Haploinsufficiency of the Insulin Receptor in the Presence of a Splice-Site Mutation in Ppp2r2a Results in a Novel Digenic Mouse Model of Type 2 Diabetes

Abstract

Insulin resistance in mice typically does not manifest as diabetes due to multiple compensatory mechanisms. Here, we present a novel digenic model of type 2 diabetes in mice heterozygous for a null allele of the insulin receptor and an N-ethyl-N-nitrosourea-induced alternative splice mutation in the regulatory protein phosphatase 2A (PP2A) subunit PPP2R2A. Inheritance of either allele independently results in insulin resistance but not overt diabetes. Doubly heterozygous mice exhibit progressive hyperglycemia, hyperinsulinemia, and impaired glucose tolerance from 12 weeks of age without significant increase in body weight. Alternative splicing of Ppp2r2a decreased PPP2R2A protein levels. This reduction in PPP2R2A containing PP2A phosphatase holoenzyme was associated with decreased serine/threonine protein kinase AKT protein levels. Ultimately, reduced insulin-stimulated phosphorylated AKT levels were observed, a result that was confirmed in Hepa1-6, C2C12, and differentiated 3T3-L1 cells knocked down using Ppp2r2a small interfering RNAs. Altered AKT signaling and expression of gluconeogenic genes in the fed state contributed to an insulin resistance and hyperglycemia phenotype. This model demonstrates how genetic changes with individually small phenotypic effects interact to cause diabetes and how differences in expression of hypomorphic alleles of PPP2R2A and potentially other regulatory proteins have deleterious effects and may therefore be relevant in determining diabetes risk.

Figure 1. Mapping of IGT10 and identification of a causative novel splice mutation.

A) Mapping of minimal candidate region on chromosome 6. B) Confirmation of intronic ENU induced mutation in Itm2b by Sanger Sequencing and diagram of intron/exon structure. C) Confirmation of intronic ENU induced mutation in Ppp2r2a by Sanger Sequencing and diagram of intron/exon structure. D) Reduction in relative expression levels of Ppp2r2a. E) Relative expression levels of Itm2b. QrtPCR data is representative of Mean ± SD, N=8 biological replicates. Statistical analysis 2 way ANOVA * p<0.05, ** p<0.01, *** p<0.001.

Figure 2. Phenotypic characterization of male IGT10 mice and controls.

A) Fasted plasma glucose levels. B) Presence of glucose in urine. C) Fasted plasma insulin levels. D) IPGTT at 12 weeks of age. E) ITT at 16 weeks of age normalized to starting glucose levels. F) Representative H&E stained pancreas sections. Black circles represent IR heterozygous null mice; Red squares double heterozygous mice, Blue triangles wildtype mice (wt/wt); Orange triangles heterozygous PPP2R2A mice. N=10-15 data expressed as mean ± SD, statistical analysis 2 way ANOVA with a Bonferroni post-test correction for multiple measures * p<0.05, ** p<0.01, *** p<0.001 vs wildtype.

Figure 3. Mis-splicing of Ppp2r2a causes a reduction in PPP2R2A and AKT protein levels and a reduction in insulin stimulated AKT phosphorylation.

A) Relative protein concentrations of PPP2R2A and AKT in unchallenged liver lysates. B) Relative protein concentrations of PPP2R2A and AKT in unchallenged skeletal muscle lysates. C) Relative protein concentrations of PPP2R2A and AKT in unchallenged WAT lysates. D-I) Relative protein concentrations of pAKT(Thr-308; D-F), pAKT(Ser-473; G-I), in liver (D,G), skeletal muscle (E,H) and WAT (F,I) isolated from mice exposed to either saline or insulin. Phosphorylated protein quantified by MSD technology and normalized to a parallel sample assayed for Tubulin on the PEGGY system. Light shaded bars saline treatment, dark shaded bars insulin treatment. Data is represented as mean ± SD, N=4 per treatment/per tissue type. Statistical analysis 2 way ANOVA * p<0.05, ** p<0.01, *** p<0.001.

Figure 4. Mis-splicing of Ppp2r2a causes a reduction in insulin stimulated AKT signaling.

A) Relative protein concentrations of GSK-3ßand p70S6K in unchallenged liver lysates. B) Relative protein concentrations of GSK-3ßand p70S6K in unchallenged skeletal muscle lysates. C) Relative protein concentrations of GSK- 3ßand p70S6K in unchallenged WAT lysates. D-I) Relative protein concentrations of pGSK-3ßD-F), pp70S6K (G-I), in liver (D,G), skeletal muscle (E,H) and WAT (F,I) isolated from mice exposed to either saline or insulin. Phosphorylated and total protein quantified on MSD, data represent percentage phosphorylation. Light shaded bars saline treatment, dark shaded bars insulin treatment. Data is represented as mean± SD, N=4 per treatment/per tissue type. Statistical analysis 2 way ANOVA * p<0.05, ** p<0.01, *** p<0.001.

Figure 5. Impaired transcriptional regulation of hepatic glucose production in Hepa 1-6 cells treated with siRNAs targeting Ppp2r2a.

Expression in serum free vs serum fed Hepa 1-6 cells. A) Foxo1, B) Pck1, C) G6pc, D) Irs-2, E) Srebflc, F) Fasn, G) Insig1, H) Ppargc1a. N=6 biological replicates. Data is represented as mean ± SD. Black bars siRNA negative control treatment, hatched bars Ppp2r2a siRNA treatment. I) Glucose production assay on control or siRNA treated Hepa 1-6 cells with insulin (white bars) or without insulin treatment (black bars) N=9 technical replicates (2 independent experiments). Statistical analysis 2 way ANOVA * p<0.05, ** p<0.01, *** p<0.001.

Figure 6. Impaired transcriptional regulation of hepatic glucose production in livers of PPP2R2A het mice.

Hepatic gene expression in free fed vs 18 hour fasted mice. A) Foxo1, B) Pck1, C) G6pc, D) Irs-2, E) Srebflc, F) Fasn, G) Insig1, H) Ppargc1a. Black bars represent IR het only, white bars IR/PPP2R2A double hets. N=8 biological replicates, data is represented as mean ± SD, statistical analysis 2 way ANOVA compared to IR het only fasted * p<0.05, **p<0.01, ***p<0.001.