Islet-1 Is essential for pancreatic β-cell function

Abstract

Islet-1 (Isl-1) is essential for the survival and ensuing differentiation of pancreatic endocrine progenitors. Isl-1 remains expressed in all adult pancreatic endocrine lineages; however, its specific function in the postnatal pancreas is unclear. Here we determine whether Isl-1 plays a distinct role in the postnatal β-cell by performing physiological and morphometric analyses of a tamoxifen-inducible, β-cell-specific Isl-1 loss-of-function mouse: Isl-1(L/L); Pdx1-CreER(Tm). Ablating Isl-1 in postnatal β-cells reduced glucose tolerance without significantly reducing β-cell mass or increasing β-cell apoptosis. Rather, islets from Isl-1(L/L); Pdx1-CreER(Tm) mice showed impaired insulin secretion. To identify direct targets of Isl-1, we integrated high-throughput gene expression and Isl-1 chromatin occupancy using islets from Isl-1(L/L); Pdx1-CreER(Tm) mice and βTC3 insulinoma cells, respectively. Ablating Isl-1 significantly affected the β-cell transcriptome, including known targets Insulin and MafA as well as novel targets Pdx1 and Slc2a2. Using chromatin immunoprecipitation sequencing and luciferase reporter assays, we found that Isl-1 directly occupies functional regulatory elements of Pdx1 and Slc2a2. Thus Isl-1 is essential for postnatal β-cell function, directly regulates Pdx1 and Slc2a2, and has a mature β-cell cistrome distinct from that of pancreatic endocrine progenitors.

Figure 1

Partial ablation of Isl-1 in Isl-1^L/L; Pdx1-CreER^Tm_{(No Tm)} animals prior to Tm administration impairs response to acute glucose challenge. A and B: Isl-1 immunohistochemistry of 8-week-old female islets (20×). In B, the dotted lines highlight major, contiguous areas of Isl-1 paucity. C: Quantification of Isl-1⁺ nuclei per pancreatic endocrine area in 8-week-old females ± SEM (n = 5 per genotype; P = 0.00627). D: Isl-1 mRNA levels relative to Actin transcript using total RNA extracts from 8-week-old female isolated islets ± SEM (n = 5 per genotype). E and F: Insulin immunohistochemistry of 8-week-old female islets (20×). G: Insulin⁺ mass of 8-week-old females ± SEM (n = 5 per genotype). H: Plasma insulin levels of 8-week-old females after intraperitoneal glucose bolus following a 16-h fast ± SEM (Isl-1^L/L_{(No Tm)}, n = 4; Isl-1^L/L; Pdx1-CreER^Tm_{(No Tm)}, n = 6). In C, D, G, and H, black bars represent Isl-1^L/L_{(No Tm)} animals and white bars represent Isl-1^L/L; Pdx1-CreER^Tm_{(No Tm)} animals. Values for C, D, and G are presented as percentage of the Isl-1^L/L_{(No Tm)} animals. Analysis with two-way Student t test, **P < 0.01. Analysis with repeated measures, two-way ANOVA with Bonferroni posttest, ††P < 0.01.

Figure 2

Maximal ablation of Isl-1 in the adult animal impairs insulin secretion. A: The Tm-administration schedule for 8-week-old female animals. Three sequential intraperitoneal injections of Tm (50 µg/g of mouse bodyweight) or Veh (sunflower seed oil) were administered at 24-h intervals followed by a 2-day chase. B: Isl-1 mRNA expression relative to Actin using total RNA extracts from isolated islets as percentage of Isl-1^L/L_IP(Tm) animals ± SEM (Isl-1^L/L_IP(Tm), n = 4; Isl-1^L/L; Pdx1-CreER^Tm_IP(Tm), n = 4; P = 0.027). C and D: Coimmunofluorescence for insulin, somatostatin, Isl-1, and DAPI (20×). D: White arrowheads indicate Isl-1⁻, somatostatin⁺ cells, and white arrows indicate residual Isl-1⁺, insulin⁺ cells. E: Isl-1 ablation efficiency in glucagon⁺, somatostatin⁺, and insulin⁺ cell populations ± SEM (n = 4 per genotype). F: Random-fed plasma glucose ± SEM (Isl-1^L/L_IP(Tm), n = 5; Isl-1^L/L; Pdx1-CreER^Tm_IP(Tm), n = 8; P = 0.0006). G: Random-fed plasma insulin ± SEM (n = 8 per genotype). H: i) Intraperitoneal glucose tolerance test performed after 16-h fast. The solid line with crosshairs represents Isl-1^L/L_IP(Tm), the dashed line with crosshairs represents Isl-1^L/L; Pdx1-CreER^Tm_IP(Tm), and the dashed line with open diamonds represents Isl-1^L/L; Pdx1-CreER^Tm_IP(Veh). Values are presented as ± SEM (Isl-1^L/L_IP(Tm), n = 12; Isl-1^L/L; Pdx1-CreER^Tm_IP(Tm), n = 10; Isl-1^L/L; Pdx1-CreER^Tm_IP(Veh), n = 7). H: ii) Integrated area under intraperitoneal glucose tolerance test curves as percentage of Isl-1^L/L_IP(Tm) ± SEM. I: Plasma insulin levels in response to acute intraperitoneal glucose bolus after 16-h fast ± SEM (Isl-1^L/L_IP(Tm), n = 10; Isl-1^L/L; Pdx1-CreER^Tm_IP(Tm), n = 8). In B and E–I, black bars represent Isl-1^L/L_IP(Tm), black and white striped bars represent Isl-1^L/L; Pdx1-CreER^Tm_IP(Veh), and white bars represent Isl-1^L/L; Pdx1-CreER^Tm_IP(Tm). Analysis with two-way Student t test, *P < 0.05; ***P < 0.001. Analysis with one-way ANOVA and Tukey posttest, #P < 0.05; ###P < 0.001. Analysis with repeated measures, two-way ANOVA, and Bonferroni posttest, †P < 0.05; ††P < 0.01; †††P < 0.001; ††††P < 0.0001. Gcg⁺, glucagon⁺; Ins⁺, insulin⁺; Ins, insulin; IP, intraperitoneal; NS, no significance; SFSO, sunflower seed oil; Sst, somatostatin; Sst⁺, somatostatin⁺.

