Deubiquitinating Enzyme USP2 Alleviates Muscle Atrophy by Stabilizing PPAR-γ Free

As life expectancy increases, sarcopenia, characterized by the progressive loss of skeletal muscle mass, strength, and function, has emerged as an important public health concern, affecting 10–27% of adults and increasing the risks of falls, fractures, frailty, morbidity, and mortality among aging populations worldwide (1). Our previous study demonstrated that sarcopenia increases mortality risk among middle-aged and older adults (2). Additionally, our Mendelian randomization study of European populations revealed a causal association between sarcopenia and any death (2).

Insulin resistance is a core feature of metabolic disorders such as type 2 diabetes and plays a crucial role in sarcopenia and sarcopenic obesity (3–5). Insulin resistance compromises muscle insulin sensitivity, impedes glucose utilization and storage, accelerates muscle protein breakdown, fosters chronic inflammation, disrupts metabolic homeostasis, and impairs mitochondrial function in adipose and muscle tissues, ultimately leading to skeletal muscle atrophy (3–5). Our previous study revealed the crucial role of miR-193b in diabetic muscle atrophy (5). Deficiency of miR-193b in skeletal muscle improved insulin sensitivity in diabetic mice and alleviated muscle loss and weakness by activating the PDK1/AKT/mTOR/S6K signaling pathway. Therefore, improving insulin sensitivity is an important strategy for treating sarcopenia, particularly diabetic sarcopenia and sarcopenic obesity.

Sarcopenia primarily involves protein degradation, and the ubiquitin-specific protease (USP) pathway is the key mechanism involved in various types of sarcopenia and muscle wasting (6–8). USP family members regulate the degradation rates of critical proteins, such as the E3 ubiquitin ligases MAFbx/atrogin-1 and MuRF1 (the latter also is known as TRIM63), which are atrogenes known to induce muscle atrophy (9–12). The modulatory roles of USPs in insulin resistance are broad and diverse. In metabolic diseases such as diabetes mellitus (DM) and its complications, USPs such as USP22, USP2, USP9X, USP20, and USP33 modulate insulin signaling in various tissues through deubiquitination, affecting β-cell function, adipose tissue inflammation, and skeletal muscle metabolism (13). For instance, USP21 deficiency promotes mitochondrial activation and enhances obesity and insulin sensitivity in the skeletal muscle (13,14). Moreover, USP9X is a key determinant of the antidiabetic effect of calorie restriction by stabilizing the AMPKα2 subunit in the skeletal muscle (13,15).

USP2 is recognized as an important contributor to insulin resistance and regulates hepatic glucose tolerance, insulin sensitivity, glucocorticoid signaling, and sympathetic nervous system activation in the ventromedial hypothalamus (16,17). Despite increasing interest in USP2, its precise role in sarcopenia and its underlying mechanisms in muscle atrophy and insulin resistance within skeletal muscle tissues remain poorly understood. In this study, we established DM- and dexamethasone (DEX)-induced muscle atrophy models in wild-type (WT) and USP2-deficient mice to investigate whether USP2 is involved in sarcopenia and to analyze the underlying mechanisms. By unraveling the molecular networks involving USP2, we aimed to enhance our understanding of the pathophysiology of sarcopenia and identify potential intervention strategies.

The detailed methods are available in the Supplementary Material.

Usp2 global knockout (KO) mice (C57BL/6JGpt-Usp2^{em7Cd1220d392in5}/Gpt; strain no. T052377) were purchased from Gempharmatech Co. Ltd (Jiangsu, Nanjing, China). Usp2^flox/flox mice (C57BL/6JCya-Usp2em1flox/Cya; strain no. S-CKO-11600) were purchased from Cyagen Biosciences (Guangzhou, Guangdong, China). The sample size for animal studies was determined based on a survey of data from published research or preliminary studies (18), and no mice were excluded from the statistical analysis. These mice were housed in specific-pathogen-free units at the Animal Center of Shenzhen People’s Hospital. They were maintained under a 12-h light cycle from 8 a.m. to 8 p.m., at a temperature of 23 ± 1°C and humidity of 60–70%. The mice were provided a standard rodent diet and had free access to water in plastic bottles. Prior to the experiments, the mice were allowed to acclimate to their housing environment for a minimum of 7 days. Up to five mice were housed per plastic cage, which contained corn cob bedding material. The mice were treated in a blinded manner, and randomization was performed before administering the treatments. At the end of the experiment, all mice were anesthetized and euthanized in a CO₂ chamber. Blood or muscle samples were then collected.

