Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You and Your Team.

Learn More →

Trip12, a HECT domain E3 ubiquitin ligase, targets Sox6 for proteasomal degradation and affects fiber type-specific gene expression in muscle cells

Trip12, a HECT domain E3 ubiquitin ligase, targets Sox6 for proteasomal degradation and affects... Background: A sophisticated level of coordinated gene expression is necessary for skeletal muscle fibers to obtain their unique functional identities. We have previously shown that the transcription factor Sox6 plays an essential role in coordinating muscle fiber type differentiation by acting as a transcriptional suppressor of slow fiber-specific genes. Currently, mechanisms regulating the activity of Sox6 in skeletal muscle and how these mechanisms affect the fiber phenotype remain unknown. Methods: Yeast two-hybrid screening was used to identify binding partners of Sox6 in muscle. Small interfering RNA (siRNA)-mediated knockdown of one of the Sox6 binding proteins, Trip12, was used to determine its effect on Sox6 activity in C2C12 myotubes using quantitative analysis of fiber type-specific gene expression. Results: We found that the E3 ligase Trip12, a HECT domain E3 ubiquitin ligase, recognizes and polyubiquitinates Sox6. Inhibiting Trip12 or the 26S proteasome activity resulted in an increase in Sox6 protein levels in C2C12 myotubes. This control of Sox6 activity in muscle cells via Trip12 ubiquitination has significant phenotypic outcomes. Knockdown of Trip12 in C2C12 myotubes led to upregulation of Sox6 protein levels and concurrently to a decrease in slow fiber-specific Myh7 expression coupled with an increased expression in fast fiber-specific Myh4. Therefore, regulation of Sox6 cellular levels by the ubiquitin-proteasome system can induce identity-changing alterations in the expression of fiber type-specific genes in muscle cells. Conclusions: Based on our data, we propose that in skeletal muscle, E3 ligases have a significant role in regulating fiber type-specific gene expression, expanding their importance in muscle beyond their well-established role in atrophy. Keywords: Sox6, Skeletal muscle, Fiber type differentiation, Trip12, E3 ubiquitin ligase, HECT domain, Ubiquitin- proteasome system Background type [4], culminating in the specific biochemical charac- Mammalian skeletal muscles consist of functionally het- teristics of either fast fiber or slow fiber type. Despite its erogeneous myofibers, which can be broadly classified significance to muscle physiology and muscle degenera- into two groups, slow-twitch and fast-twitch fibers. They tive diseases (e.g., exercise physiology [5]; Duchenne differ in contraction speed, metabolic capacity, fatigue muscular dystrophy [6,7]; inflammatory atrophy [8]), resistance, sensitivity to calcium, and a variety of other knowledge regarding the molecular mechanisms orches- attributes [1-3]. This physiologic diversity among muscle trating this coordinated expression exists only in vague fibers is the outward manifestation of the concerted outlines. expression of hundreds of genes unique to each fiber Our recent studies demonstrated that the transcription factor Sox6 directly suppresses transcription of slow fiber-specific genes during mouse muscle development * Correspondence: nhagiwara@ucdavis.edu † [9,10]. Loss of a functional Sox6 protein in skeletal Equal contributors muscle results in a dramatic increase in slow fibers in Division of Cardiovascular Medicine, Department of Internal Medicine, University of California, Davis, One Shields Avenue, Davis, CA 95616, USA © 2013 An et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. An et al. Skeletal Muscle 2013, 3:11 Page 2 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 fetal as well as in adult mice [9-12]. These observations in addition to its well-established roles in muscle indicate that Sox6 is one of the key factors regulating the atrophy [15]. fiber type differentiation of mammalian skeletal muscles. To uncover how Sox6 activity is controlled in Methods muscle, we searched for regulatory proteins for Sox6 Yeast two-hybrid screening by performing yeast two-hybrid screening. In this The MATCHMAKER Two-Hybrid System 3 (Clontech, manuscript, we report identification of the Trip12 E3 Mountain View, CA, USA) was used following the manu- ligase as a negative regulator of Sox6 activity. E3 ubi- facture’s protocol. Construction of SOX6 coiled-coil (CC) quitin ligases determine substrate specificity for ubi- domain containing bait plasmid (Figure 1A) was reported quitin modification in the ubiquitin-proteasome previously [16]. AH-109 yeast cells were first transformed system [13] and are involved at all levels of transcrip- with the bait plasmid and subsequently with the human tional regulation [14]. We discovered that in C2C12 heart cDNA prey plasmid library. Screening was myotubes, Sox6 protein levels are controlled by the performed on medium-stringency plates (three selection ubiquitin-proteasome system in which Sox6 proteins markers, histidine, leucine, and tryptophan) following the are polyubiquitinated by Trip12 and degraded by the manufacturer’s recommendation. Colonies grown to the 26S proteasome. Significant to the orchestrated regula- size of 2–3mmin5days were streakedonhigh- tion of muscle fiber phenotype, we found that the sup- stringency plates (histidine, adenine, β-gal). Seventy-five pression of Trip12 levels in C2C12 myotube cultures large colonies were obtained in this screening, and prey results in three interlocking events: increased levels of cDNA plasmids were isolated using the YEASTMAKER Sox6 protein, downregulation of the slow MyHC Yeast Plasmid Isolation Kit (Clontech) for sequencing. isoform (Myh7,MyHC-I/β), and upregulation of the Twenty-five of them contained in-frame human protein fast MyHC isoform (Myh4, MyHC-IIb). Our new find- sequences recorded in GenBank. Of these, the clone ing suggests that the ubiquitin-proteasome system containing the C-terminus HECT domain of TRIP12 plays a significant role in muscle fiber differentiation (Figure 1B) was selected for further analysis. Figure 1 SOX6 coiled-coil domain physically interacts with TRIP12 HECT domain. (A) Schematic representation of the human SOX6 protein [17]. The coiled-coil (CC) domain containing the leucine-zipper (LZ) motif and Q-box (aa 143–304) was used as bait for screening a human heart cDNA library. The amino acid sequence of the CC domain is 100% conserved between mice and humans. (B) Schematic representation of the full-length human TRIP12 protein. The partial TRIP12 cDNA clone identified by yeast hybrid screening contained the most C-terminal HECT domain (aa 1614–2040). (C) Diagrams of the bait and prey expression vectors used for Co-IP. The numbers indicate the amino acid regions used for each vector construct. The SOX6 CC-domain was tagged with c-Myc at the C-terminus. The TRIP12 HECT domain was tagged with HSV at the C-terminus. (D) Co-IP assays were performed using HEK293 cell lysate transfected with the bait and prey expression vectors depicted in (C). Input lanes contained 2% (10 μg) of un-manipulated lysate. Rabbit polyclonal antibodies used for pull down are listed under IP. Mouse monoclonal antibodies used for Western blot (WB) are indicated below each panel. GST antibody was used as a negative control. An et al. Skeletal Muscle 2013, 3:11 Page 3 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 Mice Quantification of bands was performed using digital Mice were maintained and killed according to the ani- scans of the exposed films and ImageJ software (http:// mal protocol approved by the Institutional Animal Care imagej.nih.gov/ij/) according to the user guide. Only and Use Committee at the University of California, films with unsaturated intensity of bands were used for Davis, which adheres to the National Institutes of Health densitometric analysis. All densitometry measurements guidelines. were performed using at least three independent samples, and statistical significance between control and test sam- ples was determined by the two-tailed Student’s t-tests. Cell culture HEK293 (human embryonic kidney cell line) and C2C12 Co-immunoprecipitation (Co-IP) (mouse skeletal myoblast cell line) were maintained in Interaction between the human SOX6 CC domain and growth medium (GM) consisting of Dulbecco’s modified the TRIP12 HECT domain was confirmed by Co-IP. Eagle’s medium (DMEM), 10% fetal bovine serum (FBS), Human SOX6 CC-myc (tagged at the C-terminus) was 100 U/ml penicillin, and 100 μg/ml streptomycin at 37°C constructed by inserting the SOX6 CC cDNA sequence in a 5% CO -humidified incubator. To induce myotube into pcDNA3.1/myc-His (Invitrogen, Carlsbad, CA, differentiation, C2C12 cells were incubated in differenti- USA). To generate a C-terminus tagged TRIP12 HECT ation medium (DM) consisting of DMEM, 2% horse domain construct (TRIP12 HECT-HSV), the TRIP12 serum, 100 U/ml penicillin, and 100 μg/ml streptomycin. HECT domain cDNA sequence was cloned into pTriEx-1.1 (Novagen, Madison, WI, USA). SOX6 CC-myc and TRIP12 Western blotting and densitometric analysis HECT-HSV (Figure 1C) were cotransfected into HEK293 Brain, heart, lung, kidney, liver, and skeletal muscle cells using GenJet Plus DNA In Vitro Tranfection (gastrocnemius) were harvested from a 2-month-old Reagent (SignaGen Laboratories, Rockville, MD, USA) wild-type (WT) female mouse, and testis was harvested and incubated in GM for 48 h. Cells were lysed in Buffer from a 2-month-old WT male mouse, respectively. Tis- A (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM sue samples were homogenized in RIPA buffer (1X PBS, EDTA, 1% Nonidet P40) supplemented with protease 1% Nonidet P40, 0.5% sodium deoxycholate, 0.1% SDS) and phosphatase inhibitor cocktails (Thermo Scientific) supplemented with protease and phosphatase inhibitor and clarified by centrifugation at 17,900g at 4°C for 10 cocktails (Thermo Scientific, Waltham, MA, USA) and min. IP was performed by incubating 500 μgofprotein clarified by centrifugation at 17,900g at 4°C for 15 min. (in 500 μl Buffer A) with 2 μg rabbit polyclonal antibody Cultured cell samples were processed as described at targeting c-Myc (A00172, GenScript USA Inc., Piscataway, each experimental section below. Protein samples were NJ, USA) or HSV (A00624, GenScript USA Inc.) for 1 h at resolved on 7.5% or 4-15% gradient SDS-PAGE gels, 4°C on a rocking platform. Anti-GST rabbit polyclonal transferred to nitrocellulose membranes, and processed antibody (ab9085, Abcam) was used as a negative control. for incubation with appropriate antibodies. Signals were The mixture was then supplemented with 20 μl Protein detected on X-ray films using Pierce ECL Western Blot- A-agarose beads (Roche, Indianapolis, IN, USA) and incu- ting Substrate (Thermo Scientific) or Western Lighting bated overnight at 4°C on a rotating platform. Agarose Plus (PerkinElmer, Waltham, MA, USA). For detecting beads were washed five times with Buffer A, and 25 μlof polyubiquitinated Sox6 protein in vitro and in vivo, 2X SDS-PAGE loading buffer was added to the washed WesternBright Sirius Chemiluminescent HRP substrate agarose beads to extract immunoprecipitated protein. (Advansta, Menlo Park, CA, USA) was used. The follow- After incubating in boiling water for 5 min, extracted ing primary antibodies were used: Sox6 (ab30455, Abcam, protein was separated on a 4-15% gradient SDS-PAGE gel Cambridge, MA, USA: rabbit polyclonal), TRIP12 (A301- and analyzed by Western blotting using the antibodies de- 814A, Bethyl Laboratories, Inc., Montgomery, TX, USA: scribed in the previous section. Co-IP of endogenous Sox6 rabbit polyclonal), β-actin (sc-1616, Santa Cruz Biotech- and Trip12 proteins was performed using nuclear frac- nology, Inc., Santa Cruz, CA, USA: goat polyclonal), TBP tions of differentiating C2C12 cells. Nuclear fractions were (NB500-700, Novus Biologicals, Littleton, CO, USA: obtained from C2C12 cells in 15-cm plates differentiated mouse monoclonal), HA (H3663, Sigma-Aldrich: mouse for 48 h in DM using NE-PER Nuclear and Cytoplasmic monoclonal), c-Myc (R950-25, Invitrogen, Carlsbad, Extraction Reagents (Thermo Scientific). Protease and CA, USA: mouse monoclonal), HSV (69171-4, Novagen, phosphatase inhibitor cocktails (Thermo Scientific) were Madison, WI, USA: mouse monoclonal), anti- added to appropriate reagents. Prior to IP, the NaCl con- DYKDDDDK (clone L5, 637301, BioLegend, San Diego, centration of nuclear fractions (~400 mM) was adjusted to CA, USA: rat monoclonal), and myogenin (clone F5D, De- ~150 mM using Buffer A without NaCl. IP and Western velopmental Studies Hybridoma Bank, Iowa City, IA, blotting was performed with Rabbit TrueBlot Set USA: mouse monoclonal). (Rockland Immunochemicals, Inc., Gilbertsville, PA, USA) An et al. Skeletal Muscle 2013, 3:11 Page 4 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 using 300 μg of nuclear protein and 2 μg of rabbit poly- full-length Sox6 cDNA was cloned into pcDNA3.1/Zeo(+) clonal antibodies against Sox6 (ab30455, Abcam), TRIP12 (Invitrogen) with a FLAG tag at the C-terminus], and the (A301-814A, Bethyl Laboratories, Inc.), and GST (ab9085, indicated amount of TRIP12-HSV vector using 7.5 μlof Abcam) according to the manufacturer’s instructions. GenJet Plus DNA In Vitro Tranfection Reagent (SignaGen Immunoprecipitated protein was separated on a 7.5% Laboratories). pcDNA3.1/Zeo(+) and pTriEx-1.1 were SDS-PAGE gel and was analyzed by Western blotting used as empty vectors for Sox6-FLAG and TRIP12-HSV using the same Sox6 and TRIP12 antibodies. vectors, respectively. 