Pinch Loss Ameliorates Obesity, Glucose Intolerance, and Fatty Liver by Modulating Adipocyte Apoptosis in Mice

Obesity, or excess of body fat, is a global public health problem and a major risk factor of cardiometabolic diseases and traits, such as dyslipidemia, insulin resistance, diabetes, nonalcoholic steatohepatitis, and cardiovascular diseases (1,2). In obesity, the adipose tissue expansion occurs by both adipocyte hypertrophy (increase in size) of the existing adipocytes and hyperplasia (increase in number) via recruiting and differentiating preadipocytes (3). The relationship between adipose cell size and obesity is nonlinear; when the increase in cell size reaches a plateau, generation of new adipocytes is triggered. Adipocyte hyperplasia is protective against metabolic dysfunction, whereas adipocyte hypertrophy is a marker of adipose tissue dysfunction and insulin resistance (3–5). Adipocyte hypertrophy may cause insulin resistance by mechanisms involving multiple biological pathways, such as hypoxia, oxidative stress, inflammation, angiogenesis, and cross talk with immune cells such as macrophages (6).

Although obesity is associated with a higher risk of developing comorbidities, not every obese individual will develop related complications (7–9), which implies the complexity of adipose tissue and its interaction with metabolic abnormalities. Previous studies have shown that both antiapoptotic and prosurvival pathways are activated in obese adipose tissue (10,11). Wang et al. (10) reported that proper adipocyte death could ameliorate obesity and systemic insulin resistance, while several other studies have shown that inappropriate cell death affects immunity and induces inflammation, leading to metabolic diseases (12–14). Collectively, these results suggest that maintaining adipocyte apoptosis at a proper level is critical for adipose development and homeostasis. However, key signals that modulate adipocyte apoptosis remain incompletely understood.

The focal adhesion (FA) proteins activate integrins and regulate fundamental cellular processes, such as cell adhesion, migration, proliferation, survival, mechanosensation, and transduction (15–17). Recent studies demonstrate that key FA proteins, such as Kindlin-2 and Talin, play important roles in regulation of the development and homeostasis of organs and tissues, such as the heart, intestine, pancreas, kidney, and skeleton (18–27). Mammals have two functional Pinch proteins, Pinch1 and Pinch2. They are components of the FA complex and contain five LIM domains, which mediate critical protein-protein interactions. For example, Pinch interacts with integrin-linked kinase (ILK) through the LIM1 domain (28). ILK binds to Parvin and, together with Pinch, forms a complex known as IPP (ILK-PINCH-Parvin), which is essential for the stability of these proteins (29–31). Pinch1 deficiency in mice is embryonically lethal, whereas Pinch2 deletion does not cause any obvious phenotypes (32). Previous studies of Pinch proteins have primarily focused on their functions in malignancies. For example, Pinch1 is upregulated in tumor stromal cells in patients with colon, breast, and lung carcinomas and may play a critical role in tumor invasion and metastasis (33). Cumulative evidence suggests that Pinch proteins are critically involved in the control of organ and tissue development and function. Liang et al. (34) were the first to demonstrate a pivotal role of Pinch1/2 in the control of heart development and function. Furthermore, Donthamsetty et al. (35) reported that Pinch1 loss in hepatocytes resulted in stiff liver and liver cancer. We recently reported that expression of Pinch1/2 in osteocytes regulated bone mass and bone mineral density through indirect promotion of osteoblast formation and function in mice (36,37). However, the roles of Pinch proteins in adipose tissue and metabolism are unclear.

The aim of this study was to investigate whether and how Pinch proteins regulate adipose mass and metabolism. We find that the expression of Pinch1 protein is positively correlated with obesity in humans and dramatically upregulated by a high-fat diet (HFD) and in ob/ob mice. Of particular significance, we find that, under HFD, but not normal chow diet (NCD), feeding conditions, mice with loss of Pinch1 in adipocytes and Pinch2 globally display reduced white adipose tissue (WAT) mass and ameliorated glucose intolerance and fatty liver. We demonstrate that Pinch loss favors fat and glucose metabolism by control of adipocyte apoptosis at a proper level.