DEX-induced muscle atrophy models were developed following previously described procedures (19). Briefly, 8-week-old, male, WT, and USP2KO mice were randomly assigned to three groups (n = 6 in each group). The control (Ctrl) group received a polyethylene glycol (PEG) solution (10 mL/kg body weight, 30% in 0.9% saline), whereas the DEX group was administered DEX (25 mg/kg dissolved in PEG 400 solution) intervention via intraperitoneal injection for 10 days. Mouse body weight was recorded daily, and grip strength was measured three times using a grip strength test meter. Twenty-four hours after the last intraperitoneal injection, the mice were euthanized. The gastrocnemius (Gas), and tibialis anterior (TA) muscles were isolated for further analysis.

Male C57BL/6 J mice (aged 8 ± 0.5 weeks old; weight: 24 ± 1 g) were randomly split into two groups (n = 6/group): a Ctrl group and a high-fat diet plus streptozotocin (HFD/STZ) group. The HFD/STZ mice were fed an HFD (Research Diets, catalog no. MD12033) for 4 weeks, fasted for 12 h, then given 50 mg/kg STZ (Sigma-Aldrich, catalog no. S0130; dissolved in 50 mmol/L citric acid buffer, pH 4.5) intraperitoneally for 5 days. After STZ administration, the mice were maintained on the HFD for 12 weeks. Ctrl mice got the same volume of citric acid buffer and a normal diet (10% fat calories). Fasting blood glucose (FBG) concentration was measured 9 days after the first STZ injection using a glucometer (ACCU-CHEK). The model was deemed successful if two consecutive readings exceeded 16.7 mmol/L.

To overexpress USP2, both hind limbs of mice were injected in situ with Gas using adeno-associated virus 9 (AAV9)-Ctrl or AAV9-Usp2 (0.5–1.5 × 10¹¹ vg/mL in 50 μL of saline; GeneChem, Shanghai) in relevant groups. Mice were then grouped as Ctrl, Ctrl + USP2 overexpression, HFD/STZ, and HFD/STZ + USP2 overexpression (n = 6/group) for experiments.

All data were generated from at least three independent experiments. Each value was presented as the mean ± SD. All raw data were initially subjected to a normal distribution and analysis by one-sample Kolmogorov-Smirnov nonparametric test using SPSS 22.0 software. For animal and cellular experiments, a two-tailed unpaired Student t test was performed to compare the two groups. One-way ANOVA, followed by the Bonferroni post hoc test, was used to compare more than two groups. To avoid bias, all statistical analyses were performed blindly. In figures, statistical significance is indicated at P < 0.05, P < 0.01, and P < 0.001.

All data and resources relevant to this study are available upon request to the corresponding author. The corresponding author will assist with providing access to any materials, data sets, or additional information necessary to reproduce or further investigate the research findings.

Reanalysis of publicly available transcriptomic data (Gene Expression Omnibus series GSE156249) revealed that USP2 expression was downregulated in the skeletal muscle of patients with diabetes compared with that in Ctrl group of healthy study participants (Supplementary Fig. 1A). We investigated the changes in endogenous USP2 expression in diabetes-induced muscle atrophy mouse models and found that the Gas and TA muscles of diabetic mice (STZ + HFD; spontaneous diabetes mouse model) had lower USP2 levels than their respective Ctrl groups (Supplementary Fig. 1B and C). We conducted Western blotting and quantitative PCR (qPCR) assays on the Gas and TA muscles of diabetic mice and found that the expression levels of USP2 were significantly lower than those in healthy Ctrl mice (Supplementary Fig. 1D). Taken together, these results suggest that USP2 is involved in the pathogenesis of muscle atrophy.

