MOB1 deletion in murine mature adipocytes ameliorates obesity and diabetes

Abstract

There is currently a global epidemic of obesity and obesity-related diseases such as type 2 diabetes due to decreased physical activity, excessive food intake, and/or genetic predisposition. The Hippo-YAP1 pathway has attracted attention as a potential therapeutic target because YAP1/TAZ activation in murine immature adipocytes in vitro suppresses their differentiation by inhibiting PPARγ activity. However, the role of YAP1 activation in mature adipocytes in vivo remains unclear. MOB1, whose expression is increased in obesity, is the hub of the Hippo core molecule complex and negatively regulates YAP1/TAZ activation. Therefore, we generated aMob1DKO mutant mice, which feature deficiency of Mob1a/b specifically in mature adipocytes. Compared to controls, aMob1DKO mice subjected to a high-fat diet showed beneficial changes consistent with resistance to diet-induced obesity. The mutants exhibited increases in basal lipolysis, "beiging," and energy expenditure, as well as suppression of ROS production and inflammation in white adipose tissue. Insulin sensitivity and glucose tolerance were improved, and ectopic fat accumulation was reduced. Most of these changes were dependent on the YAP1 activation observed in mature white adipose tissue of aMob1DKO mice. FGF21, which improves lipid metabolism, was upregulated directly via YAP1 activation, and many of the phenotypes seen in aMob1DKO mice were also dependent on FGF21. Thus, the aMob1DKO mouse is an interesting model for the study of the metabolic effects of diet-induced obesity and protection against diabetes. Our work suggests that a YAP1-FGF21 axis exists in adipocytes that may be a potential therapeutic target for obesity.

Keywords: MOB1-YAP1-FGF21-OPA1 axis; adipocytes; diabetes; obesity.

Fig. 1.

Adipocyte-specific MOB1-deficient mice are resistant to diet-induced obesity. (A, Left) Changes in body weight over time of 5- to 6-wk-old Cont and aMob1DKO male mice that were fed on NCD or HFD for 30 wk. NCD-Cont (n = 16), NCD-aMob1DKO (NCD-DKO) (n = 10), HFD-Cont (n = 12), and HFD-aMob1DKO (HFD-DKO) (n = 9) mice. ^∗P < 0.05. Unless otherwise indicated, all data of this type are the mean ± SEM. Right: Representative images of intact HFD-Cont and HFD-aMob1DKO mice after 30 wk of HFD feeding. (B, Left) Tissue weights of iWAT, gWAT, BAT, and liver from HFD-Cont (n = 12) and HFD-aMob1DKO (n = 12) male mice after 30 wk of HFD feeding. ^∗∗P < 0.01. Right: Representative images of the gross tissues analyzed in the Left panel. (Scale bar, 1 cm.) (C) Representative images of HE-stained histological sections of the indicated tissues from the mice in B. (Scale bar, 60 μm.) (D) Percentages of fat droplets of the indicated sizes in iWAT of the mice in B. ^∗P < 0.05, ^∗∗P < 0.01. Each area of 200 fat droplets was measured in each mouse using ImageJ (n = 5/group). (E) Total number of adipocytes in iWAT (n = 5), gWAT (n = 4), and BAT (n = 3) samples from HFD-Cont and HFD-aMob1DKO mice after 22 to 28 wk of HFD feeding. (F) Fasting serum triglyceride concentrations in the HFD-Cont (n = 5) and HFD-aMob1DKO (n = 4) mice in B. ∗P < 0.05. (G) Mean daily food intake per gram body weight after 6 to 8 wk (Left, n = 6) or 30 wk (Right, n = 12) HFD feeding. (H) Locomotion of HFD-Cont and HFD-aMob1DKO mice (n = 9/group) after 17 wk of HFD feeding.

Fig. 2.

