Báo cáo khoa học: The myostatin-induced E3 ubiquitin ligase RNF13 negatively regulates the proliferation of chicken myoblasts pot

The myostatin-induced E3 ubiquitin ligase RNF13 negatively regulates the proliferation of chicken myoblasts Qiang Zhang 1, *, Kun Wang 2, *, Yong Zhang 1 , Jiao Meng 1 , Fang Yu 2 , Yan Chen 2 and Dahai Zhu 1 1 National Laboratory of Medical Molecular Biology, Department of Biochemistry and Molecular Biology, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, China 2 Molecular and Cellular Developmental Biology Laboratory, Harbin Institute of Technology, China Introduction In the vertebrate embryo, skeletal myogenesis initiates near the paraxial mesoderm and is regulated by a number of signaling molecules secreted by neighboring tissues [1]. Various factors, such as Pax-3, Shh, Wnt, bHLH transcription factors, MEF2 family members, transforming growth factor-b (TGF-b) and myostatin, play crucial roles in coordinately regulating the specifi- cation, proliferation and fusion of myocytes [2–4]. Myostatin was originally discovered in a screen for novel mammalian members of the TGF-b superfamily, and was identified as a negative regulator that inhib- ited myoblast proliferation and differentiation [5]. Targeted deletion of the mouse myostatin ⁄ GDF-8 gene and a naturally occurring internal deletion in the bovine myostatin gene in Belgian Blue cattle have been shown to result in larger animals with a widespread increase in skeletal muscle mass [5,6]. Similarly, block- ade of endogenous myostatin using intraperitoneal injection of blocking antibodies in the mdx muscular dystrophy mouse model results in an increase in Keywords myoblast; myostatin; RNF13; skeletal muscle; ubiquitin ligase Correspondence D. Zhu, National Laboratory of Medical Molecular Biology, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences & Peking Union Medical College, Tsinghua University, 5 Dong Dan San Tiao, Beijing 100005, China Fax: +86 10 6510 5083 Tel: +86 10 6529 6949 E-mail: dhzhu@pumc.edu.cn *These authors contributed equally to this work (Received 24 September 2009, revised 14 November 2009, accepted 17 November 2009) doi:10.1111/j.1742-4658.2009.07498.x The ubiquitin ligase RING finger protein 13 gene (RNF13) was first identi- fied in a screen for genes whose expression is regulated by myostatin in chicken fetal myoblasts. In this study, we demonstrate that the RNF13 gene is broadly expressed in many chicken tissues. The expression of RNF13 gradually decreases during skeletal myogenesis, and myostatin up-regulates RNF13 expression at both the transcriptional and translational levels. Interestingly, ectopic expression of RNF13 inhibits cell proliferation and suppresses the expression of the myogenic genes MyoD and Caveolin-3 in muscle cells. Moreover, recently, we have reported that RNF13 is a RING- type E3 ubiquitin ligase. In this report, we provide experimental evidence to show that mutations disrupting the RING finger abolish the growth- suppressive activity of RNF13, indicating that its E3 ligase activity is required for the inhibition of cell proliferation. Taken together, our find- ings show that RNF13 functions as an E3 ubiquitin ligase to negatively regulate cell proliferation. Abbreviations CFM, chicken fetal myoblast; c-RZF, chicken RING zinc finger; RNF13, RING finger protein 13; siRNA, small interference RNA; TGF-b,~transforming growth factor-b 466 FEBS Journal 277 (2010) 466–476 ª 2009 The Authors Journal compilation ª 2009 FEBS muscle mass, size and strength, together with a signifi- cant decrease in muscle degeneration [7]. A base pair alteration in the myostatin gene that reduces the level of the myostatin protein as a result of mis-splicing has also been identified in a 4-year-old child with enlarged muscle mass who developed unusual strength [8]. By contrast, systemic administration of exogenous myost- atin to adult mice induces severe muscle and fat loss, analogous to human cachexia syndromes [9]. In addi- tion, the muscle atrophy observed in chronic illnesses, HIV infection and the aging process has been associ- ated with increased expression of myostatin [10–12]. Taken together, these genetic analyses have established that the myostatin gene, which is almost exclusively expressed in skeletal muscle cells, is an important nega- tive regulator of skeletal muscle proliferation and differentiation. These findings suggest that the manipu- lation of the levels of myostatin or the downstream effectors of myostatin signaling transduction may pos- sess both therapeutic value for the treatment of muscle diseases and economic value for agricultural applica- tions [13]. Similar to members of the TGF-b superfamily, myo- statin uses Activin type IIB, Activin type IB or TGF-b type I receptors to induce the phosphorylation of intracellular Smad2 and Smad3 to exert its inhibitory function during myogenesis [14–16]. Very recently, Smad7 has been reported to be a negative feedback inhibitor of myostatin signaling [17], and the p38 mito- gen-activated protein kinase pathway has been reported to be involved in myostatin-induced transcrip- tional regulation [18]. The treatment of C2C12 mouse myoblasts with myostatin results in a block of the G1 to S transition, with an associated degradation of the cyclin D1 protein [19], an increase in the cyclin-depen- dent kinase inhibitor