We first generated USP2 KO (USP2KO) mice (Fig. 1A) and confirmed that they had a body weight similar to that of WT mice (Fig. 1B), with a similar heart weight to tibia length ratio and daily food intake (data not shown). Furthermore, there were no evident changes in the protein levels of MURF-1 and myosin heavy chain (MYHC) in the Gas of USP2KO mice compared with those in WT mice (Fig. 1A). Similarly, the Gas, TA, and soleus muscle weights; grip strength; and exhaustive running distance and time were not significantly different between USP2KO and WT mice (Fig. 1C–E). Basic histological evaluation of the glycolytic Gas muscles revealed that both USP2KO and WT mice had a normal muscle structure and comparable myofiber size distribution profile (Fig. 1F).

After establishing a negative correlation between USP2 expression and muscle atrophy in vivo, we used a diabetes-induced in vivo model to investigate whether USP2KO aggravates muscle atrophy. USP2 levels in the Gas were lower in diabetic WT mice than in Ctrl WT mice (Fig. 1A). In addition, diabetes was associated with significantly reduced MYHC levels, the ratio of Gas (TA or soleus) muscle mass to body weight, grip strength, and exhaustive running distance and time; in contrast, MURF-1 levels were increased in WT mice (Fig. 1A–E). Moreover, diabetic WT mice had more small myofibers than did Ctrl WT mice (Fig. 1F). Notably, the USP2KO augment exacerbated these effects of diabetes on the Gas muscle, as evidenced by a lower level of MYHC protein, the ratio of Gas (TA or soleus) muscle mass to body weight, grip strength, exhaustive running distance and time, a higher level of MURF-1 protein, and more small myofibers in USP2KO diabetic mice compared with WT diabetic mice (Fig. 1B–E). In addition, USP2 deficiency worsened insulin resistance (assessed by the insulin tolerance test and glucose tolerance test), while increasing the levels of FBG and fasting blood insulin (FBI) in diabetic mice (Fig. 1G). Similar results were also observed for mice with DEX-induced skeletal muscle atrophy (Supplementary Fig. 2A–F).

To further investigate the role of USP2 in skeletal muscles, we generated skeletal muscle–specific USP2-deletion mice by crossing Usp2^flox/flox mice with Hsa-Cre mice. Importantly, the phenotypes of skeletal muscle–specific USP2-deletion mice were consistent with those of the global USP2KO mice (Fig. 2A–H). Furthermore, we found that USP2 was relatively highly expressed in skeletal muscles of mice, whereas its expression was barely detectable in white adipose tissue (WAT) (Fig. 2I). Taken together, these findings suggest the loss of USP2 promotes muscle loss and weakness in diabetic or DEX-induced mice.

Compared with the Ctrl group, Usp2 expression was significantly increased in the Gas muscles after injection of AAV9 vectors expressing USP2 transfection, but not in kidney, liver, lung, spleen, and WAT (Fig. 3A and Supplementary Fig. 3). Usp2 overexpression was associated with slightly increased body weight of diabetic mice and significantly decreased levels of FBG and FBI (Fig. 3B and C). Consistently, the MYHC protein level and the ratio of the Gas muscle mass to body weight were significantly increased, whereas MURF-1 protein levels in the Gas muscles significantly decreased in diabetic mice after USP2 overexpression (Fig. 3A and D). Consistently, Usp2 overexpression resulted in enhanced skeletal muscle performance and endurance in diabetic mice (as assessed by exhaustive running distance and time) (Fig. 3E). Finally, diabetic mice had a normal muscle structure but fewer small myofibers in the Gas muscles after Usp2 overexpression (Fig. 3F). Collectively, Usp2 overexpression in the skeletal muscle attenuated muscle loss in diabetic mice.

Next, RNA sequencing was performed on the Gas muscles of WT and USP2KO mice. We identified 269 differentially expressed genes (DEGs) in the Gas of WT and USP2KO mice (Fig. 4A). Among these DEGs, 66 were significantly upregulated and 203 were significantly downregulated (P < 0.05; Supplementary Table 1). An ingenuity pathway analysis revealed that the insulin signaling pathway, insulin resistance pathway, peroxisome proliferator–activated receptor (PPAR) signaling pathway, and type 2 DM pathways were the top pathways affected by Usp2 KO (Supplementary Table 2).