24 h after incubation, 10 μM MG132 (EMD Millipore) was added to the medium to inhibit Tagged protein purification and in vitro ubiquitination theproteasomeactivity, and cells were incubated for assay another 6 h. Cells were collected and lysed in Buffer B Full-length human SOX6 cDNA [17] was cloned into (see the previous section) containing protease inhibitor pcDNA3.1/myc-his to produce a His-tagged SOX6-myc cocktail and 10 mM N-ethylmaleimide (an inhibitor for expression vector, while full-length human TRIP12 cDNA deubiquitinating enzymes), and briefly sonicated using (I.M.A.G.E. clone ID 40083165, ATCC, Manassas, VA, Bioruptor (Diagenode, Denville, NJ, USA). Lysate was USA) was cloned into pTriEx-1.1 to produce a His-tagged cleared by centrifugation at 17,900g for 10 min at 4°C, TRIP12-HSV expression vector. HEK293 cells were and protein was quantified using BCA Protein Assay transfected with each of the tagged protein expression Reagent (Thermo Scientific). After pre-clearing 500 μg vectors and grown for 48 h. Cells were harvested in of total protein using 2 μgofrat IgG2a, κ (BioLegend), Buffer B (50 mM Tris, pH 7.4, 150 mM NaCl, 1 mM and 20 μl of Protein G-agarose beads (Roche), protein EGTA, 1% Nonidet P40, 0.3% sodium deoxycholate) samples were mixed with 5 μg of anti-DYKDDDDK supplemented with 10 mM imidazole, protease, and (FLAG) tag antibody, and the mixture was incubated for phosphatase inhibitor cocktails (Thermo Scientific), and 2 h at 4°C on a rocking platform. Then 20 μlof Protein His-tagged SOX6 and TRIP12 proteins were purified at G-agarose beads was added, and the mixture was incu- 4°C using MagneHis Protein Purification System bated overnight at 4°C on a rocking platform. Agarose (Promega Corp., Madison, WI, USA) following the beads were washed five times with RIPA buffer, and 25 manufacturer’s instructions. After purification, buffer μl of 2X SDS-PAGE loading buffer was added to the was changed to Buffer A (see the previous section) washed agarose beads to extract immunoprecipitated supplemented with protease and phosphatase inhibitor protein. After incubating in boiling water for 5 min, cocktails (Thermo Scientific) using Amicon Ultra-0.5, extracted protein was separated on a 7.5% SDS-PAGE Ultracel-30 Membrane, 30 kDa (EMD Millipore, Billerica, gel and was analyzed by Western blotting using anti- MA, USA). HA antibody to detect polyubiquitinated Sox6-FLAG Ubiquitination of SOX6 by TRIP12 in vitro was deter- protein. The membranes were subsequently incubated mined using the purified proteins and substrates as de- with anti-DYKDDDDK (FLAG) tag antibody to detect scribed previously [18] with minor modifications. Briefly, Sox6-FLAG protein. Input samples were also analyzed 100 ng of ubiquitin-activating enzyme E1 (Enzo Life using TRIP12 antibody to detect both endogenous Sciences, Plymouth Meeting, PA, USA), 250 ng of E2 TRIP12 and overexpressed TRIP12-HSV proteins. (UbcH5a, Enzo Life Sciences), 0.8 μg of ubiquitin (Enzo Life Sciences), and 300 ng of purified SOX6 were mixed in 20 μl reaction buffer (25 mM Tris-HCl, pH 7.5, 50 siRNA experiments mM NaCl, 5 mM ATP, 10 mM MgCl , 1 mM DTT, 1X C2C12 cells were plated in six-well plates at a density of protease and phosphatase inhibitor cocktails) with or 2×10 cells/well and incubated in GM for 24 h. siRNAs without 3 μg of purified TRIP12 and were incubated at (pre-designed DsiRNAs from Integrated DNA Technolo- 37°C for 0, 1, and 2 h. Reactions were terminated by the gies, Coralville, IA, USA) were transfected using 10 μlof addition of 20 μl 2X SDS-PAGE loading buffer. Detec- TransIT-TKO (Mirus Bio LLC, Madison, WI, USA) at tion of ubiquitinated SOX6-myc protein was performed 25 nM final concentration. Then 24 h after transfection, by Western blot analysis using anti-c-Myc antibody cells were rinsed with PBS once and were incubated in (R950-25, Invitrogen). DM for 48 h. To extract protein, cells were rinsed twice with ice- In vivo ubiquitination assay cold PBS and were treated with trichloroacetic acid HEK293 cells were plated in 60-mm culture dishes at a (TCA) as described previously [19]. Briefly, cells were in- density of 2 × 10 cells/plate and incubated for 24 h. cubated in 10% TCA for 30 min on ice and were collected Cells were cotransfected with 0.5 μg of HA-Ub vector into 1.5-ml tubes. After centrifugation at 17,900g for 5 (obtained from Dr. Aldrin V. Gomes at University of min at 4°C, cell pellets were briefly sonicated in 1X SDS- California, Davis), 0.5 μg of Sox6-FLAG vector [mouse PAGE loading buffer without bromophenol blue (BPB) An et al. Skeletal Muscle 2013, 3:11 Page 5 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 and 2-mercaptoethanol (2-ME). After quantification, a Results hint of BPB and 2-ME (5%) were added, and the protein SOX6 interacts with TRIP12, an E3 ubiquitin ligase samples were incubated in boiling water for 5 min. An expressed in muscle equal amount of protein was separated on a 7.5% or 4- During the formation of healthy muscle tissue, Sox6 15% gradient SDS-PAGE gel and analyzed by Western functions to suppress transcription of a wide variety of blotting using Sox6, TBP, TRIP12, and β-actin slow fiber-specific genes [9,12]. To identify the proteins antibodies. that interact with Sox6 and regulate the muscle fiber For RNA extraction, cells were lysed in TRIzol reagent, phenotype, we performed yeast two-hybrid screening. and total RNA was extracted using Direct-zol RNA For the bait, the coiled-coil domain (CC domain) includ- MiniPrep (Zymo Research, Irvine, CA, USA) with DNase ing the leucine zipper (LZ) motif and glutamine-rich I treatment steps according to the manufacturer’s instruc- domain (Q-box) of human SOX6 was used (Figure 1A) tions. After synthesizing cDNA using High Capacity [17]. This choice of bait was based on the following ra- cDNA Reverse Transcription Kit (Applied Biosystems, tionales: (1) the coiled-coil domain is a known protein- Carlsbad, CA, USA), quantitative PCR (qPCR) was protein interacting domain shown to interact with performed on ABI Prism 7900HT Sequence Detection multiple proteins [16,22-26], and (2) the amino acid System (Applied Biosystems) using PrimeTime qPCR sequence and location of this domain are 100% con- Assays (Integrated DNA Technologies) or TaqMan Gene served between the mouse Sox6 and human SOX6 Expression Assay (Applied Biosystems) and SensiFast [17,27]. With this bait, we screened a cDNA library of Probe Hi-ROX Kit (Bioline, Taunton, MA, USA). Results adult human heart, where SOX6 is moderately were normalized to the β-actin (Actb) transcript level, and expressed [17]. One of the candidates isolated was a clone all statistical analyses were performed using the two-tailed containing the C-terminus part of TRIP12 (Figure 1B). Student's t-tests. For the time course experiments using Currently, what is known about this protein can be differentiating C2C12 cells, Huwe1 and Tbp were used as summarized succinctly: TRIP12 belongs to the HECT reference genes instead of Actb because reference gene family of E3 ubiquitin ligases, a family characterized by analysis using GeNorm [20] revealed that the combination their highly conserved catalytic subunit located at the of these two genes represented the most stably expressed C-terminus (HECT domain) [28]. Originally identified genes in differentiating C2C12 cells among eight genes as a protein that interacts with the ligand-binding do- tested (Actb, Gapdh, Huwe1, Lrrc40, Rpl37a, Rpl41, Tbp, main of the thyroid hormone and retinoid receptors and Ubb: data not shown). Additional analysis using [29], its function as an E3 ubiquitin ligase was reported NormFinder [21] also showed that these two genes are soon after for multiple targets [18,30-32]. Additionally, stably expressed in differentiating C2C12 cells (data not it was shown to be essential for mammalian develop- shown). To use these two genes as a reference, raw thresh- ment, as evidenced by a report showing that targeted in- old cycle (C ) values of these two genes were averaged activation results in embryonic lethality in mice [33]. according to the instructions of NormFinder, and the Although Trip12’s function in muscle has not been mean C values were used for further calculations. In- reported, we hypothesized that the presumed regulation formation on siRNAs, PrimeTime qPCR Assays, and of Sox6 activity by Trip12 would significantly affect TaqMan Gene Expression Assay used is provided in muscle differentiation and therefore we decided to Additional file 1: Table S1. investigate. To confirm the physical interaction between the SOX6 bait and the TRIP12 prey proteins, we first performed Treatment with cycloheximide and MG132 Co-IP using lysates from HEK293 cells cotransfected C2C12 cells were plated in six-well plates at a density of with expression vectors for the bait and prey tagged with 3×10 cells/well and incubated in GM for 24 h. Cells c-Myc and HSV epitopes at their C-termini, respectively were rinsed with PBS once and incubated in DM for 24 (Figure 1C). As shown in the upper panel of Figure 1D, the h (for time-course experiments) or 48 h (for MG132 SOX6 bait protein was successfully co-immunoprecipitated treatment). For time-course experiments, 100 μg/ml with the TRIP12 prey protein when an antibody for cycloheximide (CHX) was added to DM, and cells were the TRIP12 prey (anti-HSV antibody) was used for IP. incubated for 0, 6, 12, and 24 h. For MG132 treatment, However, the TRIP12 prey protein was not co- 1 μM MG132 was added to DM together with 100 μg/ immunoprecipitated when the SOX6 bait protein was ml CHX, and cells were incubated for 6 h. As a negative pulled down with a c-Myc antibody (Figure 1D, lower control, the same volume of dimethyl sulfoxide (DMSO) panel). The same results were observed when the was added. Protein was extracted using TCA as de- SOX6 bait protein was fused with a FLAG tag at the scribed above and analyzed by Western blotting using N-terminus and anti-FLAG antibody was used for IP appropriate antibodies. (data not shown). These results suggest that both the An et al. Skeletal Muscle 2013, 3:11 Page 6 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 N- and C-termini of the SOX6 bait protein are physic- terminus of endogenous Sox6 protein, where an epitope ally blocked by the TRIP12 prey protein or unknown for the Sox6 antibody used is located, is physically protein(s) in HEK293 cells and therefore are not avail- blocked by Trip12 or unknown protein(s). Since the able for Co-IP of the TRIP12 prey protein by the pull-down using both tagged-proteins (Figure 1D) and SOX6 bait. the endogenous proteins (Figure 2B) gave us similar re- Next, since little is known about Trip12 expression in sults, i.e., the antibodies detecting Sox6 proteins did not the mouse, we investigated expression of the Trip12 pro- pull down Trip12 proteins, binding of Trip12 to Sox6 tein in adult mouse tissues using Western blotting. The may result in masking a large surface area of the Sox6 Trip12 band (the full-length mouse protein is approxi- protein. mately 224 kDa) was detected in various tissues includ- ing skeletal muscle (Figure 2A). The specificity of the TRIP12 polyubiquitinates Sox6 antibody and the actual size of Trip12 band detected In the ubiquitin-proteasome system, E3 ubiquitin ligases were confirmed using Trip12 siRNA, which specifically such as Trip12 facilitate the transfer of ubiquitin to spe- decreased the signal of the ~250 kDa band (Figure 2A). cific substrates [13]. To examine whether Sox6 is a sub- Since Sox6 protein is expressed in both adult heart and strate of TRIP12 E3 ligase activity and if this interaction skeletal muscle [17,34] as well as in the nuclear fraction results in polyubiquitination of Sox6, we performed both of C2C12 cells (unpublished data), we examined whether in vitro and an in vivo ubiquitination assay. the endogenous Sox6 and Trip12 proteins also directly We first performed an in vitro ubiquitination assay interact. Co-IP using nuclear fractions of differentiating using purified recombinant proteins. The process of C2C12 cells showed that endogenous Sox6 was co- ubiquitination of a target protein involves three enzymes, immunoprecipitated with endogenous Trip12 when a ubiquitin-activating enzyme, E1, a ubiquitin-conjugating Trip12 antibody was used for IP (Figure 2B, left panel), enzyme, E2, and a ubiquitin-protein ligase, E3 [13,28]. In indicating physical interaction of these proteins in addition, E4 enzymes have recently been shown to en- C2C12 cells. However, endogenous Trip12 was not co- hance the polyubiquitination reaction in some cases [35]. immunoprecipitated when Sox6 antibody was used for To reconstitute the ubiquitination reaction, an equal IP (Figure 2B, right panel), suggesting that the C- amount of purified SOX6-myc (substrate) was combined Figure 2 Trip12 protein expression in adult mouse tissues and interaction between endogenous Sox6 and Trip12 proteins. (A) Western blot analysis was performed to determine Trip12 protein expression in adult mouse tissues. Each lane contained 30 μg of protein except for the C12C12 samples, which contained 15 μg of protein per lane. Relevant protein size markers (kDa) are indicated to the left. β-actin was used as a loading control. (B) Co-IP of endogenous Sox6 and Trip12 proteins using nuclear fractions of differentiating C2C12 cells. Input lane contained 5% (15 μg) of pre-cleared nuclear protein. Antibodies