p21 Waf1 ⁄ Cip1 and hypophosphory- lation of the Rb protein [20]. Myostatin-inhibited myoblast differentiation is also associated with extra- cellular signal-regulated kinase 1 ⁄ 2 activation and down-regulation of MyoD, myogenin, Myf5 and Pax-3 gene expression [21–24]. Collectively, these cellular and biochemical studies suggest that myostatin signaling may possess cross-talk with the pathways regulating muscle cell proliferation and differentiation. To identify the components involved in myostatin signaling during skeletal muscle development, we car- ried out a screen in chicken fetal myoblasts (CFMs) for genes whose expression is down- or up-regulated by myostatin treatment [25]. In addition to genes known to be regulated by myostatin, including muscle creatine kinase and troponin C, this screen identified several genes that have not been linked previously to the myostatin pathway or to muscle development, such as the RING finger protein 13 gene (RNF13) and a bcl-2-related anti-apoptotic gene (Nr-13) [25]. In this study, we report that chicken RNF13 (chicken RING zinc finger, c-RZF), whose human homolog RNF13 has been shown previously to possess RING-type ubiquitin ligase activity [26], is another myostatin- induced gene in CFMs. Furthermore, our results showed that RNF13 suppresses myoblast prolife- ration in a ubiquitin ligase-dependent manner. Taken together, our findings provide new starting points for the identification of the downstream components of myostatin signaling transduction during myogenesis. Results RNF13 is induced by myostatin Previously, during the screening for genes with altered expression in CFMs in response to myostatin treat- ment [25], we used differential display PCR to identify a clone, designated 16-D, with an mRNA level that was increased by approximately 10-fold in CFMs on treatment with recombinant myostatin for 24 h (Fig. 1A). Sequence analysis revealed that clone 16-D encodes a protein with multiple functional domains, including a RING finger in the C-terminal region of the protein. Clone 16-D corresponds to the previously reported c-RZF protein, which was identified by sub- tractive hybridization for genes with expression enriched in the chicken embryo brain by binding of the substrate adhesion molecule cytotactin ⁄ tenascin [27]. In addition, our recent study demonstrated that RNF13 is a novel RING-type ubiquitin ligase and is overexpressed in pancreatic cancer [26]. To confirm the induction of RNF13 by myostatin, we isolated total RNA from CFMs treated with myo- statin at different time points and assayed the expres- sion pattern of RNF13 by northern blot analysis. As shown in Fig. 1B, the steady-state level of RNF13 mRNA was increased by two-fold after 6 h of myosta- tin treatment and by 2.75-fold after 12 h of treatment. Furthermore, western blot analysis demonstrated that treatment of CFMs with myostatin resulted in a 1.6- fold increase in the steady-state level of the RNF13 protein (Fig. 1C). Genomic structure of the RNF13 gene Sequence alignments indicated that the chicken RNF13 gene is highly conserved across species and has closely related homologs in frogs, zebrafish, mice, rats and humans (Fig. 2A). A database search also identified homologous genes in Drosophila (NP_731080, 41% Q. Zhang et al. E3 ubiquitin ligase RNF13 suppresses cell proliferation FEBS Journal 277 (2010) 466–476 ª 2009 The Authors Journal compilation ª 2009 FEBS 467 identical over 248 residues), Caenorhabditis elegans (CAA20925, 31% identical over 224 residues) and Arabidopsis (NP_975001, 32% identical over 255 resi- dues). Phylogenetic analysis indicated that RNF13 was also conserved between vertebrates and invertebrates during evolution, especially in the PA, RING finger and transmembrane domains and in the nuclear locali- zation signals (Fig. 2B). However, no obvious homolo- gous gene related to RNF13 by primary sequence has been found in fission or budding yeast. These observa- tions suggest that the RNF13 gene evolved after the origin of metazoans. To determine the genomic structure of the RNF13 gene, we screened a chicken cosmid library and isolated two positive clones. DNA sequencing analysis showed that these two clones covered the entire coding region of the RNF13 gene. The coding region of the RNF13 gene contains nine exons separated by eight in- trons that vary from 150 to 5440 bp in length (Table 1). A database search (http://genome.ucsc.edu/) indicated that the RNF13 gene is located on chicken chromosome 9, which is syntenic to human Chr3 q25.1 and mouse chr3 qD. Both the number and location of the introns in the RNF13 gene are conserved in humans, mice and chickens (Fig. 2C). RNF13 is broadly expressed in many different tissues Tranque et al. [27] have analyzed previously the expres- sion of the RNF13 ⁄ c-RZF gene in four embryonic chicken tissues (liver, brain, heart and proventriculus A BC Fig. 1. Myostatin induces RNF13 expression in CFMs. (A) Identification of clone 16-D (designated cRNF13) as a myostatin-regulated gene using differential-display PCR (DD-PCR). Clone 16-D was induced after myostatin stimulation. P and M represent NaCl ⁄ P i - and myostatin- treated CFMs, respectively. (B) Isolated CFMs were treated with 0.5 lgÆmL )1 recombinant myostatin (MSTN) or an equal amount of NaCl ⁄ P i for 6 or 12 h. Total RNA was prepared from myostatin- and