MOB1-deficient mice show reduced inflammation and fibrosis, and improved glucose metabolism. (A, Left) Representative images of F4/80-immunostained iWAT from HFD-Cont (n = 4) and HFD-aMob1DKO (n = 4) mice after 30 wk of HFD feeding. Blue arrowheads indicate F4/80⁺ macrophages. Yellow arrowheads indicate crown-like structures (CLSs). (Scale bar, 100 μm.) Middle and Right: Mean percentage of F4/80⁺ macrophages (Middle), and mean percentage of CLSs (Right), quantified from multiple histological sections of iWAT from the mice in the Left panel. *P < 0.05. (B and C) mRNA levels of the indicated genes related to macrophage markers and inflammatory cytokines (B), and collagens and other fibrosis-related genes (C), in iWAT of HFD-aMob1DKO mice after 20 wk of HFD feeding. mRNA levels were determined by microarray analysis. Data are expressed as a ratio compared to levels in the iWAT of HFD-Cont mice (n = 4; 20 wk HFD feeding). *P < 0.05, **P < 0.01. (D) Glucose tolerance test of 16 h-fasted HFD-Cont (n = 11) and HFD-aMob1DKO (n = 7) mice after 30 wk of HFD feeding. *P < 0.05, **P < 0.01. (E) Insulin tolerance test of 6 h-fasted HFD-Cont (n = 11) and HFD-aMob1DKO (n = 6) mice after 30 wk of HFD feeding. **P < 0.01. (F) Serum insulin (Left), leptin (Middle), and adiponectin (Right) levels in overnight-fasted HFD-Cont (n = 5) and HFD-aMob1DKO (n = 3) mice after 30 wk of HFD feeding. **P < 0.01.

Fig. 3.

PKA activation and increased basal lipolysis are induced by MOB1 deficiency. (A) Glycerol (Left) and free FAs (Right) released from iWAT or gWAT of HFD-Cont and HFD-aMob1DKO mice (20 to 25 wk of HFD feeding) after incubation ex vivo with (+) or without (−) 10 μM CL316,243 (n = 7/group). *P < 0.05, **P < 0.01. (B) Free glycerol levels in the serum of HFD-Cont and HFD-aMob1DKO mice that were i.p-injected (+; n = 5) or not (−; n = 8) with 1 mg/kg CL316,243 after 20 wk of HFD feeding. *P < 0.05. (C) mRNA levels of the indicated lipolysis-positive and -negative regulator genes in iWAT of HFD-aMob1DKO mice (n = 4) after 20 wk of HFD feeding. mRNA levels were determined by microarray analysis. Data are expressed as a ratio compared to levels in iWAT of HFD-Cont mice (n = 4; 20 wk HFD feeding). *P < 0.05, **P < 0.01. (D, Left and E) Immunoblots to detect ATGL, pHSL (S660), pHSL (S563), HSL, pPLIN, PLIN, and CGI-58 proteins (D), and pPKA substrate proteins (E), in iWAT from HFD-Cont and HFD-aMob1DKO mice after 20 wk of HFD feeding. HSP90, loading control. (D, Right) Densitometric relative quantification of the indicated proteins in D, Left panel (n = 9). *P < 0.05, **P < 0.01. Data in D, Left and E are representative of nine and three independent experiments, respectively. (F) Quantification of PKA activity in iWAT of HFD-Cont and HFD-aMob1DKO mice (n = 5/group) after 20 wk of HFD feeding. Data are expressed as a ratio compared to levels in the HFD-Cont group. *P < 0.05.

Fig. 4.