used for pull down are listed under IP, and antibodies used for Western blot (WB) are indicated below each panel. GST antibody was used as a negative control. An et al. Skeletal Muscle 2013, 3:11 Page 7 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 with the representative E1 and E2 enzymes along with the presence of Trip12 siRNA, intensity of the ~224 kDa full-length TRIP12 E3 ligase (see the Methods section). As band (full-length mouse Trip12) in C2C12 myotubes de- shown in Figure 3A, addition of TRIP12 resulted in a creased to almost an undetectable level. Using this greater level of ubiquitination of SOX6-myc protein in a Trip12 siRNA, the result of Trip12 inhibition on Sox6 time-dependent manner detected as an increasing upward protein levels was determined at a time point (48 h after smear (lanes 4, 5, and 6). To test if this process occurred switching to DM) where Sox6 protein levels are known in vivo,we performed in vivo ubiquitination assays. to be decreasing in differentiating C2C12 myotube HEK293 cells were cotransfected with the full-length cultures (see Additional file 2: Figure S1). TRIP12-HSV, Sox6-FLAG, and HA-tagged ubiquitin As can be seen in Figure 4B, reduction of Trip12 led (HA-Ub) expression vectors. Immunoprecipitation was to a threefold increase in the amount of Sox6 protein, performed with FLAG antibody to concentrate Sox6- whereas there was no change in protein levels of the nu- FLAG protein, followed by Western blotting using HA clear transcription factor TATA-box binding protein antibody to detect ubiquitinated Sox6-FLAG protein. (Tbp) (used as a control). Therefore, Sox6 is a specific As TRIP12 protein expression was doubled and tripled, substrate to Trip12, by which it is polyubiquitinated and the amounts of high-molecular-weight products, which degraded. Whether a Trip12-mediated change in Sox6 are the sign of multiple additions of HA-Ub molecules, protein level had any significant biological effects was also increased (Figure 3B), reflecting the addition of initially tested by analysis of changes in mRNA expres- multiple ubiquitin molecules to the Sox6-FLAG sub- sion of Myh7, a direct Sox6 target gene [9,10]. When strate. As would be expected from a fixed pool of sub- Trip12 siRNA treatment effectively reduced Trip12 strate, the non-ubiquitinated Sox6-FLAG protein levels mRNA levels to approximately one third of the control decreased (Figure 3B), indicating that more Sox6-FLAG level (Figure 4C), Myh7 mRNA levels were reduced by 40%, was polyubiquitinated as Trip12 levels increased. whileSox6mRNAlevelswereunchanged (Figure4C), confirming the suppressive role of Sox6 on the Myh7 Knockdown of Trip12 in C2C12 myotubes results in a transcription at the post-transcriptional level. concurrent increase in Sox6 protein levels and a decrease of Myh7 transcription Since polyubiquitination frequently leads to degradation Sox6 protein is degraded by the 26S proteasome of the target substrate, we next tested the effect of The final stage of the ubiquitin-proteasome system is siRNA-induced Trip12 knockdown on Sox6 protein degradation of the polyubiquitinated substrate by the levels in muscle cells. As shown in Figure 4A, in the 26S proteasome [13]. Therefore, we determined the Figure 3 TRIP12 ubiquitinates Sox6 in vitro and in vivo. (A) In vitro ubiquitination of SOX6 protein was performed using purified SOX6-myc as the substrate and TRIP12 along with the combination of enzymes as indicated in the figure (see the Methods section for details). Lanes 4–6 contain 3 μg of TRIP12-HSV. Western blot (WB) was performed using c-Myc antibody to detect the degree of ubiquitination of Sox6-myc. Asterisk indicates non-specific bands detected in all reactions. (B) Sox6 is ubiquitinated by TRIP12 in vivo. HEK293 cells were cotransfected with plasmid DNAs encoding HA-Ub, Sox6-FLAG, and increasing amounts (0.5, 1, and 1.5 μg) of TRIP12-HSV, and lysates were immunoprecipitated (IP) with anti-DYKDDDDK (FLAG) antibody, and then processed for Western blotting (WB) using anti-HA antibody. Asterisk indicates non-specific bands detected in all IP samples, although it is possible that these bands also contain ubiquitinated Sox6-FLAG protein of lower molecular weights (with ~1 to 4 ubiquitin moieties) in Sox6-FLAG-transfected samples. The same membrane was subsequently incubated with anti-DYKDDDDK (FLAG) antibody to detect Sox6-FLAG protein; 10 μg (2%) of input protein samples (lysates) was also subjected to Western blotting using anti-TRIP 12 antibody to detect both endogenous TRIP12 and overexpressed TRIP12-HSV proteins. An et al. Skeletal Muscle 2013, 3:11 Page 8 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 Figure 4 Trip12 controls Sox6 protein level in C2C12 cells. (A) siRNA-mediated knockdown of Trip12 resulted in an increase of Sox6 protein levels in C2C12 cells. C2C12 cells were transfected with siRNA for either EGFP (negative control) or Trip12 in triplicate, and lysates were analyzed by Western blotting using a 7.5% gel. (B) Densitometric analysis of the Western blot in (A) shows approximately 3-fold increase in the Sox6 protein level in Trip12 siRNA-treated cells, while the Tbp protein level was not affected. Data are normalized for those from EGFP siRNA- transfected cells and represented as mean ± SD (n = 3). (C) Trip12 knockdown lowered mRNA level of Myh7, a known Sox6 target. Total RNA was extracted from mock- or siRNA-transfected C2C12 cells, and mRNA levels of Trip12 and Myh7 were quantified by reverse transcription-quantitative PCR (RT-qPCR). Data are normalized for those from EGFP siRNA-transfected cells and represented as mean ± SD (n = 3). **p < 0.005. effect of proteasome inhibition on the cellular Sox6 pro- protein appeared slightly more stable than Sox6 tein levels. (Figure 5B). The Trip12 half-life estimated here is in A standard approach for examining proteasome- agreement with a similar turnover rate of the human mediated degradation of proteins is to use a combination TRIP12 protein recently estimated in HeLa cells using a of a proteasome inhibitor such as MG132 and the pro- proteomic approach [37]. tein synthesis inhibitor cycloheximide (CHX) [36]. Prior Although the Sox6 half-life was found to be relatively to this experiment, the stability of the Sox6 protein long (~24 h), we first tried a conventional proteasome along with Trip12 and Tbp in C2C12 myotubes was de- inhibitor (MG132) to examine proteasomal degradation termined. C2C12 myotube cultures (induced for 24 h in of Sox6. Since extended treatment with proteasome in- DM) were treated with CHX to stop new protein synthe- hibitors is toxic to the cells [38], C2C12 cells were sis, following which, the stability of the Sox6, Trip12, treated with CHX in the absence or presence of MG132 and Tbp proteins was examined using Western blot. only for 6 h, and the Sox6 protein level was examined by After 24 h of CHX treatment, the myotubes looked Western blotting. As shown in Additional file 3: Figure healthy and morphologically normal. As shown in S2B, protein levels of myogenin, a short-lived transcrip- Figure 5A and summarized in Figure 5B, Sox6 and Tbp tion factor [39], were reduced by CHX alone and were appeared more stable than Trip12. The protein half-life recovered by MG132 addition, verifying the effect of of Trip12 was estimated as ~9 h, whereas Sox6 protein’s MG132 in our experimental condition. To conduct a half-life was estimated as ~24 h (Figure 5B). The Tbp detailed examination of the response of Sox6 protein to An et al. Skeletal Muscle 2013, 3:11 Page 9 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 Figure 5 Sox6 is degraded by proteasome in C2C12 cells. (A) Time course of Trip12, Sox6, and Tbp protein levels in C2C12 cells in the presence of cycloheximde (CHX) analyzed by Western blotting using a 4-15% gradient gel (n = 3). (B) Densitometric analysis of the Western blot in (A). Band intensity of each protein was normalized to that of 0 h in CHX and represented as mean ± SD (n = 3). (C) Validation of Psmd1 knockdown by siRNA. Total RNA was extracted from mock- or siRNA-transfected C2C12 cells, and the mRNA level of Psmd1 (the gene encoding a regulatory subunit of the 26S proteasome) was quantified by RT-qPCR. Data are normalized to EGFP siRNA-transfected cells and represented as mean ± SD (n = 3). (D) A Western blot using a 7.5% gel showing an increase in the Sox6 protein level in Psmd1 siRNA-transfected C2C12 cells. (E) Densitometric analysis of Western blotting results shows a ~4 fold increase in the Sox6 protein level in Psmd1 siRNA-treated C2C12 cells. A smaller increase was observed for Tbp protein. Data are normalized for those from EGFP siRNA-transfected cells and represented as mean ± SD (n = 3). (F) Psmd1 knockdown reduced the mRNA level of Myh7, a known target of Sox6. Data are normalized to EGFP siRNA-transfected cells and represented as mean ± SD (n = 3). **p < 0.005. MG132, we used a gel condition optimized for resolving inhibitor did not indicate proteasomal degradation of proteins of 75–100 kD (7.5%) and detected two closely Sox6 in an unambiguous manner as expected given its sized bands reacting with the Sox6 antibody (Additional long half-life, we next used a siRNA targeting a prote- file 3: Figure S2B). Although MG132 did not change the asome subunit to confirm proteasomal degradation of overall amount of the Sox6 proteins (total intensity of Sox6. We chose to knockdown Psmd1, a subunit of the lid the two bands; densitometric analysis also showed no domain of the 26S proteasome, because knockdown of statistical difference between these samples: data not Psmd1 effectively inhibits proteasome activity without sig- shown), there was a significant increase in the upper nificant toxicity [40]. Using Psmd1 siRNA, we successfully band along with a significant decrease in the lower band obtained close to 50% reduction in Psmd1 mRNA levels (Additional file 3: Figure S2B), indicating that there is a (Figure 5C). Under this condition, levels of Sox6 proteins discrete shift in the Sox6 protein size in the presence of significantly increased (Figures 5D and 5E). As shown in MG132. We found that this size difference was caused Figure 5F, Psdm1 siRNA treatment of C2C12 myotubes by phosphorylation of the Sox6 protein (Additional file 3: also resulted in significant reduction of Myh7 mRNA Figure S2C). These results suggest a possibility that stabil- levels. Taken together, these results indicate that Sox6 is ity of the Sox6 protein could be regulated by the balance degraded by the 26S proteasome and that inhibition of the between phosphorylated and non-phosphorylated Sox6 26S proteasome activity by siRNA resulted in increased proteins. We are currently investigating this possibility. protein levels of Sox6 and increased suppression of its Because treatment with a conventional proteasome target gene Myh7 in C2C12 myotubes. An et al. Skeletal Muscle 2013, 3:11 Page 10 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 Knockdown of Trip12 expression results in opposite and Myog mRNA levels as well as a significant increase in changes in expression of fast and slow MyHC genes in Myh4 (day 4 in DM), a mirror image of their expression C2C12 myotubes patterns in the Sox6 knockout muscle [9,12], suggesting Because both Trip12 knockdown and 26S proteasome that Trip12 regulation of Sox6 can have a fundamental inhibition in myotube cultures resulted in increased impact on the fiber type identity. Sox6 protein levels (Figures 4A, 4B, 5D and 5E), and de- creased mRNA levels of the Sox6 target Myh7 (Figures 4C Discussion and 5F), we hypothesized that knockdown of Trip12 We have previously reported that a primary mechanism would result in fiber type-specific gene expression changes in fiber type-specific gene expression involves the coor- opposite to those observed in the Sox6 KO mouse, i.e., an dinated suppression of hundreds of slow fiber-specific increase in fast fiber-specific gene expression and a genes by the transcription factor Sox6 [9]. Muscle- decrease in slow fiber-specific gene expression [9,12]. specific knockout of Sox6 results in a dramatic increase To test this hypothesis, C2C12 myoblasts were in slow MyHC-β (Myh7) expression coupled with a sig- transfected with Trip12 siRNA or EGFP siRNA (a negative nificant decrease in fast MyHC-IIb (Myh4) expression control) and the mRNA levels of Myh7 (slow isoform), [9,12]. Muscle-specific loss of Sox6 effectively shifts myogenin (Myog: preferentially expressed in slow muscle muscle tissue into the slower fiber phenotype leading to [41]), and Myh4 (fast MyHC-IIb isoform) were compared the prediction that regulation of Sox6 activity in skeletal in differentiating C2C12 myotube cultures. Brown et al. muscle could be a focal point of fiber type differenti- [42] showed that in differentiating C2C12 cells, Myh7 ation. Our present study expands our investigations into increases after day 1 and then begins to decline after day the mechanisms of fiber type determination by looking 4; conversely, Myh4 begins to increase after day 4 in differ- at the regulation of Sox6 by the ubiquitin-proteasome entiation medium. In this experiment, we normalized pathway. The identification of the relatively unexplored mRNA expression levels in Trip12 siRNA transfected E3 ligase Trip12 as a Sox6 partner protein led us to in- myotubes against the control siRNA transfected vestigate the impact of Trip12 activity on cellular Sox6 myotubes. Changes outside the control expression levels protein levels. Using