Pinch1^f/f and Pinch2^−/− mice have been previously described (34). Adipoq-Cre transgenic mice harboring the Adipoq-Cre BAC transgene express Cre recombinase under control of the mouse adiponectin (Adipoq) promoter/enhancer regions within the BAC transgene (purchased from The Jackson Laboratory) (38). Caspase8^fl/fl mice were also used as indicated (39). All mice used in this study were crossed with normal C57BL/6 mice for at least 10 generations. For consistency and to minimize use of animals, male mice were used for experiments, except when indicated otherwise. Mice were group housed at 22 ± 2°C and exposed to a 12-h/12-h light-dark period. Male mice were fed with NCD (fat, 10 kcal%; protein, 20 kcal%; carbohydrates, 70 kcal% (D12450; Research Diets) or HFD (fat, 60 kcal%; protein, 20 kcal%; carbohydrates, 20 kcal%) (D12492; Research Diets) for 12 weeks, and body weight was recorded weekly. All animal experiments were approved by the Institutional Animal Care and Use Committee of Southern University of Science and Technology.

Insulin tolerance (ITT) and glucose tolerance tests (GTT) were performed as we previously described (14,40). In brief, to determine glucose tolerance, 16-h fasted mice were intraperitoneally administered glucose (2 g/kg of body weight). Blood glucose from tail vein blood was quantified at the indicated times after glucose administration. To determine insulin sensitivity, 6-h fasted mice were intraperitoneally administered insulin (0.75 units/kg of body weight). Blood glucose from tail vein blood was quantified at the indicated times.

Serum levels of insulin were assayed by ELISA kits (catalog number EZRMI-13K; Millipore). Nonesterified fatty acids (catalog number AQJ0023; FUJIFILM Wako Pure Chemical, Osaka, Japan) and triglyceride (catalog number AQE0070; FUJIFILM Wako Pure Chemical) concentrations were measured using kits coupling an enzymatic reaction according to the manufacturer’s instructions.

For the total lysates, tissues were homogenized in radioimmunoprecipitation assay buffer supplemented with a cocktail of protease inhibitors and phosphatase inhibitors (catalog number PPC2020; Sigma-Aldrich). Equal amounts of proteins (20–30 μg) were separated by SDS-PAGE and transferred to polyvinylidene fluoride membranes. The membrane was blocked in 5% nonfat milk in Tris-buffered saline/Tween 20 buffer; probed with primary antibodies, followed by incubation with secondary antibodies conjugated with horseradish peroxidase; and visualized using a Western Blotting Detection Kit (catalog number 1705060; Bio-Rad Laboratories). Antibodies used in this study are listed in Supplementary Table 1.

Pieces of epididymal adipose tissues from wild-type (WT) and knockout (KO) mice were fixed in 10% formalin for 24 h at room temperature. The tissues were then embedded into paraffin, blocked, and sectioned at 5 μm for hematoxylin-eosin (H-E) staining and immunohistochemical staining. For H-E staining, the sections were deparaffinized and rehydrated and the nuclei stained with hematoxylin for 15 min. Sections were then rinsed in running tap water and stained with eosin for 1 min, dehydrated, and mounted. Whole-slide digital images were collected at magnification 200× with a microscope scanner. For immunohistochemical staining, the sections were blocked with blocking buffer containing 5% goat serum, 2% BSA, 0.2% Triton X-100, and 0.1% sodium azide in 1× PBS for 1 h after deparaffinization and antigen retrieval. Samples were then incubated with the indicated primary antibodies diluted in blocking buffer overnight. After washing with PBS, the samples were incubated with secondary antibodies for 45 min at room temperature. Images for control and KO samples were captured.

