IER3IP1 Mutations Cause Neonatal Diabetes Due to Impaired Proinsulin Trafficking

Abstract

IER3IP1 mutations are linked to the development of microcephaly, epilepsy, and early-onset diabetes syndrome 1. However, the underlying molecular mechanisms of cell dysfunction are unknown. Using targeted genome editing, we generated specific IER3IP1 mutations in human embryonic stem cell lines that were differentiated into pancreatic islet lineages. Loss of IER3IP1 resulted in a threefold reduction in endoplasmic reticulum-to-Golgi trafficking of proinsulin in stem cell-derived β-cells, leading to β-cell dysfunction both in vitro and in vivo. Loss of IER3IP1 also triggered increased markers of endoplasmic reticulum stress, indicating the pivotal role of the endoplasmic reticulum-to-Golgi trafficking pathway for β-cell homeostasis and function.

Figure 1

Depletion of IER3IP1 leads to a decline in β-cell number and a reduction in their insulin content. A: IER3IP1 expression at the SC, definitive endoderm (DE), pancreatic progenitors (PP), and functional SC-islets stages of the pancreatic-directed differentiation in vitro from bulk RNA sequencing (n = 4 independent experiments). Values are presented as fragments per kilobase million (FPKM). B: Schematic presentation of the CRISPR-Cas9–mediated genome editing strategy of the hESCs to create the IER3IP1^V21G knock-in model harboring the patient mutation and the IER3IP1^−/− KO model. C: Schematic presentation of the seven-stage differentiation protocol. D: Immunohistochemistry analysis of the SC-islets for INS, GCG, and DNA (Hoechst nuclear stain) (representative of n = 4 independent experiments). Scale bar = 50 µm. E: Quantification of SC-islets cell composition using flow cytometry for INS and GCG (n = 6–7 independent experiments). F: Dynamic insulin secretion responses to perifusion with 3 mmol/L and 17 mmol/L glucose, 50 ng/mL exendin 4 (EX4) and 30 mmol/L KCL. Values are presented as absolute insulin secretion values (mU/L) (n = 3 independent experiments). G: hINS content of the SC-islets presented as ng of hINS normalized to average DNA content (n = 3–6 independent experiments). H: Glucagon content of the SC-islets presented as fold change relative to WT cells (n = 3–6 independent experiments). I: Ratio of human (h)PROINS content to hINS content of the SC-islets (n = 3–6 independent experiments). Data are represented as mean ± SEM. Welch one-way ANOVA test, *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 2

Disruption of IER3IP1 result in proinsulin accumulation in the ER. A: Schematic presentation of the ER-to-Golgi trafficking pathway implemented in insulin biosynthesis. Proinsulin is presented as coiled lines. B: Immunohistochemistry analysis of the SC-islets for INS, PROINS, and DNA (Hoechst nuclear stain) (representative of n = 3–5 independent experiments). Scale bar = 10 µm. C: Quantification of the percentage of INS⁺ cells showing increased accumulation of PROINS (n = 3–5 independent experiments). D: Representative immunofluorescent images for PROINS, the ER marker PDI, the colocalized volume of PROINS and PDI, and DNA (Hoechst nuclear stain) using confocal microscopy (representative of n = 5 independent experiments) Scale bar = 10 µm. E: Quantification of the percentage of PROINS volume colocalizing with PDI volume using Imaris imaging software (n = 5 independent experiments). F: Representative immunofluorescent images for PROINS, the Golgi marker GM130, the colocalized volume of PROINS and GM130), and DNA (Hoechst nuclear stain) using confocal microscopy (representative of n = 5 independent experiments). Scale bar = 10 µm. G: Quantification of the percentage of PROINS volume colocalizing with GM130 volume using Imaris (n = 5 independent experiments). Data are represented as mean ± SEM. Welch one-way ANOVA test, *P < 0.05; **P < 0.01.

Figure 3

IER3IP1 mutation impairs ER-to-Golgi trafficking. A: Schematic presentation of ProCpepRUSH trafficking. B: Representative immunofluorescent images for ProCpepRush, the Golgi marker GM130, and DNA (Hoechst nuclear stain) using confocal microscopy at 0, 15, 30, and 60 min postbiotin addition (representative of n = 3 independent experiments). Scale bar = 5 µm. C: Quantification of the percentage of ProCpepRUSH volume colocalizing with GM130 volume using Imaris imaging software (n = 3 independent experiments). Data are represented as mean ± SEM. Welch one-way ANOVA test, **P < 0.01.

Figure 4

IER3IP1 mutations elevate endoplasmic reticulum stress in human β-cells. A: Relative gene expression levels of IER3IP1, pancreatic hormonal markers, and ER stress markers of the SC-islets analyzed by RT-qPCR (n = 6–8 independent experiments). B: Immunohistochemistry analysis of the SC-islets for INS, BiP, and DNA (Hoechst nuclear stain) (representative of n = 3–5 independent experiments) Scale bar = 25 µm. C: Quantification of the percentage of INS⁺ cells showing increased level of BiP (n = 3–5 independent experiments). D: Electron micrographs of SC–β-cells. Scale bars = 1 µm. The yellow arrows denote ER structures, representative of several cells from n = 3 independent experiments. E: Quantification of the ER structures width measured by Fiji (n = 30 cells per genotype from n = 3 independent experiments). Data are represented as mean ± SEM. Welch one-way ANOVA test, *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 5

IER3IP1 mutant β-cells show impaired functionality in vivo. A: Schematic presentation of the in vivo workflow. IHC, immunohistochemistry; ipGTT, intraperitoneal glucose tolerance test. B: Follow-up of SC-islets in vivo functionality by measuring the levels of circulating human C-peptide of implanted mice (n = 3–6 mice per genotype, measurements from ad libitum mice). C: Follow-up of SC-islets in vivo functionality by measuring the blood glucose levels of implanted mice (n = 3–6 mice per genotype, measurements from ad libitum mice). D: Levels of circulating human C-peptide during an ipGTT of fasted implanted mice (n = 4–6 mice per genotype). E: Quantification of the ability of the implanted cells to secrete human C-peptide by measuring the area under the curve (AUC) of D. F: Blood glucose measurements during an ipGTT of fasted implanted mice (n = 4–6 mice per genotype). G: Quantification of the ability of the implanted cells to control the blood glucose levels of the implanted mice during an ipGTT by measuring the AUC of F. Data are represented as mean ± SEM. Welch one-way ANOVA test; ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 6

Elevated ER stress and accumulation of proinsulin in IER3IP1 mutant β-cells. A: Immunohistochemistry analysis of 3-month-old grafts for INS, GCG, and DNA (Hoechst nuclear stain) (representative of n = 3 grafts per genotype). Scale bar = 50 µm. B: Quantification of the percentage of INS⁺ and GCG⁺ cells from A (n = 3 mice per genotype). C: Immunohistochemistry analysis of 3-month-old grafts for INS, proinsulin (PROINS), and the nuclear stain Hoechst (blue) (representative of 3 grafts per genotype, scale bar = 50 µm). D: Quantification of the relative number of INS⁺ showing high accumulation of PROINS from C (n = 3 mice per genotype). E: Immunohistochemistry analysis of 3-month-old grafts for insulin (INS), BiP, and the nuclear stain Hoechst (blue) (representative of n = 3 grafts per genotype). Scale bar = 50 µm. F: Quantification of the relative number of INS⁺ showing high level of BiP from E (n = 3 mice per genotype). Data are represented as mean ± SEM. Welch’s one-way ANOVA test, *P < 0.05; ** P < 0.01.