Advertisement
Research ArticleMuscle biology Free access | 10.1172/JCI143737
1Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, The Institute for Advanced Studies, Frontier Science Center for Immunology and Metabolism, Wuhan University, Wuhan, China.
2State Key Laboratory of Pharmaceutical Biotechnology and MOE Key Laboratory of Model Animals for Disease Study, Department of Spine Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Jiangsu Key Laboratory of Molecular Medicine, Chemistry and Biomedicine Innovation Center (ChemBIC), Model Animal Research Center, Nanjing University Medical School, Nanjing University, Nanjing, China.
3Division of Ophthalmology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
4Key Laboratory of Animal Genetics, Breeding and Reproduction of Shaanxi Province, College of Animal Science and Technology, Northwest A&F University, Yangling, China.
5Cellular Signaling Laboratory, Key Laboratory of Molecular Biophysics of Ministry of Education, Huazhong University of Science and Technology, Wuhan, China.
Address correspondence to: Yong Liu, College of Life Sciences, Wuhan University, 299 Bayi Road, Wuhan 430072, China. Phone: 86.27.68753463; Email: liuyong31279@whu.edu.cn. Or to: Zhenji Gan, Model Animal Research Center of Nanjing University, 12 Xuefu Road, Pukou, Nanjing 210061, China. Phone: 86.25.58641546; Email: ganzj@nju.edu.cn.
Find articles by He, S. in: JCI | PubMed | Google Scholar
1Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, The Institute for Advanced Studies, Frontier Science Center for Immunology and Metabolism, Wuhan University, Wuhan, China.
2State Key Laboratory of Pharmaceutical Biotechnology and MOE Key Laboratory of Model Animals for Disease Study, Department of Spine Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Jiangsu Key Laboratory of Molecular Medicine, Chemistry and Biomedicine Innovation Center (ChemBIC), Model Animal Research Center, Nanjing University Medical School, Nanjing University, Nanjing, China.
3Division of Ophthalmology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
4Key Laboratory of Animal Genetics, Breeding and Reproduction of Shaanxi Province, College of Animal Science and Technology, Northwest A&F University, Yangling, China.
5Cellular Signaling Laboratory, Key Laboratory of Molecular Biophysics of Ministry of Education, Huazhong University of Science and Technology, Wuhan, China.
Address correspondence to: Yong Liu, College of Life Sciences, Wuhan University, 299 Bayi Road, Wuhan 430072, China. Phone: 86.27.68753463; Email: liuyong31279@whu.edu.cn. Or to: Zhenji Gan, Model Animal Research Center of Nanjing University, 12 Xuefu Road, Pukou, Nanjing 210061, China. Phone: 86.25.58641546; Email: ganzj@nju.edu.cn.
Find articles by Fu, T. in: JCI | PubMed | Google Scholar
1Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, The Institute for Advanced Studies, Frontier Science Center for Immunology and Metabolism, Wuhan University, Wuhan, China.
2State Key Laboratory of Pharmaceutical Biotechnology and MOE Key Laboratory of Model Animals for Disease Study, Department of Spine Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Jiangsu Key Laboratory of Molecular Medicine, Chemistry and Biomedicine Innovation Center (ChemBIC), Model Animal Research Center, Nanjing University Medical School, Nanjing University, Nanjing, China.
3Division of Ophthalmology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
4Key Laboratory of Animal Genetics, Breeding and Reproduction of Shaanxi Province, College of Animal Science and Technology, Northwest A&F University, Yangling, China.
5Cellular Signaling Laboratory, Key Laboratory of Molecular Biophysics of Ministry of Education, Huazhong University of Science and Technology, Wuhan, China.
Address correspondence to: Yong Liu, College of Life Sciences, Wuhan University, 299 Bayi Road, Wuhan 430072, China. Phone: 86.27.68753463; Email: liuyong31279@whu.edu.cn. Or to: Zhenji Gan, Model Animal Research Center of Nanjing University, 12 Xuefu Road, Pukou, Nanjing 210061, China. Phone: 86.25.58641546; Email: ganzj@nju.edu.cn.
Find articles by Yu, Y. in: JCI | PubMed | Google Scholar
1Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, The Institute for Advanced Studies, Frontier Science Center for Immunology and Metabolism, Wuhan University, Wuhan, China.
2State Key Laboratory of Pharmaceutical Biotechnology and MOE Key Laboratory of Model Animals for Disease Study, Department of Spine Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Jiangsu Key Laboratory of Molecular Medicine, Chemistry and Biomedicine Innovation Center (ChemBIC), Model Animal Research Center, Nanjing University Medical School, Nanjing University, Nanjing, China.
3Division of Ophthalmology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
4Key Laboratory of Animal Genetics, Breeding and Reproduction of Shaanxi Province, College of Animal Science and Technology, Northwest A&F University, Yangling, China.
5Cellular Signaling Laboratory, Key Laboratory of Molecular Biophysics of Ministry of Education, Huazhong University of Science and Technology, Wuhan, China.
Address correspondence to: Yong Liu, College of Life Sciences, Wuhan University, 299 Bayi Road, Wuhan 430072, China. Phone: 86.27.68753463; Email: liuyong31279@whu.edu.cn. Or to: Zhenji Gan, Model Animal Research Center of Nanjing University, 12 Xuefu Road, Pukou, Nanjing 210061, China. Phone: 86.25.58641546; Email: ganzj@nju.edu.cn.
Find articles by Liang, Q. in: JCI | PubMed | Google Scholar
1Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, The Institute for Advanced Studies, Frontier Science Center for Immunology and Metabolism, Wuhan University, Wuhan, China.
2State Key Laboratory of Pharmaceutical Biotechnology and MOE Key Laboratory of Model Animals for Disease Study, Department of Spine Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Jiangsu Key Laboratory of Molecular Medicine, Chemistry and Biomedicine Innovation Center (ChemBIC), Model Animal Research Center, Nanjing University Medical School, Nanjing University, Nanjing, China.
3Division of Ophthalmology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
4Key Laboratory of Animal Genetics, Breeding and Reproduction of Shaanxi Province, College of Animal Science and Technology, Northwest A&F University, Yangling, China.
5Cellular Signaling Laboratory, Key Laboratory of Molecular Biophysics of Ministry of Education, Huazhong University of Science and Technology, Wuhan, China.
Address correspondence to: Yong Liu, College of Life Sciences, Wuhan University, 299 Bayi Road, Wuhan 430072, China. Phone: 86.27.68753463; Email: liuyong31279@whu.edu.cn. Or to: Zhenji Gan, Model Animal Research Center of Nanjing University, 12 Xuefu Road, Pukou, Nanjing 210061, China. Phone: 86.25.58641546; Email: ganzj@nju.edu.cn.
Find articles by Li, L. in: JCI | PubMed | Google Scholar
1Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, The Institute for Advanced Studies, Frontier Science Center for Immunology and Metabolism, Wuhan University, Wuhan, China.
2State Key Laboratory of Pharmaceutical Biotechnology and MOE Key Laboratory of Model Animals for Disease Study, Department of Spine Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Jiangsu Key Laboratory of Molecular Medicine, Chemistry and Biomedicine Innovation Center (ChemBIC), Model Animal Research Center, Nanjing University Medical School, Nanjing University, Nanjing, China.
3Division of Ophthalmology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
4Key Laboratory of Animal Genetics, Breeding and Reproduction of Shaanxi Province, College of Animal Science and Technology, Northwest A&F University, Yangling, China.
5Cellular Signaling Laboratory, Key Laboratory of Molecular Biophysics of Ministry of Education, Huazhong University of Science and Technology, Wuhan, China.
Address correspondence to: Yong Liu, College of Life Sciences, Wuhan University, 299 Bayi Road, Wuhan 430072, China. Phone: 86.27.68753463; Email: liuyong31279@whu.edu.cn. Or to: Zhenji Gan, Model Animal Research Center of Nanjing University, 12 Xuefu Road, Pukou, Nanjing 210061, China. Phone: 86.25.58641546; Email: ganzj@nju.edu.cn.
Find articles by Liu, J. in: JCI | PubMed | Google Scholar |
1Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, The Institute for Advanced Studies, Frontier Science Center for Immunology and Metabolism, Wuhan University, Wuhan, China.
2State Key Laboratory of Pharmaceutical Biotechnology and MOE Key Laboratory of Model Animals for Disease Study, Department of Spine Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Jiangsu Key Laboratory of Molecular Medicine, Chemistry and Biomedicine Innovation Center (ChemBIC), Model Animal Research Center, Nanjing University Medical School, Nanjing University, Nanjing, China.
3Division of Ophthalmology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
4Key Laboratory of Animal Genetics, Breeding and Reproduction of Shaanxi Province, College of Animal Science and Technology, Northwest A&F University, Yangling, China.
5Cellular Signaling Laboratory, Key Laboratory of Molecular Biophysics of Ministry of Education, Huazhong University of Science and Technology, Wuhan, China.
Address correspondence to: Yong Liu, College of Life Sciences, Wuhan University, 299 Bayi Road, Wuhan 430072, China. Phone: 86.27.68753463; Email: liuyong31279@whu.edu.cn. Or to: Zhenji Gan, Model Animal Research Center of Nanjing University, 12 Xuefu Road, Pukou, Nanjing 210061, China. Phone: 86.25.58641546; Email: ganzj@nju.edu.cn.
