Elevated androgen levels induce hyperinsulinemia through increase in Ins1 transcription in pancreatic beta cells in female rats

Abstract

Hyperandrogenism is associated with hyperinsulinemia and insulin resistance in adult females. We tested whether androgens dysregulate pancreatic beta cell function to induce hyperinsulinemia through transcriptional regulation of insulin gene (Ins) in the islets. Adult female Wistar rats implanted with dihydrotestosterone (DHT; 7.5-mg, 90-d release) or placebo pellets were examined after 10 weeks. DHT exposure increased plasma DHT levels by 2-fold similar to that in polycystic ovary syndrome in women. DHT exposure induced hyperinsulinemia with increased HOMA-IR index in fasting state and glucose intolerance and exaggerated insulin responses following glucose tolerance test. DHT females had no change in islet number, size and beta cell proliferation/apoptosis but exhibited significant mitochondrial dysfunction (higher ADP/ATP ratio, decreased mtDNA copy number, increased reactive oxygen production and downregulation of mitochondrial biogenesis) and enhanced glucose-stimulated insulin secretion. Ins expression was increased in DHT islets. In vitro incubation of control islets with DHT dose dependently stimulated Ins transcription. Analysis of Ins1 gene revealed a putative androgen responsive element in the promoter. Chromatin-immunoprecipitation assays showed that androgen receptors bind to this element in response to DHT stimulation. Furthermore, reporter assays showed that the promoter element is highly responsive to androgens. Insulin-stimulated glucose uptake in skeletal muscle was decreased with associated decrease in IRβ expression in DHT females. Our studies identified a novel androgen-mediated mechanism for the control of Ins expression via transcriptional regulation providing a molecular mechanism linking elevated androgens and hyperinsulemia. Decreased IRβ expression in the skeletal muscles may contribute, in part, to glucose intolerance in this model.

Figure 1.

Changes in plasma DHT levels (A), fasting glucose (B), fasting insulin (C), C-peptide (D), HOMA-IR (E), and HOMA-IS (F) in female rats treated for 10 weeks with DHT or control pellets. All data are expressed as means ± SE of six animals in each group; *P < 0 .05 vs. control.

Figure 2.

Blood glucose levels and insulin responses following intraperitoneal glucose tolerance test in control and DHT rats. Female rats were treated with DHT or control pellets for 10 weeks and following overnight fasting, intraperitoneal glucose tolerance tests were performed. Blood samples were collected at 0, 30, 60, 90, and 120 min following intraperitoneal glucose (2 g/kg) administration for measurement of blood glucose (A) and insulin levels (B). The overall plasma glucose levels and insulin responses are expressed as the AUC calculated by the trapezoidal method (bottom panel). Results are expressed as means ± SE of six animals in each group; *P < .05 vs. control.

Figure 3.

Pancreatic histology in control and DHT rats. Pancreatic sections from control and DHT-treated females were stained with hematoxylin and eosin. Representative islets from both groups are shown with original magnification ×40 (A). The number of islets and mean islet area were determined as described in Materials and Methods. Insulin staining intensity—red fluorescence (B) in control and DHT islets were determined using four clusters per pancreatic section and three pancreatic sections per animal. Values are presented as the mean ± SEM of n = 5 animals in each group; *P < 0.05 vs. control.

Figure 4.

Mitochondrial function is control and DHT islets. Pancreatic islets were isolated from female rat treated with DHT or control pellet for 10 weeks. ADP/ATP ratio (A), mitochondrial copy number (B), and ROS generation (C) were measured as described in Materials and Methods. PGC-1α mRNA expressions in islets isolated from control and DHT-treated rats were determined by qRT-PCR and normalized with β-actin (D). Results are presented as the mean ± SEM of n = 5 animals in each group; *P < 0.05 vs. control.

Figure 5.

In vitro glucose-stimulated insulin secretion (A) and Ins gene expression (B) in control and DHT islets. Pancreatic islets were isolated from control and DHT-treated rats using collagenase digestion method. Following recovery and incubation, islets were transferred to KRH buffer with low (2.8 mM) and high (16.7 mM) glucose. After 1 h, supernatant was collected and used for insulin measurement by ELISA. The mRNA expression of Ins1 and Ins2 was done by quantitative RT-PCR and normalized with β-actin. Values are presented as the mean ± SEM of n = 5 animals in each group; *P < 0.05 vs. respective control.

Figure 6.

Insulin gene expression in islets following in vitro exposure to DHT. Pancreatic islets were isolated from control female rats using collagenase digestion method. After recovery islets were treated with vehicle and increasing concentration of DHT for 24 h. Ins1 (A) and Ins2 (B) mRNA expressions were determined by quantitative RT-PCR and results were normalized with β-actin. Results are presented as the mean ± SEM of n = 5 animals in each group; *P < 0.05 vs. vehicle.

Figure 7.

Identification for putative androgen receptor binding sites in the ins1 gene promoter. Bioinformatic prediction using ConSite web-based prediction tool shows putative ARE sites in the promoter of rat ins1 gene (A). The ChIP assay shows DHT-induced AR interaction with a putative response element ARE1 (B) indicating the presence of an active ARE on ins1 promoter. Data represented as the mean ± SEM of four independent experiments with *P < 0.05 vs. vehicle.

Figure 8.

Functional effect of ARE1 in enhancing ins1 gene expression. Reporter assay showing luciferase activity in HEK cells transfected with reporter plasmid containing ARE1 showing enhanced luciferase activity when incubated with DHT. This increase was suppressed when treated with DHT in the presence of flutamide. Data represent mean ± SEM of three independent experiments with *P < 0.05 vs. vehicle.

Figure 9.

IRβ protein expression and glucose uptake in control and DHT rats. IRβ levels were determined in gastrocnemius muscle. Representative western blots for IRβ (A, upper panel) and blot density obtained from densitometric scanning of IRβ normalized to GAPDH (A, lower panel) are shown. Basal and insulin-stimulated (1 mU/ml) glucose uptake was assessed in soleus muscles using 2-deoxy-D-[1-3 H]-Glucose and D-[14C]-mannitol (B). Results are presented as the mean ± SEM with n = 4 animals in each group; *P < 0.05 vs. respective controls and ^#P < 0.05 vs. without insulin control.