siRNA of Trip12, we were able to are reflected as greater or lesser than 1. As shown in show that the knockdown of Trip12 expression in Figure 6 and Additional file 4: Figure S3, transfection of C2C12 myotubes resulted in (1) an increase in Sox6 pro- Trip12 siRNA caused a significant decrease in both Myh7 tein levels, (2) the simultaneous downregulation of both Figure 6 Trip12 modulates mRNA levels of fiber-type-specific genes in differentiating C2C12 cells. C2C12 cells were transfected with siRNA for either EGFP (negative control) or Trip12, and medium was switched to DM 24 h after transfection to induce differentiation into myotubes. Total RNA was then extracted every 24 h, and the time course of mRNA levels of Trip12, Myh4 (MyHC-IIb), Myh7 (MyHC-I/β), and myogenin (Myog) were quantified by RT-qPCR using Huwe1 and Tbp as reference genes (see the Methods section for details). Data are normalized for those from EGFP siRNA-transfected cells and represented as mean ± SD (n=3). No expression change between EGFP siRNA- and Trip12 siRNA-transfected samples is expressed as 1 on the graph. *p < 0.05, **p < 0.01, ***p < 0.005. An et al. Skeletal Muscle 2013, 3:11 Page 11 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 the slow fiber-specific Myh7 and preferentially-slow at the carboxyl terminus and is evolutionarily highly myogenin, and (3) upregulation of Myh4 (fast), thus conserved [28,45]. Amino acid sequences lying outside of suggesting that Trip12, by controlling Sox6 protein the HECT domain vary among the family members lead- levels, plays a critical role in regulating muscle fiber ing to suggestions that this N-terminus part of the protein type-specific gene expression. This observation signifi- primarily determines substrate specificity [28,46]. In the cantly expands the role of E3 ligases in skeletal muscle ubiquitination of target proteins, many E3 ubiquitin li- beyond their traditionally conceived role in atrophy and, gases possess the capability of polyubiquitination, while possibly, into the realm of fundamental developmental some are only capable of monoubiquitination when alone, processes in the muscle. and require an E4 ligase for attachment of additional The role of Sox6 as a transcriptional suppressor of ubiquitin molecules [35]. Our in vitro ubiquitination assay slow fiber-specific isoform genes in skeletal muscle has results show that TRIP12 can polyubiquitinate Sox6 in the been reported in mice and zebrafish [9,10,43]. These re- absence of an E4 (Figure 3A). Interestingly, the length of ports indicate that the function of Sox6 during skeletal smear bands detected was different between in vitro and muscle development is likely conserved through verte- in vivo:while in vitro ubiquitination assay showed smear brate evolution since the loss of Sox6 activity in both bands up to ~250 kDa (Figure 3A), a higher molecular species results in the increased mRNA expression of weight of smear bands was detected in in vivo slow fiber-specific genes [9,12,43]. Sox6 is also expressed ubiquitination assay (Figure 3B). These results suggest that in the adult heart at a moderate level [17,34]. During there might be additional factor(s) in vivo (e.g., an E4 lig- heart development, it has been shown that Sox6 is ase), which assists further elongation of polyubiquitin expressed at a high level in proliferating cardiomyocyte chains by Trip12. TRIP12, a more recently recognized E3 progenitor cells, and a reduced Sox6 expression level ubiquitin ligase, was identified based on its sequence causes the progenitors to exit from the cell cycle and homology to the yeast UFD4 E3 ubiquitin ligase [32]. The differentiate [44]. Therefore, control of the cellular levels E3 ligase activity of TRIP12 HECT domain was subse- of the Sox6 protein is potentially a vital mechanism to quently demonstrated [18]. The TRIP12 substrates known regulate development of both cardiac and skeletal to date include APP1-BP1 (amyloid beta precursor muscle. protein-binding protein 1) [32], the SWI/SNF chromatin To identify Sox6 interacting proteins, we performed remodeling complex subunit BAF57 [31], the tumor sup- yeast two-hybrid screening and identified TRIP12, an E3 pressor protein ARF [30], and the RING finger E3 ligase ubiquitin ligase, as a SOX6-interacting protein. In adult RNF168 [47]. Our current report adds the transcription mouse tissues, the Trip12 protein is expressed highest in factor Sox6 as a TRIP12 substrate polyubiquitinated in testis and moderately in both skeletal muscle and the muscle cells. Since the TRIP12 clone identified by yeast heart (Figure 2A). Endogenous Trip12 and Sox6 proteins two-hybrid screening contained only the HECT domain in C2C12 cells directly interact (Figure 2B), and Trip12 (Figure 1B) and the TRIP12 HECT domain alone could polyubiquitinates Sox6 both in vitro and in vivo (Figure 3). interact with the SOX6 coiled-coil domain (Figure 1D), we Ubiquitination is a complex and multifaceted regulatory speculate that the HECT domain is the minimal necessary system involving the modification of a target protein by requirement for TRIP12 to recognize SOX6 as a substrate. addition of either one or multiple ubiquitin molecules In fact, the dual role of the TRIP12 HECT domain (E3 leading to many different outcomes depending on the ligase catalytic activity and the substrate recognition) has number of ubiquitins added or the topology of the ubiqui- been previously demonstrated [18]. Park and colleagues tin chain [14,45]. The addition of one or more ubiquitin reported that the HECT domain of TRIP12, but not molecules to a protein is a multistep process involving at UBE3A (E6-AP), could recognize and ubiquitinate an E3 least three ubiquitin enzymes, E1, E2 and E3. In this art- ligase substrate on its own [32]. Our current result indi- cates that this observation holds for the recognition of icle, we focused on Trip 12, an E3 ubiquitin ligase, which polyubiquitinates Sox6. The E3 ligases’ role in the SOX6 as a substrate by TRIP12. The ability of the HECT ubiquitination chain comes at the end of a multistep domain to function as a ligase catalytic domain as well as a substrate recognition module could be a unique trait for process, where it functions to attach one or more ubiqui- tin molecules to a specific target protein [13,45]. the TRIP12 HECT domain, which will be tested in the Currently, it is estimated that more than 1,000 distinct E3 future as more information becomes available for the other HECT domain E3 ligases. ligase genes are encoded in the human genome [13]. The majority of these are classified into the RING family of E3 The addition of one or more ubiquitin molecules to a ligases with a much smaller family of genes belonging to target protein lends a tremendous flexibility to the cellu- the HECT family, of which Trip12 is a member [28,46]. lar fate of that protein. Depending on the ubiquitin code The HECT domain consists of approximately 350 amino (for review, see Komander et al. [45]), the activity, loca- acids encoding the E3 ligase catalytic domain. It is located tion, partners and even the protein’s survival can be An et al. Skeletal Muscle 2013, 3:11 Page 12 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 altered. We found that Sox6 is polyubiquitinated by outcomes on differentiation of various cell types. Although TRIP12 and is degraded by the 26S proteasome. In con- Sox6 mRNA is transcribed in multiple tissues at moderate trast to other critical myogenic regulatory factors such as to high levels [17,34], its role as a transcription factor is MyoD or myogenin, the half-life of Sox6 is surprisingly quite distinct in different cell types [22]. This is because, long (~24 h). Half-lives of MyoD, Myf5, and myogenin in part, cell type-specific Sox6 functions are dictated by its have been reported as 0.8-1.0 h, <1.0 h, and ~1.0 h, regulatory co-factors [22,24]. To achieve developmental respectively [39,48,49]. The stable nature of Sox6 may stage-specific functions, however, close monitoring of the reflect its function in muscle development; Sox6 plays a Sox6 protein expression level is also critical. In light of role in specification of myotube phenotype by regulating this, the ubiquitin-proteasome system mediated degrad- fiber type differentiation and is expressed at high levels in ation of the Sox6 protein likely plays a crucial role in tem- myotubes [10]. Therefore, the relative stability of Sox6 poral regulation of the Sox6 protein level during may be necessary for stably maintaining the fiber pheno- development. It is possible that a different set of E3 ligases type of myotubes. The ubiquitously expressed nuclear catalyzes ubiquitination of the Sox6 protein depending on protein Tbp is also stable, supporting the correlation of different cellular environments (e.g.. different cell types) or protein stability with maintenance of gene expression. different signals. Therefore, further uncovering of the In muscle, the overall stability of a multigene regulator ubiquitin-proteasome regulation of transcription factors such as Sox6 lends stability to a phenotype such as will enrich our understanding of the multi-layered mecha- muscle fiber type, but this same stability could quickly nisms of transcriptional regulation during development. become a hindrance when external signals demand a fiber type shift. Thus, the polyubiquitination of Sox6 by Conclusions the Trip12 E3 ligase leading to its degradation would We have shown here that Trip12, a HECT domain E3 allow for a swift response in the face of changing work- ubiquitin ligase, targets Sox6, a suppressor for slow loads of the muscle. Therefore, Trip12 likely functions as fiber-specific genes. In addition, we showed that Trip12 a pivot point in fiber type transition, altering the balance is involved in regulation of the fast MyHC isoform gene between slow and fast fiber gene expression. To date, expression. Based on our current data, we propose that two other mechanisms regulating Sox6 protein activities in skeletal muscle, E3 ligases have a significant role in within the cells have been reported: regulation by regulating fiber type-specific gene expression, expanding microRNA and sumoylation. The Sox6 mRNA has a their functional importance in muscle beyond their well- long (~5 kb) 3’-UTR sequence, which contains multiple established role in atrophy. microRNA seed sequences [22]. MicroRNAs are expressed in a cell type-specific manner and are known Additional files to regulate the protein levels of cell type-specific genes [50]. miR-499 is known to target the Sox6 mRNA and Additional file 1: Table S1. siRNAs, PrimeTime qPCR Assays, and suppresses Sox6 protein expression in skeletal muscle TaqMan Gene Expression Assay used in the experiments. [51-54] and in differentiating cardiomyocytes [44]. In the Additional file 2: Figure S1. Temporal changes in Sox6 expression central nervous system, it has been shown that miR-219 level in differentiating C2C12 cells. (A) RNA was prepared every 24 h from C2C12 cells grown in DM at the indicated time. RT-qPCR for Sox6 targets Sox6 in oligodendrocytes and induces the ter- was performed using TaqMan Gene Expression Assays (Applied minal differentiation of oligodendrocytes [55,56]. Post- Biosystems). Gapdh expression level was used to normalize data. Fold translational modification can also control the function increase in mRNA levels (0 h in DM = 1) was calculated at each time point. Each data set represents four independent RT-qPCR experiments of a protein. In the case of Sox6, it has been shown that (mean ± SD). (B) Crude protein extract prepared from differentiating SUMO (small ubiquitin-related modifier) reduces the C2C12 cells (0, 24, 48, and 72 h in DM) was separated on a 7% SDS-PAGE binding affinity of Sox6 to DNA, thus changing its tran- gel (100 μg per well), and Western blotting was performed using Sox6 antibody. A ~90 kDa Sox6 band was detected in all C2C12 cell extracts. scriptional activity [57]. The highest Sox6 protein expression was observed at 24 h in DM and In the current report, we have demonstrated that then rapidly dropped close to the 0 h level at 48 h in DM. β-actin was Trip12 ubiquitinates the Sox6 protein, Sox6 is degraded used as a loading control. by the 26S proteasome, and in both these instances, the Additional file 3: Figure S2. MG132 treatment shifted the intensity of the two Sox6 bands in differentiating C2C12 cells. (A) A procedure for expression of Sox6 target genes are affected. Since skel- MG132 experiments. C2C12 cells were seeded in six-well plates at a etal muscle is highly plastic in nature and in transitions 5 density of 3 × 10 cells/well and incubated in GM for 24 h. Cells were between the slow and fast fiber types [58], the speedy re- rinsed with PBS once and incubated in DM. After 48 h, 100 μg/ml CHX and 1 μM MG132 or the same volume of DMSO (with which MG132 moval of relevant transcription factors (e.g., Sox6) via stock solution was prepared) were added to DM, and cells were the ubiquitin-proteasome system could allow for a rapid incubated for another 6 h before Western blotting. (B) Western blots of response to external cues. MG132-treated cells. Three independent samples were prepared for each treatment. Tbp was used as a loading control. (C) Phosphatase treatment Taken together, regulation of Sox6 protein activity, both of C2C12 nuclear protein; 5 μg of nuclear protein prepared from post-transcriptional and post-translational, has critical An et al. Skeletal Muscle 2013, 3:11 Page 13 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 11. Hagiwara N, Ma B, Ly A: Slow and fast fiber isoform gene expression is differentiating C2C12 cells without phosphatase inhibitor cocktail was systematically altered in skeletal muscle of the Sox6 mutant, p100H. treated with 10 units of calf intestinal alkaline phosphatase (CIP) in the Dev Dyn 2005, 234:301–311. absence or presence of phosphatase inhibitor cocktail (PIC) for 10 min at 12. Quiat D, Voelker KA, Pei J, Grishin NV, Grange RW, Bassel-Duby R, Olson EN: 37°C and analyzed by Western blotting using a 7.5% gel. Positions of two Concerted regulation of myofiber-specific gene expression and muscle Sox6 bands, which are always seen on 7.5% gels but not on 4-15% performance by the transcriptional repressor Sox6. Proc Natl Acad Sci USA gradient gels, are indicated. CIP treatment shifted the position of Sox6 2011, 108:10196–10201. band from upper to lower size, indicating that the upper band consists 13. Varshavsky A: The ubiquitin system, an immense realm. Annu Rev Biochem of a phosphorylated form of Sox6. 