Endocrine Abstracts

Introduction: Human induced pluripotent stem cells (iPSCs) differentiated into β-like cells represent an exciting model to investigate the pathogenesis of type 1 diabetes (T1D). The unlimited source of iPSC-β cells enables comprehensive omics studies of gene expression (transcriptomics) and protein synthesis (translatomics). Here, we examined how β cells respond to inflammatory stimuli known to play a role in T1D, aiming to better understand disease mechanisms and uncover potential therapeutic targets. We focused on interferon alpha (IFNα), that is linked to early, preclinical stages and triggers initial immune activation, interferon gamma (IFNγ) and interleukin-1 beta (IL1β), that are produced by autoreactive T cells and macrophages during the active β cell destruction phase, and tumor necrosis factor alpha (TNFα), that is involved in amplifying inflammation and cellular stress.

Methods: iPSCs were differentiated into β cells and purified by MACS using the iPSC-β cell surface marker CD49a. The cells were exposed for 24h to the individual cytokines IFNα, IFNγ, IL1β, TNFα or to the combination IFNγ+IL1β. RNA sequencing was used to study the transcriptome and ribosome profiling to examine which RNAs are actively translated into proteins (translatome). These analyses link gene expression changes to altered protein production.

Results: MACS purification generated 80±7% pure iPSC-β cells. Among all treatments, IFNγ+IL1β had the greatest impact, changing the expression of over 5,600 genes. Of the single cytokines, IFNs had the strongest impact (especially IFNy). Pathway analyses showed that translation was affected in all conditions. The comparison between transcriptome and translatome showed common upregulation of immune reaction pathways and downregulation of metabolic pathways and regulation of insulin secretion. In iPSC-β cells cytokines inhibited glucose-stimulated insulin secretion, from 4.5-fold stimulation in control to 3.5-fold for IL1β and TNFα and 1.5-fold for IFNα, IFNγ and IFNγ+IL1β. Notably, this occurred without a reduction in insulin content, indicating that the problem lies in secretion, not production. This aligns with T1D pathophysiology, where early β cell dysfunction induces impaired insulin release before full βcell destruction occurs. When comparing our data with FACS-purified β cells from T1D organ donors, high overlap was seen in particular for IFNα, IFNγ+IL1β and IFNγ confirming the clinical relevance of this in vitro model. Using the Connectivity Map (a resource linking gene expression signatures to drug effects), we identified novel therapeutic targets that may counteract the overlapping gene signatures. Among the most promising classes of drugs were bromodomain inhibitors (known to reduce cytokine-induced inflammation), leucine-rich repeat kinase (LRRK) inhibitors and a farnesyltransferase inhibitor.

Conclusion: Transcriptome, translatome and functional analyses suggest a preferential translation of genes which increase visibility of βcells to the immune system at the expense of βcell function. This functional imbalance mimics early T1D pathology. The high transcriptome overlap between T1D patients’ β cells and the iPSC-βcells supports the use of the latter model to identify new therapeutic strategies to delay or prevent T1D.

Keywords: Induced pluripotent stem cells, βcells, transcriptome, insulin secretion, type 1 diabetes