Metabolic cage study was performed at the Brain Cognition and Brain Disease Institute, Shenzhen Institutes of Advanced Technology, using CLAMS (Columbus Instruments). Energy expenditure (EE) was calculated from Vo₂ and respiratory exchange ratio (RER) using the Lusk equation: EE in kcal/h = (3.815 + 1.232 × RER) × Vo₂ in mL/min (41).

WAT were dissected and cut into small pieces and digested with collagenase A (Roche) according to a previous report (42). For stromal vascular fraction (SVF) adipogenic differentiation, the cells were induced with induction medium (10% FBS DMEM supplemented with 5 μg/mL insulin, 2 μg/mL dexamethasone, 0.5 μmol/L rosiglitazone, 0.5 mmol/L 3-isobutyl-methylxanthine, and 1 nmol/L T3) for 3 d and switched to differentiation media (induction medium plus 5 μg/mL insulin) for another 7 d. Media were changed every 2 d.

Human adipose tissues were obtained from Chinese people who were undergoing surgery in the People’s Hospital of Guiyang (Guiyang, China). Written informed consent was obtained from all participants, and the experiments were conducted according to the principles outlined in the Declaration of Helsinki.

Total RNA was extracted from cells or tissues using Trizol reagent (catalog number 12183555; Invitrogen) according to the manufacturer’s instructions. RNA was treated with RNase-free DNase I to remove contaminating genomic DNA according to the manufacturer’s protocol. The purity and concentration of total RNA were measured by a spectrophotometer (Nanodrop 3000; Thermo Fisher Scientific) at 260 and 280 nm. Total RNA (2 μg) from each sample was reverse transcribed using the High-Capacity cDNA Reverse Transcription Kit (catalog number 4374967; Applied Biosystems) according to the manufacturer’s instructions. Real-time PCR analysis was carried out with a Bio-Rad Laboratories system using SYBR Green Master Mix and gene-specific primers. Primer sequences are listed in Supplementary Table 2. The relative changes in gene expression were normalized to Gapdh as internal control.

The sample size of this study was determined based on our previous experience in similar studies. Mice used in experiments of this study were randomly grouped. Statistical analyses were completed using Prism GraphPad. Two-tailed unpaired Student t test (two groups) and one-way ANOVA (multiple groups) were used. Values are presented as means ± SEM. Differences with P < 0.05 were considered statistically significant.

The data generated and analyzed in this project are included in the manuscript and its Supplementary Material. Additional data and resources are available from the corresponding author upon reasonable request.

As an initial effort to determine the potential role of Pinch in adipose tissue, we measured the expression levels of PINCH1 mRNA in adipose tissues from liposuction surgery in human participants with different BMI by quantitative RT-PCR (qRT-PCR) analysis. Results revealed that PINCH1 mRNA levels were markedly increased in obese participants and positively correlated with BMI (Fig. 1A). On the basis of this observation, we next determined the expression of Pinch1 in adipose tissues in HFD-induced obese mice. Results from qRT-PCR analysis showed that HFD feeding for 12 weeks dramatically increased the Pinch1 mRNA level in both epididymal WAT (eWAT) and subcutaneous WAT (subWAT) (Fig. 1B and C). Results from Western blot analysis showed dramatic upregulation of Pinch1 and Pinch2 protein expression in eWAT from HFD-fed obese mice compared with that from NCD-fed lean mice (Fig. 1D and E). Furthermore, we found that the expression levels of Pinch1 mRNA and protein were dramatically upregulated in WAT of ob/ob mice relative to that of control lean mice (Fig. 1B, C, and F). Consistent with results from qRT-PCR and Western blot analyses, in situ immunofluorescence staining revealed a dramatic increase in expression of Pinch1 protein in adipocytes in eWAT from HFD-fed obese mice (Fig. 1G and H) and ob/ob mice (Fig. 1G and H). Finally, expression of Pinch1 protein was increased during 3T3-L1 cell differentiation (Fig. 1I).