Find articles by Zhang, X. in: JCI | PubMed | Google Scholar
1Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, The Institute for Advanced Studies, Frontier Science Center for Immunology and Metabolism, Wuhan University, Wuhan, China.
2State Key Laboratory of Pharmaceutical Biotechnology and MOE Key Laboratory of Model Animals for Disease Study, Department of Spine Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Jiangsu Key Laboratory of Molecular Medicine, Chemistry and Biomedicine Innovation Center (ChemBIC), Model Animal Research Center, Nanjing University Medical School, Nanjing University, Nanjing, China.
3Division of Ophthalmology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
4Key Laboratory of Animal Genetics, Breeding and Reproduction of Shaanxi Province, College of Animal Science and Technology, Northwest A&F University, Yangling, China.
5Cellular Signaling Laboratory, Key Laboratory of Molecular Biophysics of Ministry of Education, Huazhong University of Science and Technology, Wuhan, China.
Address correspondence to: Yong Liu, College of Life Sciences, Wuhan University, 299 Bayi Road, Wuhan 430072, China. Phone: 86.27.68753463; Email: liuyong31279@whu.edu.cn. Or to: Zhenji Gan, Model Animal Research Center of Nanjing University, 12 Xuefu Road, Pukou, Nanjing 210061, China. Phone: 86.25.58641546; Email: ganzj@nju.edu.cn.
Find articles by Zhou, Q. in: JCI | PubMed | Google Scholar
1Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, The Institute for Advanced Studies, Frontier Science Center for Immunology and Metabolism, Wuhan University, Wuhan, China.
2State Key Laboratory of Pharmaceutical Biotechnology and MOE Key Laboratory of Model Animals for Disease Study, Department of Spine Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Jiangsu Key Laboratory of Molecular Medicine, Chemistry and Biomedicine Innovation Center (ChemBIC), Model Animal Research Center, Nanjing University Medical School, Nanjing University, Nanjing, China.
3Division of Ophthalmology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
4Key Laboratory of Animal Genetics, Breeding and Reproduction of Shaanxi Province, College of Animal Science and Technology, Northwest A&F University, Yangling, China.
5Cellular Signaling Laboratory, Key Laboratory of Molecular Biophysics of Ministry of Education, Huazhong University of Science and Technology, Wuhan, China.
Address correspondence to: Yong Liu, College of Life Sciences, Wuhan University, 299 Bayi Road, Wuhan 430072, China. Phone: 86.27.68753463; Email: liuyong31279@whu.edu.cn. Or to: Zhenji Gan, Model Animal Research Center of Nanjing University, 12 Xuefu Road, Pukou, Nanjing 210061, China. Phone: 86.25.58641546; Email: ganzj@nju.edu.cn.
Find articles by Guo, Q. in: JCI | PubMed | Google Scholar
1Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, The Institute for Advanced Studies, Frontier Science Center for Immunology and Metabolism, Wuhan University, Wuhan, China.
2State Key Laboratory of Pharmaceutical Biotechnology and MOE Key Laboratory of Model Animals for Disease Study, Department of Spine Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Jiangsu Key Laboratory of Molecular Medicine, Chemistry and Biomedicine Innovation Center (ChemBIC), Model Animal Research Center, Nanjing University Medical School, Nanjing University, Nanjing, China.
3Division of Ophthalmology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
4Key Laboratory of Animal Genetics, Breeding and Reproduction of Shaanxi Province, College of Animal Science and Technology, Northwest A&F University, Yangling, China.
5Cellular Signaling Laboratory, Key Laboratory of Molecular Biophysics of Ministry of Education, Huazhong University of Science and Technology, Wuhan, China.
Address correspondence to: Yong Liu, College of Life Sciences, Wuhan University, 299 Bayi Road, Wuhan 430072, China. Phone: 86.27.68753463; Email: liuyong31279@whu.edu.cn. Or to: Zhenji Gan, Model Animal Research Center of Nanjing University, 12 Xuefu Road, Pukou, Nanjing 210061, China. Phone: 86.25.58641546; Email: ganzj@nju.edu.cn.
Find articles by Xu, D. in: JCI | PubMed | Google Scholar
1Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, The Institute for Advanced Studies, Frontier Science Center for Immunology and Metabolism, Wuhan University, Wuhan, China.
2State Key Laboratory of Pharmaceutical Biotechnology and MOE Key Laboratory of Model Animals for Disease Study, Department of Spine Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Jiangsu Key Laboratory of Molecular Medicine, Chemistry and Biomedicine Innovation Center (ChemBIC), Model Animal Research Center, Nanjing University Medical School, Nanjing University, Nanjing, China.
3Division of Ophthalmology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
4Key Laboratory of Animal Genetics, Breeding and Reproduction of Shaanxi Province, College of Animal Science and Technology, Northwest A&F University, Yangling, China.
5Cellular Signaling Laboratory, Key Laboratory of Molecular Biophysics of Ministry of Education, Huazhong University of Science and Technology, Wuhan, China.
Address correspondence to: Yong Liu, College of Life Sciences, Wuhan University, 299 Bayi Road, Wuhan 430072, China. Phone: 86.27.68753463; Email: liuyong31279@whu.edu.cn. Or to: Zhenji Gan, Model Animal Research Center of Nanjing University, 12 Xuefu Road, Pukou, Nanjing 210061, China. Phone: 86.25.58641546; Email: ganzj@nju.edu.cn.
Find articles by Chen, Y. in: JCI | PubMed | Google Scholar
1Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, The Institute for Advanced Studies, Frontier Science Center for Immunology and Metabolism, Wuhan University, Wuhan, China.
2State Key Laboratory of Pharmaceutical Biotechnology and MOE Key Laboratory of Model Animals for Disease Study, Department of Spine Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Jiangsu Key Laboratory of Molecular Medicine, Chemistry and Biomedicine Innovation Center (ChemBIC), Model Animal Research Center, Nanjing University Medical School, Nanjing University, Nanjing, China.
3Division of Ophthalmology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
4Key Laboratory of Animal Genetics, Breeding and Reproduction of Shaanxi Province, College of Animal Science and Technology, Northwest A&F University, Yangling, China.
5Cellular Signaling Laboratory, Key Laboratory of Molecular Biophysics of Ministry of Education, Huazhong University of Science and Technology, Wuhan, China.
Address correspondence to: Yong Liu, College of Life Sciences, Wuhan University, 299 Bayi Road, Wuhan 430072, China. Phone: 86.27.68753463; Email: liuyong31279@whu.edu.cn. Or to: Zhenji Gan, Model Animal Research Center of Nanjing University, 12 Xuefu Road, Pukou, Nanjing 210061, China. Phone: 86.25.58641546; Email: ganzj@nju.edu.cn.
Find articles by Wang, X. in: JCI | PubMed | Google Scholar |
1Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, The Institute for Advanced Studies, Frontier Science Center for Immunology and Metabolism, Wuhan University, Wuhan, China.
2State Key Laboratory of Pharmaceutical Biotechnology and MOE Key Laboratory of Model Animals for Disease Study, Department of Spine Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Jiangsu Key Laboratory of Molecular Medicine, Chemistry and Biomedicine Innovation Center (ChemBIC), Model Animal Research Center, Nanjing University Medical School, Nanjing University, Nanjing, China.
3Division of Ophthalmology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
4Key Laboratory of Animal Genetics, Breeding and Reproduction of Shaanxi Province, College of Animal Science and Technology, Northwest A&F University, Yangling, China.
5Cellular Signaling Laboratory, Key Laboratory of Molecular Biophysics of Ministry of Education, Huazhong University of Science and Technology, Wuhan, China.
Address correspondence to: Yong Liu, College of Life Sciences, Wuhan University, 299 Bayi Road, Wuhan 430072, China. Phone: 86.27.68753463; Email: liuyong31279@whu.edu.cn. Or to: Zhenji Gan, Model Animal Research Center of Nanjing University, 12 Xuefu Road, Pukou, Nanjing 210061, China. Phone: 86.25.58641546; Email: ganzj@nju.edu.cn.
Find articles by Chen, Y. in: JCI | PubMed | Google Scholar
1Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, The Institute for Advanced Studies, Frontier Science Center for Immunology and Metabolism, Wuhan University, Wuhan, China.
2State Key Laboratory of Pharmaceutical Biotechnology and MOE Key Laboratory of Model Animals for Disease Study, Department of Spine Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Jiangsu Key Laboratory of Molecular Medicine, Chemistry and Biomedicine Innovation Center (ChemBIC), Model Animal Research Center, Nanjing University Medical School, Nanjing University, Nanjing, China.
3Division of Ophthalmology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
4Key Laboratory of Animal Genetics, Breeding and Reproduction of Shaanxi Province, College of Animal Science and Technology, Northwest A&F University, Yangling, China.
5Cellular Signaling Laboratory, Key Laboratory of Molecular Biophysics of Ministry of Education, Huazhong University of Science and Technology, Wuhan, China.
Address correspondence to: Yong Liu, College of Life Sciences, Wuhan University, 299 Bayi Road, Wuhan 430072, China. Phone: 86.27.68753463; Email: liuyong31279@whu.edu.cn. Or to: Zhenji Gan, Model Animal Research Center of Nanjing University, 12 Xuefu Road, Pukou, Nanjing 210061, China. Phone: 86.25.58641546; Email: ganzj@nju.edu.cn.