2012, 81:167–176. Additional file 4: Figure S3. Relative mRNA levels of fiber-type-specific 14. Geng F, Wenzel S, Tansey WP: Ubiquitin and proteasomes in transcription. genes shown in Figure 6. Relative mRNA levels against reference genes Annu Rev Biochem 2012, 81:177–201. -ΔCt (Huwe1 and Tbp) were calculated using the formula 2 and 15. Glass DJ: Signaling pathways perturbing muscle mass. Curr Opin Clin Nutr represented as mean ± SD (n=3). Metab Care 2010, 13:225–229. 16. Cohen-Barak O, Yi Z, Hagiwara N, Monzen K, Komuro I, Brilliant MH: Sox6 regulation of cardiac myocyte development. Nucleic Acids Res 2003, Abbreviations 31:5941–5948. CC: Coiled-coil; Co-IP: Co-immunoprecipitation; CHX: Cycloheximide; 17. Cohen-Barak O, Hagiwara N, Arlt MF, Horton JP, Brilliant MH: Cloning, DM: Differentiation medium; GM: Growth medium; HECT: Homologous to characterization and chromosome mapping of the human SOX6 gene. the E6-AP carboxyl terminus; IP: Immunoprecipitation; MyHC: Myosin heavy Gene 2001, 265:157–164. chain; qPCR: Quantitative PCR; RT-qPCR: Reverse transcription-quantitative 18. Park Y, Yoon SK, Yoon JB: The HECT domain of TRIP12 ubiquitinates PCR; siRNA: Small interfering RNA; WB: Western blot. substrates of the ubiquitin fusion degradation pathway. J Biol Chem 2009, 284:1540–1549. Competing interests 19. Niikura Y, Nonaka T, Imajoh-Ohmi S: Monitoring of caspase-8/FLICE The authors declare no competing interests. processing and activation upon Fas stimulation with novel antibodies directed against a cleavage site for caspase-8 and its substrate, FLICE- Authors’ contributions like inhibitory protein (FLIP). J Biochem 2002, 132:53–62. CIA performed Co-IP, in vitro and in vivo ubiquitination assays, siRNA 20. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, experiments, cycloheximide and MG132 experiments, RT-qPCR, Western Speleman F: Accurate normalization of real-time quantitative RT-PCR blotting, and densitometric analysis and contributed to writing the data by geometric averaging of multiple internal control genes. manuscript. EG carried out yeast two-hybrid screening to identify TRIP12 and Genome Biol 2002, 3:RESEARCH0034. vector construction. NH conceived, designed, and supervised the study and 21. Andersen CL, Jensen JL, Orntoft TF: Normalization of real-time contributed to writing the manuscript. All authors read and approved the quantitative reverse transcription-PCR data: a model-based variance final manuscript. estimation approach to identify genes suited for normalization, applied to bladder and colon cancer data sets. Cancer Res 2004, 64:5245–5250. Acknowledgements 22. Hagiwara N: Sox6, jack of all trades: a versatile regulatory protein in The authors thank Dr. Aldrin V. Gomes for both technical support and vertebrate development. Dev Dyn 2011, 240:1311–1321. discussion and Mr. Adam Jenkins for helpful discussion. This work was 23. Iguchi H, Urashima Y, Inagaki Y, Ikeda Y, Okamura M, Tanaka T, Uchida A, supported by grants from the National Institutes of Health (AR055209) and Yamamoto TT, Kodama T, Sakai J: SOX6 suppresses cyclin D1 promoter American Heart Association (0635070N) to N.H. activity by interacting with beta-catenin and histone deacetylase 1, and its down-regulation induces pancreatic beta-cell proliferation. Received: 25 October 2012 Accepted: 8 April 2013 J Biol Chem 2007, 282:19052–19061. Published: 10 May 2013 24. Kamachi Y, Uchikawa M, Kondoh H: Pairing SOX off: with partners in the regulation of embryonic development. Trends Genet 2000, 16:182–187. 25. Lefebvre V, Li P, de Crombrugghe B: A new long form of Sox5 (L-Sox5), References Sox6 and Sox9 are coexpressed in chondrogenesis and cooperatively 1. Baylor SM, Hollingworth S: Intracellular calcium movements during activate the type II collagen gene. EMBO J 1998, 17:5718–5733. excitation-contraction coupling in mammalian slow-twitch and fast- 26. Ohe K, Tamai KT, Parvinen M, Sassone-Corsi P: DAX-1 and SOX6 molecular twitch muscle fibers. J Gen Physiol 2012, 139:261–272. interplay results in an antagonistic effect in pre-mRNA splicing. 2. Schiaffino S, Reggiani C: Fiber types in mammalian skeletal muscles. Dev Dyn 2009, 238:1595–1604. Physiol Rev 2011, 91:1447–1531. 27. Connor F, Wright E, Denny P, Koopman P, Ashworth A: The Sry-related 3. Zierath JR, Hawley JA: Skeletal muscle fiber type: influence on contractile HMG box-containing gene Sox6 is expressed in the adult testis and and metabolic properties. PLoS Biol 2004, 2:e348. developing nervous system of the mouse. Nucleic Acids Res 1995, 4. Chemello F, Bean C, Cancellara P, Laveder P, Reggiani C, Lanfranchi G: 23:3365–3372. Microgenomic analysis in skeletal muscle: expression signatures of 28. Rotin D, Kumar S: Physiological functions of the HECT family of ubiquitin individual fast and slow myofibers. PLoS One 2011, 6:e16807. ligases. Nat Rev Mol Cell Biol 2009, 10:398–409. 5. Seene T, Kaasik P, Umnova M: Structural rearrangements in contractile 29. Lee JW, Choi HS, Gyuris J, Brent R, Moore DD: Two classes of proteins apparatus and resulting skeletal muscle remodelling: effect of exercise dependent on either the presence or absence of thyroid hormone for training. J Sports Med Phys Fitness 2009, 49:410–423. interaction with the thyroid hormone receptor. Mol Endocrinol 1995, 6. Selsby JT, Morine KJ, Pendrak K, Barton ER, Sweeney HL: Rescue of 9:243–254. dystrophic skeletal muscle by PGC-1alpha involves a fast to slow fiber 30. Chen D, Shan J, Zhu WG, Qin J, Gu W: Transcription-independent ARF type shift in the mdx mouse. PLoS One 2012, 7:e30063. regulation in oncogenic stress-mediated p53 responses. Nature 2010, 7. Webster C, Silberstein L, Hays AP, Blau HM: Fast muscle fibers are 464:624–627. preferentially affected in Duchenne muscular dystrophy. Cell 1988, 31. Keppler BR, Archer TK: Ubiquitin-dependent and ubiquitin-independent 52:503–513. control of subunit stoichiometry in the SWI/SNF complex. J Biol Chem 8. Schakman O, Gilson H, Thissen JP: Mechanisms of glucocorticoid-induced 2010, 285:35665–35674. myopathy. J Endocrinol 2008, 197:1–10. 32. Park Y, Yoon SK, Yoon JB: TRIP12 functions as an E3 ubiquitin ligase of 9. An CI, Dong Y, Hagiwara N: Genome-wide mapping of Sox6 binding sites APP-BP1. Biochem Biophys Res Commun 2008, 374:294–298. in skeletal muscle reveals both direct and indirect regulation of muscle terminal differentiation by Sox6. BMC Dev Biol 2011, 11:59. 33. Kajiro M, Tsuchiya M, Kawabe Y, Furumai R, Iwasaki N, Hayashi Y, Katano M, 10. Hagiwara N, Yeh M, Liu A: Sox6 is required for normal fiber type Nakajima Y, Goto N, Watanabe T, et al: The E3 ubiquitin ligase activity of differentiation of fetal skeletal muscle in mice. Dev Dyn 2007, Trip12 is essential for mouse embryogenesis. PLoS One 2011, 236:2062–2076. 6:e25871. An et al. Skeletal Muscle 2013, 3:11 Page 14 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 34. Hagiwara N, Klewer SE, Samson RA, Erickson DT, Lyon MF, Brilliant MH: Sox6 57. Fernandez-Lloris R, Osses N, Jaffray E, Shen LN, Vaughan OA, Girwood D, is a candidate gene for p100H myopathy, heart block, and sudden Bartrons R, Rosa JL, Hay RT, Ventura F: Repression of SOX6 transcriptional neonatal death. Proc Natl Acad Sci USA 2000, 97:4180–4185. activity by SUMO modification. FEBS Lett 2006, 580:1215–1221. 35. Hoppe T: Multiubiquitylation by E4 enzymes: 'one size' doesn't fit all. 58. Schiaffino S, Sandri M, Murgia M: Activity-dependent signaling pathways Trends Biochem Sci 2005, 30:183–187. controlling muscle diversity and plasticity. Physiology (Bethesda) 2007, 36. Zhou P: Determining protein half-lives. Methods Mol Biol 2004, 284:67–77. 22:269–278. 37. Boisvert FM, Ahmad Y, Gierlinski M, Charriere F, Lamont D, Scott M, Barton G, Lamond AI: A quantitative spatial proteomics analysis of proteome doi:10.1186/2044-5040-3-11 turnover in human cells. Mol Cell Proteomics 2012, 11:M111 011429. Cite this article as: An et al.: Trip12, a HECT domain E3 ubiquitin ligase, targets Sox6 for proteasomal degradation and affects fiber type-specific 38. Kisselev AF, Goldberg AL: Proteasome inhibitors: from research tools to gene expression in muscle cells. Skeletal Muscle 2013 3:11. drug candidates. Chem Biol 2001, 8:739–758. 39. Vinals F, Ventura F: Myogenin protein stability is decreased by BMP-2 through a mechanism implicating Id1. J Biol Chem 2004, 279:45766–45772. 40. Silva JM, Li MZ, Chang K, Ge W, Golding MC, Rickles RJ, Siolas D, Hu G, Paddison PJ, Schlabach MR, et al: Second-generation shRNA libraries covering the mouse and human genomes. Nat Genet 2005, 37:1281–1288. 41. Hughes SM, Chi MM, Lowry OH, Gundersen K: Myogenin induces a shift of enzyme activity from glycolytic to oxidative metabolism in muscles of transgenic mice. J Cell Biol 1999, 145:633–642. 42. Brown DM, Parr T, Brameld JM: Myosin heavy chain mRNA isoforms are expressed in two distinct cohorts during C2C12 myogenesis. J Muscle Res Cell Motil 2012, 32:383–390. 43. von Hofsten J, Elworthy S, Gilchrist MJ, Smith JC, Wardle FC, Ingham PW: Prdm1- and Sox6-mediated transcriptional repression specifies muscle fibre type in the zebrafish embryo. EMBO Rep 2008, 9:683–689. 44. Sluijter JP, van Mil A, van Vliet P, Metz CH, Liu J, Doevendans PA, Goumans MJ: MicroRNA-1 and −499 regulate differentiation and proliferation in human-derived cardiomyocyte progenitor cells. Arterioscler Thromb Vasc Biol 2010, 30:859–868. 45. Komander D, Rape M: The ubiquitin code. Annu Rev Biochem 2012, 81:203–229. 46. Scheffner M, Staub O: HECT E3s and human disease. BMC Biochem 2007, 1(8 Suppl):S6. 47. Gudjonsson T, Altmeyer M, Savic V, Toledo L, Dinant C, Grofte M, Bartkova J, Poulsen M, Oka Y, Bekker-Jensen S, et al: TRIP12 and UBR5 suppress spreading of chromatin ubiquitylation at damaged chromosomes. Cell 2012, 150:697–709. 48. Lindon C, Albagli O, Domeyne P, Montarras D, Pinset C: Constitutive instability of muscle regulatory factor Myf5 is distinct from its mitosis- specific disappearance, which requires a D-box-like motif overlapping the basic domain. Mol Cell Biol 2000, 20:8923–8932. 49. Sun L, Trausch-Azar JS, Muglia LJ, Schwartz AL: Glucocorticoids differentially regulate degradation of MyoD and Id1 by N-terminal ubiquitination to promote muscle protein catabolism. Proc Natl Acad Sci USA 2008, 105:3339–3344. 50. Bartel DP: MicroRNAs: target recognition and regulatory functions. Cell 2009, 136:215–233. 51. Bell ML, Buvoli M, Leinwand LA: Uncoupling of expression of an intronic microRNA and its myosin host gene by exon skipping. Mol Cell Biol 2010, 30:1937–1945. 52. McCarthy JJ, Esser KA, Peterson CA, Dupont-Versteegden EE: Evidence of MyomiR network regulation of beta-myosin heavy chain gene expression during skeletal muscle atrophy. Physiol Genomics 2009, 39:219–226. 53. Rossi AC, Mammucari C, Argentini C, Reggiani C, Schiaffino S: Two novel/ ancient myosins in mammalian skeletal muscles: MYH14/7b and MYH15 are expressed in extraocular muscles and muscle spindles. J Physiol 2010, Submit your next manuscript to BioMed Central 588:353–364. and take full advantage of: 54. van Rooij E, Quiat D, Johnson BA, Sutherland LB, Qi X, Richardson JA, Kelm RJ Jr, Olson EN: A family of microRNAs encoded by myosin genes • Convenient online submission governs myosin expression and muscle performance. Dev Cell 2009, • Thorough peer review 17:662–673. 55. Dugas JC, Cuellar TL, Scholze A, Ason B, Ibrahim A, Emery B, Zamanian JL, • No space constraints or color figure charges Foo LC, McManus MT, Barres BA: Dicer1 and miR-219 Are required for • Immediate publication on acceptance normal oligodendrocyte differentiation and myelination. Neuron 2010, • Inclusion in PubMed, CAS, Scopus and Google Scholar 65:597–611. 56. Zhao X, He X, Han X, Yu Y, Ye F, Chen Y, Hoang T, Xu X, Mi QS, Xin M, et al: • Research which is freely available for redistribution MicroRNA-mediated control of oligodendrocyte differentiation. Neuron 2010, 65:612–626. Submit your manuscript at www.biomedcentral.com/submit http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Skeletal Muscle Springer Journals

Trip12, a HECT domain E3 ubiquitin ligase, targets Sox6 for proteasomal degradation and affects fiber type-specific gene expression in muscle cells

Skeletal Muscle , Volume 3 (1) – May 10, 2013

Loading next page...
 
/lp/springer-journals/trip12-a-hect-domain-e3-ubiquitin-ligase-targets-sox6-for-proteasomal-XmzOQtEwiC
Publisher
Springer Journals
Copyright
Copyright © 2013 by An et al.; licensee BioMed Central Ltd.