Increased beiging and β-oxidation of iWAT, and elevated energy expenditure, are induced by MOB1 deficiency. (A) mRNA levels of the indicated brown/beige marker genes and beige regulator genes in iWAT of HFD-aMob1DKO mice (n = 4) after 20 wk of HFD feeding. mRNA levels were determined by microarray analysis. Data are expressed as a ratio compared to levels in iWAT of HFD-Cont mice (n = 4; 20 wk HFD feeding). *P < 0.05, **P < 0.01. (B, Top) Immunoblots to detect UCP1, PGC1α, β3-AR, FGF21, PPARα, and OPA1 proteins in iWAT from HFD-Cont and HFD-aMob1DKO mice (n = 9/group) after 20 wk of HFD feeding. HSP90, loading control. Bottom: Densitometric relative quantification of the indicated proteins in the blots in the top panels. *P < 0.05, **P < 0.01. (C) Serum norepinephrine levels in overnight-fasted HFD-Cont (n = 17) and HFD-aMob1DKO (n = 9) mice after 20 wk of HFD feeding. n.s., not significant. (D) mRNA levels of the indicated FAO-related genes in iWAT of HFD-aMob1DKO mice (n = 4) after 20 wk of HFD feeding. mRNA levels were determined by microarray analysis. Data are expressed as the ratio compared to levels in HFD-Cont mice (n = 4; 20 wk HFD feeding). *P < 0.05, **P < 0.01. (E–G) Monitoring and quantification of VO₂ (E), VCO₂ (F), and energy expenditure (EE) (G) during light (l) and dark (d) phases in HFD-Cont and HFD-aMob1DKO mice (n = 7/group) after HFD feeding for 20 wk. For part of the dark phase experiment, the temperature was reduced to 4 °C. In E, the arrow indicates the timing of administration of the β3-agonist CL316,243 (n = 5/group). *P < 0.05, **P < 0.01.

Fig. 5.

Reduced ROS production in MOB1-deficient WAT. (A, Left) Representative flow cytometric histogram of labeled H₂DCFDA fluorescence in iWAT cells from HFD-Cont and HFD-aMob1DKO mice (n = 6/group) after 20 wk of HFD feeding. Right: Mean fluorescence intensity (MFI) of labeled H₂DCFDA in the iWAT cells in the Left panel. **P < 0.01. (B, Left) Representative images of iWAT samples from HFD-Cont (n = 5) and HFD-aMob1DKO (n = 4) mice that were analyzed after 25 wk of HFD feeding and immunostained to detect 4-HNE (Top) or 8-OHdG (Bottom). Yellow arrowheads indicate cells with 8-OHdG-positive nuclei. (Scale bars, 20 μm.) Right: MFI of 4-HNE immunostaining of adipocyte cytoplasm (Top), and the percentage of adipocytes with 8-OHdG-positive nuclei (Bottom), as quantified from multiple histologic sections of the iWAT samples in the Left panel. *P < 0.05. (C) mRNA levels of the indicated ROS generator, ROS scavenger, and ROS regulator genes in iWAT of HFD-aMob1DKO mice (n = 4) after 20 wk of HFD feeding. mRNA levels were determined by microarray analysis. Data are expressed as the ratio compared to levels in HFD-Cont mice (n = 4; 20 wk HFD feeding). *P < 0.05, **P < 0.01.

Fig. 6.

YAP1 dependence of the MOB1-deficient mouse phenotype. (A) Body weights of HFD-Cont (n = 6), HFD-aMob1DKO (n = 6), and HFD-aMob1/Yap1TKO (HFD-TKO; n = 9) male mice fed on HFD for 20 wk. **P < 0.01. (B, Left and Middle) Tissue weights of iWAT, liver, and BAT from the mice in A. **P < 0.01. Right: Representative gross images of iWAT, BAT, and liver from the mice in A. (Scale bars, 1 cm.) (C) HE-stained histological sections of iWAT, BAT, and liver from the mice in A. Black bar, 50 μm; yellow bar, 20 μm. (D) Free glycerol (Left) and free FAs (Right) released spontaneously from iWAT of HFD-Cont, HFD-aMob1DKO, and HFD-aMob1/Yap1TKO mice (n = 6/group) after 20 wk of HFD feeding. *P < 0.05, **P < 0.01. (E) mRNA levels of the indicated brown/beige marker genes and beige regulator genes in iWAT of HFD-aMob1DKO (n = 4) and HFD-aMob1/Yap1TKO (n = 3) mice after 20 wk of HFD feeding. mRNA levels were determined by microarray analysis. Data are expressed as a ratio compared to levels in iWAT of HFD-Cont mice (n = 4; 20 wk HFD feeding). *P < 0.05, **P < 0.01 for HFD-aMob1DKO vs. HFD-aMob1/Yap1TKO comparisons. (F, Top) Representative flow cytometric histogram of labeled H₂DCFDA fluorescence in iWAT cells from HFD-Cont (n = 7), HFD-aMob1DKO (n = 6), and HFD-aMob1/Yap1TKO (n = 5) mice after 20 wk of HFD feeding. Bottom: MFI of labeled H₂DCFDA in the iWAT cells in the Top panel. **P < 0.01.

