u0026thinsp;\u0026le;\u0026thinsp;1.5x10\u003csup\u003e\u0026minus;\u0026thinsp;3\u003c/sup\u003e) compared to EA. AA donors beta cells displayed upregulated exocrine genes (e.g., \u003cem\u003ePNLIP\u003c/em\u003e, \u003cem\u003ePRSS1\u003c/em\u003e) and mildly downregulated \u003cem\u003eINS\u003c/em\u003e compared to EA. However, AA donor alpha cells expressed more \u003cem\u003eINS\u003c/em\u003e, and higher HNF4A and HNF4G transcription factor binding accessibility. In contrast, EA exocrine cells expressed higher \u003cem\u003eINS\u003c/em\u003e, \u003cem\u003eGCG\u003c/em\u003e, \u003cem\u003eSST\u003c/em\u003e, and \u003cem\u003ePPY\u003c/em\u003e, with a corresponding increase of HNF4A and HNF4G binding site accessibility. These data demonstrate key differences in both islet cell composition and insulin signaling between AA and EA individuals with T2D.\u003c/p\u003e","manuscriptTitle":"Multi-omics islet profiling in type 2 diabetes reveals differential transcriptomics and chromatin accessibility by ancestry","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-16 08:26:05","doi":"10.21203/rs.3.rs-5881592/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e4c8c30e-95fc-4f34-b090-15bf4148adf8","owner":[],"postedDate":"July 16th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":50362513,"name":"Biological sciences/Genetics/Genomics/Epigenomics"},{"id":50362514,"name":"Biological sciences/Genetics/Genomics/Transcriptomics"},{"id":50362515,"name":"Health sciences/Diseases/Endocrine system and metabolic diseases/Diabetes/Type 2 diabetes mellitus"}],"tags":[],"updatedAt":"2025-09-26T14:39:07+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-16 08:26:05","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5881592","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5881592","identity":"rs-5881592","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}