Find articles by Liu, J. in: JCI | PubMed | Google Scholar
1Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, The Institute for Advanced Studies, Frontier Science Center for Immunology and Metabolism, Wuhan University, Wuhan, China.
2State Key Laboratory of Pharmaceutical Biotechnology and MOE Key Laboratory of Model Animals for Disease Study, Department of Spine Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Jiangsu Key Laboratory of Molecular Medicine, Chemistry and Biomedicine Innovation Center (ChemBIC), Model Animal Research Center, Nanjing University Medical School, Nanjing University, Nanjing, China.
3Division of Ophthalmology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
4Key Laboratory of Animal Genetics, Breeding and Reproduction of Shaanxi Province, College of Animal Science and Technology, Northwest A&F University, Yangling, China.
5Cellular Signaling Laboratory, Key Laboratory of Molecular Biophysics of Ministry of Education, Huazhong University of Science and Technology, Wuhan, China.
Address correspondence to: Yong Liu, College of Life Sciences, Wuhan University, 299 Bayi Road, Wuhan 430072, China. Phone: 86.27.68753463; Email: liuyong31279@whu.edu.cn. Or to: Zhenji Gan, Model Animal Research Center of Nanjing University, 12 Xuefu Road, Pukou, Nanjing 210061, China. Phone: 86.25.58641546; Email: ganzj@nju.edu.cn.
Find articles by Gan, Z. in: JCI | PubMed | Google Scholar |
1Hubei Key Laboratory of Cell Homeostasis, College of Life Sciences, The Institute for Advanced Studies, Frontier Science Center for Immunology and Metabolism, Wuhan University, Wuhan, China.
2State Key Laboratory of Pharmaceutical Biotechnology and MOE Key Laboratory of Model Animals for Disease Study, Department of Spine Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Jiangsu Key Laboratory of Molecular Medicine, Chemistry and Biomedicine Innovation Center (ChemBIC), Model Animal Research Center, Nanjing University Medical School, Nanjing University, Nanjing, China.
3Division of Ophthalmology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
4Key Laboratory of Animal Genetics, Breeding and Reproduction of Shaanxi Province, College of Animal Science and Technology, Northwest A&F University, Yangling, China.
5Cellular Signaling Laboratory, Key Laboratory of Molecular Biophysics of Ministry of Education, Huazhong University of Science and Technology, Wuhan, China.
Address correspondence to: Yong Liu, College of Life Sciences, Wuhan University, 299 Bayi Road, Wuhan 430072, China. Phone: 86.27.68753463; Email: liuyong31279@whu.edu.cn. Or to: Zhenji Gan, Model Animal Research Center of Nanjing University, 12 Xuefu Road, Pukou, Nanjing 210061, China. Phone: 86.25.58641546; Email: ganzj@nju.edu.cn.
Find articles by Liu, Y. in: JCI | PubMed | Google Scholar
Published July 20, 2021 - More info
Skeletal muscle can undergo a regenerative process in response to injury or disease to preserve muscle mass and function, which are critically influenced by cellular stress responses. Inositol-requiring enzyme 1 (IRE1) is an ancient endoplasmic reticulum stress sensor and mediates a key branch of the unfolded protein response. In mammals, IRE1α is implicated in the homeostatic control of stress responses during tissue injury and regeneration. Here, we show that IRE1α serves as a myogenic regulator in skeletal muscle regeneration in response to injury and muscular dystrophy. We found in mice that IRE1α was activated during injury-induced muscle regeneration, and muscle-specific IRE1α ablation resulted in impaired regeneration upon cardiotoxin-induced injury. Gain- and loss-of-function studies in myocytes demonstrated that IRE1α acts to sustain both differentiation in myoblasts and hypertrophy in myotubes through regulated IRE1-dependent decay (RIDD) of mRNA encoding myostatin, a key negative regulator of muscle repair and growth. Furthermore, in the mouse model of Duchenne muscular dystrophy, loss of muscle IRE1α resulted in augmented myostatin signaling and exacerbated the dystrophic phenotypes. These results reveal a pivotal role for the RIDD output of IRE1α in muscle regeneration, offering insight into potential therapeutic strategies for muscle loss diseases.
The endoplasmic reticulum (ER) is an essential membrane network that governs the folding and secretion of nearly one-third of the cellular proteome in eukaryotes. Increased protein folding demand or accumulation of unfolded/misfolded proteins within the ER lumen causes ER stress, activating the adaptive cell response termed the unfolded protein response (UPR) (1–3). Among the 3 ER-resident transmembrane UPR sensors (3), inositol-requiring enzyme 1 (IRE1) is the most conserved signal transducer, with dual enzyme activities, i.e., Ser/Thr protein kinase and endoribonuclease (RNase), within its cytoplasmic portion (1, 2, 4). Upon ER stress, IRE1 is activated through dimerization/oligomerization and autophosphorylation, mediating a key branch of the UPR signaling through its RNase activity (1–3, 5, 6). Activated IRE1 catalyzes the unconventional splicing of the mRNA encoding the transcription factor X-box binding protein 1 (XBP1) to generate a spliced active form of XBP1 (XBP1s), thereby driving a major UPR program involved in modulating protein folding, secretion, and ER-associated degradation (5, 6). In addition, IRE1 can also directly degrade select mRNA species in a process referred to as “regulated IRE1-dependent decay” (RIDD) (7–9). In mammals, IRE1α signaling has been implicated in regulating a wide range of biological processes, including determination of cell fate, metabolic homeostasis, and immune responses (10–14). Whereas XBP1s serves as a crucial effector of IRE1α’s RNase activation, recent studies have also revealed implications of its RIDD activity in the control of secretory cargo proteins (8, 15), insulin production in pancreatic β cells (16), hepatic lipid metabolism (17), and dendritic cell homeostasis in immunity (18). Moreover, a growing body of evidence suggests that dysregulated activation of IRE1α contributes to the pathogenesis of many chronic diseases, including metabolic disorders, cancer, and neurodegeneration (1, 11, 19–22).
Skeletal muscles have the remarkable capacity to undergo regenerative growth in response to injury or other external stimuli. Muscle repair is a highly coordinated process involving muscle damage, regeneration, and myofiber remodeling (23–25). Muscle regeneration relies on activation and expansion of myogenic stem cells, known as satellite cells, residing beneath the basal lamina of myofibers, followed by myoblast differentiation and ultimate fusion into multinucleated myotubes (26–28). At the final stage of muscle repair, myotubes undergo hypertrophy remodeling to generate mature muscle fibers and restore their contractile capacity. Multiple myogenic signaling pathways serve to control muscle regeneration and hypertrophy (27–29), mainly mediated by orchestrated activation of the transcriptional regulatory factors MYF5, MyoD, and myogenin, along with regulated expression of the myosin heavy chain (MyHC) genes (27, 28). In addition, a number of cytokines and growth factors are known to be critically involved in controlling muscle repair and growth, as exemplified by IL-6, myostatin/TGF-β, and insulin-like growth factor-1 (IGF-1) (30, 31). Impairment of muscle regeneration and growth leads to muscle atrophy or dystrophy, which represents a key pathological feature of many human diseases with muscle loss (32). For instance, Duchenne muscular dystrophy (DMD), caused by mutations in the X-linked gene dystrophin, is the most common and severe form of human muscular dystrophy and remains incurable to date (33, 34). A better understanding of the complex interplays between the intrinsic cell signaling network and microenvironmental cues will uncover the molecular regulatory mechanisms in the maintenance of muscle mass, thus facilitating the development of regenerative therapeutics for muscle degeneration disorders.
As a critical regulator of ER stress response in restoration of proteostatic homeostasis, IRE1α has been documented as promoting cell proliferation during the reparative regeneration of the liver (35) and the compensatory growth of pancreatic islets under metabolic stress (36). However, it has yet to be explored whether IRE1α plays a role in sensing skeletal muscle stress and regulating its regeneration process. In this study, we used a cardiotoxin-induced (CTX-induced) acute muscle injury model as well as the genetic mdx model of DMD in mice, and found that IRE1α acts to sustain muscle repair and growth, largely through its RIDD activity to downregulate the expression of the mRNA encoding myostatin, a critical negative regulator of muscle regeneration and growth. Our findings demonstrate the physiological importance of IRE1α’s RIDD activity in the control of muscle regeneration, and its dysregulation may represent an unrecognized mechanism linking ER stress to muscle degenerative diseases.
Activation of the IRE1α signaling branch is implicated in skeletal muscle regeneration. To investigate whether IRE1α is implicated in regulating muscle regeneration, we first examined its activation states during the regenerative response to acute skeletal muscle injury by injecting CTX into the tibialis anterior (TA) muscles of adult mice. In line with previously reported findings (37), CTX-induced myofiber degeneration triggered robust muscle regeneration responses, as indicated by prominently elevated expression of embryonic MyHC (eMyHC) at 2 days after CTX treatment (Figure 1A). Starting at 2 days after CTX-elicited injury, we observed significant increases in the phosphorylation at Ser724 within the kinase domain of IRE1α that peaked at 4 and 8 days (Figure 1A). Consistent with activated IRE1α’s RNase activity, significantly increased Xbp1 mRNA splicing and decreased abundance of Blos1 mRNA, a typical RIDD substrate of IRE1α, were detected during the regenerative response to CTX-induced injury (Figure 1B), in parallel with marked activation of the Myog and Myh8 gene expression program (Figure 1C). Notably, increased eIF2α phosphorylation and BiP protein expression levels were also seen following CTX treatment (Figure 1A). These results indicate that the IRE1α branch of the UPR is activated in response to muscle damage, and suggest that IRE1α may act as a component in the control mechanism during muscle repair and regenerative growth response.
