STAT3 dictates β-cell apoptosis by modulating PTEN in streptozocin-induced hyperglycemia

Abstract

Insufficient pancreatic β-cell mass or insulin-producing β-cells are implicated in all forms of diabetes mellitus. However, the molecular mechanisms underlying β-cell destruction are complex and not fully defined. Here we observed that activation of STAT3 is intensely and specifically inhibited in β-cells under hyperglycemic conditions. By knocking out STAT3 specifically in mouse β-cells, we found that the loss of STAT3 sensitized mice to three low doses of STZ stimulation resulting in hyperglycemia. Mechanistically, accumulating PTEN, induced by STAT3 deficiency, directly represses phosphorylation of AKT, which negatively modulates transcription factor activation, dysregulates β-cell function, positively promotes apoptotic signaling, and finally induces β-cell apoptosis. Notably, the defective secretion of insulin and β-cells apoptosis was completely rescued by PTEN ablation in STAT3-null islets or PTEN inhibitor bpv(phen) treatment. Thus our data suggest that STAT3 is a vital modulator of β-cell survival and function, highlighting a critical role for STAT3 in the negative regulation of PTEN-AKT signaling pathway associated with β-cell dysfunction and apoptosis.

Conflict of interest statement

The authors declare that they have no conflict of interest.

Figures

Fig. 1

Signal transducer and activator of transcription 3 (STAT3) activation is inhibited in damaged β-cells. a–c INS-1 cells were exposed to normal (11.1 mM) and high glucose conditions (22.2 mM, 33.3 mM) glucose for 72 h (a) or 1 mM, 2 mM streptozocin (STZ) (b) or 100 μM, 200 μM H2O2 (c) in the presence of 11.1 mM or 33.3 mM glucose for 8 h. The protein level of STAT1, pSTAT1, STAT3, pSTAT3, STAT6, and pSTAT6 were detected by western blotting. Glc glucose. β-Actin or GAPDH was used as the loading control. d–f Ten-week-old male wild-type (WT) mice were injected with 150 mg kg−1 STZ or 130 mg kg−1 STZ (i.p.) or vehicle for single dose after fasting overnight (n = 10 mice/group). i.p., intraperitoneal. d Random blood glucose levels of vehicle and STZ-induced mice (n = 5 mice/group) 24 days after vehicle or STZ injection. Cont control (vehicle), H1 hyperglycemia (≥11.1 mM), H2 hyperglycemia (≥16.7 mM). e Immunohistochemistry and quantification for STAT3 and pSTAT3 in islet sections from control mice, as well as STZ-induced hyperglycemia (H1 and H2) mice at day 24. Scale bar, 200 μm. f Islets were isolated from vehicle mice and hyperglycemic mice (≥16.7 mM) at day 24. Western blotting was used to detect the protein expression (n = 5 mice/group). g, h Four-week-old male WT mice fed a normal chow diet or a high-fat diet (HFD) for 16 weeks. g Blood glucose levels of HFD-induced and chow-fed mice (n = 5 mice/group) performed at week 16. h Immunohistochemistry and quantification for STAT3 and pSTAT3 in islet sections from control mice and HFD-induced mice at week 16. Scale bar, 200 μm. Immunoreactivity was assessed by quantitative morphometry with automated image analysis (Image-Pro plus, Version6.0). Data represent the mean ± s.d., *P < 0.05, **P < 0.01; ###P < 0.001 (Student’s t test in g, h, one-way analysis of variance with Tukey’s multiple comparisons test in d, e)

Fig. 2

Signal transducer and activator of transcription 3 (STAT3)-deficient mice are more sensitive to streptozocin (STZ) treatment. a Ten-week-old male β-STAT3KO (KO) mice (n = 21) and their littermates (n = 22) were injected with 40 mg kg−1 STZ  (i.p.) or vehicle for 3 consecutive days after fasting overnight. b Blood glucose levels of KO mice and their littermates at the indicated time points. i.p., intraperitoneal. c Glucose tolerance test (i.g.) was performed at day 21; blood glucose was measured at 0, 15, 30, 60, 90, 120, and 180 min.  i.g., intragastric. d Plasma insulin levels were measured at 0 and 30 min after glucose gavage at day 23. *P < 0.05 (STZ-KO 0 min vs. STZ-WT 0 min); ##P < 0.01 (STZ-KO 30 min vs. STZ-WT 30 min). e β-Cell mass from KO mice and their littermates injected with vehicle or STZ at day 24. f Left, the pancreas of KO mice and their littermates injected with vehicle or STZ were at day 24 and immunostained for Insulin and Glucagon, respectively. Scale bars, 200 μm (inset), 1 mm (top), 100 μm (bottom). Right, quantification of the percentage of Insulin+ α-cells and Glucagon+ β-cells (n = 8 mice/group). g Western blotting was carried out to detect apoptosis in mouse islets from KO and WT mice. GAPDH, loading control. h Immunostaining of Cleaved Caspase 3 (CC3) with Insulin in the WT and KO mice pancreas at day 24. Scale bar, 100 μm. i Quantification of the percentage of CC3+ cells and Ki67+ cells among insulin+ β-cells at day 24 (n = 8 mice/group). j Quantitative reverse transcription PCR was carried out to analyze β-cell identity genes in pancreatic islets from 3MLD-STZ-induced or vehicle-treated KO mice and WT mice at day 24 (n = 5 mice/group). Gapdh, control. All values are expressed as mean ± s.d., *P < 0.05, **P < 0.01, ***P < 0.001; ##P < 0.01, ###P < 0.001 (Student’s t test in b–d, i and j, one-way analysis of variance with Tukey’s multiple comparisons test in e, f)