Figure 3

Ablation of Isl-1 reduces pancreatic insulin content but does not increase β-cell apoptosis. A and B: TUNEL, insulin, and DAPI shown by coimmunofluorescence (40×). TUNEL assays were performed on pancreatic sections from animals 14 days after first Tm injection. White arrows point to TUNEL⁺, insulin⁺ cells, and the white arrowhead points to a TUNEL⁺, insulin⁻ cell. C: Insulin⁺, TUNEL⁺ cells as a percentage of total Nkx6.1⁺ β-cells ± SEM (n = 4 per genotype). D: Quantification of insulin⁺ mass ± SEM (Isl-1^L/L_IP(Tm), n = 6; Isl-1^L/L; Pdx1-CreER^Tm_IP(Tm), n = 8). E: Quantification of pancreatic insulin content ± SEM (Isl-1^L/L_IP(Tm), n = 6; Isl-1^L/L; Pdx1-CreER^Tm_IP(Tm), n = 8; P = 0.0482). F: Quantification of Insulin mRNA expression relative to Hprt using total RNA extracts from isolated islets ± SEM (Isl-1^L/L_IP(Tm), n = 6; Isl-1^L/L; Pdx1-CreER^Tm_IP(Tm), n = 8; P = 5.57 × 10⁻³). G: Static incubation of isolated islets in glucose at 2.5 and 16.0 mmol/L ± SEM (Isl-1^L/L_IP(Tm), n = 6; Isl-1^L/L; Pdx1-CreER^Tm_IP(Tm), n = 8). Insulin secretion was normalized to islet number. H: Insulin content of islets used in static incubations ± SEM (Isl-1^L/L_IP(Tm), n = 6; Isl-1^L/L; Pdx1-CreER^Tm_IP(Tm), n = 8; P = 0.014). I: Relative islet insulin secretion where islet insulin secretion was normalized to islet insulin content ± SEM. Values for D and F are presented as percentage of Isl-1^L/L_IP(Tm) animals. In C–I, black bars represent Isl-1^L/L_IP(Tm) and white bars represent Isl-1^L/L; Pdx1-CreER^Tm_IP(Tm). Analysis with two-way Student t test, *P < 0.05; **P < 0.01. Analysis with two-way ANOVA and Bonferroni posttest, †P < 0.05.

Figure 4

Ablating Isl-1 significantly alters the pancreatic islet transcriptome. A: Exploded pie chart representing genes identified by microarray with significant expression differences in Isl-1^L/L; Pdx1-CreER^Tm_IP(Tm) animals. The red and green partitions represent genes that were, respectively, up- and downregulated. Partitions outlined with the dashed lines represent genes in 100 kb proximity to areas of Isl-1 enrichment identified by the Isl-1 βTC3 insulinoma ChIP-Seq. Values are presented as a percentage of all 714 significantly misregulated genes identified in the microarray. B: Selected biological processes that were enriched in gene ontology analysis of the microarray data set. Values are presented as −log₁₀ (P value). C: MafA and Slc2a2 mRNA expression relative to Hprt ± SEM (Isl-1^L/L_IP(Tm), n = 6; Isl-1^L/L; Pdx1-CreER^Tm_IP(Tm), n = 8). Values are normalized by gene to the relative expression in Isl-1^L/L_IP(Tm) animals. Black bars represent Isl-1^L/L_IP(Tm), and white bars = Isl-1^L/L; Pdx1-CreER^Tm_IP(Tm). Analysis with two-way Student t test, *P < 0.05. D and E: Visualization of MafA using immunohistochemistry (20×). F and G: Visualization of Glut2 and glucagon using coimmunofluorescence (20×). H: De novo network generated from gene network analysis of Isl-1^L/L; Pdx1-CreER^Tm microarray data set. Red and green highlights indicate statistically significant up- and downregulation, respectively; intensity of highlight corresponds to fold change. White highlight indicates factors that were not identified by the microarray. Solid arrows and dashed arrows represent confirmed and suspected regulatory relationships, respectively. Gcg, glucagon; GO, gene ontology.