On the basis of the above observations, we next generated adipocyte-specific Pinch1-deficient mice by crossing mice carrying floxed alleles of Pinch1 with the Adipoq-Cre transgenic mice (Adipoq-Cre; Pinch1^fl/fl). Compared with their Cre-negative littermates (Pinch1^fl/fl), Adipoq-Cre; Pinch1^fl/fl mutant mice did not display any marked alterations in body weight, adipose mass, fat or liver histology under either NCD or HFD conditions (Supplementary Fig. 1). Furthermore, when compared with control mice, Pinch2 KO mice fed on NCD or HFD for 12 weeks did not display any marked alterations in body weight, fat (eWAT, subWAT, and brown adipose tissue [BAT]) mass, fat, or liver histology by H-E staining analysis (Supplementary Fig. 2). We wondered whether there is a functional redundancy of both factors in adipose tissue. To determine if this is the case, we deleted Pinch1 expression in adipocytes and Pinch2 globally using our routine breeding strategy with Adipoq-Cre; Pinch1^fl/fl mice and Pinch2 global KO mice (Pinch2^−/−) mice. We thus generated Adipoq-Cre; Pinch1^f/f; Pinch2^−/− mice (referred to as KO hereafter) and other genotypes for this study (Supplementary Fig. 3A). As expected, Pinch1 mRNA levels were significantly reduced in eWAT, subWAT, and BAT, but not in the liver, kidney, spleen, or heart (Fig. 2A). Western blot analysis revealed dramatic reduction in expression of Pinch1 protein in subWAT, eWAT, and BAT, but not in the liver or kidney, in KO relative to WT mice (Fig. 2B). Furthermore, Pinch1 expression was deleted in mature adipocytes, but not in adipose SV cells, from KO mice (Supplementary Fig. 3B and C). Collectively, these results validate the specific deletion of Pinch1 in adipocytes in KO mice.

We next investigated the effects of Pinch deletion on adipose tissue in mice with and without obesity. Six-week-old control and KO mice were fed with NCD or HFD for 12 weeks. Results showed that Pinch deletion significantly reduced the body weight in HFD-fed, but not NCD-fed, mice (Fig. 2C and D and Supplementary Fig. 4A and B). Furthermore, Pinch deletion prevented HFD-induced increases in both eWAT and subWAT mass (Fig. 2E and F and Supplementary Fig. 4C–E). It should be noted that HFD slightly increased BAT mass, which was not significantly decreased by Pinch deletion (Fig. 2F). Furthermore, there were no significant changes in food intake or water intake between the two genotypes with and without HFD treatment (Fig. 2G and H).

A decrease in adipose tissue mass can be attributed to decreases in adipocyte size, adipocyte number, or both. Histological analyses showed that, under NCD feeding conditions, Pinch deletion did not markedly alter adipocyte size in WAT or BAT between the two genotypes in either males and females (Fig. 2I and Supplementary Fig. 5A). However, when challenged with HFD, KO mice displayed decreased adipocyte size and altered adipocyte size distribution primarily in eWAT, but not subWAT or BAT (Fig. 2J–O and Supplementary Fig. 5B).

Obesity is closely related to glucose intolerance and insulin resistance. We next assessed the effects of Pinch deletion on glucose homeostasis and insulin sensitivity. There were no statistically significant differences in glucose tolerance, as determined by GTT, or insulin sensitivity, determined by ITT, between the two genotypes in either males or females fed on NCD (Fig. 3A and B and Supplementary Fig. 6A and B). Then, we assessed the effects of Pinch deletion under HFD challenge. The results from GTT experiments showed that Pinch deletion significantly ameliorated glucose intolerance in both males and females (Fig. 3C and E). Furthermore, Pinch deletion slightly increased insulin sensitivity in male, but not in female, mice, as revealed by ITT experiments (Fig. 3D and F). Furthermore, when compared with HFD-fed control mice, HFD-fed KO mice showed significant decreases in levels of fasting blood glucose (Fig. 3G) and serum insulin (Fig. 3H). Furthermore, Pinch deletion reduced the levels of serum adiponectin and leptin under HFD feeding conditions (Fig. 3I and J).