Fig. 3

Depletion of phosphatase and tensin homolog (PTEN) alleviates β-cell dysfunction induced by signal transducer and activator of transcription 3 (STAT3) deficiency in 3MLD-STZ mice. a Ten-week-old male knockout (KO) mice and their littermates were injected with 40 mg kg−1 streptozocin (STZ, i.p.) for 3 consecutive days after fasting overnight. i.p., intraperitoneal.  Quantitative reverse transcription PCR (qRT-PCR) quantification showing the expression of selected genes in pancreatic islets (n = 5 mice/group). Gapdh, control. b, c INS-1 cells were transfected with siSTAT3 #1 or control siRNA (siNC) and treated with or without 33.3 mM glucose for 48 h. 11.1 mM, normal condition as control; 33.3 mM, high glucose. qRT-PCR (b) and western blotting (c) were used to detect STAT3 and PTEN. GAPDH, control. ***P < 0.001 (11.1 mM + siNC vs. 11.1 mM + siSTAT3 #1); ##P < 0.01, ###P < 0.001 (33.3 mM + siNC vs. 33.3 mM + siSTAT3 #1). d Islets were isolated from different groups (wild-type (WT), KO, STZ-induced WT, and STZ-induced KO mice) at day 24, the expression of STAT3 and PTEN were examined by western blotting. e Representative immunostaining image and quantitative analysis for Insulin (red) and PTEN (green) at day 24 (n = 8 mice/group). Scale bar, 100 μm. f–h Ten-week-old male β-STAT3-PTENDKO (DKO) mice, KO mice, and their littermates were injected with 40 mg kg−1 STZ (n = 10, 15, 10) or vehicle (n = 10 mice/group) for 3 consecutive days after fasting overnight. f Random blood glucose was checked at the indicated times. g Glucose tolerance test experiment was performed at day 21 to detect the blood glucose at 0, 15, 30, 60, 90, 120, and 180 min. h Immunostaining (left) and quantification (right) of the percentage of Insulin (red) and Glucagon (green) or CC3 (green) from KO and DKO mice at day 27 (n = 5 mice/group). Scale bar, 100 μm. All values are expressed as mean ± s.d., **P < 0.01, ***P < 0.001; ##P < 0.01, ###P < 0.001 (Student’s t test in b, e–g, one-way analysis of variance with Tukey’s multiple comparisons test in a, h)

Fig. 4

Signal transducer and activator of transcription 3 (STAT3)-phosphatase and tensin homolog (PTEN) regulates β-cell survival and function through AKT activation. a 293FT cells were transfected with luciferase reporters driven by PDX1 or MAFA promoter together with siNC or siSTAT3 and treated with/without bpv(phen) as indicated for 24 h. b–e Ten-week-old male DKO mice, knockout (KO) mice, and their littermates were injected with 40 mg kg−1 streptozocin (STZ) or vehicle for 3 consecutive days after fasting overnight (n = 10 mice/group). b Quantitative reverse transcription PCR was performed to analyze β-cell identity genes in islets from DKO mice, KO mice and their littermates at day 24 (n = 5 mice/group). Gapdh, control. Values are expressed as mean ± s.e. c Representative images of immunolabeling (left) and quantification (right) for PDX1 (green) and Insulin (red) from DKO mice, KO mice and their littermates at day 24 (n = 5 mice/group). Scale bar, 100 μm. d Pancreatic islet cells isolated from the above mice were subjected to immunoblotting against STAT3, PTEN, AKT, pAKT, PDX1, Nkx6.1, Cleaved Caspase 3, and Cleaved poly ADP-ribose polymerase  (PARP) 27 days after last STZ. GAPDH, loading control. e Immunohistochemistry for pAKT in sections from mouse pancreases at day 24. Scale bar, 200 μm. f,g Ten-week old-male DKO mice (n = 5), KO mice (n = 6), and their littermates (n = 5) were injected with 40 mg kg−1 STZ or vehicle for 3 consecutive days after fasting overnight, then treated with AZD5363 (i.g., 200 mg kg−1 daily) from day 11 to day 27, and sacrificed at day 27. f Blood glucose levels of randomly fed of DKO, KO mice, and the relative wild-type littermates with or without AZD5363 treatment. The arrow represents the first day of AZD5363 administration. g Blood glucose levels were measured at day 21 by glucose tolerance test (i.g.) at 0, 15, 30, 60, 90, 120, and 180 min. Immunoreactivity was assessed by quantitative morphometry with automated image analysis (Image-Pro plus, Version6.0). i.g., intragastric. All values are expressed as mean ± s.d., *P < 0.05, **P < 0.01, ***P < 0.001; #P < 0.05, ###P < 0.001 (Student’s t test in a, f, g, one-way analysis of variance with Tukey’s multiple comparisons test in b, c)