Figure 5

Isl-1 directly regulates Pdx1. A: i) Western blot for Pdx1 and α-tubulin using protein lysates of isolated islets. A: ii) Quantification of Pdx1 normalized to α-tubulin ± SEM (n = 3 per genotype; P = 0.01222). B: Relative mRNA expression of Pdx1 to Hprt ± SEM (Isl-1^L/L_IP(Tm), n = 6; Isl-1^L/L; Pdx1-CreER^Tm_IP(Tm), n = 8). Values are normalized to relative expression in Isl-1^L/L _IP(Tm) animals. In A ii) and B, black bars represent Isl-1^L/L_IP(Tm) and white bars represent Isl-1^L/L; Pdx1-CreER^Tm_IP(Tm). C: A scaled schematic of the Pdx1 genomic locus. The known Pdx1 regulatory domains areas I, II, and III are represented by white boxes and area IV by a black box. The Pdx1 transcription start site is indicated by the arrow. UCSC mouse genome browser of the Pdx1 genomic locus annotated with βTC3 chromatin Isl-1 ChIP-Seq and vertebrate Multiz alignment in black; the Pdx1 transcription start site is underlined. Arrows highlight Isl-1 enrichment at the four described Pdx1 regulatory elements. D: Isl-1 ChIP using chromatin extracted from βTC3 cells ± SEM (n = 3). E: Isl-1 ChIP using chromatin extracted from isolated islets (n = 1). Values in both D and E are presented as fold enrichment over normal mouse IgG after first normalizing to the inactive Pepck locus. F: EMSA using recombinant Myc-tagged Isl-1 protein. The Isl-1 binding site at MafA-Reg3 was used as the radiolabeled probe. Control competition assays were performed with wild-type and mutated MafA-Reg3 oligo probes. Competition assays were also performed using oligonucleotides representing putative HBEs in Pdx1-areas I, II, and IV. Competitors were added at 100× free probe concentration. Supershifts using both antibodies against Isl-1 and Myc were observed. A protein:DNA complex was not observed in the absence of recombinant Isl-1-Myc protein. Bold, italicized enumeration corresponds to selected HBEs for mutational analysis. G: Luciferase assay for putative Isl-1 HBEs in Pdx1 enhancers using βTC3 cells. Values are normalized to empty pGL4.27 vector ± SEM (n = 3). The black and white striped bar represents empty vector. Vector inserts correspond to the selected HBEs in F. The black bars represent wild-type sequences. The white bars represent mutational ablation of the putative HBE. H: Luciferase assay for Pdx1 area II in HeLa cells. Values are normalized to empty pGL4.27 vector ± SEM (n = 3). White and black bars represent cells transfected using pGL4.27 with and without the Pdx1 area II insert, respectively. Analysis with two-way Student t test, *P < 0.05. Analysis with one-way ANOVA and Tukey posttest, ###P < 0.001. Analysis with two-way ANOVA and Bonferroni posttest, †P < 0.05, †††P < 0.001. IVT Neg, in vitro translation negative; TSS, transcription start site.

Figure 6

Isl-1 directly regulates Slc2a2. A: A schematic of the genomic Slc2a2 locus. White boxes represent putative enhancer elements, Re1 and Re2. Re1 and Re2 are ∼9 kb upstream and 39 kb downstream of the Slc2a2 transcription start site, respectively. B: Luciferase assay for putative Slc2a2 enhancer elements using βTC3 cells. Values are normalized to empty pGL4.27 vector ± SEM (n = 3). Black and white bars represent βTC3 cells with and without overexpression of Isl-1-Myc, respectively. C: Slc2a2-Re2 sequence alignment for mouse, rat, and human. Base pairs that differ from the mouse sequence are underlined. Putative HBEs are bold and italic in the mouse sequence and bold in both rat and human sequences. The five putative HBEs are enumerated 1–5. D: Luciferase assay for putative Isl-1 binding sites in Slc2a2-Re2 using βTC3 cells. Values are normalized to pGL4.27 vector containing wild-type Slc2a2-Re2 ± SEM (n = 3). The black bar represents wild-type Slc2a2-Re2. The white bars represent vectors containing Slc2a2-Re2 with mutational ablation of the respective putative HBE. Analysis with one-way ANOVA and Tukey posttest, ###P < 0.001. Analysis with two-way ANOVA and Bonferroni posttest, ††††P < 0.0001. TSS, transcription start site; WT, wild type.