Injury-induced IRE1α activation affects skeletal muscle regeneration and myostatin signaling. (A–C) Muscle injury induces IRE1α activation. TA muscle of adult male mice was injected with CTX to induce acute muscle injury (n = 5 mice at each time point). (A) Immunoblot analysis of the phosphorylation of IRE1α and eIF2α, and protein expression of BiP and eMyHC in muscle extracts. Each lane represents 1 individual mouse. Averaged p-IRE1α/IRE1α and p-eIF2α/eIF2α ratios are shown from densitometric quantification. (B and C) Quantitative RT-PCR analysis of Xbp1 mRNA splicing and the abundance of mRNAs encoding the indicated genes. (D–J) IRE1α abrogation impairs muscle regeneration and enhances myostatin signaling. TA muscles of male Ern1fl/fl and Myod1-Cre control and Ern1fl/fl Myod1-Cre mice were subjected to CTX-induced injuries (n = 5 mice at each time point). (D) TA muscle weight relative to tibia length. (E) Representative H&E staining of TA muscles from mice of the indicated genotypes. (F) Percentage of regenerated myofibers in the indicated cross-sectional areas of TA muscles at 12 days after CTX injection. Myofibers containing centralized nuclei were quantified by ImageJ from 500 myofibers of the TA muscle in each mouse. (G) Representative laminin (green) and eMyHC (red) immunostaining of TA muscles at 8 days after CTX injection. (H) Quantification of the percentage of eMyHC+ myofibers within laminin staining. (I) Quantitative RT-PCR analysis of Xbp1 mRNA splicing and Mstn mRNA abundance in TA muscles. Data are shown as relative to the value for Ern1fl/fl TA muscle at day 0 after normalization to Gapdh as the internal control. (J) Immunoblot analysis of TA muscle lysates. Averaged p-Smad/Smad ratios are shown after normalization to the value for Ern1fl/fl TA muscle at day 0. All data are presented as mean ± SEM. Significance was calculated by 1-way (A–C and H) or 2-way (D, F, I, and J) ANOVA with Bonferroni’s multiple-comparison test. *P < 0.05, **P < 0.01, ***P < 0.001 vs. day 0 or Ern1fl/fl. #P < 0.05, ###P < 0.001 vs. Myod1-Cre. Scale bars: 100 μm.
To determine the importance of IRE1α in skeletal muscle regeneration, we used Ern1fl/fl mice with 2 loxP sites flanking exon 2 of the Ern1 gene (38) and Myod1-Cre mice to generate skeletal muscle–specific IRE1α-knockout (referred to as Ern1fl/fl Myod1-Cre) mice, which were born at normal Mendelian ratios and appeared to be grossly normal. Immunoblot analysis showed substantial reductions in IRE1α protein levels from multiple muscle types in Ern1fl/fl Myod1-Cre mice relative to their Ern1fl/fl littermates or Myod1-Cre controls, whereas no changes were detected in their hearts (Supplemental Figure 1A; supplemental material available online with this article; https://doi.org/10.1172/JCI143737DS1). As expected, markedly reduced Xbp1 mRNA splicing was detected in the Ern1fl/fl Myod1-Cre muscles relative to their control groups (Supplemental Figure 1B). Ern1fl/fl Myod1-Cre mice exhibited no overt differences in their body weight, muscle weight, or myofiber morphology (Supplemental Figure 2, A–D), and showed no significant alterations in the abundance of Atf4 or Chop mRNA (Supplemental Figure 2E) or of Myh1, Myh2, or Myh4 mRNA (Supplemental Figure 2F) in their TA muscles. This suggests that ablation of IRE1α neither provokes ER stress, nor affects muscle development. When subjected to CTX-induced injury in their TA muscles, animals of all 3 genotypes showed markedly reduced TA muscle weight, presumably due to extensive myofiber degeneration, at 4 days after CTX injection (Figure 1D). Whereas gradual restoration of TA muscle mass was seen in all 3 genotypes at 8 and 12 days after CTX injection (Figure 1D), a significant reduction in the weight of regenerating TA muscle was observed in Ern1fl/fl Myod1-Cre mice relative to their Ern1fl/fl or Myod1-Cre counterparts at 12 days after CTX-induced injury (Figure 1D). This indicates that loss of IRE1α in muscle cells resulted in impaired regenerative response to acute muscle injury. Indeed, histological analysis revealed that at 8 days after CTX injection, while TA muscles from all 3 genotypes showed newly formed myofibers containing centralized nuclei that replaced the damaged myofibers (Figure 1E), a significantly lower percentage of large regenerated myofibers was found in Ern1fl/fl Myod-Cre TA muscles than in those of Ern1fl/fl or Myod1-Cre control mice at 8 or 12 days after CTX-induced injury (Figure 1, E and F, and Supplemental Figure 3A). This defect of reparative regeneration in Ern1fl/fl Myod1-Cre TA muscles was further reflected by analysis of muscle sections stained for eMyHC (Figure 1, G and H, and Supplemental Figure 3B). At 8 days after CTX injection, whereas Ern1fl/fl or Myod1-Cre TA muscles exhibited strongly expressed eMyHC in most myofibers, Ern1fl/fl Myod1-Cre TA muscles showed marked reductions in eMyHC expression in many myofibers, along with eMyHC-positive fibers displaying apparent heterogeneity in size (Figure 1, G and H). At 12 days after CTX injection, however, Ern1fl/fl or Myod1-Cre TA muscles had weaker eMyHC signals and larger myofibers, but Ern1fl/fl Myod1-Cre TA muscles displayed accumulation of eMyHC in smaller myofibers, likely reflecting an earlier stage of delayed regeneration (Supplemental Figure 3B). Thus, loss of muscle IRE1α leads to impaired skeletal muscle regeneration after acute injury.
Skeletal muscle regeneration in response to injury involves highly orchestrated myogenic transcriptional network and cytokine signaling. Interestingly, we detected markedly decreased Xbp1 mRNA splicing but significantly increased abundance of Mstn mRNA, encoding myostatin (Figure 1I), along with a considerable reduction in Igf1 mRNA expression (Supplemental Figure 3C), in TA muscles of Ern1fl/fl Myod1-Cre mice relative to their Ern1fl/fl counterparts during the regenerative response at 4 and 8 days after CTX injection. Given that myostatin is a secreted TGF-β family member protein that functions as a potent negative regulator of myogenic differentiation and muscle growth (30), this suggests that IRE1α ablation affected the signaling pathways in the control of muscle regeneration. Indeed, we observed prominently elevated phosphorylation of Smad3, the downstream transcription factor of the myostatin signaling cascade (39), in Ern1fl/fl Myod1-Cre TA muscles (Figure 1J). In accordance with poorer muscle regeneration, this enhanced myostatin signaling was associated with marked reductions in eMyHC protein levels (Figure 1J). Together, these results demonstrate that IRE1α abrogation leads to impaired muscle regeneration and enhanced myostatin signaling in response to acute muscle injury, indicating that IRE1α may function as a myogenic regulator of muscle repair and growth.
IRE1α RNase regulates myostatin expression and myoblast differentiation. Because skeletal muscle regeneration involves the activation, proliferation, and differentiation of muscle satellite cells (MuSCs) (28), we first examined whether IRE1α ablation affected MuSCs in Ern1fl/fl Myod1-Cre mice. Using a FACS-based method (40) for quantitative analysis of either quiescent or activated MuSCs (VCAM1+CD45–CD31–Sca1– cells), we observed no significant differences in the number of MuSCs from CTX-injured TA muscles of Ern1fl/fl, Myod1-Cre, and Ern1fl/fl Myod1-Cre mice (Supplemental Figure 4, A and B). This suggests that the satellite cell niche was presumably not affected by muscle IRE1α deficiency. We then investigated the requirement for IRE1α in myogenesis using primary myoblasts isolated from Ern1fl/fl Myod1-Cre and Myod1-Cre mice. During the early period of myoblast differentiation into myotubes, we observed significantly induced IRE1α phosphorylation (Figure 2, A and B), in parallel with increased Xbp1 mRNA splicing and markedly induced Myh4 expression (Supplemental Figure 5A). Consistently, loss of IRE1α in primary myoblasts resulted in significantly upregulated expression of Mstn mRNA (Figure 2C), along with elevated Smad3 phosphorylation and reduced MyHC protein level (Figure 2, D and E). At 3 days after differentiation, increased myostatin protein was detected in the culture medium of Ern1-mKO (i.e. Ern1fl/fl Myod1-Cre) myotubes (Figure 2F); and most Myod1-Cre myoblasts became elongated and fused with expression of the differentiation marker MyHC (Figure 2, G–I). By contrast, Ern1-mKO myoblasts showed significant decreases in cell fusion and multinuclear myotube formation (Figure 2, G–I), with fewer and shorter myotubes observed after differentiation for 5 days (Supplemental Figure 5B). Moreover, Ern1-mKO myocytes had significant reductions in the mRNA and protein expression levels of the differentiation markers myogenin and MyHC when compared with their Myod1-Cre controls (Figure 2, J–L). These results suggest that IRE1α ablation leads to upregulation of myostatin expression and impairment of myoblast differentiation in a cell-autonomous manner.