The 269 DEGs identified between the WT and USP2KO mice were input into ChIP-X Enrichment Analysis, version 3 (ChEA3), a web-based transcription factor (TF) enrichment analysis tool that includes six primary reference gene set libraries from multiple sources (20). The score of PPAR-γ ranked second among all TFs predicted by ChEA3, based on the input genes (Supplementary Table 3). PPAR-γ is closely associated with insulin resistance in skeletal muscle and plays a key role in the glucose uptake pathway in skeletal muscle through modulation of glucose transporter 4 (GLUT4) and insulin receptor substrate 1 expression (21). Therefore, we hypothesized that Usp2 depletion in the diabetes- and DEX-induced muscle atrophy models might destabilize PPAR-γ protein. Destabilization of PPAR-γ, in turn, reduces its downstream transcriptional targets that contribute to insulin resistance and muscle atrophy.

To further investigate whether USP2 directly interacts with PPAR-γ, we performed endogenous and exogenous co-immunoprecipitation (Co-IP) analysis. USP2 interacted with PPAR-γ in C2C12 cells and the Gas muscle (Fig. 4B). WT USP2 overexpression induced an increase in the protein level of PPAR-γ and decreased PPAR-γ ubiquitination, but not USP2 C276A (an enzyme-activity dead mutation) overexpression (Fig. 4C). According to the GPS database (https://cplm.biocuckoo.cn/), PPAR-γ has five known ubiquitination sites that include lysine (K): K184, K185, K268, K293, and K462. Hence, we mutated all five Ks to arginine (R), which mimicked the deubiquitination of the protein. K184R and the K185 mutant had an increased PPAR-γ protein level compared with PPAR-γ WT, PPAR-γ K268R, PPAR-γ K293R, and PPAR K462R in USP2 knockdown (Fig. 4D). DKR (both K184R and K185R) had a lower ubiquitination level of PPAR-γ compared with WT PPAR-γ, which was not further affected by USP2 knockdown (Fig. 4E). In addition, molecular docking simulations showed both PPAR-γ K184 and K185 at the contact surface between PPAR-γ and USP2 (Fig. 4F). Our results strongly support the notion that USP2 is an important deubiquitinating enzyme for PPAR-γ.

In mice with diabetes- or DEX-induced muscle atrophy, Usp2 KO downregulated PPAR-γ protein expression in the Gas muscles (Fig. 5A and B). Usp2 overexpression increased PPAR-γ protein levels in the Gas muscles of diabetic mice (Fig. 5C). We performed siRNA-mediated targeted Usp2 (siUsp2) knockdown and selected the optimal TNF-α dose (50 μmol/L) to mimic muscle atrophy in C2C12 myotubes to further explore if Usp2 inhibition could aggravate the TNF-α–induced muscle atrophy phenotype (Fig. 5D).

Based on protein expression, we confirmed Usp2 knockdown (Fig. 5D). Expression of PPAR-γ, GLUT4, and IRS1 was downregulated in the TNF-α–treated cells, which was further enhanced by siUsp2 transfection (Fig. 5D). Furthermore, the mRNA levels of ubiquitin ligases (e.g., atrogin-1, muscle ubiquitin ligase of SCF complex in atrophy-1 [MUSA1], and F-box protein 31), which induce muscle atrophy (12,22), were increased in the TNF-α–treated cells, and Usp2 KO augmented these effects (Fig. 5D). Immunofluorescence staining demonstrated that Usp2 knockdown amplified the TNF-α–triggered decrease in MYHC staining, cell fusion, and multinucleated myotube formation (Fig. 5E). We also observed that adenovirus-mediated Usp2 (AdUsp2) overexpression increased PPAR-γ, GLUT4, and IRS1 expression under TNF-α stimulation (Fig. 5F). Usp2 overexpression also suppressed the TNF-α–induced expression of atrogenes (Fig. 5F).

Next, to investigate the role of USP2 in the development of muscle atrophy, we created an in vitro DEX-induced model by challenging cells with synthetic DEX. Usp2 knockdown reduced the expression of PPAR-γ, GLUT4, and IRS1, and increased atrogene expression under DEX stimulation (Supplementary Fig. 4A). In contrast, Usp2 overexpression induced the expression of PPAR-γ, GLUT4, and IRS1, and reduced the expression of atrogenes under DEX stimulation (Supplementary Fig. 4B). Collectively, these results suggest USP2 prevents TNF-α– and DEX-induced myotube atrophy.