NaCl ⁄ P i -stimulated CFMs and was size fractionated in pairs by electrophoresis on a 1.2% formaldehyde agarose gel. A radioactively labeled RNF13 cDNA fragment was used as a probe for northern blot hybridization. The ethidium bromide staining of 18S and 28S ribosomal RNAs was used as an equal loading control. The same experiment was performed three times for statistical analysis. RNF13 band intensities were quantified by densitometry. The fold increase in RNF13 mRNA was calcu- lated by dividing the normalized RNF13 band intensity (RNF13 ⁄ 28S rRNA) at each time point by the normalized band intensity of the NaCl ⁄ P i -treated sample at 6 h. (C) CFMs were stimulated with 0.5 lgÆmL )1 myostatin or an equal amount of NaCl ⁄ P i for 24 h. A total of 30 lg of cell lysate for each sample was separated on a 12% SDS-PAGE gel, and then transferred onto a poly(vinylidene difluoride) mem- brane for western blot analysis with the anti-RNF13 monoclonal IgG. The membrane was stripped and reprobed with an anti-tubulin IgG for equal loading detection. The fold increase in RNF13 protein was calculated by dividing the normalized RNF13 band intensity (RNF13 ⁄ tubulin) indicated in (B) by the normalized band intensity of the NaCl ⁄ P i -treated sample. E3 ubiquitin ligase RNF13 suppresses cell proliferation Q. Zhang et al. 468 FEBS Journal 277 (2010) 466–476 ª 2009 The Authors Journal compilation ª 2009 FEBS smooth muscle), and detected the expression of RNF13 mRNA in both brain and heart tissues. In our previous report, the expression of RNF13 was found in all human tissues studied [26]. In order to gain an insight into RNF13 function in chickens, we isolated both total RNA and protein from skeletal muscle, cardiac muscle, liver, kidney, intestine, proventriculus, spleen and brain of newly hatched chickens, and measured the steady- state expression of RNF13 in terms of both mRNA and protein by northern and western blot analysis, respec- tively (Fig. 3A). RNF13 gene expression was lowest, but still readily detectable, in the liver, with moderate expression in the heart, intestine and spleen, and high expression in skeletal muscle, kidney, proventriculus A B C Fig. 2. Conservation analysis of RNF13 homologs. (A) Amino acid sequences of RNF13 and its homologs in humans, mice, rats, Xenopus and zebrafish were obtained from GENBANK and aligned using CLUSTALX software. (B) Amino acid sequence-based phylogenetic trees for RNF13 and homologous proteins were constructed with MEGA version 2.1 based on P distance. (C) Genomic organization of the RNF13 gene. Exons are numbered and indicated by filled black boxes in the RNF13 gene and homologous genes (RNF13) in humans and mice. Introns and noncoding regions are indicated by lines. Table 1. Genomic organization of the cRNF13 gene and the donor– acceptor junctions. Exon Exon size (bp) 5¢ splice donor Intron size (bp) 3¢ splice acceptor 159 CAGgtgagggagc 4815 atattaaaagATG 2 113 GCGgtaagtgtaa 3809 tttcttgcagTAT 382 AAGgtattttata 2137 tgtttttcagGGA 4 126 AAGgtaatacaca 2723 tggcttacagGTT 588 ACAgtaagtacag 3474 ttttatacagTTG 691 GGGgtaagttacc 150 ctctctacagTGG 7 106 ATGgtaagttcct 2708 atgcttccagATC 894 AAGgtaagtggat 5440 tgtgctgtagGAG 981 ATGgtaagcagct 660 tttctaacagCGT 10 365 TGAgactactcat Q. Zhang et al. E3 ubiquitin ligase RNF13 suppresses cell proliferation FEBS Journal 277 (2010) 466–476 ª 2009 The Authors Journal compilation ª 2009 FEBS 469 A B C D Fig. 3. Temporal and spatial expression patterns of RNF13. (A) Spatial expression pattern of RNF13. Newly hatched male White Leghorn chicks were maintained on a standard diet. Total RNA and protein from the pectoralis muscle (Sm), cardiac muscle (H), liver (Li), kidney (K), intestine (I), proventriculus (P), spleen (Sp) and brain (B) were isolated and analyzed by northern and western blots, respectively. (B) Expression pattern analysis of RNF13 during skeletal muscle development. The pectoralis muscles of White Leghorn chickens were obtained from different developmental stages (10-, 12-, 14-, 16- and 18-day embryos, as well as from chicks 1 day and 1, 2, 3, 5 and 7 weeks after hatching). The transcript and protein levels of RNF13 were determined by northern and western analysis, respectively. The ethidium bromide staining of 18S and 28S ribosomal RNAs and immunoblotting of tubulin were used as equal loading controls. (C) Morphology of CFMs cultured in vitro for 1, 3 and 5 days. (D) RNF13 expression decreased during in vitro myogenesis. CFMs were isolated and maintained in culture medium for 1, 3 and 5 days to induce spontaneous differentiation. The transcript and protein levels of RNF13 were determined during in vitro myogenesis. Immunoblots for myosin heavy chain (MHC) and tubulin were used as a myogenic differentiation marker and an equal loading control, respectively. E3 ubiquitin ligase RNF13 suppresses cell proliferation Q. Zhang et al. 470 FEBS Journal 277 (2010) 466–476 ª 2009 The Authors Journal compilation ª 2009 FEBS and brain. Broad expression of the RNF13 gene in all eight tissues tested was confirmed by western blot anal- ysis. Our results indicate that the RNF13 gene is broadly expressed in many different tissues, and that the function of RNF13 may not be restricted to muscle cells, which is consistent with its presence in plants, worms and flies. RNF13 is down-regulated during skeletal muscle development