IRE1α ablation increases myostatin expression and impairs myoblast differentiation. (A–E) Primary myoblasts isolated from TA muscle of male Myod1-Cre and Ern1fl/fl Myod1-Cre (KO or Ern1-mKO) mice were differentiated for up to 3 days as indicated (n = 3–5 independent experiments). (A and B) Immunoblot analysis of IRE1α phosphorylation and MyHC expression (A) at the indicated time of differentiation, and quantification of averaged p-IRE1α/IRE1α ratios (B). (C) Quantitative RT-PCR analysis of the mRNA abundance of Mstn. (D and E) Immunoblot analysis of cell lysates (D) using the indicated antibodies, and averaged p-Smad3/Smad3 and MyHC/tubulin levels (E). (F–L) Primary TA myoblasts were differentiated into myotubes for 3 days (n = 4–6 independent experiments). (F) ELISA analysis of myostatin protein in culture medium. (G–I) Representative images (G) of immunostaining of MyHC (green) and DAPI staining of nuclei (blue), with quantification of the fusion index (H) and the number of nuclei per MyHC+ fiber (I). (J–L) Immunoblot analysis (J) and quantification (K) of MyoG and MyHC in differentiated myocytes, and quantitative RT-PCR analysis (L) of their mRNA expression levels. All data are presented as mean ± SEM. Significance was calculated by unpaired 2-tailed Student’s t test (F, H, I, K, and L), 1-way ANOVA (B), or 2-way ANOVA (C and E) with Bonferroni’s multiple-comparison test. *P < 0.05, **P < 0.01, ***P < 0.001 vs. Myod1-Cre group, or vs. day 0 for B. Scale bars: 100 μm.
To further test whether IRE1α’s RNase activity is required for myoblast differentiation, we examined the effects of restored expression of WT or mutant IRE1α using adenoviruses on the defective differentiation of Ern1-mKO myoblasts. Immunostaining of MyHC showed that overexpression of IRE1α-WT effectively restored the differentiation capacity (Figure 3, A and B) as well as the protein and mRNA expression levels of Myog and Myh4 in Ern1-mKO myoblasts (Figure 3, C–F) to an extent similar to that in Myod1-Cre controls, whereas overexpression of the IRE1α kinase-dead (KD, K599A) or RNase-dead (RD, K907A) mutant versions (41) lost these effects (Figure 3, A–F). Supporting the role of IRE1α in suppression of myostatin, overexpression of IRE1α-WT, but not its kinase- or RNase-deficient mutants, substantially blunted the upregulated expression of Mstn in Ern1-mKO myocytes, which resembled the changes of the typical RIDD substrate, Blos1 (Figure 3F). Notably, adenoviral overexpression of XBP1s did not significantly affect the differentiation capacity or Mstn expression in Ern1-mKO myocytes (Figure 3, A–F). These data demonstrate an IRE1α RNase-dependent, XBP1-independent regulation of myoblast differentiation, suggesting that IRE1α acts as a negative regulator of myostatin to control myoblast differentiation during muscle regeneration.
IRE1α RNase blunts myostatin signaling and promotes myoblast differentiation. Primary Myod1-Cre and Ern1-mKO myoblasts were infected with control recombinant adenovirus or adenoviruses expressing human WT or mutant (KD, kinase dead) IRE1α protein or XBP1s and then differentiated for 3 days (n = 4 independent experiments). (A) Representative images of anti-MyHC and DAPI staining. (B) Quantification of the fusion index. (C) Immunoblot analysis of the indicated proteins. (D) Quantification of MyoG/tubulin and MyHC/tubulin after normalization to the value of control. (E and F) Quantitative RT-PCR analysis of Xbp1 mRNA splicing (E) and the mRNA abundance of the indicated genes (F). All data are presented as mean ± SEM. Significance was calculated by 1-way ANOVA with Bonferroni’s multiple-comparison test. **P < 0.01, ***P < 0.001 vs. Myod1-Cre group. ###P < 0.001 vs. Ern1-mKO + control adenovirus. Scale bars: 100 μm.
IRE1α downregulates myostatin expression and promotes myotube hypertrophy. Because myostatin is also known as an autocrine factor to suppress hypertrophy in myotubes (30, 42, 43), we next tested whether IRE1α could impact myotube hypertrophy remodeling by downregulating myostatin expression. In differentiated C2C12 myotubes, adenoviral expression of a short hairpin RNA (shRNA) directed against IRE1α (Sh-Ern1), in comparison with the scramble shRNA control (Sh-Con), resulted in substantially decreased Xbp1 mRNA splicing, but significantly increased mRNA abundance of Blos1 and Mstn (Figure 4A), accompanied by a significant reduction in the mRNA level of Myh4 but not Igf1 (Figure 4A). In line with enhanced myostatin signaling and impaired myotube growth, knockdown of IRE1α expression led to significantly higher production of myostatin protein in the medium (Figure 4B) and increased phosphorylation of Smad3 (Figure 4C), along with decreases in S6K phosphorylation and MyHC protein levels. Moreover, this was in parallel with significantly reduced diameters of myotubes (Figure 4D) and total cellular protein contents (Figure 4E). This is consistent with the phenotypes of smaller myofibers in mice with muscle IRE1α abrogation following CTX-induced injury. Conversely, when compared with the control, adenoviral IRE1α overexpression in differentiated C2C12 myotubes resulted in significantly elevated Xbp1 mRNA splicing (Figure 4F), reduced mRNA abundance of Blos1 and Mstn (Figure 4F), and lower myostatin protein production (Figure 4G) and signaling (Figure 4H), along with enlarged myotubes (Figure 4I) and higher cellular protein contents (Figure 4J). Notably, adenoviral shRNA knockdown of XBP1 expression caused a slight reduction in Mstn mRNA level (Supplemental Figure 6A), while showing no significant effect on myotube growth (Supplemental Figure 6B). In addition, adenoviral overexpression of IRE1α-WT, but not its KD or RD mutant versions, significantly increased Xbp1 mRNA splicing and decreased the mRNA abundance of Mstn and Blos1 in C2C12 myotubes (Figure 4K). These in vitro analyses via loss- and gain-of-function strategies in myotubes strongly suggest that IRE1α downregulation of myostatin expression also contributes to the control of myotube hypertrophy remodeling. Since myostatin was reported to impact myoblast proliferation (44), we also evaluated whether IRE1α exerts a regulatory action in this regard. MTT analyses showed that neither deficiency nor overexpression of IRE1α had significant effects on the proliferation rate of primary or C2C12 myoblasts (Supplemental Figure 7, A–C). Moreover, FACS analyses showed that knockdown of IRE1α expression in C2C12 myoblasts did not affect cell death, which was increased upon differentiation stimuli (Supplemental Figure 7D). Thus, IRE1α-directed regulation of myostatin is implicated in both myocyte differentiation and hypertrophy growth, consistent with our observation in vivo that loss of IRE1α led to impairment in muscle regeneration upon injury.
IRE1α downregulates myostatin expression and promotes hypertrophy of differentiated myotubes. (A–J) C2C12 myoblast cells were differentiated for 4 days into myotubes and then infected for 48 hours with adenoviruses expressing a scramble control shRNA (Sh-Con) or shRNA directed against IRE1α (Sh-Ern1) (A–E), or with empty control adenovirus (Ad-Con) or that expressing human IRE1α (Ad-IRE1α) (F–J) (n = 4 independent experiments). (A and F) Quantitative RT-PCR analysis of Xbp1 mRNA splicing and the mRNA abundance of the indicated genes. (B and G) ELISA analysis of myostatin protein in culture medium. (C and H) Immunoblot analysis of the indicated proteins from myotube extracts. Averaged MyHC/actin, p-S6K/S6K, and p-Smad3/Smad3 ratios were normalized to the value of Sh-Con or Ad-Con myotubes. (D and I) MyHC immunostaining of myotubes. Myotube diameters were quantified using ImageJ software. (E and J) Total cellular protein content relative to genomic DNA was measured in myotubes. (K) C2C12 myotubes were infected with empty control or adenoviruses expressing the WT or indicated mutant IRE1α protein. Quantitative RT-PCR analysis of Xbp1 mRNA splicing and the mRNA abundance of Blos1 and Mstn (n = 4 independent experiments). All data represent mean ± SEM. Significance was calculated by unpaired 2-tailed Student’s t test (A–J) or 1-way ANOVA (K) with Bonferroni’s multiple-comparison test. *P < 0.05, **P < 0.01, ***P < 0.001 vs. Sh-Con or Ad-Con. Scale bars: 100 μm.