To directly demonstrate that PPAR-γ is necessary for USP2 to regulate muscle atrophy and insulin signaling, we infected C2C12 cells with AdUsp2 and/or siPPAR-γ. Knockdown of PPAR-γ reduced and Usp2 overexpression increased the levels of PPAR-γ, IRS1, and GLUT4 in C2C12 cells (Fig. 6A). Forkhead box subgroup O (FoxO) signaling is central to muscle atrophy (23). Consistently, we found that PPAR-γ knockdown increased and Usp2 overexpression decreased the expression of FoxO signaling pathway-related genes, including cyclin G2 (Ccng2), cyclin-dependent kinase inhibitor 1B (Cdkn1b), retinoblastoma-like protein 2 (Rbl2), and Bcl-2 interacting protein 3 (Bnip3) (Fig. 6B). In line with our hypothesis, PPAR-γ knockdown alone showed results similar to the siPPAR-γ + AdUsp2 transfected cells: IRS1 and GLUT4 protein levels decreased after knockdown of PPAR-γ even with Usp2 overexpression (Fig. 6A). Consistent with these results, Usp2 overexpression did not inhibit the PPAR-γ deficiency–mediated upregulation of FoxO target genes and atrogenes (Fig. 6B). Immunofluorescence staining demonstrated that PPAR-γ knockdown blocked the Usp2 overexpression–triggered increase in MYHC staining and cell fusion (Fig. 6C).

In line with the outcomes observed in C2C12 cells treated with TNF-α, Usp2 overexpression in these DEX-treated cells typically led to the upregulation of IRS1 and GLUT4, and the downregulation of FoxO target genes and atrogenes expression (Fig. 6D and E). However, when PPAR-γ was deficient, this Usp2 overexpression-mediated regulatory cascade was disrupted (Fig. 6D and E). PPAR-γ plays an important role in regulating triglyceride and glycogen metabolism in skeletal muscle (24,25). In diabetic mice, we found that the triglyceride content in skeletal muscle was increased, whereas the glycogen content was decreased (Supplementary Fig. 5A and B). Moreover, USP2KO exacerbated these changes (Supplementary Fig. 5A). In contrast, USP2 overexpression inhibited these changes (Supplementary Fig. 5B). Above all, these data demonstrate that the insulin signaling and muscle atrophy regulated by USP2 require PPAR-γ in skeletal muscle cells.

Next, we sought to determine whether USP2 affects insulin sensitivity and muscle atrophy in diabetic mice, by modulating PPAR-γ stability. We administered AAV9 Usp2 to WT diabetic mice to express the USP2 protein via in situ Gas muscle injection. The mice were treated with AAV expressing shRNA for PPAR-γ (AAV-shPPAR-γ) via in situ injection into the Gas muscle. PPAR-γ protein was significantly reduced in the Gas muscle of mice after AAV-shPPAR-γ transfection (Fig. 7A). Compared with the Ctrl group, PPAR-γ–deficient diabetic mice had a reduced ratio of TA and Gas muscle weight to body weight, as well as decreased exhaustive running distance and time, and insulin sensitivity (assessed by FBG and FBI), and had a smaller cross-sectional diameter of the muscle fibers (Fig. 7A–E). Importantly, these changes induced by PPAR-γ deficiency were not affected after AAV-Usp2 transfection (Fig. 7A–E). Taken together, these in vivo findings suggest PPAR-γ is essential for USP2 in regulating insulin sensitivity and muscle atrophy.