We further performed metabolic cage experiments to determine energy metabolism in the two genotypes of mice after HFD challenge. The results showed that the oxygen consumption rate and carbon dioxide production rate were slightly but significantly increased in KO mice compared with control mice (Fig. 3K and L). The RER and EE were increased in KO versus control mice (Fig. 3M and N).

Given that hepatic steatosis is closely associated with obesity and insulin resistance, we next determined the effect of adipocyte Pinch deletion on hepatic lipid deposition under HFD conditions. As expected, HFD promoted marked lipid accumulation in the liver (Fig. 4A and B). Pinch deletion in adipocytes greatly decreased lipid accumulation induced by HFD (Fig. 4A and B). Pinch deletion reduced the mass of liver, but not that of the spleen, heart, or kidney, in HFD-fed mice (Fig. 4C). Pinch deletion decreased the levels of liver triglyceride and serum total cholesterol without affecting the level of serum triglyceride (Fig. 4D–F). The mRNA levels of fatty acid synthesis genes, including fatty acid synthase (Fas), acetyl-CoA carboxylase (Acc), sterol regulatory element-binding protein-1c (Srebp-1c), and CD36, were decreased in the livers of KO mice compared with those of control mice (Fig. 4G). In contrast, expression of the lipolysis genes, including adipose triglyceride lipase (Atgl) and hormone-sensitive lipase (Hsl), was increased in the livers of HFD-fed KO mice compared with HFD-fed control mice (Fig. 4G). Hepatic steatosis is often accompanied by macrophage infiltration, which facilitates inflammation and insulin resistance. The mRNA levels of inflammatory genes, including Cd68, Mcp-1, and Il-1β, were downregulated in the livers of HFD-fed KO mice relative to those of HFD-fed control mice (Fig. 4H). Collectively, these results indicate loss of Pinch in adipocytes ameliorates HFD-induced hepatic steatosis in mice.

To determine whether loss of adipose tissue in KO mice was due to enhanced adipocyte cell apoptosis, we examined the expression of the adipocyte marker protein Perilipin 1. Under NCD conditions, WAT adipocytes from control and KO mice showed well-defined Perilipin staining. In contrast, eWAT adipocytes from HFD-fed KO mice displayed a number of Perilipin-negative cells (Fig. 5A–C). Further analysis of WAT with TUNEL staining revealed an increased number of apoptotic cells in WAT of HFD-fed KO mice, especially in eWAT (Fig. 5D–G). Western blot analyses showed that Pinch deletion increased expression of several proapoptotic proteins, including activated Caspase-3, activated Caspase-8, receptor-interacting protein kinase 3 (Rip3) (a necroptosis marker), and cleaved poly(ADP-ribose) polymerase (Parp) in eWAT of HFD-fed mice (Fig. 5H and I). The expression levels of apoptosis-related proteins, including Bax, Bcl-2, and cytochrome c (Cc), were increased in eWAT of HFD-fed KO mice relative to that of HFD-fed control mice (Fig. 5H and I). Results from Ki67 staining revealed that the adipocyte proliferation rate was increased in eWAT in HFD-fed KO versus HFD-fed control mice (Supplementary Fig. 7).