As a myostatin-induced gene, the expression pattern of RNF13 during chicken skeletal myogenesis was analyzed at both the mRNA and protein levels. Pectoralis muscles were collected from chickens at different developmental stages. Northern blot analysis indicated that the RNF13 gene was highly expressed in embryonic skeletal muscle, whereas its expression was dramatically decreased around hatching and was almost undetectable from 1 to 7 weeks after hatching. Consistently, the RNF13 protein was persistently expressed in skeletal muscle tissue dur- ing embryonic development and for 2 weeks after hatch- ing, but was completely undetectable after 3 weeks of postnatal development (Fig. 3B). Primary and second- ary myogenesis occur during embryonic development, and muscle satellite cell activation is the major biologi- cal event during postnatal myogenesis. Our data suggest that RNF13 expression may be associated with primary and secondary myogenesis during embryonic develop- ment. Isolated CFMs can spontaneously differentiate from myoblasts into myotubes under normal culture conditions, providing an in vitro myogenesis system (muscle cell proliferation and differentiation) in which to test this possibility. CFMs were isolated and cultured for 1, 3 and 5 days, and their differentiation was con- firmed by morphological examination (Fig. 3C). Total RNA and protein were extracted from proliferating myoblasts and differentiated myotubes, and analyzed for RNF13 expression by northern and western blot analysis, respectively. Both RNF13 mRNA and protein levels decreased in myocytes during in vitro differentia- tion (Fig. 3D). Thus, as determined during both in vivo and in vitro myogenesis, RNF13 expression was high in proliferating myoblasts, decreased during myogenesis and was undetectable in differentiated skeletal muscles. Overexpression of RNF13 inhibits cell proliferation, and RNF13 E3 ligase activity is required for this inhibition We next tested the functional role of RNF13 during myoblast proliferation in CFMs. At 36 h after trans- fection with plasmids expressing sense RNF13 (pS-RNF13), antisense RNF13 (pAS-RNF13) and RNF13 small interference RNA (siRNA), cell numbers were counted. Ectopic expression of RNF13, but not of its antisense control or siRNA, reduced the number of cells by approximately 50% (Fig. 4A, B). To con- firm the growth-suppressive activity of RNF13,we evaluated DNA replication by measuring [ 3 H]thymi- dine incorporation after ectopic expression of RNF13. As shown in Fig. 4C, an almost 30% reduction in the [ 3 H]thymidine incorporation rate was evident in the cells transfected with wild-type RNF13 compared with the cells transfected with the antisense and control vec- tors, indicating that the decreased cell number follow- ing overexpression of RNF13 was most probably a result of the inhibitory effects of RNF13 on CFM pro- liferation, as opposed to increased cell death. We have recently reported that RNF13 is a RING- type E3 ligase, and its C-terminal RING finger domain is essential for its ubiquitin ligase activity [26]. Therefore, we investigated whether the E3 ligase activity of RNF13 is required for its growth-suppressive activity. We ectopi- cally expressed an RNF13 RING domain deletion mutant, RING D , and two point mutants, C258A ⁄ H260A and W270A, in CFMs and compared their ability to inhi- bit cell proliferation relative to wild-type RNF13 by mea- suring [ 3 H]thymidine incorporation. The three mutants lacked the growth-suppressive activity of wild-type RNF13 (Fig. 4C), suggesting that an intact RING finger is required for RNF13 inhibition of CFM proliferation, probably as a result of the E3 ligase activity of this domain. The E3 ligase activity of RNF13 was then assayed in CFMs, and the results presented in Fig. 4D show that wild-type RNF13 possesses higher E3 ligase activity than the two mutant forms of full-length RNF13. These findings provide experimental evidence to support the concept that the antiproliferative activity of RNF13 is dependent on its E3 ubiquitin ligase activity. Given that RNF13 is widely expressed in many dif- ferent tissues, we tested whether its inhibition of prolif- eration was muscle specific or general. NIH ⁄ 3T3 fibroblasts and KYSE180 esophageal cancer cells were used as nonmuscle cells for the RNF13 functional assay. As shown in Fig. 4E, RNF13 inhibited the pro- liferation of both NIH ⁄ 3T3 and KYSE180 cells, and the results strongly suggest that RNF13 may have a general functional role in controlling cell proliferation. Expression of MyoD and Caveolin-3 decreased in RNF13-overexpressing CFMs Very little is known about the molecular mecha- nisms underlying myostatin function in the negative regulation of both muscle cell differentiation and pro- Q. Zhang et al. E3 ubiquitin ligase RNF13 suppresses cell proliferation FEBS Journal 277 (2010) 466–476 ª 2009 The Authors Journal compilation ª 2009 FEBS 471 liferation. The suppressive effect of RNF13 overexpres- sion on myoblast proliferation and the up-regulation of RNF13 by myostatin suggest that myostatin inhibits myoblast proliferation, in part by up-regulating RNF13 expression in skeletal muscle tissue. Therefore, we examined whether the expression of two important myogenic regulatory factors, MyoD and caveolin-3, was regulated by RNF13 in CFMs. As shown in Fig. 4F, the ectopic