Myostatin mediates IRE1α’s regulation of muscle cell differentiation and growth. We then determined whether myostatin is required for IRE1α-mediated effect on myoblast differentiation using Mstn-KO C2C12 cells in which Mstn was abrogated by a CRISPR/Cas9–based strategy (45). Immunostaining of MyHC in differentiated myotubes showed slightly increased fusion index and myotube diameter in Mstn-KO cells relative to the WT control cells (Figure 5, A–C). Whereas knockdown of IRE1α expression resulted in lower cell fusion and smaller myotube formation in parallel with higher myostatin production (Figure 5D) in WT myoblasts, loss of myostatin diminished these effects of IRE1α deficiency in Mstn-KO cells (Figure 5, A–D). Consistently, immunoblotting and quantitative reverse transcription PCR (qRT-PCR) analyses showed that ablation of myostatin restored the protein and mRNA expression of the myogenic markers myogenin and MyHC in the face of IRE1α knockdown (Figure 5, E–G), with no significant alterations observed in Igf1 mRNA level (Figure 5G). Moreover, the promoting effects of IRE1α overexpression on muscle cell differentiation and growth were also diminished in Mstn-KO cells (Supplemental Figure 8, A–E). These data revealed the importance of myostatin suppression, at least in large part, in mediating IRE1α regulation of myocyte differentiation and growth. Together, our results support a model in which IRE1α suppresses myostatin expression, thereby promoting myoblast differentiation and myotube hypertrophy growth during skeletal muscle regeneration.
Myostatin ablation reverses the effect of IRE1α deficiency on muscle cell differentiation and growth. WT and Mstn-KO C2C12 myoblast cells were infected for 24 hours with Sh-Con or Sh-Ern1 adenoviruses and then differentiated for 3–4 days in myogenic medium. (A–C) Representative images of MyHC immunostaining (A), and quantification of the fusion index (B) and myotube diameters (C) (n = 4 independent experiments). (D) ELISA analysis of myostatin protein in culture medium (n = 3 independent experiments). (E and F) Immunoblot analysis of the indicated proteins (E), and quantification of averaged MSTN/tubulin, MyoG/tubulin, MyHC/tubulin, and p-Smad3/Smad3 levels (F) (n = 3 independent experiments). (G) Quantitative RT-PCR analysis of Xbp1 mRNA splicing and the mRNA abundance of the indicated genes (n = 4 independent experiments). All results represent mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 by 2-way ANOVA with Bonferroni’s multiple-comparison test. Ud, undetectable. Scale bars: 100 μm.
Myostatin mediates IRE1α’s regulatory action during skeletal muscle regeneration. To determine the importance of myostatin in mediating IRE1α’s regulatory effect in vivo on muscle regeneration, we generated the recombinant adeno-associated virus 2/9 (AAV2/9) expressing the N-terminal propeptide (MyoPPT) of myostatin, which is capable of inhibiting the biological activity and function of mature myostatin (46–48). Systemic delivery of MyoPPT (49, 50) to neonatal Ern1fl/fl and Ern1fl/fl Myod1-Cre mice, relative to the GFP control, resulted in efficient MyoPPT expression in their TA muscles and did not influence IRE1α’s Xbp1 mRNA splicing activity (Figure 6, A and B); and in accordance with reported findings (49, 50), blocking myostatin signaling by MyoPPT caused significant increases in their body weight and TA muscle weight in both groups (Figure 6, C and D). Importantly, at 8 days after CTX-induced injury, MyoPPT not only promoted the reparative regeneration of TA muscles of the Ern1fl/fl control group, but also prominently rescued the regenerative defect of Ern1fl/fl Myod1-Cre TA muscles, as evidenced by significantly increased TA muscle weight (Figure 6E), more newly formed myofibers containing centralized nuclei (Figure 6F), increased eMyHC expression (Figure 6G), and higher percentages of large regenerated myofibers (Figure 6H) and eMyHC-positive fibers (Figure 6I) when compared with the GFP control group (Figure 6, E–I). These results further demonstrate in vivo that myostatin is a crucial player in mediating IRE1α ablation–associated impairment of skeletal muscle regeneration upon acute injury.
Inhibition in vivo of myostatin signaling rescues the impairment of muscle regeneration resulting from IRE1α deficiency. Male Ern1fl/fl or Ern1fl/fl Myod1-Cre mice were injected intraperitoneally on postnatal day 3 (P3) and P6 with adeno-associated virus 2/9 (AAV2/9) expressing GFP (AAV-GFP) or MyoPPT (AAV-MyoPPT). TA muscles of indicated mice were injected with PBS or CTX at 7 weeks of age and analyzed at 8 days after injection. (A and B) Quantitative RT-PCR analysis of the mRNA abundance of MyoPPT (A) and Xbp1 mRNA splicing (B) in PBS-treated uninjured TA muscles (n = 6 mice per group). (C) Body weight of mice of the indicated groups. (D) Weight of PBS-treated uninjured TA muscles. (E) Weight of CTX-injured TA muscles from mice of the indicated groups (n = 5–7 mice per group). (F and G) Representative H&E staining (F) and laminin (green) and eMyHC (red) immunostaining (G) of injured TA muscles (n = 5 mice per group). (H) Percentage of regenerated myofibers in the indicated cross-sectional areas of TA muscles. Myofibers containing centralized nuclei were quantified by ImageJ from 500 myofibers in each mouse (n = 5 mice per group). (I) Quantification of the percentage of eMyHC+ myofibers within laminin staining (n = 5 mice per group). All data are presented as mean ± SEM. Significance was calculated by 2-way ANOVA with Bonferroni’s multiple-comparison test. *P < 0.05, **P < 0.01, ***P < 0.001 vs. Ern1fl/fl + AAV-GFP. #P < 0.05, ###P < 0.001 vs. Ern1fl/fl Myod1-Cre + AAV-GFP. Scale bars: 100 μm.
IRE1α suppresses myostatin expression through its RIDD activity. We next asked whether the mRNA expression of Mstn is subjected to the control of IRE1α’s RIDD activity. To determine whether Mstn mRNA represents a valid RIDD substrate, we conducted bioinformatics analysis combined with prediction of mRNA secondary structures. Indeed, we identified a highly conserved region in mammalian mRNAs encoding myostatin (spanning nucleotides 335–349 downstream of the putative transcription start site in mouse Mstn mRNA) that possesses the postulated RIDD consensus sequence (CNGCNN) (9) within a predicted hairpin structure containing a potential IRE1α cleavage site as found in Xbp1 mRNA (Figure 7A). Then, we tested whether this region of Mstn mRNA could be directly cleaved by purified recombinant IRE1α protein, using a synthetic 14-nucleotide Mstn mRNA molecule labeled with 5′-FAM fluorophore and 3′-black hole quencher (BHQ) as the substrate in an in vitro biochemical assay (Figure 7B). Incubation with IRE1α of this Mstn mRNA substrate abolished the quenching effect of BHQ, resulting in a dramatic increase in its fluorescent output signal (Figure 7B). By contrast, addition of 4μ8C, a pharmacological inhibitor of IRE1α RNase activity (51, 52), markedly reduced this cleavage-based fluorescent signal (Figure 7B). To affirm IRE1α’s action at the putative cleavage site, we generated a synthetic mutant Mstn mRNA substrate by replacing the conserved GC with CA, which completely abolished the cleavage-based fluorescent signal in comparison with its WT version (Figure 7B). Furthermore, quantitative RT-PCR analysis in HEK293T cells showed that cotransfected expression of IRE1α resulted in significantly decreased abundance of the WT but not the GC-to-CA mutant Mstn mRNA spanning the entire coding region of myostatin (Figure 7C). Interestingly, RT-PCR analyses showed that this Mstn mRNA could be immunoprecipitated along with the WT but not the truncation mutant IRE1α protein lacking its RNase domain from cotransfected HEK293T cells (Figure 7D); by contrast, deletion of the stem-loop structure of the Mstn mRNA abolished its ability to associate with IRE1α protein (Figure 7D). This indicates the structural involvement of IRE1α RNase domain and the stem-loop of the Mstn mRNA substrate in this particular RIDD machinery. We further determined whether IRE1α could affect the stability of Mstn mRNA using C2C12 myotubes following treatment with actinomycin D. Knockdown of IRE1α resulted in a decrease in the degradation rate of Mstn mRNA (Figure 7E), whereas IRE1α overexpression significantly accelerated it (Figure 7F). Collectively, these results demonstrate that Mstn mRNA could indeed serve as a bona fide RIDD substrate of IRE1α.
Myostatin mRNA is a RIDD target of IRE1α. (A) Sequence alignment of Mstn mRNAs from the indicated species with the putative conserved RIDD region, with the predicted stem-loop secondary structure shown for human and mouse Mstn mRNA. The potential IRE1α cleavage site is indicated by an arrow. (B) Fluorescence-based analysis of IRE1α cleavage of the synthetic WT or mutant (mut, GC to CA) Mstn mRNA (n = 3 independent experiments). (C) Quantitative RT-PCR analysis of the abundance of Mstn mRNA in HEK293T cells cotransfected with plasmids expressing the entire coding region of WT or mutant myostatin (GC to CA) together with empty vector or human IRE1α expression plasmid (n = 4 independent experiments). (D) HEK293T cells were cotransfected with plasmids expressing the WT or stem-loop deletion (ΔSL) mutant myostatin together with vector (Vec) control or plasmids expressing FLAG-tagged human WT or deletion mutant (KR, kinase and RNase domain; R, RNase domain) IRE1α proteins. Representative RT-PCR analysis of Mstn mRNA from immunoprecipitates (IP) using anti-FLAG antibody or from total cellular RNA (Input). Shown also are representative immunoblots of immunoprecipitated IRE1α proteins and IgG control (n = 3 independent experiments). (E and F) The stability of the Mstn mRNA was determined by quantitative RT-PCR in C2C12 myotubes infected with Sh-Con or Sh-Ern1 adenoviruses (E), or Ad-Con or Ad-IRE1α adenoviruses (F), after treatment with actinomycin D (ActD) for the indicated time intervals. Data are shown relative to the value at time 0 of ActD treatment (set as 1) after normalization to the GAPDH mRNA level as internal control (n = 3 independent experiments). All data are shown as mean ± SEM. Significance was calculated by 2-way ANOVA with Bonferroni’s multiple-comparison test. *P < 0.05, **P < 0.01, ***P < 0.001. ###P < 0.001. Scale bars: 100 μm.