By performing a reanalysis of the database obtained from The Signaling Pathways Project (a web server for functional enrichment analysis of TF ChIP-seq peaks) (26), we identified that myogenic differentiation 1 (MYOD1) binds to the USP2 promoter in skeletal muscles (Fig. 8A). MYOD1 plays a critical role in muscle differentiation, development, and regeneration (27,28). Studies have consistently reported downregulation of MYOD1 expression during skeletal muscle atrophy, leading to decreased muscle function (29,30). In line with this, MYOD1 protein levels were reduced in the skeletal muscles of diabetic mice and in mice with DEX-induced diabetes (Fig. 8B). Hence, we first determined whether MYOD1 regulates USP2 expression in C2C12 cells and found that Myod1 overexpression increased, whereas Myod1 knockdown inhibited Usp2 protein levels in C2C12 cells (Fig. 8C). Thus, we performed assays with several luciferase (Luc) reporter constructs containing the human USP2 promoter and found that MYOD1 overexpression enhanced the transcription of –3,002 Luc and –2,190 Luc, but not –1,522 Luc or –501 Luc, in HEK293T cells (Fig. 8D). Therefore, the MYOD1 binding site within the USP2 promoter was between 3,002 and 1,522 bp.

Two putative MYOD1 binding motif sequences (MBE) were predicted in the USP2 promoter region (JASPAR analysis), and ChIP-qPCR assays further demonstrated the binding of MYOD1 to the promoter of the USP2 (Fig. 8E). Furthermore, TNF-α and DEX treatment inhibited the binding of MYOD1 to the promoter of USP2 in the C2C12 cells (Fig. 8F). Taken together, these findings demonstrate that MYOD1 transcriptionally activates USP2 expression by binding to its promoter region.

In this study, we demonstrated that USP2 plays a crucial role in the pathogenesis of muscle atrophy. USP2 is downregulated in the Gas and TA muscles of diabetic and DEX-treated mice. Mechanistic studies demonstrated that USP2 protects against muscle atrophy by increasing PPAR-γ signaling activity through deubiquitinating and stabilization of PPAR-γ, thereby resulting in increased insulin sensitivity. Additionally, MYOD1 induces the transcription of USP2, which is downregulated during muscle atrophy. The global deletion of USP2 aggravated diabetes- and DEX-induced muscle atrophy and insulin resistance in mice. Specific overexpression of USP2 in the Gas muscle of diabetic mice mitigated the progression of muscle atrophy and preserved insulin sensitivity. Taken together, these results revealed a previously unrecognized yet critical role of USP2 in the regulation of PPAR-γ signaling and in the pathogenesis of muscle atrophy.

Our mechanistic investigations revealed that USP2 prevents muscle atrophy by regulating the stability of PPAR-γ, a key TF implicated in muscle metabolism and insulin sensitivity (31–33). Recent studies have elucidated the critical role of PPAR-γ in regulating muscle atrophy and regeneration (34–37). Constitutive overexpression of PPAR-γ in mouse skeletal muscle reduces myosteatosis and boosts oxidative myofiber content and insulin sensitivity by inducing adiponectin production in muscle cells (34,35). Exercise-induced activation of PPAR-γ and suppression of miR-29b attenuate angiotensin II–induced muscle atrophy (36). Additionally, muscle injuries trigger a prostacyclin-PPAR-γ/fatty acid oxidation spike, promoting muscle regeneration and controlling myogenesis (37). Clinical evidence from a study of 16-week pioglitazone treatment in patients with metabolic syndrome revealed no significant changes in muscle cross-sectional area (38,39). However, the treatment reduced skeletal muscle insulin resistance and intramyocellular lipid deposition, and improved skeletal muscle fatty acid metabolism, thereby enhancing whole-body aerobic capacity and skeletal muscle energy metabolism (38,39). The PPAR-γ agonist rosiglitazone attenuates inflammation, enhances the number of oxidative fibers, and enlarges the mitochondrial area in the muscles of aging mice (40). The mitochondrial dysfunction/NLRP3 inflammasome axis can contribute to muscle atrophy via PPAR-γ under specific mitochondrial dysfunction and inflammatory conditions (41). Exercise may attenuate angiotensin II–induced muscle atrophy by targeting the PPAR-γ/miR29b pathway, suggesting PPAR-γ’s link to atrophy development (36). Also, PPAR-γ can inhibit NLRP3 inflammasome activation, indicating its anti-inflammatory effect on atrophy related to chronic inflammation (42). Research indicates that PPAR-γ modulates FOXO3 activity through a SIRT6-mediated mechanism (43). A decline in the PI3K/AKT pathway activates FoxO factors in atrophying myotubes, whereas IGF-1 treatment or AKT overexpression suppresses this activation (44). Because PPAR-γ is crucial for maintaining skeletal-muscle insulin sensitivity, it may influence FOXO3 transcriptional activity by regulating SIRT6 or the PI3K/AKT pathway (31). In this study, USP2 deficiency reduced PPAR-γ expression and increased its protein ubiquitination (Fig. 4). In contrast, USP2 overexpression regulated PPAR-γ stability and activity, affecting muscle atrophy and insulin sensitivity (Fig. 5). Data also showed the USP2-PPAR-γ interaction was key in mediating the effects of USP2 on muscle atrophy and insulin signaling (Figs. 6 and 7). By identifying USP2 as a novel PPAR-γ deubiquitinase and protective factor, we have broadened the PPAR-γ regulatory network and revealed the potential of targeting the USP2–PPAR-γ axis for skeletal muscle and insulin signaling disorders.