expression of RNF13 significantly decreased the expression of both MyoD and Caveolin-3, AB CD EF Fig. 4. Forced expression of RNF13 revealed anti-proliferative effects on CFMs, which required the E3 ligase activity of RNF13. (A) CFMs were transfected by electroporation with the control vector pcDNA3, the overexpression vector pS-RNF13, the antisense vector pAS-RNF13, with cRNF13 siRNA or with scramble siRNA, and were allowed to grow for 24 or 48 h. Each transfection was repeated three times indepen- dently, and the number of cells was counted. (B) Proliferating CFMs were transfected with pcDNA3 (control vector), pS-RNF13, pAS-RNF13, cRNF13 siRNA or scramble siRNA for 48 h. (C) The E3 ligase activity of RNF13 was required for the inhibition of CFM proliferation. CFMs were transfected with pcDNA3, pS-RNF13, C258A ⁄ H260A, W270A, RING D , pAS-RNF13, cRNF13 siRNA or scramble siRNA independently, and [ 3 H]thymidine incorporation was measured as indicated in the Materials and methods section. (D) CFMs were co-transfected with an HA-ubiquitin expression vector and a control vector, Myc-RNF13 (W.T., pS-RNF13), Myc-RNF13 (C258A ⁄ H260A) or Myc-RNF13(W270A), and were cultured for another 48 h. Myc-tagged RNF13 proteins were immunoprecipitated with an anti-Myc IgG and analyzed by western blot with an anti-HA IgG. (E) NIH ⁄ 3T3 cells and KYSE180 human esophageal cancer cells were transfected with empty vector and Myc-RNF13 (W.T.) and cultured for another 24 h. For the proliferation assay, the transfected cells were cultured for 6 h in the presence of [ 3 H]thymidine, and the [ 3 H]thymidine incorporation assay was then performed as described in the Materials and methods section. Ectopically expressed RNF13 proteins were detected with an anti-Myc IgG. (F) At 24 h after transfection with the indicated plasmids, total RNA was prepared and the expression levels of MyoD, Caveolin-3 and RNF13 were measured using semi-quantitative RT-PCR. G3PDH was used as an internal control. E3 ubiquitin ligase RNF13 suppresses cell proliferation Q. Zhang et al. 472 FEBS Journal 277 (2010) 466–476 ª 2009 The Authors Journal compilation ª 2009 FEBS indicating that high levels of RNF13 expression inhibit cell proliferation by decreasing myogenic gene expres- sion. These results suggest that RNF13 is a potential mediator of myostatin function in the negative regula- tion of muscle cell proliferation. Discussion It has been well documented that myostatin, a TGF-b superfamily member, is a negative regulator of skeletal muscle growth [5]. Compared with its well-established physiological function in controlling muscle growth, the molecular mechanisms of myostatin action remain elusive. We used CFMs as a model system to identify several genes with altered expression in myostatin-stim- ulated cells, including those related to myogenic differ- entiation, transcriptional regulation and apoptosis [25]. In this article, we have provided evidence that the E3 ubiquitin ligase RNF13 plays a role in the negative regu- lation of cell proliferation. Developmental expression analysis showed that the levels of RNF13 mRNA and protein decreased in skeletal muscle during myogenesis both in vivo and in vitro. Most importantly, overexpres- sion of RNF13 inhibited the proliferation of CFMs with a corresponding decrease in the expression of two repre- sentative myogenic genes, MyoD and Caveolin-3, and disruption of the RING finger of RNF13 abolished its ability to inhibit cell proliferation. Given that RNF13, like myostatin, negatively regulates muscle cell prolifera- tion and inhibits myogenic gene expression, RNF13 may function as a downstream effector of the myostatin signaling pathway. During myogenesis, specific patterns of gene expres- sion usually suggest potential roles in myocyte prolifer- ation and ⁄ or differentiation. We showed that RNF13 is highly expressed in skeletal muscle during chicken embryo development, but is down-regulated during postnatal myogenesis. Primary and secondary myogen- esis occur during embryonic development, and our data suggest that RNF13 expression may be positively associated with primary and secondary myogenesis during embryonic development in terms of the regula- tion of skeletal muscle cell proliferation. The results of our in vitro myogenesis experiments using CFMs pro- vide additional evidence to support this concept. Taken together, this specific pattern of RNF13 expres- sion suggests that RNF13 plays different roles in the regulation of myocyte proliferation and differentiation during embryonic and postnatal development. Recently, the expression of RNF13 has been reported to increase in adult mouse tissues compared with embryonic mouse tissues, and has been reported to be up-regulated in B35 neuroblastoma cells stimulated to undergo differentiation [28]. Different expression pat- terns in skeletal muscle and brain suggest that RNF13 has specific functions during development. At the same time, it should be noted that, unlike myostatin, which is predominantly expressed in skeletal muscle cells and is not present in C. elegans, plants or flies, the RNF13 gene is broadly expressed in many other tissues and is evolutionarily conserved among metazoans. Therefore, we tested the functional role of RNF13 in the regula- tion of cell proliferation in nonmuscle cells, and our results demonstrate that RNF13 