We further validated that pharmacological inhibition by 4μ8C of IRE1α RNase activity in primary myoblasts was able to, in a dose-dependent fashion, suppress Xbp1 mRNA splicing and elevate the mRNA abundance of Blos1 and Mstn (Figure 8A), leading to increased activation of myostatin signaling, reduced protein expression of myogenic markers, and prominently lower myotube formation (Figure 8, B–E). Consistently, 4μ8C inhibition of IRE1α RNase activity in differentiated C2C12 myotubes also resulted in decreased Xbp1 mRNA splicing and elevated mRNA abundance of Mstn (Figure 8F), enhanced myostatin signaling and lower levels of myogenic markers (Figure 8G), and decreased diameters of myotubes (Figure 8, H and I). Together, these data clearly demonstrate that IRE1α could employ its RNase-dependent RIDD activity to blunt the mRNA expression of myostatin and exert its myogenic effect during the differentiation of myoblasts and hypertrophic growth of myotubes.
Pharmacological inhibition of IRE1α RNase activity impairs muscle cell differentiation and growth. (A–E) Primary myoblasts isolated from WT mice were cultured in growth medium (GM), and subsequently incubated in differentiation medium (DM) with or without 4μ8C for 3 days (n = 4 independent experiments). (A) Quantitative RT-PCR analysis of Xbp1 mRNA splicing and the mRNA abundance of Blos1 and Mstn. (B) Immunoblot analysis of the expression of the indicated proteins. (C) Quantification of MyoG/tubulin, MyHC/tubulin, and p-Smad3/Smad3 levels. (D) MyHC immunostaining. (E) Quantification of the fusion index. (F and G) Differentiated C2C12 myotubes were treated for 24 hours with DMSO versus 5 μM or 10 μM 4μ8C. (F) Quantitative RT-PCR analysis of Xbp1 mRNA splicing and the Mstn mRNA abundance (n = 4 independent experiments). (G) Immunoblot analysis of the indicated proteins (n = 3 independent experiments). (H and I) C2C12 myoblasts infected by EGFP-expressing adenovirus were differentiated for 4 days and likewise treated with 4μ8C (n = 4 independent experiments). (H) Representative images of myotubes by fluorescence microscopy. (I) Quantification of myotube diameters using ImageJ software. All data are shown as mean ± SEM. Significance was calculated by 1-way ANOVA with Bonferroni’s multiple-comparison test. *P < 0.05, **P < 0.01, ***P < 0.001. Scale bars: 100 μm.
Loss of IRE1α exacerbates the dystrophic phenotypes in mdx mice. DMD is the most common and severe form of muscular dystrophy, and it remains incurable to date (33, 34). The mdx mouse model of DMD exhibits extensive muscle degeneration and regeneration due to a null mutation in the Dmd gene, coding for the dystrophin protein, that causes higher sensitivity of myofibers to contractile stress (53). We wondered whether the IRE1α-myostatin regulatory circuit also plays a protective role in the setting of muscular dystrophy. To this end, we generated mdx/Ern1fl/fl Myod1-Cre mice with muscle-specific IRE1α deletion in the mdx background through genetic intercrossing. As expected, we detected significantly increased abundance of TA muscle Mstn mRNA as well as higher circulating myostatin protein in mdx/Ern1fl/fl Myod1-Cre mice relative to their mdx/Ern1fl/fl littermates (Figure 9, A and B). This was associated with elevated Smad3 phosphorylation and reduced eMyHC protein level (Figure 9, C and D). Supporting a defective IRE1α-mediated suppression of myostatin and weakened muscle regeneration, mdx/Ern1fl/fl Myod1-Cre mice at 6 weeks, as well as at 21 weeks, of age had grossly smaller gastrocnemius (Gas) and TA muscles, and the weight of these muscles was significantly lower than in their mdx/Ern1fl/fl counterparts (Figure 9, E and F, and Supplemental Figure 9A). Histological analyses revealed more severe muscle dystrophic characteristics in mdx/Ern1fl/fl Myod1-Cre animals, as indicated by higher variations in their fiber size, apparently elevated number of small fibers, and more infiltration of inflammatory immune cells as a result of degenerated myofibers (Figure 9G and Supplemental Figure 9B). Immunostaining of laminin in muscle sections showed many smaller regenerating myofibers in Gas and TA muscles of mdx/Ern1fl/fl Myod1-Cre mice at 6 or 21 weeks of age compared with their mdx/Ern1fl/fl counterparts (Figure 9H and Supplemental Figure 9C), with a significantly lower percentage of larger regenerating myofibers observed in TA muscles (Figure 9I and Supplemental Figure 9D). This indicates that loss of IRE1α results in impaired maturation of myofibers during the regenerative growth of mdx/Ern1fl/fl Myod1-Cre muscles, similar to its effect on reparative muscle growth following CTX-induced acute injury. Using Evans blue dye staining to assess muscle fiber integrity, we further detected more pronounced damage with higher dye uptake in mdx/Ern1fl/fl Myod1-Cre muscles (Figure 9, J and K). In addition, we observed no overt changes in the serum levels of creatine kinase (CK), a hallmark indicator of damaged muscles, in Ern1fl/fl Myod1-Cre mice, but found marked elevations in serum CK levels in mdx/Ern1fl/fl mice relative to their Ern1fl/fl controls (Figure 9L). Moreover, mdx/Ern1fl/fl Myod1-Cre mice displayed more pronounced increases in their serum CK levels relative to mdx/Ern1fl/fl mice (24,109 ± 3797 U/L vs. 14,437 ± 2719 U/L, n = 5 mice per genotype, P = 0.0013). To further evaluate the impact of IRE1α deficiency on muscle functionality and performance, we challenged the animals with treadmill exercise and measured the running time and distance before exhaustion. Whereas Ern1fl/fl Myod1-Cre and Ern1fl/fl mice displayed no significant difference in their exercise performance, mdx/Ern1fl/fl animals showed a significantly shorter running time and distance (Figure 9M), and mdx/Ern1fl/fl Myod1-Cre mice exhibited a more pronounced impairment in their running ability (mdx/Ern1fl/fl, 651 ± 88 m, vs. mdx/Ern1fl/fl Myod1-Cre, 364 ± 90 m; n = 5 mice per genotype, P = 0.004). These results demonstrate that ablation of IRE1α aggravates the severity of muscle dystrophy stemming from the lack of dystrophin in mdx mice, further establishing the IRE1α-myostatin axis as an important regulatory loop during regeneration in the setting of muscle dystrophy.
Loss of IRE1α exacerbates the dystrophic phenotypes in mdx mice. (A) Quantitative RT-PCR analysis of Xbp1 mRNA splicing and the mRNA abundance of Mstn in TA muscles from male mdx/Ern1fl/fl and mdx/Ern1fl/fl Myod1-Cre mice at 6 weeks of age (n = 5–6 mice per genotype). (B) ELISA analysis of serum myostatin levels (n = 5–6 mice per genotype). (C) Immunoblot analysis of TA muscle lysates using the indicated antibodies. (D) Averaged p-Smad3/Smad3 and eMyHC/tubulin ratios after normalization to the value in mdx muscle (n = 5–6 mice per genotype). (E and F) Representative pictures (E) and weight (F) of gastrocnemius (Gas) and TA muscles from mdx/Ern1fl/fl and mdx/Ern1fl/fl Myod1-Cre mice at 6 weeks of age (n = 5–6 mice per genotype). (G and H) Representative images of H&E staining of Gas and TA muscles (G) and laminin (green) immunostaining of Gas and TA muscles (H) (n = 5 mice per genotype). (I) Percentage of myofibers in the indicated cross-sectional areas of TA muscle. Quantification was conducted by ImageJ from 500 myofibers of the TA muscle from each mouse (n = 5 mice per genotype). (J and K) Analysis of Evans blue dye uptake of Gas and TA muscles (n = 5 mice per genotype). (J) Representative images of Gas and TA muscles. (K) Representative fluorescent micrographs of muscle sections. (L and M) Serum creatine kinase (CK) activity (L) and mean running time and distance on a motorized treadmill (M) for mice of the indicated genotypes at 6 weeks of age (n = 5 mice per genotype). All data are shown as mean ± SEM. Significance was calculated by unpaired 2-tailed Student’s t test (A, B, D, and F), 2-way ANOVA (I), or 1-way ANOVA (L and M) with Bonferroni’s multiple-comparison test. *P < 0.05, **P < 0.01, ***P < 0.001. Scale bars: 100 μm.