Inflammation plays a central role in the pathogenesis of muscle-wasting disorders, including sarcopenia, cachexia, and diabetes-induced muscle atrophy (45–47). Among the key mediators of inflammation, TNF-α emerges as a pivotal regulator of muscle homeostasis by directly suppressing specific force production and promoting muscle protein degradation (47,48). TNF-α treatment downregulated USP2 (Fig. 5D and F). Usp2 knockdown aggravated and AdUsp2 overexpression suppressed TNF-α–induced MYHC staining reduction and myotube impairment (Fig. 5F). These results highlight the protective role of USP2 against TNF-α–induced muscle atrophy, suggesting inflammation disrupts the USP2-PPAR-γ–mediated insulin pathway, aligning with prior studies on TNF-α in muscle atrophy. Our study further showed that TNF-α–induced inflammation significantly reduced MYOD1 binding to the USP2 promoter (Fig. 8F). Thus, targeting USP2 to combat inflammation-related muscle wasting has important therapeutic potential and may offer new strategies to alleviate inflammation-induced muscle atrophy and enhance insulin sensitivity in diabetes.

In conclusion, our study demonstrated the protective role of USP2 against muscle atrophy by stabilizing PPAR-γ through deubiquitination, particularly in diabetic mice and mice with DEX-induced diabetes, suggesting its therapeutic potential. Through enhancing PPAR-γ signaling via deubiquitination, USP2 improved insulin sensitivity and preserved muscle mass. These findings highlight the potential of targeting the USP2–PPAR-γ axis for therapeutic interventions aimed at combating muscle atrophy and enhancing insulin sensitivity, thereby presenting novel avenues for addressing metabolic disorders and sarcopenia.

First, our experimental design solely incorporated male mice, thereby lacking the comparative insights that could have been gleaned from their female counterparts. Second, we did not explore the pharmacologic induction of USP2 overexpression. Such an investigation could have provided a more clinically relevant perspective, bridging the gap between basic research and potential therapeutic applications, and offering valuable translational insights into future intervention strategies.

Acknowledgments. Sequencing service was provided by Bioyi Biotechnology Co., Ltd. Wuhan, China.

Funding. This work was supported by the National Natural Science Foundation of China (grants 82370876 to S.Y.; 82170842 and 82371572 to Z.L.; and 82171556 to L.K.); the Shenzhen Medical Academy of Research and Translation, Shenzhen, China (grant A2303031 to S.Y.); the Shenzhen Science and Technology Program, Shenzhen, China (grant JCYJ20240813103959020 to L.X.); Shenzhen Sustainable Development Science and Technology Special Project, China (grant KCXFZ20201221173600001 to Z.L.); and Key Program Topics of Shenzhen Basic Research, China (grant JCYJ20220818102605013 to L.K.).

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

Author Contributions. S.Y. contributed to the conception and the study design. T.L., G.Y., L.X., L.L., and Y.L. conducted experiments. L.X. contributed to the acquisition, analysis, and interpretation of data. S.Y. and Z.L. drafted the or reviewed the manuscript critically for important intellectual content. L.K. analyzed the data and reviewed the article critically for important intellectual content. All authors gave their approval of the version to be published. Z.L. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.