We next sought to investigate the mechanism by which Pinch regulates adipocyte apoptosis. We determined the expression levels of several key pro- and antiapoptotic genes by qRT-PCR analysis using RNAs isolated from eWAT of WT and KO mice fed on HFD for 12 weeks. The results showed that expression of the antiapoptotic genes Bcl2 and Bcl2l2 was downregulated, whereas that of the proapoptotic genes Bim and Bax was upregulated in KO versus WT WAT (Fig. 6A). The upregulation of Bim expression in KO WAT was confirmed at the protein level by Western blotting (Fig. 6B and C). The upregulation of Bim, activated Caspase-8, and Rip3 proteins was observed in primary adipocytes from KO eWAT (Fig. 6D and E). Furthermore, we differentiated adipocytes from SVF of WT and KO mice and observed increased expression of Bim, cleaved Caspase-8, and Rip3 proteins in cultures from KO mice compared with those from WT mice (Fig. 6F and G). qRT-PCR analysis showed upregulation of Bim mRNA expression in differentiated adipocytes from KO versus WT SVF cultures (Fig. 6H). Notably, KO SVF cultures displayed a dramatic increase in cell apoptosis (Fig. 6I and J). We next used lentivirus to knock down expression of Bim through shRNA in differentiated adipocytes and found that shRNA knockdown of Bim expression blocked adipocyte apoptosis caused by Pinch deficiency (Fig. 6K and L). Bim knockdown reduced the expression levels of activated Caspase-8 and Rip3 proteins caused by Pinch deficiency (Fig. 6M and N). Of note, shRNA knockdown of Bim expression did not alter the level of Bax protein in adipocytes (Supplementary Fig. 8). Collectively, these results suggest that Pinch loss promotes adipocyte apoptosis probably by activating the Bim/Caspase pathway.

We finally tested our hypothesis that enhanced adipocyte apoptosis plays a major role in promoting the alterations observed in KO mice. We deleted expression of Caspase-8 in adipocytes (Supplementary Fig. 9) and determined its effects on HFD-induced obesity, glucose intolerance, and fatty liver in KO mice. We found that, under HFD feeding conditions, the decreases in both eWAT and subWAT mass caused by Pinch deletion were reversed by ablation of Caspase-8 in adipocytes (Fig. 7A–C). Furthermore, results from GTT and ITT experiments revealed that Caspase-8 ablation abolished the improving effects of Pinch deficiency on glucose intolerance and insulin sensitivity (Fig. 7D and E). The abnormal morphology of eWAT adipocytes in KO mice was largely reversed by Caspase-8 ablation, as revealed by H-E staining (Fig. 7F and G). Results from TUNEL staining showed that Caspase-8 ablation inhibited the Pinch deletion–induced adipocyte apoptosis in KO mice (Fig. 7H and I). Immunofluorescence staining showed that Caspase-8 ablation reversed the decrease in Perilipin 1–expressing cells in eWAT of KO mice (Fig. 7J and K). Finally, the ameliorated fatty liver caused by Pinch loss was abolished by deletion of Caspase-8 (Fig. 7L).

In this study, we for the first time to our knowledge demonstrate that Pinch1/2 deficiency in adipocytes dramatically reduces WAT mass, favors glucose intolerance, and ameliorates fatty liver in obese mice. Specifically, we demonstrate that, under HFD feeding conditions, deletion of Pinch in adipocytes in mice largely ameliorates metabolic abnormalities by decreasing body weight, WAT mass, and fasting blood glucose level and improving glucose intolerance and fat accumulation in the liver. Interestingly, under NCD feeding conditions, deletion of Pinch does not markedly affect the above parameters regarding fat or glucose metabolism. We demonstrate that the expression level of Pinch1 protein is drastically upregulated in adipose tissues by HFD feeding and in ob/ob mice, suggesting a role of Pinch upregulation in the pathogenesis of obesity. In support of this notion, we observe a strong positive correlation between the level of Pinch1 expression and obesity in humans. Collectively, these important findings highlight a requirement to investigate the potential involvement of alteration in Pinch expression in adipose tissue in the pathogenesis of obesity and type 2 diabetes in humans. Our results also suggest that targeting Pinch expression in adipocytes may be a useful therapy for these diseases.