Fig. 7.

Activation of a YAP1-FGF21-OPA1 axis and FGF21 dependence of the MOB1-deficient mouse phenotype. (A, Left) Immunoblot to detect FGF21 and OPA1 proteins in iWAT from HFD-aMob1DKO and HFD-aMob1/Yap1TKO mice (n = 6/group) after 20 wk of HFD feeding. HSP90, loading control. Right: Densitometric relative quantification of the indicated proteins in the blots in the Left panels *P < 0.05, **P < 0.01. (B) qPCR to detect mRNA levels of the indicated genes in 3T3L1 cells overexpressing YAP5SA (Dox+) or not (Dox−). Data are expressed as the fold increase in mRNA expression in Dox+ cells over Dox− cells. Data are from four independent trials. *P < 0.05, **P < 0.01. (C, Left) Immunoblot to detect FGF21, OPA1, and YAP1 proteins in 3T3L1 cells overexpressing YAP5SA (Dox+) or not (Dox−) (n = 4/group). Exo, exogenous; endo, endogenous. HSP90, loading control. Right: Densitometric relative quantification of the indicated proteins in the blots in the Left panel. *P < 0.05, **P < 0.01. (D) Left: Immunoblot to detect OPA1 proteins in 3T3L1 cells overexpressing YAP5SA (Dox+) in the presence of either Scramble siRNA, Fgf21siRNA #1, or Fgf21siRNA #2 (n = 6/group). HSP90, loading control. Right: Densitometric relative quantification of OPA1 proteins in the blots in the Left panel. *P < 0.05. (E) Quantitative evaluation of ChIP assays used to detect binding of the YAP1 complex to TBS-containing DNA upstream of the Fgf21 gene in 3T3L1 cells overexpressing YAP5SA (Dox+) (n = 3/group). Immunoprecipitation by control IgG and DNA downstream of the Fgf21 gene lacking a TBS served as a negative control. DNA upstream of the Ctgf gene containing a TBS was used as a positive control. Data are relative enrichment values. *P < 0.05, **P < 0.01. (F) Quantitation of relative Fgf21-luciferase reporter activity in 293T cells overexpressing YAP5SA (“YAP5SA”) or not (“control”) when the luciferase vector contained Fgf21 promoter DNA (−289 to −1) plus WT TBS1 and TBS2; the same Fgf21 DNA (−289 to −1) plus mutated TBS1; the same Fgf21 DNA (−289 to −1) plus mutated TBS2; or control DNA downstream of Fgf21 (+1,813 ~+ 2,101) (n = 3/group). Please see SI Appendix, Fig. S11 A and B. *P < 0.05. (G) Changes in body weight of HFD-aMob1DKO (n = 7/group) and HFD-Cont (n = 6/group) male mice that were fed HFD for 20 wk and then injected i.p. on day 0 with either neutralizing anti-FGF21 antibody or normal IgG. Data are representative of two independent experiments. *P < 0.05. (H) Basal lipolysis as determined by free glycerol (Left) and free FAs (Right) spontaneously released from iWAT of the antibody-injected HFD-aMob1DKO (n = 7/group) and HFD-Cont (n = 6/group) mice in G). **P < 0.01. Data are representative of two independent experiments. (I) ROS production as determined by MFI of labeled H2DCFDA in iWAT of the antibody-injected HFD-aMob1DKO (n = 7/group) and HFD-Cont (n = 6/group) mice in G. *P < 0.05. Data are representative of two independent experiments.