Subject
Life Sciences; Cell Biology; Developmental Biology; Biochemistry, general; Systems Biology; Biotechnology
eISSN
2044-5040
DOI
10.1186/2044-5040-3-11
pmid
23663701
Publisher site
See Article on Publisher Site

Abstract

Background: A sophisticated level of coordinated gene expression is necessary for skeletal muscle fibers to obtain their unique functional identities. We have previously shown that the transcription factor Sox6 plays an essential role in coordinating muscle fiber type differentiation by acting as a transcriptional suppressor of slow fiber-specific genes. Currently, mechanisms regulating the activity of Sox6 in skeletal muscle and how these mechanisms affect the fiber phenotype remain unknown. Methods: Yeast two-hybrid screening was used to identify binding partners of Sox6 in muscle. Small interfering RNA (siRNA)-mediated knockdown of one of the Sox6 binding proteins, Trip12, was used to determine its effect on Sox6 activity in C2C12 myotubes using quantitative analysis of fiber type-specific gene expression. Results: We found that the E3 ligase Trip12, a HECT domain E3 ubiquitin ligase, recognizes and polyubiquitinates Sox6. Inhibiting Trip12 or the 26S proteasome activity resulted in an increase in Sox6 protein levels in C2C12 myotubes. This control of Sox6 activity in muscle cells via Trip12 ubiquitination has significant phenotypic outcomes. Knockdown of Trip12 in C2C12 myotubes led to upregulation of Sox6 protein levels and concurrently to a decrease in slow fiber-specific Myh7 expression coupled with an increased expression in fast fiber-specific Myh4. Therefore, regulation of Sox6 cellular levels by the ubiquitin-proteasome system can induce identity-changing alterations in the expression of fiber type-specific genes in muscle cells. Conclusions: Based on our data, we propose that in skeletal muscle, E3 ligases have a significant role in regulating fiber type-specific gene expression, expanding their importance in muscle beyond their well-established role in atrophy. Keywords: Sox6, Skeletal muscle, Fiber type differentiation, Trip12, E3 ubiquitin ligase, HECT domain, Ubiquitin- proteasome system Background type [4], culminating in the specific biochemical charac- Mammalian skeletal muscles consist of functionally het- teristics of either fast fiber or slow fiber type. Despite its erogeneous myofibers, which can be broadly classified significance to muscle physiology and muscle degenera- into two groups, slow-twitch and fast-twitch fibers. They tive diseases (e.g., exercise physiology [5]; Duchenne differ in contraction speed, metabolic capacity, fatigue muscular dystrophy [6,7]; inflammatory atrophy [8]), resistance, sensitivity to calcium, and a variety of other knowledge regarding the molecular mechanisms orches- attributes [1-3]. This physiologic diversity among muscle trating this coordinated expression exists only in vague fibers is the outward manifestation of the concerted outlines. expression of hundreds of genes unique to each fiber Our recent studies demonstrated that the transcription factor Sox6 directly suppresses transcription of slow fiber-specific genes during mouse muscle development * Correspondence: nhagiwara@ucdavis.edu † [9,10]. Loss of a functional Sox6 protein in skeletal Equal contributors muscle results in a dramatic increase in slow fibers in Division of Cardiovascular Medicine, Department of Internal Medicine, University of California, Davis, One Shields Avenue, Davis, CA 95616, USA © 2013 An et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. An et al. Skeletal Muscle 2013, 3:11 Page 2 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 fetal as well as in adult mice [9-12]. These observations in addition to its well-established roles in muscle indicate that Sox6 is one of the key factors regulating the atrophy [15]. fiber type differentiation of mammalian skeletal muscles. To uncover how Sox6 activity is controlled in Methods muscle, we searched for regulatory proteins for Sox6 Yeast two-hybrid screening by performing yeast two-hybrid screening. In this The MATCHMAKER Two-Hybrid System 3 (Clontech, manuscript, we report identification of the Trip12 E3 Mountain View, CA, USA) was used following the manu- ligase as a negative regulator of Sox6 activity. E3 ubi- facture’s protocol. Construction of SOX6 coiled-coil (CC) quitin ligases determine substrate specificity for ubi- domain containing bait plasmid (Figure 1A) was reported quitin modification in the ubiquitin-proteasome previously [16]. AH-109 yeast cells were first transformed system [13] and are involved at all levels of transcrip- with the bait plasmid and subsequently with the human tional regulation [14]. We discovered that in C2C12 heart cDNA prey plasmid library. Screening was myotubes, Sox6 protein levels are controlled by the performed on medium-stringency plates (three selection ubiquitin-proteasome system in which Sox6 proteins markers, histidine, leucine, and tryptophan) following the are polyubiquitinated by Trip12 and degraded by the manufacturer’s recommendation. Colonies grown to the 26S proteasome. Significant to the orchestrated regula- size of 2–3mmin5days were streakedonhigh- tion of muscle fiber phenotype, we found that the sup- stringency plates (histidine, adenine, β-gal). Seventy-five pression of Trip12 levels in C2C12 myotube cultures large colonies were obtained in this screening, and prey results in three interlocking events: increased levels of cDNA plasmids were isolated using the YEASTMAKER Sox6 protein, downregulation of the slow MyHC Yeast Plasmid Isolation Kit (Clontech) for sequencing. isoform (Myh7,MyHC-I/β), and upregulation of the Twenty-five of them contained in-frame human protein fast MyHC isoform (Myh4, MyHC-IIb). Our new find- sequences recorded in GenBank. Of these, the clone ing suggests that the ubiquitin-proteasome system containing the C-terminus HECT domain of TRIP12 plays a significant role in muscle fiber differentiation (Figure 1B) was selected for further analysis. Figure 1 SOX6 coiled-coil domain physically interacts with TRIP12 HECT domain. (A) Schematic representation of the human SOX6 protein [17]. The coiled-coil (CC) domain containing the leucine-zipper (LZ) motif and Q-box (aa 143–304) was used as bait for screening a human heart cDNA library. The amino acid sequence of the CC domain is 100% conserved between mice and humans. (B) Schematic representation of the full-length human TRIP12 protein. The partial TRIP12 cDNA clone identified by yeast hybrid screening contained the most C-terminal HECT domain (aa 1614–2040). (C) Diagrams of the bait and prey expression vectors used for Co-IP. The numbers indicate the amino acid regions used for each vector construct. The SOX6 CC-domain was tagged with c-Myc at the C-terminus. The TRIP12 HECT domain was tagged with HSV at the C-terminus. (D) Co-IP assays were performed using HEK293 cell lysate transfected with the bait and prey expression vectors depicted in (C). Input lanes contained 2% (10 μg) of un-manipulated lysate. Rabbit polyclonal antibodies used for pull down are listed under IP. Mouse monoclonal antibodies used for Western blot (WB) are indicated below each panel. GST antibody was used as a negative control. An et al. Skeletal Muscle 2013, 3:11 Page 3 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 Mice Quantification of bands was performed using digital Mice were maintained and killed according to the ani- scans of the exposed films and ImageJ software (http:// mal protocol approved by the Institutional Animal Care imagej.nih.gov/ij/) according to the user guide. Only and Use Committee at the University of California, films with unsaturated intensity of bands were used for Davis, which adheres to the National Institutes of Health densitometric analysis. All densitometry measurements guidelines. were performed using at least three independent samples, and statistical significance between control and test sam- ples was determined by the two-tailed Student’s t-tests. Cell culture HEK293 (human embryonic kidney cell line) and C2C12 Co-immunoprecipitation (Co-IP) (mouse skeletal myoblast cell line) were maintained in Interaction between the human SOX6 CC domain and growth medium (GM) consisting of Dulbecco’s modified the TRIP12 HECT domain was confirmed by Co-IP. Eagle’s medium (DMEM), 10% fetal bovine serum (FBS), Human SOX6 CC-myc (tagged at the C-terminus) was 100 U/ml penicillin, and 100 μg/ml streptomycin at 37°C constructed by inserting the SOX6 CC cDNA sequence in a 5% CO -humidified incubator. To induce myotube into pcDNA3.1/myc-His (Invitrogen, Carlsbad, CA, differentiation, C2C12 cells were incubated in differenti- USA). To generate a C-terminus tagged TRIP12 HECT ation medium (DM) consisting of DMEM, 2% horse domain construct (TRIP12 HECT-HSV), the TRIP12 serum, 100 U/ml penicillin, and 100 μg/ml streptomycin. HECT domain cDNA sequence was cloned into pTriEx-1.1 (Novagen, Madison, WI, USA). SOX6 CC-myc and TRIP12 Western blotting and densitometric analysis HECT-HSV (Figure 1C) were cotransfected into HEK293 Brain, heart, lung, kidney, liver, and skeletal muscle cells using GenJet Plus DNA In Vitro Tranfection (gastrocnemius) were harvested from a 2-month-old Reagent (SignaGen Laboratories, Rockville, MD, USA) wild-type (WT) female mouse, and testis was harvested and incubated in GM for 48 h. Cells were lysed in Buffer from a 2-month-old WT male mouse, respectively. Tis- A (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM sue samples were homogenized in RIPA buffer (1X PBS, EDTA, 1% Nonidet P40) supplemented with protease 1% Nonidet P40, 0.5% sodium deoxycholate, 0.1% SDS) and phosphatase inhibitor cocktails (Thermo Scientific) supplemented with protease and phosphatase inhibitor and clarified by centrifugation at 17,900g at 4°C for 10 cocktails (Thermo Scientific, Waltham, MA, USA) and min. IP was performed by incubating 500 μgofprotein clarified by centrifugation at 17,900g at 4°C for 15 min. (in 500 μl Buffer A) with 2 μg rabbit polyclonal antibody Cultured cell samples were processed as described at targeting c-Myc (A00172, GenScript USA Inc., Piscataway, each experimental section below. Protein samples were NJ, USA) or HSV (A00624, GenScript USA Inc.) for 1 h at resolved on 7.5% or 4-15% gradient SDS-PAGE gels, 4°C on a rocking platform. Anti-GST rabbit polyclonal transferred to nitrocellulose membranes, and processed antibody (ab9085, Abcam) was used as a negative control. for incubation with appropriate antibodies. Signals were The mixture was then supplemented with 20 μl Protein detected on X-ray films using Pierce ECL Western Blot- A-agarose beads (Roche, Indianapolis, IN, USA) and incu- ting Substrate (Thermo Scientific) or Western Lighting bated overnight at 4°C on a rotating platform. Agarose Plus (PerkinElmer, Waltham, MA, USA). For detecting beads were washed five times with Buffer A, and 25 μlof polyubiquitinated Sox6 protein in vitro and in vivo, 2X SDS-PAGE loading buffer was added to the washed WesternBright Sirius Chemiluminescent HRP substrate agarose beads to extract immunoprecipitated protein. (Advansta, Menlo Park, CA, USA) was used. The follow- After incubating in boiling water for 5 min, extracted ing primary antibodies were used: Sox6 (ab30455, Abcam, protein was separated on a 4-15% gradient SDS-PAGE gel Cambridge, MA, USA: rabbit polyclonal), TRIP12 (A301- and analyzed by Western blotting using the antibodies de- 814A, Bethyl Laboratories, Inc., Montgomery, TX, USA: scribed in the previous section. Co-IP of endogenous Sox6 rabbit polyclonal), β-actin (sc-1616, Santa Cruz Biotech- and Trip12 proteins was performed using nuclear frac- nology, Inc., Santa Cruz, CA, USA: goat polyclonal), TBP tions of differentiating C2C12 cells. Nuclear fractions were (NB500-700, Novus Biologicals, Littleton, CO, USA: obtained from C2C12 cells in 15-cm plates differentiated mouse monoclonal), HA (H3663, Sigma-Aldrich: mouse for 48 h in DM using NE-PER Nuclear and Cytoplasmic monoclonal), c-Myc (R950-25, Invitrogen, Carlsbad, Extraction Reagents (Thermo Scientific). Protease and CA, USA: mouse monoclonal), HSV (69171-4, Novagen, phosphatase inhibitor cocktails (Thermo Scientific) were Madison, WI, USA: mouse monoclonal), anti- added to appropriate reagents. Prior to IP, the NaCl con- DYKDDDDK (clone L5, 637301, BioLegend, San Diego, centration of nuclear fractions (~400 mM) was adjusted to CA, USA: rat monoclonal), and myogenin (clone F5D, De- ~150 mM using Buffer A without NaCl. IP and Western velopmental Studies Hybridoma Bank, Iowa City, IA, blotting was performed with Rabbit TrueBlot Set USA: mouse monoclonal). (Rockland Immunochemicals, Inc., Gilbertsville, PA, USA) An et al. Skeletal Muscle 2013, 3:11 Page 4 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 using 300 μg of nuclear protein and 2 μg of rabbit poly- full-length Sox6 cDNA was cloned into pcDNA3.1/Zeo(+) clonal antibodies against Sox6 (ab30455, Abcam), TRIP12 (Invitrogen) with a FLAG tag at the C-terminus], and the (A301-814A, Bethyl Laboratories, Inc.), and GST (ab9085, indicated amount of TRIP12-HSV vector using 7.5 μlof Abcam) according to the manufacturer’s instructions. GenJet Plus DNA In Vitro Tranfection Reagent (SignaGen Immunoprecipitated protein was separated on a 7.5% Laboratories). pcDNA3.1/Zeo(+) and pTriEx-1.1 were SDS-PAGE gel and was analyzed by Western blotting used as empty vectors for Sox6-FLAG and TRIP12-HSV using the same Sox6 and TRIP12 antibodies. vectors, respectively. 