Mechanistically, in the current study, we demonstrate that loss of Pinch reduces fat mass and favors glucose and liver fat metabolism by promoting adipocyte apoptosis. This notion is supported by several lines of evidence from this study and those by others. Pinch loss greatly accelerates adipocyte cell apoptosis, primarily in eWAT and to a less extent in subWAT. Our in vitro studies reveal that Pinch loss promotes adipocyte apoptosis by activating the Bim/Caspase-8 pathway. Most importantly, we demonstrate that deleting expression of Caspase-8 in adipocytes reverses the ameliorating effects of Pinch loss on obesity, glucose intolerance, and fatty liver in HFD-induced obese mice. In support of our findings, Montanez et al. (43) showed that disruption of the Pinch1 gene led to sustained activity of Jnk, a proapoptotic factor. Furthermore, Gkretsi et al. (44) reported that loss of Ilk, a Pinch-interacting protein, in hepatocytes accelerated apoptosis in mice. Liang et al. (45) reported that disruption of Pinch1 in cardiac tissue promoted cell apoptosis. Results from this study suggest that maintaining adipocyte apoptosis at a proper level may play an important role in prevention of obesity and favors glucose and liver fat metabolism.

We find that Pinch loss slightly but significantly increases EE in the mutant mice under HFD feeding conditions. While the underlying mechanism remains to be determined, this result suggests that accelerated EE may contribute, at least in part, to the ameliorating effect of Pinch loss on metabolism in the mutant mice.

Cumulative evidence suggests that genomic sequences that encode potential miRNAs, small open reading frames that express micropeptides or other small genomic elements, have important physiological functions (46,47). Thus, one limitation of this study is that we cannot completely exclude the possibility that the phenotypes observed in the mutant mice are in part caused by deletion of these potential genomic sequences in addition to those encoding the Pinch proteins. Future study will explore if this is the case.

It should be noted that Pinch loss decreases adipocyte size and alters cell size distribution mainly in eWAT, which should in part contribute to the low fat mass phenotype in the mutant mice.

It is intriguing that deficiency of adipocyte Pinch1 promotes HFD-induced adipocyte death primarily in eWAT (also known as visceral adipose tissue), with much less adipocyte apoptosis observed in subWAT. It is plausible that visceral and subcutaneous adipocytes have different developmental origins (48) and that visceral adipocytes are more susceptible to HFD-induced cell death than subcutaneous adipocytes. Consistent with this notion, the adipocyte death rate was much higher in eWAT than in subWAT after HFD feeding (49). Future studies will determine the underlying mechanisms that account for the different effects of Pinch1 ablation on visceral versus subcutaneous adipocytes.

In the current study, we find that Pinch1 deletion in adipocytes and Pinch2 global KO greatly ameliorate HFD-induced obesity, glucose intolerance, and fatty liver, while loss of either Pinch1 or Pinch2 does not generate these improvements. Interestingly, a similar functional redundancy of Pinch1/2 was observed in their regulation of heart formation and skeletal homeostasis (34,37). The underlying mechanism requires further investigation.

In summary, we demonstrate that adipocyte Pinch plays an important role in regulating adipocyte apoptosis and glucose and liver fat metabolism. Pinch ablation results in adipocyte apoptosis and ameliorates metabolic dysfunction under HFD challenge. Thus, we demonstrate a critical novel role of Pinch1/2 in adipocytes in regulation of adipose mass, glucose, and liver fat metabolism.

Acknowledgments. The authors thank Dr. Tingting Geng at National University of Singapore for critical reading of this manuscript. The authors acknowledge the assistance of the Core Research Facilities of Southern University of Science and Technology (SUSTech).

Funding. This work was supported in part by National Key Research and Development Program of China grants 2019YFA0906004 and 2019YFA0906001; National Natural Science Foundation of China grants 81991513, 82022047, 81630066, 81870532, and 81972100; Guangdong Provincial Science and Technology Innovation Council grant 2017B030301018; and Shenzhen Municipal Science and Technology Innovation Council grants JCYJ20180302174117738, JCYJ20180302174246105, and KQJSCX20180319114434843.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. H.G. and G.X. designed the study. H.G., Y.Z., Z.D., S.L., X.H., W.T., X. Zhou, X. Zou, J.S., F.Y., X.B., C.L., H.C., and G.X. conducted the study and performed data collection, analysis, and interpretation. H.G. and G.X. drafted the manuscript. H.C. and G.X. are the guarantors of this work and, as such, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.