Human expandable pancreatic progenitor-derived β cells ameliorate diabetes

Abstract

An unlimited source of human pancreatic β cells is in high demand. Even with recent advances in pancreatic differentiation from human pluripotent stem cells, major hurdles remain in large-scale and cost-effective production of functional β cells. Here, through chemical screening, we demonstrate that the bromodomain and extraterminal domain (BET) inhibitor I-BET151 can robustly promote the expansion of PDX1⁺NKX6.1⁺ pancreatic progenitors (PPs). These expandable PPs (ePPs) maintain pancreatic progenitor cell status in the long term and can efficiently differentiate into functional pancreatic β (ePP-β) cells. Notably, transplantation of ePP-β cells rapidly ameliorated diabetes in mice, suggesting strong potential for cell replacement therapy. Mechanistically, I-BET151 activates Notch signaling and promotes the expression of key PP-associated genes, underscoring the importance of epigenetic and transcriptional modulations for lineage-specific progenitor self-renewal. In summary, our studies achieve the long-term goal of robust expansion of PPs and represent a substantial step toward unlimited supplies of functional β cells for biomedical research and regenerative medicine.

Fig. 1.. A chemical screen identified that I-BET151 can promote hPSC-derived pancreatic progenitor expansion.

(A) Schematic representation of the pancreatic differentiation process from hPSCs to PPs and chemical screen for PP expansion. (B) Flowchart of the chemical screening process. After replication and confirmation, I-BET151 was found as the hit compound. (C) Chemical structure of I-BET151. (D) Representative immunofluorescent staining of PPs treated with or without I-BET151 for PDX1, NKX6.1, and nuclei. Scale bar, 100 μm. (E and F) Representative flow cytometry dot plots (E) and population percentages (F) of cells stained for PDX1 and NKX6.1. N = 6. (G) RT-qPCR analysis of NKX6.1, PDX1, HNF6, and SOX9 gene expression in PPs treated with or without I-BET151. N = 3. (H) Volcano plot of differentially expressed genes (|log₂FC| > 1, FDR < 0.05) in samples treated with I-BET151 versus dimethyl sulfoxide (DMSO). Red, up-regulated genes (620 genes); blue, down-regulated genes (2209 genes). X and Y axes represent log₂(counts + 1). (I) GO terms of up-regulated and down-regulated genes in samples treated with I-BET151 versus DMSO. (J) Heatmap of differentially expressed genes in samples treated with I-BET151 versus DMSO. (K) I-BET151 strengthened the gene regulatory network of PPs (lines, the square of two genes’ expression’s Pearson correlation coefficient > 0.7; red lines, positive coexpression; blue lines, negative coexpression). Circle, PP marker genes; square, genes in Notch signaling pathway; triangle, cell cycle–related genes; octagon, EP markers. All data are expressed as means ± SD. Statistical significance was calculated using two-tailed Student’s t test, *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Fig. 2.. Long-term expansion and detailed characterization of ePPs.

(A) Schematic of the long-term expansion of ePPs. (B) ePPs could be passaged more than 35 passages (N = 3). (C) Volcano plot comparing global gene expression patterns between ePP-P9 and ePP-P21. The red dots and blue dots respectively represent up-regulated (log₂FC > 1) and down-regulated (log₂FC < −1) differentially expressed genes in ePP-P21 versus ePP-P9 with statistical significance (FDR < 0.05). X and Y axes represent log₂(TPM + 0.01). (D) Karyotype of ePP-P24 indicates the chromatin stability of ePPs after long-term expansion. (E) Representative immunofluorescent staining of ePPs for PDX1, FOXA2, SOX9, NKX6.1, and nuclei (Hoechst). Scale bar, 100 μm. (F and G) Representative flow cytometry dot plots (F) and population percentages (G) of PDX1⁺NKX6.1⁺ and SOX9⁺FOXA2⁺ cells demonstrated that ePPs were almost homogenous (N = 4). Data are expressed as means ± SD. (H) Representative immunofluorescent staining of ePPs for PDX1, Ki67, and nuclei. Scale bar, 100 μm. (I) Principal component analysis (PCA) of RNA-seq data. PP1, PDX1⁺ pancreatic progenitors; PP2, PDX1⁺NKX6.1⁺ pancreatic progenitors; ePP-P9, ePPs at passage 9; ePP-P21, ePPs at passage 21. Circle, data cited from a recently published paper from Melton laboratory (10); triangle, data of the current study. (J) Transcriptome analysis revealed different gene expression among hPSC, DE, PP1, PP2, ePP-P9, ePP-P21, and EP. M represents published data from Melton laboratory (10). Representative GO terms and genes are also shown. (K) Heatmap of PP- and EP-related marker genes in the samples shown in (I). Both early- and late-passage ePPs expressed PP-related marker genes including PDX1, NKX6.1, SOX9, HNF6, and MNX1 but not EP-related marker genes such as NEUROG3, NKX2.2, and NEUROD1. JNK, c-Jun N-terminal kinase.

