Single-Cell RNA-seq Analysis Reveals a Positive Correlation between Ferroptosis and Beta-Cell Dedifferentiation in Type 2 Diabetes

Abstract

Insulin deficiency in patients with type 2 diabetes mellitus (T2D) is associated with beta-cell dysfunction, a condition increasingly recognized to involve processes such as dedifferentiation and apoptosis. Moreover, emerging research points to a potential role for ferroptosis in the pathogenesis of T2D. In this study, we aimed to investigate the potential involvement of ferroptosis in the dedifferentiation of beta cells in T2D. We performed single-cell RNA sequencing analysis of six public datasets. Differential expression and gene set enrichment analyses were carried out to investigate the role of ferroptosis. Gene set variation and pseudo-time trajectory analyses were subsequently used to verify ferroptosis-related beta clusters. After cells were categorized according to their ferroptosis and dedifferentiation scores, we constructed transcriptional and competitive endogenous RNA networks, and validated the hub genes via machine learning and immunohistochemistry. We found that ferroptosis was enriched in T2D beta cells and that there was a positive correlation between ferroptosis and the process of dedifferentiation. Upon further analysis, we identified two beta clusters that presented pronounced features associated with ferroptosis and dedifferentiation. Several key transcription factors and 2 long noncoding RNAs (MALAT1 and MEG3) were identified. Finally, we confirmed that ferroptosis occurred in the pancreas of high-fat diet-fed mice and identified 4 proteins (NFE2L2, CHMP5, PTEN, and STAT3) that may participate in the effect of ferroptosis on dedifferentiation. This study helps to elucidate the interplay between ferroptosis and beta-cell health and opens new avenues for developing therapeutic strategies to treat diabetes.

Keywords

• ferroptosis
• dedifferentiation
• type 2 diabetes mellitus
• single-cell sequencing
• machine learning
• bioinformatics