24 h after incubation, 10 μM MG132 (EMD Millipore) was added to the medium to inhibit Tagged protein purification and in vitro ubiquitination theproteasomeactivity, and cells were incubated for assay another 6 h. Cells were collected and lysed in Buffer B Full-length human SOX6 cDNA [17] was cloned into (see the previous section) containing protease inhibitor pcDNA3.1/myc-his to produce a His-tagged SOX6-myc cocktail and 10 mM N-ethylmaleimide (an inhibitor for expression vector, while full-length human TRIP12 cDNA deubiquitinating enzymes), and briefly sonicated using (I.M.A.G.E. clone ID 40083165, ATCC, Manassas, VA, Bioruptor (Diagenode, Denville, NJ, USA). Lysate was USA) was cloned into pTriEx-1.1 to produce a His-tagged cleared by centrifugation at 17,900g for 10 min at 4°C, TRIP12-HSV expression vector. HEK293 cells were and protein was quantified using BCA Protein Assay transfected with each of the tagged protein expression Reagent (Thermo Scientific). After pre-clearing 500 μg vectors and grown for 48 h. Cells were harvested in of total protein using 2 μgofrat IgG2a, κ (BioLegend), Buffer B (50 mM Tris, pH 7.4, 150 mM NaCl, 1 mM and 20 μl of Protein G-agarose beads (Roche), protein EGTA, 1% Nonidet P40, 0.3% sodium deoxycholate) samples were mixed with 5 μg of anti-DYKDDDDK supplemented with 10 mM imidazole, protease, and (FLAG) tag antibody, and the mixture was incubated for phosphatase inhibitor cocktails (Thermo Scientific), and 2 h at 4°C on a rocking platform. Then 20 μlof Protein His-tagged SOX6 and TRIP12 proteins were purified at G-agarose beads was added, and the mixture was incu- 4°C using MagneHis Protein Purification System bated overnight at 4°C on a rocking platform. Agarose (Promega Corp., Madison, WI, USA) following the beads were washed five times with RIPA buffer, and 25 manufacturer’s instructions. After purification, buffer μl of 2X SDS-PAGE loading buffer was added to the was changed to Buffer A (see the previous section) washed agarose beads to extract immunoprecipitated supplemented with protease and phosphatase inhibitor protein. After incubating in boiling water for 5 min, cocktails (Thermo Scientific) using Amicon Ultra-0.5, extracted protein was separated on a 7.5% SDS-PAGE Ultracel-30 Membrane, 30 kDa (EMD Millipore, Billerica, gel and was analyzed by Western blotting using anti- MA, USA). HA antibody to detect polyubiquitinated Sox6-FLAG Ubiquitination of SOX6 by TRIP12 in vitro was deter- protein. The membranes were subsequently incubated mined using the purified proteins and substrates as de- with anti-DYKDDDDK (FLAG) tag antibody to detect scribed previously [18] with minor modifications. Briefly, Sox6-FLAG protein. Input samples were also analyzed 100 ng of ubiquitin-activating enzyme E1 (Enzo Life using TRIP12 antibody to detect both endogenous Sciences, Plymouth Meeting, PA, USA), 250 ng of E2 TRIP12 and overexpressed TRIP12-HSV proteins. (UbcH5a, Enzo Life Sciences), 0.8 μg of ubiquitin (Enzo Life Sciences), and 300 ng of purified SOX6 were mixed in 20 μl reaction buffer (25 mM Tris-HCl, pH 7.5, 50 siRNA experiments mM NaCl, 5 mM ATP, 10 mM MgCl , 1 mM DTT, 1X C2C12 cells were plated in six-well plates at a density of protease and phosphatase inhibitor cocktails) with or 2×10 cells/well and incubated in GM for 24 h. siRNAs without 3 μg of purified TRIP12 and were incubated at (pre-designed DsiRNAs from Integrated DNA Technolo- 37°C for 0, 1, and 2 h. Reactions were terminated by the gies, Coralville, IA, USA) were transfected using 10 μlof addition of 20 μl 2X SDS-PAGE loading buffer. Detec- TransIT-TKO (Mirus Bio LLC, Madison, WI, USA) at tion of ubiquitinated SOX6-myc protein was performed 25 nM final concentration. Then 24 h after transfection, by Western blot analysis using anti-c-Myc antibody cells were rinsed with PBS once and were incubated in (R950-25, Invitrogen). DM for 48 h. To extract protein, cells were rinsed twice with ice- In vivo ubiquitination assay cold PBS and were treated with trichloroacetic acid HEK293 cells were plated in 60-mm culture dishes at a (TCA) as described previously [19]. Briefly, cells were in- density of 2 × 10 cells/plate and incubated for 24 h. cubated in 10% TCA for 30 min on ice and were collected Cells were cotransfected with 0.5 μg of HA-Ub vector into 1.5-ml tubes. After centrifugation at 17,900g for 5 (obtained from Dr. Aldrin V. Gomes at University of min at 4°C, cell pellets were briefly sonicated in 1X SDS- California, Davis), 0.5 μg of Sox6-FLAG vector [mouse PAGE loading buffer without bromophenol blue (BPB) An et al. Skeletal Muscle 2013, 3:11 Page 5 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 and 2-mercaptoethanol (2-ME). After quantification, a Results hint of BPB and 2-ME (5%) were added, and the protein SOX6 interacts with TRIP12, an E3 ubiquitin ligase samples were incubated in boiling water for 5 min. An expressed in muscle equal amount of protein was separated on a 7.5% or 4- During the formation of healthy muscle tissue, Sox6 15% gradient SDS-PAGE gel and analyzed by Western functions to suppress transcription of a wide variety of blotting using Sox6, TBP, TRIP12, and β-actin slow fiber-specific genes [9,12]. To identify the proteins antibodies. that interact with Sox6 and regulate the muscle fiber For RNA extraction, cells were lysed in TRIzol reagent, phenotype, we performed yeast two-hybrid screening. and total RNA was extracted using Direct-zol RNA For the bait, the coiled-coil domain (CC domain) includ- MiniPrep (Zymo Research, Irvine, CA, USA) with DNase ing the leucine zipper (LZ) motif and glutamine-rich I treatment steps according to the manufacturer’s instruc- domain (Q-box) of human SOX6 was used (Figure 1A) tions. After synthesizing cDNA using High Capacity [17]. This choice of bait was based on the following ra- cDNA Reverse Transcription Kit (Applied Biosystems, tionales: (1) the coiled-coil domain is a known protein- Carlsbad, CA, USA), quantitative PCR (qPCR) was protein interacting domain shown to interact with performed on ABI Prism 7900HT Sequence Detection multiple proteins [16,22-26], and (2) the amino acid System (Applied Biosystems) using PrimeTime qPCR sequence and location of this domain are 100% con- Assays (Integrated DNA Technologies) or TaqMan Gene served between the mouse Sox6 and human SOX6 Expression Assay (Applied Biosystems) and SensiFast [17,27]. With this bait, we screened a cDNA library of Probe Hi-ROX Kit (Bioline, Taunton, MA, USA). Results adult human heart, where SOX6 is moderately were normalized to the β-actin (Actb) transcript level, and expressed [17]. One of the candidates isolated was a clone all statistical analyses were performed using the two-tailed containing the C-terminus part of TRIP12 (Figure 1B). Student's t-tests. For the time course experiments using Currently, what is known about this protein can be differentiating C2C12 cells, Huwe1 and Tbp were used as summarized succinctly: TRIP12 belongs to the HECT reference genes instead of Actb because reference gene family of E3 ubiquitin ligases, a family characterized by analysis using GeNorm [20] revealed that the combination their highly conserved catalytic subunit located at the of these two genes represented the most stably expressed C-terminus (HECT domain) [28]. Originally identified genes in differentiating C2C12 cells among eight genes as a protein that interacts with the ligand-binding do- tested (Actb, Gapdh, Huwe1, Lrrc40, Rpl37a, Rpl41, Tbp, main of the thyroid hormone and retinoid receptors and Ubb: data not shown). Additional analysis using [29], its function as an E3 ubiquitin ligase was reported NormFinder [21] also showed that these two genes are soon after for multiple targets [18,30-32]. Additionally, stably expressed in differentiating C2C12 cells (data not it was shown to be essential for mammalian develop- shown). To use these two genes as a reference, raw thresh- ment, as evidenced by a report showing that targeted in- old cycle (C ) values of these two genes were averaged activation results in embryonic lethality in mice [33]. according to the instructions of NormFinder, and the Although Trip12’s function in muscle has not been mean C values were used for further calculations. In- reported, we hypothesized that the presumed regulation formation on siRNAs, PrimeTime qPCR Assays, and of Sox6 activity by Trip12 would significantly affect TaqMan Gene Expression Assay used is provided in muscle differentiation and therefore we decided to Additional file 1: Table S1. investigate. To confirm the physical interaction between the SOX6 bait and the TRIP12 prey proteins, we first performed Treatment with cycloheximide and MG132 Co-IP using lysates from HEK293 cells cotransfected C2C12 cells were plated in six-well plates at a density of with expression vectors for the bait and prey tagged with 3×10 cells/well and incubated in GM for 24 h. Cells c-Myc and HSV epitopes at their C-termini, respectively were rinsed with PBS once and incubated in DM for 24 (Figure 1C). As shown in the upper panel of Figure 1D, the h (for time-course experiments) or 48 h (for MG132 SOX6 bait protein was successfully co-immunoprecipitated treatment). For time-course experiments, 100 μg/ml with the TRIP12 prey protein when an antibody for cycloheximide (CHX) was added to DM, and cells were the TRIP12 prey (anti-HSV antibody) was used for IP. incubated for 0, 6, 12, and 24 h. For MG132 treatment, However, the TRIP12 prey protein was not co- 1 μM MG132 was added to DM together with 100 μg/ immunoprecipitated when the SOX6 bait protein was ml CHX, and cells were incubated for 6 h. As a negative pulled down with a c-Myc antibody (Figure 1D, lower control, the same volume of dimethyl sulfoxide (DMSO) panel). The same results were observed when the was added. Protein was extracted using TCA as de- SOX6 bait protein was fused with a FLAG tag at the scribed above and analyzed by Western blotting using N-terminus and anti-FLAG antibody was used for IP appropriate antibodies. (data not shown). These results suggest that both the An et al. Skeletal Muscle 2013, 3:11 Page 6 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 N- and C-termini of the SOX6 bait protein are physic- terminus of endogenous Sox6 protein, where an epitope ally blocked by the TRIP12 prey protein or unknown for the Sox6 antibody used is located, is physically protein(s) in HEK293 cells and therefore are not avail- blocked by Trip12 or unknown protein(s). Since the able for Co-IP of the TRIP12 prey protein by the pull-down using both tagged-proteins (Figure 1D) and SOX6 bait. the endogenous proteins (Figure 2B) gave us similar re- Next, since little is known about Trip12 expression in sults, i.e., the antibodies detecting Sox6 proteins did not the mouse, we investigated expression of the Trip12 pro- pull down Trip12 proteins, binding of Trip12 to Sox6 tein in adult mouse tissues using Western blotting. The may result in masking a large surface area of the Sox6 Trip12 band (the full-length mouse protein is approxi- protein. mately 224 kDa) was detected in various tissues includ- ing skeletal muscle (Figure 2A). The specificity of the TRIP12 polyubiquitinates Sox6 antibody and the actual size of Trip12 band detected In the ubiquitin-proteasome system, E3 ubiquitin ligases were confirmed using Trip12 siRNA, which specifically such as Trip12 facilitate the transfer of ubiquitin to spe- decreased the signal of the ~250 kDa band (Figure 2A). cific substrates [13]. To examine whether Sox6 is a sub- Since Sox6 protein is expressed in both adult heart and strate of TRIP12 E3 ligase activity and if this interaction skeletal muscle [17,34] as well as in the nuclear fraction results in polyubiquitination of Sox6, we performed both of C2C12 cells (unpublished data), we examined whether in vitro and an in vivo ubiquitination assay. the endogenous Sox6 and Trip12 proteins also directly We first performed an in vitro ubiquitination assay interact. Co-IP using nuclear fractions of differentiating using purified recombinant proteins. The process of C2C12 cells showed that endogenous Sox6 was co- ubiquitination of a target protein involves three enzymes, immunoprecipitated with endogenous Trip12 when a ubiquitin-activating enzyme, E1, a ubiquitin-conjugating Trip12 antibody was used for IP (Figure 2B, left panel), enzyme, E2, and a ubiquitin-protein ligase, E3 [13,28]. In indicating physical interaction of these proteins in addition, E4 enzymes have recently been shown to en- C2C12 cells. However, endogenous Trip12 was not co- hance the polyubiquitination reaction in some cases [35]. immunoprecipitated when Sox6 antibody was used for To reconstitute the ubiquitination reaction, an equal IP (Figure 2B, right panel), suggesting that the C- amount of purified SOX6-myc (substrate) was combined Figure 2 Trip12 protein expression in adult mouse tissues and interaction between endogenous Sox6 and Trip12 proteins. (A) Western blot analysis was performed to determine Trip12 protein expression in adult mouse tissues. Each lane contained 30 μg of protein except for the C12C12 samples, which contained 15 μg of protein per lane. Relevant protein size markers (kDa) are indicated to the left. β-actin was used as a loading control. (B) Co-IP of endogenous Sox6 and Trip12 proteins using nuclear fractions of differentiating C2C12 cells. Input lane contained 5% (15 μg) of pre-cleared nuclear protein. Antibodies used for pull down are listed under IP, and antibodies used for Western blot (WB) are indicated below each panel. GST antibody was used as a negative control. An et al. Skeletal Muscle 2013, 3:11 Page 7 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 with the representative E1 and E2 enzymes along with the presence of Trip12 siRNA, intensity of the ~224 kDa full-length TRIP12 E3 ligase (see the Methods section). As band (full-length mouse Trip12) in C2C12 myotubes de- shown in Figure 3A, addition of TRIP12 resulted in a creased to almost an undetectable level. Using this greater level of ubiquitination of SOX6-myc protein in a Trip12 siRNA, the result of Trip12 inhibition on Sox6 time-dependent manner detected as an increasing upward protein levels was determined at a time point (48 h after smear (lanes 4, 5, and 6). To test if this process occurred switching to DM) where Sox6 protein levels are known in vivo,we performed in vivo ubiquitination assays. to be decreasing in differentiating C2C12 myotube HEK293 cells were cotransfected with the full-length cultures (see Additional file 2: Figure S1). TRIP12-HSV, Sox6-FLAG, and HA-tagged ubiquitin As can be seen in Figure 4B, reduction of Trip12 led (HA-Ub) expression vectors. Immunoprecipitation was to a threefold increase in the amount of Sox6 protein, performed with FLAG antibody to concentrate Sox6- whereas there was no change in protein levels of the nu- FLAG protein, followed by Western blotting using HA clear transcription factor TATA-box binding protein antibody to detect ubiquitinated Sox6-FLAG protein. (Tbp) (used as a control). Therefore, Sox6 is a specific As TRIP12 protein expression was doubled and tripled, substrate to Trip12, by which it is polyubiquitinated and the amounts of high-molecular-weight products, which degraded. Whether a Trip12-mediated change in Sox6 are the sign of multiple additions of HA-Ub molecules, protein level had any significant biological effects was also increased (Figure 3B), reflecting the addition of initially tested by analysis of changes in mRNA expres- multiple ubiquitin molecules to the Sox6-FLAG sub- sion of Myh7, a direct Sox6 target gene [9,10]. When strate. As would be expected from a fixed pool of sub- Trip12 siRNA treatment effectively reduced Trip12 strate, the non-ubiquitinated Sox6-FLAG protein levels mRNA levels to approximately one third of the control decreased (Figure 3B), indicating that more Sox6-FLAG level (Figure 4C), Myh7 mRNA levels were reduced by 40%, was