Fig. 3.. Efficient generation of functional ePP-β cells from ePPs.

(A) Schematic indicating the differentiation of ePPs to ePP-β cells. (B) Representative immunofluorescence staining of ePP-β cells. Scale bars, 50 μm (low magnification) and 10 μm (high magnification). (C) FACS results (N = 3). (D) PCA of RNA-seq data. PP2, PDX1⁺NKX6.1⁺ pancreatic progenitors; ePP-P9, ePPs at passage 9; ePP-P21, ePPs at passage 21; ePP-β, ePP-derived β cells; SC-β, hPSC-derived β cells; β cell, human pancreatic islet β cells. Circle, data cited from a recently published paper (10); triangle, data of the current study. (E) The expression of β cell genes. M represents published data (10). (F) Volcano plot comparing global gene expression patterns. Significant up-regulated genes (log₂FC > 1, FDR < 0.05) are shown in red (2173 genes), and significant down-regulated genes (log₂FC < −1, FDR < 0.05) are shown in blue (1400 genes). X and Y axes represent log₂(TPM + 0.01). (G) KEGG pathways enrichment. (H) Representative transmission electron micrographs of mitochondria in ePP-β cells. Scale bar, 500 nm. (I) Mitochondrial DNA (mtDNA) content. mtDNA content was measured by the ratio of mtDNA/nuclear DNA using qPCR for the mitochondrial 16S rRNA gene and the nuclear β2 microglobulin gene (N = 3). (J) Representative immunofluorescence staining of ePP-β cells. Scale bars, 50 μm (low magnification) and 10 μm (high magnification). (K) Representative transmission electron micrographs of insulin granules in ePP-β cells. Scale bar, 1 μm. (L) Total insulin and C-peptide content in ePP-β cells and human islets (N = 4). (M) GSIS analysis of ePP-β cells and human islets (N = 4). All data are expressed as means ± SD. Statistical significance was calculated using two-tailed Student’s t test: ns, P > 0.05; *P < 0.05; **P < 0.01; ***P < 0.001. cAMP, cyclic adenosine 3′,5′-monophosphate; ECM, extracellular matrix.

Fig. 4.. ePP-β cells can rapidly ameliorate diabetes.

(A) Schematic showing transplantation of ePP-β cells into immunodeficient diabetic mice. (B and C) In vivo glucose challenge test 3 days (B) and 12 weeks (C) after transplantation. The levels of human C-peptide in mouse serum were measured after overnight fasting (yellow) and 1 hour after an intraperitoneal glucose injection (orange). The x axis represents the number of individual animals (N = 5). (D) Intraperitoneal glucose tolerance test (IPGTT) was performed 30 days after transplantation with ePP-β cells (N = 3), nontransplanted diabetic SCID beige mice (N = 4), and nontransplanted wild-type (WT) SCID beige mice (N = 3). Area under the curve (AUC; left) was determined for each group (right). (E) Blood glucose level of STZ-treated control mice (no transplant, blue) and mice transplanted with ePP-β cells (orange). Tx, transplantation; Nx, nephrectomy. Control group (N = 5) and experimental group (N = 4). (F) Representative immunofluorescent images of ePP-β grafts stained with C-peptide, NKX6.1, PDX1, SST, GCG, and nuclei. Scale bar, 10 μm. All data are expressed as means ± SD. Statistical significance was calculated using two-tailed Student’s t test, *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Fig. 5.. Molecular mechanisms of I-BET151.

(A) Differential BRD4-binding peaks in PP cells cultured in EF6 or EF6I medium. (B) Transcription factor (TF) binding motifs enriched at up-regulated BRD4-binding peaks induced by I-BET151. (C) GO categories of BRD4-binding peaks specifically enhanced by I-BET151. (D) Differential H3K27ac peaks in PP cells cultured in EF6 or EF6I medium. (E) TF binding motifs enriched at up-regulated H3K27ac peaks induced by I-BET151. (F) GO categories of H3K27ac peaks specifically enhanced by I-BET151. (G) Differential ATAC peaks in PP cells cultured in EF6 or EF6I medium. (H) TF binding motifs enriched at up-regulated ATAC peaks induced by I-BET151. (I) GO categories of enriched, more accessible genes induced by I-BET151. (J) Browser tracks showing BRD4-binding peaks, H3K27ac peaks, ATAC peaks, and RNA abundance at PDX1, NKX6.1, SOX9, HES1, and HEY1 loci 7 days after I-BET151 treatment. EF6 group, peaks in brighter colors; EF6I group, peaks in darker colors.