High glucose represses β-klotho expression and impairs fibroblast growth factor 21 action in mouse pancreatic islets: involvement of peroxisome proliferator-activated receptor γ signaling

Abstract

Circulating fibroblast growth factor 21 (FGF21) levels are elevated in diabetic subjects and correlate directly with abnormal glucose metabolism, while pharmacologically administered FGF21 can ameliorate hyperglycemia. The pancreatic islet is an FGF21 target, yet the actions of FGF21 in the islet under normal and diabetic conditions are not fully understood. This study investigated the effects of high glucose on islet FGF21 actions in a diabetic mouse model by investigating db/db mouse islet responses to exogenous FGF21, the direct effects of glucose on FGF21 signaling, and the involvement of peroxisome proliferator-activated receptor γ (PPARγ) in FGF21 pathway activation. Results showed that both adult db/db mouse islets and normal islets treated with high glucose ex vivo displayed reduced β-klotho expression, resistance to FGF21, and decreased PPARγ expression. Rosiglitazone, an antidiabetic PPARγ ligand, ameliorated these effects. Our data indicate that hyperglycemia in type 2 diabetes mellitus may lead to FGF21 resistance in pancreatic islets, probably through reduction of PPARγ expression, which provides a novel mechanism for glucose-mediated islet dysfunction.

FIG. 1.

Hepatic FGF21 mRNA and circulating levels of FGF21 are increased in db/db mice by 12 weeks of age. A: mRNA expression as quantified by standard quantitative RT-PCR. B: Circulating FGF21 levels as determined by ELISA of serum from a terminal bleed. Data are means ± SEs. **P < 0.01; ***P < 0.001 vs. age-matched m+/db group; #P < 0.05 vs. 4-week db/db group (n = 5 to 6).

FIG. 2.

β-Klotho is downregulated in islets of adult db/db mice. mRNA levels of FGFRs and β-klotho (A) were determined in 12-week-old m+/db and db/db mice. mRNA (B) and protein (C) levels of β-klotho in islets were analyzed in 4-, 8-, and 12-week-old mice. For the Western blot, β-actin was used as a loading control. Densitometry is shown and was calculated as β-klotho/β-actin. Data are means ± SEs. *P < 0.05; **P < 0.01; ***P < 0.001 vs. age-matched m+/db group; #P < 0.05; ##P < 0.01 vs. 4-week db/db group (n = 4 to 5).

FIG. 3.

FGF21-induced FRS2 phosphorylation is attenuated in islets from adult db/db mice relative to m+/db mice. Islets were isolated from the mice at 4 (A), 8 (B), and 12 weeks (C) of age and treated with FGF21 (at the indicated concentrations) for 10 min. Phosphorylated (pFRS2) and total FRS2 expression were quantitated in Western blots. Densitometry is shown and calculated as pFRS2/total FRS2. Data are means ± SEs. *P < 0.05; **P < 0.01 vs. 0 nmol/L group; #P < 0.05; ##P < 0.01 vs. 10 nmol/L group (n = 4).

FIG. 4.

FGF21-induced expression of immediate early genes is attenuated in islets of adult db/db mice relative to m+/db mice. Islets were isolated from the mice at 4 (A and B), 8 (C and D), and 12 weeks (E and F) of age and treated with FGF21 (at the indicated concentrations) for 1 h. Egr1 (A, C, and E) and cFos (B, D, and F) mRNA levels were analyzed. Data are means ± SEs. *P < 0.05; **P < 0.01; ***P < 0.001 vs. 0 nmol/L group; #P < 0.05; ##P < 0.01 vs. 50 nmol/L group (n = 5 to 6).

FIG. 5.

High glucose reduces β-klotho expression and impairs FGF21-induced FRS2 phosphorylation in islets. A: Islets were isolated from lean mice and treated with 5.6 or 28 mmol/L glucose for 24, 48, or 72 h. **P < 0.01 vs. time-matched 5.6 mmol/L group; #P < 0.05 vs. 24-h 28 mmol/L group (n = 4). B: Islets were isolated from lean mice and pretreated with 5.6 mmol/L or 28 mmol/L glucose for 72 h, then were treated with FGF21 (at the indicated concentrations) for 10 min. *P < 0.05; **P < 0.01 vs. 0 nmol/L group; #P < 0.05 vs. 10 nmol/L group (n = 4). Data are means ± SEs. pFRS2, phosphorylated FRS2.

FIG. 6.

Rosiglitazone reverses the downregulation of β-klotho and impairment of FGF21-induced FRS2 phosphorylation in islets subjected to high glucose treatment and in adult db/db mouse islets, and these effects are blocked by the PPARγ antagonist GW9662. A: Islets were isolated from lean mice and treated with 5.6 or 28 mmol/L glucose, 20 μmol/L rosiglitazone (Rosi), with or without 20 μmol/L GW9662 (GW) for 72 h. *P < 0.05; **P < 0.01 vs. 28 mmol/L group; #P < 0.05 vs. 28 mmol/L + Rosi group (n = 4). B: Pretreated islets were exposed to 0 or 100 nmol/L FGF21 for 10 min for signaling analyses. *P < 0.05; ***P < 0.001 vs. 0 nmol/L group (n = 4). C: Isolated islets of adult m+/db and db/db mice were treated with 20 μmol/L Rosi, with or without 20 μmol/L GW, for 72 h. **P < 0.01; ***P < 0.001 vs. db/db group; #P < 0.05 vs. db/db + Rosi group (n = 4). D: Pretreated islets were exposed to 0 or 100 nmol/L FGF21 for 10 min for signaling analyses. *P < 0.05; **P < 0.01 vs. 0 nmol/L group (n = 4). Data are means ± SEs. pFRS2, phosphorylated FRS2.

FIG. 7.

PPARγ expression is reduced in high glucose–treated islets and adult db/db mouse islets, which is reversed by rosiglitazone. A: Isolated islets from lean mice were treated with 5.6 or 28 mmol/L glucose, 20 μmol/L rosiglitazone (Rosi), with or without 20 μmol/L GW9662 (GW), for 72 h. *P < 0.05 vs. 28 mmol/L group; #P < 0.05 vs. 28 mmol/L + Rosi group (n = 6). B: Isolated islets from 12-week-old m+/db and db/db mice were treated with 20 μmol/L Rosi with or without 20 μmol/L GW for 72 h. *P < 0.05; **P < 0.01 vs. db/db group (n = 6). C: Islet PPARγ expression was examined in m+/db and db/db mice at 4, 8, and 12 weeks of age. ***P < 0.001 vs. age-matched m+/db group; ###P < 0.001 vs. 4-week db/db group (n = 4). β-Actin was used as a loading control. Densitometry is shown and calculated as PPARγ/β-actin. Data are means ± SEs.