1522-P: Integrating Transcriptome of Human Pancreatic Cells and Spatial Tissue Architecture during Type 1 Diabetes Progression at Single-Cell Resolution

• Published in: Diabetes (New York, N.Y.) Vol. 74; no. Supplement_1; p. 1
• Main Authors: BATE, THOMAS S.R., LUO, XIN, MOO, KEAGAN G., FENG, FAN, HOPKIRK, ALEXANDER L., PATEL, KEVAL, LEE, KAI, ESKAROS, ADEL, ORCHARD, PETER, ROBERTSON, CASSIE, CARTAILLER, JP, SAUNDERS, DIANE C., POWERS, ALVIN C., LIU, JIE, PARKER, STEPHEN C., BRISSOVA, MARCELA
• Format: Journal Article
• Language: English
• Published: New York American Diabetes Association 20.06.2025
• Subjects: Algorithms; Autoantibodies; Beta cells; Cell interactions; Diabetes; Diabetes mellitus (insulin dependent); Gene expression; Molecular modelling; Organ donors; Pancreas; Therapeutic targets; Transcriptomes
• ISSN: 0012-1797; 1939-327X
• DOI: 10.2337/db25-1522-P

Introduction and Objective: Type 1 diabetes (T1D) is a complex autoimmune disorder characterized by the loss of pancreatic islet beta cells. Analyses of pancreatic tissues from organ donors and MRI scans of patients reveal significant changes in both endocrine and exocrine compartments as T1D progresses. To explore the molecular mechanisms behind these changes, we combined single-cell RNA sequencing (scRNA-seq) with co-detection by indexing (CODEX) multiplex tissue imaging, using data from the Human Pancreas Analysis Program (HPAP). Methods: We analyzed transcriptome data from 37 donors: 18 controls without diabetes (ND), 10 autoantibody-positive donors (AAB), and 9 donors with T1D. For a subset of 15 donors, CODEX imaging data and transcriptome data were both available: 6 ND, 4 AAB, and 5 T1D donors. We employed new pipelines for scRNA-seq and CODEX cell annotations, and for pseudo-bulk differential expression analysis. Results: Our pseudo-bulk approach identified differentially expressed genes in the major cell types of the endocrine (alpha, N=16,577 and beta, N=11,446 cells) and exocrine compartments (ductal, N=10,430 and acinar, N=20,058 cells) in both AAB and T1D groups compared to the ND group. To spatially contextualize these changes, we aligned scRNA-seq data with CODEX imaging data (alpha, N=81,243; beta, N=80,292; ductal, N=910,371; acinar, N=393,893 cells) using the cell-cell similarity alignment algorithm CelLink. This algorithm enables the imputation of gene expression within annotated cells in CODEX images by utilizing common marker genes across the two data modalities and developing a pipeline to provide new insights into spatial cell-cell interactions in T1D progression. Conclusion: Our integrative high-dimensional data analysis shows that the pancreatic endocrine and exocrine compartments change before the clinical T1D onset. Further validation of these findings could help identify therapeutic targets within the pancreas to treat or prevent T1D.