Fig 2: GC-induced glucose intolerance is compromised in MKO mice.A, male 8-week-old WT and MKO mice were treated with 10 mg/kg of PBS or DEX in drinking water for 1 week. On the last day, mice were injected intraperitoneally with insulin (1 unit/body weight) for 10 min, and then various tissues were collected. ELISA kits were used to monitor the level of Akt and phosphor-Akt in eWAT, liver, and GA muscle. The results are presented as relative pAkt/AKt level. Error bars represent the SD, n = 3 to 9 and *p = 0.05 comparing no insulin treatment to insulin treatment. B, male 8-week-old WT mice and MKO mice were treated with 10 mg/kg of DEX for 7 days. On the last day, mice were fasted for 15 h, and the IPGTT was performed. Relative area under curve (AUC) for IPGTT results (relative to PBS-treated WT mice). Error bars represent the SD, n = 6–12 and *p = 0.05. C, plasma insulin level was measured before glucose injection (0 min time point). Error bars represent the SD, n = 6 to 12 and *p = 0.05. D, ITT was performed in mice as described in Methods. ITT results were depicted as percentage of initial plasma glucose level (the plasma glucose level before insulin injection). Error bars represent the SD, n = 3 to 7. Relative area under curve (AUC) for ITT results (relative to PBS-treated WT mice) is shown. Error bars represent the SD, n = 6 to 11 and *p = 0.05. DEX, dexamethasone; IPGTT, intraperitoneal glucose tolerance test; ITT, insulin tolerance test.

Fig 3: Mechanism of action of chrysin against TGF-ß1-mediated EMT signaling chrysin inhibits CsA-induced TGF-ß1/Smad-dependent increase in Snail expression, preventing a decrease in E-cadherin expression and subsequent localization of ß-catenin into the cytoplasm. Chrysin also inhibits Akt/GSK-3ß-mediated signaling, inducing the degradation of cytosolic ß-catenin. Action on both of these pathways prevents cytosolic accumulation of ß-catenin thereby inhibiting EMT and preventing the development and progression of fibrosis.

Fig 4: Effect of chrysin on CsA- or TGF-ß1-induced EMT-signaling. LLC-PK1 cells were treated with vehicle (Veh), 4.2 µM CsA, 5 ng/mL TGF-ß1 or 25 µM chrysin (Chr), or combinations. (A) Protein expression for TGF-ß1, phospho-Smad3, total Smad 3 and Snail were determined by Western blot analysis. Phosphorylation of Akt (B) and GSK-3ß (C) was determined by ELISA. Representative Western blots are shown, and data are represented as mean ± SEM of three individual experiments. Statistical significance was determined by one-way ANOVA followed by Tukey’s test for multiple comparisons. * indicates p < 0.05 vs. control; † indicates p < 0.05 vs. chrysin co-treatment. The effects of chrysin on CsA- or TGF-ß1-induced localization of ß-catenin (panels D–I) were also visualized. Release of membrane-bound ß-catenin is indicated using white arrowheads. Treatments: (D) control, (E) 5 ng/mL TGF-ß1, (F) 5 ng/mL TGF-ß1 + 25 µM chrysin, (G) 4.2 µM CsA, (H) 4.2 µM CsA + 25 µM chrysin and (I) 25 µM chrysin. Scale bar indicates 100 µm.

Fig 5: Overexpression of DUSP4 attenuated Dex-induced glucose but not insulin resistance.a Western blot of DUSP4 shows overexpression in liver but not gastrocnemius muscle (GA) or epidydimal white adipose tissue (eWAT), b IPGTT in Ehmt2K182R/K182R mice expressing GFP (Pink) or overexpressing mouse Dusp4 (mDusp4) (Blue) (GFP n = 4, GFP+ n = 3, mDusp4 n = 4, mDusp4+ n = 4 biologically independent mice), c plasma insulin (GFP n = 4, GFP+ n = 6, mDusp4 n = 3, mDusp4+ n = 4 biologically independent mice), d ITT (GFP n = 4, GFP+ n = 3, mDusp4 n = 3, mDusp4 + n = 5 biologically independent mice), e pAKT/AKT ratios in liver, gastrocnemius muscle (GA), and epididymal white adipose tissue (eWAT) measured with ELISA (GFP n = 4, GFP+ n = 3, mDusp4 n = 3, mDusp4+ n = 4 biologically independent mice), f Gene expression (GFP+ n = 4, mDusp4+ n = 5 biologically independent mice) and g western blot n = 3 biologically independent mice for Ehmt2K182R/K182R mice expressing GFP or overexpressing mouse Dusp4 treated with Dex, h Western blot of DUSP4 shows knockdown in liver but not gastrocnemius muscle or epidydimal white adipose tissue (i) IPGTT in WT mice with shRNA-Scramble (Black) or Dusp4 (Red) (Scr n = 4, Scr+ n = 4, shDusp4 n = 4, shDusp4+ n = 5 biologically independent mice), j Plasma Insulin (Scr n = 4, Scr+ n = 4, shDusp4 n = 4, shDusp4+ n = 5 biologically independent mice), k ITT (Scr n = 3, Scr+ n = 5, shDusp4 n = 3, shDusp4+ n = 4 biologically independent mice), l pAKT/AKT ratios in liver, gastrocnemius muscle (GA), and epididymal white adipose tissue (eWAT) measured with ELISA (Scr n = 3, Scr+ n = 3, shDusp4 n = 3, shDusp4+ n = 4 biologically independent mice), m Gene expression (Scr+ n = 4, shDusp4+ n = 5 biologically independent mice) and n Western blot n = 3 for biologically independent mice WT mice with shRNA-Scr or shDusp4 treated with Dex. Statistical tests used were two-way ANOVA with a Holm-Šídák’s multiple comparison test, or an unpaired two-tailed t test, The center lines depict the mean. Error bars represent SEM for the tolerance tests and standard deviation for the rest. Source data are provided as a Source Data file.