inhibits the prolifera- tion of NIH ⁄ 3T3 fibroblast cells and KYSE180 human esophageal cancer cells, suggesting that RNF13 may have a general role in controlling cell proliferation, rather than acting specifically in muscle cells. This hypothesis is consistent with the broad expression pat- tern of RNF13 in many tissues and its evolutionary conservation in both vertebrates and invertebrates. To our knowledge, this is the first demonstration that RNF13 and its E3 ubiquitin ligase activity are required for the inhibition of myoblast proliferation, and is the first identification of a myostatin function that may involve a nonmuscle-specific E3 ligase. The key question concerning the mechanism and function of the myostatin–RNF13 pathway that remains to be answered involves the substrate(s) of RNF13 E3 ligase. In addition to the RING finger, RNF13 contains a PA domain whose function has yet to be firmly estab- lished, although it has been implicated in the media- tion of protein–protein interactions [29,30]. Therefore, we speculate that RNF13, similar to other RING E3 ligases, recruits specific proteins via the PA domain and presents them to the C-terminally situated RING finger for ubiquitination. Therefore, the identification of the protein(s) that interact(s) with RNF13, in partic- ular with the PA domain, will be important for the understanding of both the mechanism of action of RNF13 and the myostatin signaling pathway during muscle growth, as well as for the possible clarification of the function of RNF13 in other cellular processes. Materials and methods Cell culture and treatment CFMs were prepared from the pectoralis muscles of 10-day- old White Leghorn chicken embryos, and were cultured in Dulbecco’s modified Eagle’s medium as described previously [25]. At 12–18 h post-plating, CFMs were treated with puri- fied recombinant myostatin, as described previously [31], at a final concentration of 0.5 lgÆmL )1 for the indicated amounts of time. For myogenic differentiation experiments, CFMs were cultured for 5 days, and the cells were collected at days Q. Zhang et al. E3 ubiquitin ligase RNF13 suppresses cell proliferation FEBS Journal 277 (2010) 466–476 ª 2009 The Authors Journal compilation ª 2009 FEBS 473 1, 3 and 5 for RNA and protein extraction. COS-7 and NIH ⁄ 3T3 cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum. KYSE180 cells were cultured in 1640 medium supplemented with 10% fetal bovine serum. Animals and tissue collection Male adult White Leghorn chickens were maintained on a standard diet in accordance with the school’s Animal Care and Use Committee policy, and were killed by cervical dis- location. Various tissues (pectoralis muscle, heart, liver, kidney, intestine, proventriculus, spleen and brain) were collected from the chickens and were stored at )70 °C for RNA and protein isolation. In addition, the pectoralis mus- cles were obtained from chickens in different developmental stages (10-, 12-, 14-, 16- and 18-day embryos, as well as from chicks 1 day and 1, 2, 3, 5 and 7 weeks after hatch- ing), and were stored at )70 °C until use. Plasmid construction RNF13 cDNA was PCR amplified with the forward primer p574 (5¢-ATAAGAATGCGGCCGCAATGGTGCTGTCA ATAGGAATGCTG-3¢), which includes a NotI site, and the reverse primer p575 (5¢-TGCTCTAGAATTCATTCAC TTCTCGTCATGAGT-3¢), which includes a XhoI site. The PCR products were subsequently digested and cloned into the pcDNA3.0 vector in the sense direction to express RNF13; this vector is referred to as pS-RNF13. The anti- sense RNF13 construct was PCR cloned into the pcDNA3.0 vector with the forward primer p576 (5¢-ATA AGAATGCGGCCGCAATTCATTCACTTCTCGTCATG AGT-3¢) and the reverse primer p577 (5¢-TGCTCTAGA ATGGTGCTGTCAATAGGAATGCTG-3¢), containing a NotI site and a XhoI site, respectively, and this vector is referred to as pAS-RNF13. For the construction of a Myc- tagged RNF13 expression plasmid (W.T.), RNF13 cDNA was PCR cloned into the pcDNA4.0-Myc-His vector at the EcoRI and XhoI sites with the forward primer p10 (5¢- GGAATTCATATGCTGCTGTCAATAGGAATGCTG-3¢) and the reverse primer p154 (5¢-CCGCTCGAGAACAGT ATTTGTCACTCTGTAATCCCT-3¢). For the construction of the expression plasmid for the Myc-tagged RNF13 RING domain deletion mutant (RING D ), a DNA fragment containing the PA domain was cloned into the pcDNA4.0- Myc-His vector at the EcoRI and XhoI sites with the forward primer p10 (see above) and the reverse primer p155 (5¢-CCGCTCGAGCTCCAAAGGAAGACTGAACT CTGGGAT-3¢). Two point mutations (C258A ⁄ H260A and W270A) in the RING finger domain of RNF13 were introduced by PCR overlap extension with the external primers p10 ⁄ p154 (see above). To generate the C258A ⁄ H260A mutation, the internal primers p161 (5¢-CTCAGAATCCTTCCATGC(GC C)TCTCAT(GCT)GCGTATCACTGCAAG-3¢) and p162 (5¢-GCAGTGATACGCATG(AGC)AGAGCA(GGC)TGG AAGGATTCTGAGCTT-3¢) were used, and to generate the W270A mutation, the internal primers p157 (5¢-GCAA GTGCGTGGACCCATGG(GCG)CTGACAAAAACA-3¢) and p158 (5¢-GTTTTTGTCAGCCA(CGC)TGGGTCCAC GCACTTGCA-3¢) were used. The final PCR products were cloned into pcDNA4.0-Myc-His and were confirmed by sequencing the entire RNF13 cDNA using standard DNA sequencing procedures (ABI 377 instrument, Perkin-Elmer, Waltham, MA, USA). Transfection and proliferation assay CFM cells (2 · 10 5 ) were transiently transfected by electro- poration [350 V, 16 ms, one pulse