Skeletal muscle relies critically on its self-regeneration for preserving muscle mass and ensuring muscle function in response to injury and disease. Muscle regenerative capacity declines with age and is severely compromised in many muscular dystrophies (39). Delineation of the molecular mechanisms involved in regulating skeletal muscle regeneration and repair is of great importance to development of therapeutic approaches for muscle degenerative diseases. Herein, using 2 mouse models, we have uncovered a critical role for the UPR sensor IRE1α in promoting skeletal muscle regeneration in response to CTX-induced acute muscle injury or genetic muscle dystrophy. Ablation of IRE1α specifically in skeletal muscles results in impaired muscle regenerative repair or worsened progression of muscle dystrophy. Importantly, our data demonstrate that IRE1α employs its RNase-dependent RIDD action to downregulate the level of mRNA encoding myostatin, a key inhibitory regulator of muscle growth and repair. Our findings reveal a physiological function of the RIDD output of IRE1α enzyme in the control of muscle regeneration and growth (Figure 10), suggesting that IRE1α may serve as an important modifier in the disease progression of muscular dystrophy or muscle loss.
IRE1α regulates skeletal muscle regeneration through RIDD suppression of myostatin. Upon injury-induced or dystrophy-associated stress, IRE1α RNase is activated in muscle cells to decay the mRNA encoding myostatin, thereby promoting myoblast differentiation and myotube growth during the skeletal muscle regeneration response.
IRE1α is the most ancient ER stress sensor, conveying a critical signaling response through its RNase activity to generate an active transcription factor, XBP1s (3, 5). While IRE1α is best known to exert its prosurvival actions through XBP1s when coping with ER stress, emerging lines of evidence also indicate that activated IRE1α RNase activity may have a broad range of functions, as reflected by an increasing number of identified putative RIDD substrates (8, 15). However, the physiologically relevant RIDD actions of IRE1α have yet to be established in vivo. In this study, we provide physiological evidence for a RIDD-mediated mechanism by which IRE1α regulates myostatin production and regenerative maturation of myofibers in response to muscle injury or degenerative dystrophy. It is notable that ablation of IRE1α did not cause changes in muscle and myofiber sizes or myostatin expression levels in Ern1fl/fl Myod1-Cre mice in the absence of injury or dystrophy. Moreover, IRE1α phosphorylation is relatively low in muscle cells under basal conditions, but is highly induced during myoblast differentiation or upon muscle injury. This suggests that activation of the IRE1α-myostatin regulatory circuit is coupled to the stress condition induced by muscle injury or damage, conceivably reflecting the multifunctional property of IRE1α in utilizing its RIDD activity to handle various types of stress conditions in a tissue- or cell type–specific fashion.
It is worth noting that we used Myod1-Cre mice to create mice with IRE1α deletion specifically in skeletal muscles. Because of the activated expression pattern of Myod1 during myogenesis (54), IRE1α abrogation most likely occurred in muscle stem cells (i.e., satellite cells) at particular differentiation stages as well as in mature muscle fibers. Our results from analyses of muscle cell differentiation and growth in vitro suggest that IRE1α can promote both myoblast differentiation and myotube hypertrophy growth. Thus, the impairment of injury-induced regeneration in Ern1fl/fl Myod1-Cre mice can be largely attributed to the cell-autonomous action of IRE1α in myocyte differentiation and growth. This is in accordance with the fact that myostatin can regulate both differentiation in myoblasts and hypertrophy in myotubes (55). Moreover, inhibition of myostatin has been shown to improve the hallmarks of the muscular dystrophy in mdx mice (56, 57). This also supports our model that the lack of IRE1α-myostatin regulatory axis contributes to the more severe dystrophic phenotypes in mdx/Ern1fl/fl Myod1-Cre mice.
XBP1 has been recently reported to be dispensable for satellite cell–mediated muscle regeneration, whereas the PERK/eIF2α arm of the UPR may play an important role in this process (58). Documented studies have also suggested that myostatin can suppress satellite cell activation and self-renewal (43, 59, 60). While our results indicate that IRE1α deficiency caused no significant changes in the number of TA muscle satellite cells upon acute injury, it has yet to be more thoroughly interrogated whether IRE1α can regulate satellite cells through a myostatin-mediated mechanism or cell-intrinsically in an XBP1-independent manner, e.g., by affecting the STAT3 pathway (35, 61), which was reported to critically regulate the proliferation and self-renewal of muscle stem cells (62).
Myostatin, GDF11, and activin A are the 3 circulating factors that belong to the TGF-β family in the negative control of skeletal muscle mass (30). It remains to be dissected whether IRE1α also contributes in the regulation of GDF11 or activin A production. In addition, even though IRE1α did not directly, or through myostatin signaling, affect the expression of IGF1, a potent inducer of muscle hypertrophy (30), in differentiated myotubes, and XBP1s showed a merely marginal effect on myotube growth in vitro, it warrants further investigation whether IRE1α can simultaneously act to positively regulate Igf1 expression through the XBP1s pathway in vivo.
In summary, our results have unmasked a crucial role of IRE1α in regulating skeletal muscle regeneration through RIDD control of myostatin expression. Given its well-established central role in suppressing muscle regeneration and growth (30, 42, 43, 59, 60, 63–65), myostatin has been shown to represent a valuable target for improving muscular dystrophy in mouse models. In this scenario, targeted modulation of the RIDD output of IRE1α may represent a potential avenue for therapeutic intervention in DMD or other skeletal muscle degenerative diseases.
Detailed methods can be found in Supplemental Methods.
Animal studies. C57BL/6J mice were obtained from Shanghai Laboratory Animal Co. Ltd. Generation of Ern1fl/fl mice has been described previously in detail (38). Ern1fl/fl mice were intercrossed with mice expressing Cre recombinase under the control of a Myod1 promoter (The Jackson Laboratory, stock 014140) to create mice with muscle-specific deletion of IRE1α. The mdx mice were purchased from The Jackson Laboratory (stock 000476). Male Ern1fl/fl mice were first bred with homozygote female Dmdmdx/mdx mice to generate Ern1fl/+ Dmdmdx/Y males and Ern1fl/+ Dmdmdx/+ females. Male Ern1fl/+ Dmdmdx/Y mice were then bred to female Dmdmdx/mdx mice to generate Ern1fl/+ Dmdmdx/mdx females and Ern1fl/+ Dmdmdx/Y males, which were subsequently intercrossed to obtain Ern1fl/fl Dmdmdx/mdx females. Female Ern1fl/fl Dmdmdx/mdx mice were finally bred with male Ern1fl/fl Myod1-Cre mice to produce Dmdmdx/Y/Ern1fl/fl Myod1-Cre or Dmdmdx/Y/Ern1fl/fl males. Male offspring were genotyped, randomly assigned, and used at 5–21 weeks of age. Littermate controls were used in all cases, and histological analysis was performed by investigators who were blinded to the experimental groups.
Statistics. All data represent the mean SEM, with P less than 0.05 considered as a statistically significant difference. Differences between groups were tested using the unpaired 2-tailed Student’s t test, or by ANOVA with Bonferroni’s post hoc test when more than 2 groups were compared, as indicated in the figure legends. All data points were used in statistical analyses. Statistical analyses were carried out using Prism 8 software (GraphPad Software).
Study approval. All animal studies were carried out in strict accordance with the institutional guidelines for the humane treatment of animals, and all experimental protocols were approved by the IACUCs at the College of Life Sciences, Wuhan University, and the Model Animal Research Center of Nanjing University.
SH, ZG, and YL conceived and designed the studies. SH performed most of the experiments and analyzed the data. TF, YY, QL, LL, Jing Liu, XZ, QZ, QG, DX, and Yong Chen conducted some of the animal studies and cellular experiments. SH, ZG, and YL analyzed data. XW, Yulin Chen, and Jianmiao Liu provided assistance with the experimental design, reagents and data analysis. SH, ZG, and YL wrote the manuscript.
We thank Dahai Zhu, Chinese Academy of Medical Sciences, and Zhen-Guo Wu, Hong Kong University of Science and Technology, for insightful discussions. This work was supported by grants from the Ministry of Science and Technology of China (National Key R&D Program of China 2018YFA0800700 and 2016YFA0500100) to YL and ZG, and from the National Natural Science Foundation of China (31690102, 91857204, 32021003, 31922033, 91857105, 31771291, 91739303, and 32071136) to YL, ZG, Jianmiao Liu, and TF; and by the Natural Science Foundation of Jiangsu Province (BK20170014, to ZG) and Fundamental Research Funds for the Central Universities (021414380511 and 2042020kf1056, to ZG and YL).
Address correspondence to: Yong Liu, College of Life Sciences, Wuhan University, 299 Bayi Road, Wuhan 430072, China. Phone: 86.27.68753463; Email: liuyong31279@whu.edu.cn. Or to: Zhenji Gan, Model Animal Research Center of Nanjing University, 12 Xuefu Road, Pukou, Nanjing 210061, China. Phone: 86.25.58641546; Email: ganzj@nju.edu.cn.
Conflict of interest: The authors have declared that no conflict of interest exists.
Copyright: © 2021, American Society for Clinical Investigation.
Reference information: J Clin Invest. 2021;131(17):e143737.https://doi.org/10.1172/JCI143737.