Single-nucleus RNA sequencing of human pancreatic islets identifies novel gene sets and distinguishes β-cell subpopulations with dynamic transcriptome profiles

Abstract

Abstract Background Single-cell RNA sequencing (scRNA-seq) provides valuable insights into human islet cell types and their corresponding stable gene expression profiles. However, this approach requires cell dissociation that complicates its utility in vivo. On the other hand, single-nucleus RNA sequencing (snRNA-seq) has compatibility with frozen samples, elimination of dissociation-induced transcriptional stress responses, and affords enhanced information from intronic sequences that can be leveraged to identify pre-mRNA transcripts. Methods We obtained nuclear preparations from fresh human islet cells and generated snRNA-seq datasets. We compared these datasets to scRNA-seq output obtained from human islet cells from the same donor. We employed snRNA-seq to obtain the transcriptomic profile of human islets engrafted in immunodeficient mice. In both analyses, we included the intronic reads in the snRNA-seq data with the GRCh38-2020-A library. Results First, snRNA-seq analysis shows that the top four differentially and selectively expressed genes in human islet endocrine cells in vitro and in vivo are not the canonical genes but a new set of non-canonical gene markers including ZNF385D, TRPM3, LRFN2, PLUT (β-cells); PTPRT, FAP, PDK4, LOXL4 (α-cells); LRFN5, ADARB2, ERBB4, KCNT2 (δ-cells); and CACNA2D3, THSD7A, CNTNAP5, RBFOX3 (γ-cells). Second, by integrating information from scRNA-seq and snRNA-seq of human islet cells, we distinguish three β-cell sub-clusters: an INS pre-mRNA cluster (β3), an intermediate INS mRNA cluster (β2), and an INS mRNA-rich cluster (β1). These display distinct gene expression patterns representing different biological dynamic states both in vitro and in vivo. Interestingly, the INS mRNA-rich cluster (β1) becomes the predominant sub-cluster in vivo. Conclusions In summary, snRNA-seq and pre-mRNA analysis of human islet cells can accurately identify human islet cell populations, subpopulations, and their dynamic transcriptome profile in vivo.