using an electroporator (Bio-Rad, Hercules, CA, USA)] with 10 lg of plasmid DNA. NIH ⁄ 3T3 and KYSE180 cells were transfected with LipofectamineÔ 2000 (Invitrogen, Carlsbad, CA, USA) using 10 lg of plasmid DNA. The transfected cells were seeded in six-well plates and were maintained in Dulbecco’s modified Eagle’s medium for 36 h. During the last 6 h, 5 lCi of [ 3 H]thymidine was added to the culture medium. The cells were washed twice with NaCl ⁄ P i and were trypsi- nized to prepare the cells for radioactivity measurements in scintillation vials. The assay was performed in triplicate and was repeated three times. Expression, purification and production of a monoclonal antibody against RNF13 A C-terminally truncated RNF13 (Arg215–Val381) construct containing the RING finger domain was PCR cloned with the primers p436 (5¢-CGGAATTCCGTGCAAGACAGAC ACAGGGCAAGAAGG-3¢) and p437 (5¢-CCCAAGCTTT CAATCCCTTTCATCATTGGGCTGGAGCTG-3¢), which contained an EcoRI site and a HindIII site, respectively. The PCR product was purified, digested and ligated into pET28b (Novagen, Darmstadt, Germany), confirmed by sequencing and introduced into Escherichia coli strain BL21 (DE3). The recombinant His6-tagged truncated RNF13 protein was expressed and purified using nickel nitrilotriacetic acid agarose (Qiagen, Hilden, Germany), according to the manu- facturer’s protocol. For monoclonal antibody production, 5–6-week-old female BALB ⁄ C mice were immunized with the purified recombinant protein using a standard procedure. The culture supernatant or ascites from hybridomas were used for western blot analysis. Genomic cloning and sequence analysis of the RNF13 gene Six positive clones were isolated from a genomic chicken cosmid library (Clontech, Mountain View, CA, USA) E3 ubiquitin ligase RNF13 suppresses cell proliferation Q. Zhang et al. 474 FEBS Journal 277 (2010) 466–476 ª 2009 The Authors Journal compilation ª 2009 FEBS according to the manufacturer’s standard procedure, and two of these clones were used as templates for DNA sequencing using primer walking to obtain a genomic clone. Sequence data were analyzed and assembled with dnastar software (http://www.dnastar.com/), and the genomic sequences were submitted to GenBank (accession number: AY787020). Amino acid alignments were analyzed by clustalx. For phylogenetic analysis, a neighbor-joining tree was constructed with mega version 2.1 based on the P distance. We used 1000 bootstrap replications to estimate the support rate. RNA extraction and northern blot analysis Total RNA was isolated from CFMs and various chicken tissues with TRIzol Reagent (Invitrogen), as described pre- viously [25]. A total of 15 lg of RNA from each sample was analyzed by northern blot analysis using [a- 32 P]dCTP RNF13 cDNA, labeled using a random-priming labeling kit (Amersham Pharmacia Biotech, Piscataway, NJ, USA), as described previously [25]. The radioactive signals were visualized using X-ray film (Kodak). Semi-quantitative RT-PCR RNA was isolated from transfected cells and reverse tran- scribed at 42 °C for 1 h in a 10 lL reaction mixture with SuperScript IIÔ reverse transcriptase (Invitrogen). We used 100 ng of first-strand cDNA as a template for PCR amplifi- cation of MyoD, Caveolin-3, RNF13 and G3PDH with the following primers: MyoD (p66, 5 ¢-AGTCGCCCCCATGG ACTTACT-3¢;p67,5¢-TTATAGCACTTGGTAGATTGG-3 ¢), Caveolin-3 (p207, 5¢-ATGGCTGAGGAGCAGAGAGAG CTGGAG-3¢; p208, 5¢-GTGGTTGTGGTGCTGCTGGGA TTTAGG-3¢), RNF13 (p10, 5¢-ATGCTGCTGTCAATAG GAATGCTG-3¢; p11, 5¢-ATTCATTCACTTCTCGTCATG AGT-3¢) and G3PDH (p243, 5¢-CACCAACTGGGACGA CAT-3¢; p244, 5¢-CATACTCCTGCTTGCTGATC-3¢). A 50 lL PCR mixture was subjected to one cycle of 94 °C for 1 min, 25 cycles of 94 °C for 30 s, 58 °C for 30 s and 72 °C for 1 min, and one cycle of 72 °C for 7 min, in a model PE9700 thermocycler from Perkin-Elmer. The PCR prod- ucts were separated on a 0.8% agarose gel and were ana- lyzed using quantity one software (Bio-Rad). Western blot analysis The CFM cells were lysed in lysis buffer (50 mm Tris ⁄ HCl, pH 7.5, 150 mm NaCl, 0.5% Nonidet P40, 50 mm NaF, 1mm Na 3 VO 4 ,5mm b-glycerophosphate, 1 mm dithiothre- itol, 1 mm phenylmethylsulfonyl fluoride). Equal amounts of total protein were separated by 12% SDS-PAGE, trans- ferred onto a poly(vinylidene difluoride) membrane and probed with either anti-RNF13 (ascites at 1 : 5000 dilution) or anti-tubulin (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) IgG. The proteins were visualized using an ECL detection system (Amersham Biosciences). Acknowledgements This work was supported by grants from the National Basic Research Program of China (2005CB 522405, 2005CB522505, 2007CB946903, 2009CB 941602, 2009CB825403), the National Natural Science Foundation of China (30721063) and the Chinese National Programs for High Technology Research and Development (2006AA10A121, 2007AA02Z109). We greatly appreciate the assistance of Dr Yue Xiong (University of North Carolina at Chapel Hill), Dr Yeguang Chen (Tsinghua University) and Dr Jian Zhang (Institute of Genetics and Developmental Biol- ogy, Chinese Academy of Science) during the prepara- tion of the manuscript. References 1 Cossu G, Tajbakhsh S & Buckingham M (1996) How is myogenesis initiated in the embryo? Trends Genet 12, 218–223. 2 Black BL & Olson EN (1998) Transcriptional control of muscle development by myocyte enhancer factor-2 (MEF2) proteins. Annu Rev Cell Dev Biol 14, 167–196. 