Fig. 5

bpv(phen) rescues β-cell survival and function disrupted by 3MLD-STZ-treatment in β-STAT3KO mice. a Ten-week-old male knockout (KO) mice (n = 16) and their littermates (n = 15) were injected with 40 mg kg−1 streptozocin (STZ, i.p.) or vehicle for 3 consecutive days after fasting overnight, daily injected with 2 mg kg−1 bpv(phen, i.p.) or vehicle from day 11 to day 27. i.p., intraperitoneal. b Blood glucose levels were checked at the indicated times of KO mice or wild-type (WT) littermates. The arrow represents the first day of bpv(phen) administration. c Body weight monitored every 3 days. d Blood glucose was analyzed after 0, 15, 30, 60, 90, 120, and 180 min in glucose tolerance test (i.g.) experiment at day 21. e Insulin secretion was measured before (0 min) and 30 min after glucose gavage at day 23. ***P < 0.001 (STZ-KO 0 min vs. STZ-KO-bpv 0 min); ###P < 0.001 (STZ-KO 30 min vs. STZ-KO-bpv 30 min). f β-cells mass from KO mice and their littermates at day 27. g Representative images (left) of α-cell (green) and β-cell (red) staining in islets from different groups and quantification (right) at day 27 (n = 5 mice/group). Scale bar, 100 μm. h Islets were isolated from WT and KO male mice treated with STZ or combined with bpv(phen) (n = 5 mice/group) at day 27. Western blotting was used to detect Nkx6.1, PDX1, Cleaved Caspase 3, and Cleaved poly ADP-ribose polymerase  (PARP). GAPDH, loading control. i Quantification of the percentage of Ki67+ pancreatic β-cells from KO and WT mice (n = 5 mice/group). j Immunohistochemical staining of pAKT in pancreatic sections. Scale bar, 200 μm. k, l Representative images (k) of PDX1 (green) and Insulin (red) staining in islets from different groups and quantification analysis (l) (n = 5 mice/group). Scale bar, 100 μm. m Quantitative reverse transcription PCR analysis of β-cell identity gene expression in islets from KO mice, bpv(phen)-injected KO mice, and their control mice (n = 5 mice/group) at day 27. Gapdh, control. Immunoreactivity was assessed by quantitative morphometry with automated image analysis (Image-Pro plus, Version6.0). All values are expressed as mean ± s.d., *P < 0.05, **P < 0.01, ***P < 0.001; #P < 0.05, ##P < 0.01, ###P < 0.001 (Student’s t test in b–e, one-way analysis of variance with Tukey’s multiple comparisons test in f, g, i, l, m)

Fig. 6

Disruption of AKT abolishes alleviated hyperglycemia by deleted phosphatase and tensin homolog (PTEN) in knockout (KO) mice. a–e Ten-week old-male KO mice and their littermates were injected with 40 mg kg−1 streptozocin  (STZ, i.p.) for 3 consecutive days after fasting overnight, then treated with bpv(phen) (i.p., 2 mg kg−1 once daily) alone or with AZD5363 (i.g., 200 mg kg−1 once daily) and at day 27 (n = 9 mice/group). a Random-fed blood glucose was monitored every 3 days of KO mice or the respective wild-type (WT) littermates. The arrow represents the first day of bpv(phen) administration or with AZD5363. b Plasma glucose was measured after 0, 15, 30, 60, 90, 120, and 180 min in glucose tolerance test (i.g.) experiment at day 21. c Representative staining and quantification analysis for Insulin (red) and Glucagon (green) in islets from different groups at day 27 (n = 5 mice/group). Scale bar, 100 μm. d Representative images and statistics of anti-CC3 (green) and anti-Insulin (red) immunostaining (n = 5 mice/group). Scale bar, 100 μm. e Immunofluorescence and percentage of anti-PDX1 (green) and anti-Insulin (red) immunostaining (n = 5 mice/group). Scale bar, 100 μm. f Left, in the normal milieu, activated signal transducer and activator of transcription 3 (STAT3) inhibits PTEN and promotes AKT phosphorylation, thereby prevents β-cells from apoptosis and triggers β-cell-related gene transcriptions, which is involved in β-cell function. Right, the proposed model for the role of STAT3-PTEN signaling axis in β-cells. In hyperglycemic stimuli or β-cells suffering from oxidative/DNA damage, which inhibits STAT3 activation, PTEN is heavily accumulated, and AKT activation is impaired that then triggers Cleavage of Caspase 3 and acceleration of β-cell death. β-Cell-related gene transcription factors are also impaired, further disturbing β-cell survival and function. i.p., intraperitoneal;  i.g., intragastric. All values are expressed as mean ± s.d., ***P < 0.001 (Student’s t test)