3 Buckingham ME (1994) Muscle: the regulation of myo- genesis. Curr Opin Genet Dev 4, 745–751. 4 Liu D, Kang JS & Derynck R (2004) TGF-b-activated Smad3 represses MEF2-dependent transcription in myo- genic differentiation. EMBO J 23, 1557–1566. 5 McPherron AC, Lawler AM & Lee SJ (1997) Regula- tion of skeletal muscle mass in mice by a new TGF-beta superfamily member. Nature 387, 83–90. 6 Grobet L, Martin LJ, Poncelet D, Pirottin D, Brouwers B, Riquet J, Schoeberlein A, Dunner S, Me ´ nissier F, Massabanda J et al. (1997) A deletion in the bovine myostatin gene causes the double-muscled phenotype in cattle. Nat Genet 17, 71–74. 7 Bogdanovich S, Krag TO, Barton ER, Morris LD, Whittemore LA, Ahima RS & Khurana TS (2002) Functional improvement of dystrophic muscle by myo- statin blockade. Nature 420, 418–421. 8 Schuelke M, Wagner KR, Stolz LE, Hu ¨ bner C, Riebel T, Ko ¨ men W, Braun T, Tobin JF & Lee SJ (2004) Myosta- tin mutation associated with gross muscle hypertrophy in a child. N Engl J Med 26, 2682–2688. 9 Zimmers TA, Davies MV, Koniaris LG, Haynes P, Esquela AF, Tomkinson KN, McPherron AC, Wolfman NM & Lee SJ (2002) Induction of cachexia in Q. Zhang et al. E3 ubiquitin ligase RNF13 suppresses cell proliferation FEBS Journal 277 (2010) 466–476 ª 2009 The Authors Journal compilation ª 2009 FEBS 475 [...]... protein 13 is an endosomal integral membrane E3 ubiquitin ligase whose RING finger domain is released to the cytoplasm by proteolysis FEBS J 276, 1860–1877 Mahon P & Bateman A (2000) The PA domain: a protease-associated domain Protein Sci 9, 1930–1934 Luo X & Hofmann K (2001) The protease-associated domain: a homology domain associated with multiple classes of proteases Trends Biochem Sci 26, 147–148... Zhu D (2009) RNF13: a novel RING-type ubiquitin ligase overexpressed in pancreatic cancer Cell Res 19, 348–357 Tranque P, Crossin KL, Cirelli C, Edelman GM & Mauro VP (1996) Identification and characterization of a RING zinc finger gene (C-RZF) expressed in chicken embryo cells Proc Natl Acad Sci USA 93, 3105–3109 Bocock JP, Carmicle S, Chhotani S, Ruffolo MR, Chu H & Erickson AH (2009) The PA-TM-RING... myostatinmediated inhibition of myogenesis Dev Biol 270, 19–30 Amthor H, Huang R, McKinnell I, Christ B, Kambadur R, Sharma M & Patel K (2002) The regulation and action of myostatin as a negative regulator of muscle development during avian embryogenesis Dev Biol 251, 241–257 Yang W, Zhang Y, Ma G, Zhao X, Chen Y & Zhu D (2005) Identification of gene expression modifications in myostatin-stimulated myoblasts Biochem.. .E3 ubiquitin ligase RNF13 suppresses cell proliferation 10 11 12 13 14 15 16 17 18 19 20 476 Q Zhang et al mice by systemically administered myostatin Science 296, 1486–1488 Reardon KA, Davis J, Kapsa RM, Choong P & Byrne... R, Asa S, Mamita M et al (1998) Organization of the human myostatin gene and expression in healthy men and HIV-infected men with muscle wasting Proc Natl Acad Sci USA 95, 14938–14943 Ivey FM, Roth SM, Ferrell RE, Tracy BL, Lemmer JT, Hurlbut DE, Martel GF, Siegel EL, Fozard JL, Jeffrey Metter E et al (2000) Effects of age, gender, and myostatin genotype on the hypertrophic response to heavy resistance... Biol 23, 7230–7242 Zhu X, Topouzis S, Liang LF & Stotish RL (2004) Myostatin signaling through Smad2, Smad3 and Smad4 is regulated by the inhibitory Smad7 by a negative feedback mechanism Cytokine 26, 262–272 Philip B, Lu Z & Gao Y (2005) Regulation of GDF-8 signaling by the p38 MAPK Cell Signal 17, 365–375 Yang W, Zhang Y, Li Y, Wu Z & Zhu D (2007) Myostatin induces cyclin D1 degradation to cause cell... (2004) Regulation of muscle mass by myostatin Annu Rev Cell Dev Biol 20, 61–86 Sharma M, Kambadur R, Matthews KG, Somers WG, Devlin GP, Conaglen JV, Fowke PJ & Bass JJ (1999) Myostatin, a transforming growth factor-beta superfamily member, is expressed in heart muscle and is upregulated in cardiomyocytes after infarct J Cell Physiol 180, 1–9 Lee SJ & McPherron AC (2001) Regulation of myostatin activity... proteases Trends Biochem Sci 26, 147–148 Yang W, Wang K, Chen Y, Zhang Y, Huang B & Zhu D (2003) Functional characterization of recombinant myostatin and its inhibitory role to chicken muscle development Acta Biochim Biophys Sin 35, 1016–1022 FEBS Journal 277 (2010) 466–476 ª 2009 The Authors Journal compilation ª 2009 FEBS ... beta pathway and is antagonized by insulin-like growth factor 1 J Biol Chem 282, 3799– 3808 Thomas M, Langley B & Berry C (2000) Myostatin, a negative regulator of muscle growth, functions by 21 22 23 24 25 26 27 28 29 30 31 inhibiting myoblast proliferation J Biol Chem 275, 40235–40243 Yang W, Chen Y, Zhang Y, Wang X, Yang N & Zhu D (2006) Extracellular signal-regulated kinase 1 ⁄ 2 mitogen-activated . result of the E3 ligase activity of this domain. The E3 ligase activity of RNF13 was then assayed in CFMs, and the results presented in Fig. 4D show that wild-type RNF13 possesses higher E3 ligase activity. The myostatin-induced E3 ubiquitin ligase RNF13 negatively regulates the proliferation of chicken myoblasts Qiang Zhang 1, *, Kun Wang 2, *, Yong. nonmuscle-specific E3 ligase. The key question concerning the mechanism and function of the myostatin RNF13 pathway that remains to be answered involves the substrate(s) of RNF13 E3 ligase. In addition to the