polyubiquitinated as Trip12 levels increased. whileSox6mRNAlevelswereunchanged (Figure4C), confirming the suppressive role of Sox6 on the Myh7 Knockdown of Trip12 in C2C12 myotubes results in a transcription at the post-transcriptional level. concurrent increase in Sox6 protein levels and a decrease of Myh7 transcription Since polyubiquitination frequently leads to degradation Sox6 protein is degraded by the 26S proteasome of the target substrate, we next tested the effect of The final stage of the ubiquitin-proteasome system is siRNA-induced Trip12 knockdown on Sox6 protein degradation of the polyubiquitinated substrate by the levels in muscle cells. As shown in Figure 4A, in the 26S proteasome [13]. Therefore, we determined the Figure 3 TRIP12 ubiquitinates Sox6 in vitro and in vivo. (A) In vitro ubiquitination of SOX6 protein was performed using purified SOX6-myc as the substrate and TRIP12 along with the combination of enzymes as indicated in the figure (see the Methods section for details). Lanes 4–6 contain 3 μg of TRIP12-HSV. Western blot (WB) was performed using c-Myc antibody to detect the degree of ubiquitination of Sox6-myc. Asterisk indicates non-specific bands detected in all reactions. (B) Sox6 is ubiquitinated by TRIP12 in vivo. HEK293 cells were cotransfected with plasmid DNAs encoding HA-Ub, Sox6-FLAG, and increasing amounts (0.5, 1, and 1.5 μg) of TRIP12-HSV, and lysates were immunoprecipitated (IP) with anti-DYKDDDDK (FLAG) antibody, and then processed for Western blotting (WB) using anti-HA antibody. Asterisk indicates non-specific bands detected in all IP samples, although it is possible that these bands also contain ubiquitinated Sox6-FLAG protein of lower molecular weights (with ~1 to 4 ubiquitin moieties) in Sox6-FLAG-transfected samples. The same membrane was subsequently incubated with anti-DYKDDDDK (FLAG) antibody to detect Sox6-FLAG protein; 10 μg (2%) of input protein samples (lysates) was also subjected to Western blotting using anti-TRIP 12 antibody to detect both endogenous TRIP12 and overexpressed TRIP12-HSV proteins. An et al. Skeletal Muscle 2013, 3:11 Page 8 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 Figure 4 Trip12 controls Sox6 protein level in C2C12 cells. (A) siRNA-mediated knockdown of Trip12 resulted in an increase of Sox6 protein levels in C2C12 cells. C2C12 cells were transfected with siRNA for either EGFP (negative control) or Trip12 in triplicate, and lysates were analyzed by Western blotting using a 7.5% gel. (B) Densitometric analysis of the Western blot in (A) shows approximately 3-fold increase in the Sox6 protein level in Trip12 siRNA-treated cells, while the Tbp protein level was not affected. Data are normalized for those from EGFP siRNA- transfected cells and represented as mean ± SD (n = 3). (C) Trip12 knockdown lowered mRNA level of Myh7, a known Sox6 target. Total RNA was extracted from mock- or siRNA-transfected C2C12 cells, and mRNA levels of Trip12 and Myh7 were quantified by reverse transcription-quantitative PCR (RT-qPCR). Data are normalized for those from EGFP siRNA-transfected cells and represented as mean ± SD (n = 3). **p < 0.005. effect of proteasome inhibition on the cellular Sox6 pro- protein appeared slightly more stable than Sox6 tein levels. (Figure 5B). The Trip12 half-life estimated here is in A standard approach for examining proteasome- agreement with a similar turnover rate of the human mediated degradation of proteins is to use a combination TRIP12 protein recently estimated in HeLa cells using a of a proteasome inhibitor such as MG132 and the pro- proteomic approach [37]. tein synthesis inhibitor cycloheximide (CHX) [36]. Prior Although the Sox6 half-life was found to be relatively to this experiment, the stability of the Sox6 protein long (~24 h), we first tried a conventional proteasome along with Trip12 and Tbp in C2C12 myotubes was de- inhibitor (MG132) to examine proteasomal degradation termined. C2C12 myotube cultures (induced for 24 h in of Sox6. Since extended treatment with proteasome in- DM) were treated with CHX to stop new protein synthe- hibitors is toxic to the cells [38], C2C12 cells were sis, following which, the stability of the Sox6, Trip12, treated with CHX in the absence or presence of MG132 and Tbp proteins was examined using Western blot. only for 6 h, and the Sox6 protein level was examined by After 24 h of CHX treatment, the myotubes looked Western blotting. As shown in Additional file 3: Figure healthy and morphologically normal. As shown in S2B, protein levels of myogenin, a short-lived transcrip- Figure 5A and summarized in Figure 5B, Sox6 and Tbp tion factor [39], were reduced by CHX alone and were appeared more stable than Trip12. The protein half-life recovered by MG132 addition, verifying the effect of of Trip12 was estimated as ~9 h, whereas Sox6 protein’s MG132 in our experimental condition. To conduct a half-life was estimated as ~24 h (Figure 5B). The Tbp detailed examination of the response of Sox6 protein to An et al. Skeletal Muscle 2013, 3:11 Page 9 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 Figure 5 Sox6 is degraded by proteasome in C2C12 cells. (A) Time course of Trip12, Sox6, and Tbp protein levels in C2C12 cells in the presence of cycloheximde (CHX) analyzed by Western blotting using a 4-15% gradient gel (n = 3). (B) Densitometric analysis of the Western blot in (A). Band intensity of each protein was normalized to that of 0 h in CHX and represented as mean ± SD (n = 3). (C) Validation of Psmd1 knockdown by siRNA. Total RNA was extracted from mock- or siRNA-transfected C2C12 cells, and the mRNA level of Psmd1 (the gene encoding a regulatory subunit of the 26S proteasome) was quantified by RT-qPCR. Data are normalized to EGFP siRNA-transfected cells and represented as mean ± SD (n = 3). (D) A Western blot using a 7.5% gel showing an increase in the Sox6 protein level in Psmd1 siRNA-transfected C2C12 cells. (E) Densitometric analysis of Western blotting results shows a ~4 fold increase in the Sox6 protein level in Psmd1 siRNA-treated C2C12 cells. A smaller increase was observed for Tbp protein. Data are normalized for those from EGFP siRNA-transfected cells and represented as mean ± SD (n = 3). (F) Psmd1 knockdown reduced the mRNA level of Myh7, a known target of Sox6. Data are normalized to EGFP siRNA-transfected cells and represented as mean ± SD (n = 3). **p < 0.005. MG132, we used a gel condition optimized for resolving inhibitor did not indicate proteasomal degradation of proteins of 75–100 kD (7.5%) and detected two closely Sox6 in an unambiguous manner as expected given its sized bands reacting with the Sox6 antibody (Additional long half-life, we next used a siRNA targeting a prote- file 3: Figure S2B). Although MG132 did not change the asome subunit to confirm proteasomal degradation of overall amount of the Sox6 proteins (total intensity of Sox6. We chose to knockdown Psmd1, a subunit of the lid the two bands; densitometric analysis also showed no domain of the 26S proteasome, because knockdown of statistical difference between these samples: data not Psmd1 effectively inhibits proteasome activity without sig- shown), there was a significant increase in the upper nificant toxicity [40]. Using Psmd1 siRNA, we successfully band along with a significant decrease in the lower band obtained close to 50% reduction in Psmd1 mRNA levels (Additional file 3: Figure S2B), indicating that there is a (Figure 5C). Under this condition, levels of Sox6 proteins discrete shift in the Sox6 protein size in the presence of significantly increased (Figures 5D and 5E). As shown in MG132. We found that this size difference was caused Figure 5F, Psdm1 siRNA treatment of C2C12 myotubes by phosphorylation of the Sox6 protein (Additional file 3: also resulted in significant reduction of Myh7 mRNA Figure S2C). These results suggest a possibility that stabil- levels. Taken together, these results indicate that Sox6 is ity of the Sox6 protein could be regulated by the balance degraded by the 26S proteasome and that inhibition of the between phosphorylated and non-phosphorylated Sox6 26S proteasome activity by siRNA resulted in increased proteins. We are currently investigating this possibility. protein levels of Sox6 and increased suppression of its Because treatment with a conventional proteasome target gene Myh7 in C2C12 myotubes. An et al. Skeletal Muscle 2013, 3:11 Page 10 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 Knockdown of Trip12 expression results in opposite and Myog mRNA levels as well as a significant increase in changes in expression of fast and slow MyHC genes in Myh4 (day 4 in DM), a mirror image of their expression C2C12 myotubes patterns in the Sox6 knockout muscle [9,12], suggesting Because both Trip12 knockdown and 26S proteasome that Trip12 regulation of Sox6 can have a fundamental inhibition in myotube cultures resulted in increased impact on the fiber type identity. Sox6 protein levels (Figures 4A, 4B, 5D and 5E), and de- creased mRNA levels of the Sox6 target Myh7 (Figures 4C Discussion and 5F), we hypothesized that knockdown of Trip12 We have previously reported that a primary mechanism would result in fiber type-specific gene expression changes in fiber type-specific gene expression involves the coor- opposite to those observed in the Sox6 KO mouse, i.e., an dinated suppression of hundreds of slow fiber-specific increase in fast fiber-specific gene expression and a genes by the transcription factor Sox6 [9]. Muscle- decrease in slow fiber-specific gene expression [9,12]. specific knockout of Sox6 results in a dramatic increase To test this hypothesis, C2C12 myoblasts were in slow MyHC-β (Myh7) expression coupled with a sig- transfected with Trip12 siRNA or EGFP siRNA (a negative nificant decrease in fast MyHC-IIb (Myh4) expression control) and the mRNA levels of Myh7 (slow isoform), [9,12]. Muscle-specific loss of Sox6 effectively shifts myogenin (Myog: preferentially expressed in slow muscle muscle tissue into the slower fiber phenotype leading to [41]), and Myh4 (fast MyHC-IIb isoform) were compared the prediction that regulation of Sox6 activity in skeletal in differentiating C2C12 myotube cultures. Brown et al. muscle could be a focal point of fiber type differenti- [42] showed that in differentiating C2C12 cells, Myh7 ation. Our present study expands our investigations into increases after day 1 and then begins to decline after day the mechanisms of fiber type determination by looking 4; conversely, Myh4 begins to increase after day 4 in differ- at the regulation of Sox6 by the ubiquitin-proteasome entiation medium. In this experiment, we normalized pathway. The identification of the relatively unexplored mRNA expression levels in Trip12 siRNA transfected E3 ligase Trip12 as a Sox6 partner protein led us to in- myotubes against the control siRNA transfected vestigate the impact of Trip12 activity on cellular Sox6 myotubes. Changes outside the control expression levels protein levels. Using siRNA of Trip12, we were able to are reflected as greater or lesser than 1. As shown in show that the knockdown of Trip12 expression in Figure 6 and Additional file 4: Figure S3, transfection of C2C12 myotubes resulted in (1) an increase in Sox6 pro- Trip12 siRNA caused a significant decrease in both Myh7 tein levels, (2) the simultaneous downregulation of both Figure 6 Trip12 modulates mRNA levels of fiber-type-specific genes in differentiating C2C12 cells. C2C12 cells were transfected with siRNA for either EGFP (negative control) or Trip12, and medium was switched to DM 24 h after transfection to induce differentiation into myotubes. Total RNA was then extracted every 24 h, and the time course of mRNA levels of Trip12, Myh4 (MyHC-IIb), Myh7 (MyHC-I/β), and myogenin (Myog) were quantified by RT-qPCR using Huwe1 and Tbp as reference genes (see the Methods section for details). Data are normalized for those from EGFP siRNA-transfected cells and represented as mean ± SD (n=3). No expression change between EGFP siRNA- and Trip12 siRNA-transfected samples is expressed as 1 on the graph. *p < 0.05, **p < 0.01, ***p < 0.005. An et al. Skeletal Muscle 2013, 3:11 Page 11 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 the slow fiber-specific Myh7 and preferentially-slow at the carboxyl terminus and is evolutionarily highly myogenin, and (3) upregulation of Myh4 (fast), thus conserved [28,45]. Amino acid sequences lying outside of suggesting that Trip12, by controlling Sox6 protein the HECT domain vary among the family members lead- levels, plays a critical role in regulating muscle fiber ing to suggestions that this N-terminus part of the protein type-specific gene expression. This observation signifi- primarily determines substrate specificity [28,46]. In the cantly expands the role of E3 ligases in skeletal muscle ubiquitination of target proteins, many E3 ubiquitin li- beyond their traditionally conceived role in atrophy and, gases possess the capability of polyubiquitination, while possibly, into the realm of fundamental developmental some are only capable of monoubiquitination when alone, processes in the muscle. and require an E4 ligase for attachment of additional The role of Sox6 as a transcriptional suppressor of ubiquitin molecules [35]. Our in vitro ubiquitination assay slow fiber-specific isoform genes in skeletal muscle has results show that TRIP12 can polyubiquitinate Sox6 in the been reported in mice and zebrafish [9,10,43]. These re- absence of an E4 (Figure 3A). Interestingly, the length of ports indicate that the function of Sox6 during skeletal smear bands detected was different between in vitro and muscle development is likely conserved through verte- in vivo:while in vitro ubiquitination assay showed smear brate evolution since the loss of Sox6 activity in both bands up to ~250 kDa (Figure 3A), a higher molecular species results in the increased mRNA expression of weight of smear bands was detected in in vivo slow fiber-specific genes [9,12,43]. Sox6 is also expressed ubiquitination assay (Figure 3B). These results suggest that in the adult heart at a moderate level [17,34]. During there might be additional factor(s) in vivo (e.g., an E4 lig- heart development, it has been shown that Sox6 is ase), which assists further elongation of polyubiquitin expressed at a high level in proliferating cardiomyocyte chains by Trip12. TRIP12, a more recently recognized E3 progenitor cells, and a reduced Sox6 expression level ubiquitin ligase, was identified based on its sequence causes the progenitors to exit from the cell cycle and homology to the yeast UFD4 E3 ubiquitin ligase [32]. The differentiate [44]. Therefore, control of the cellular levels E3 ligase activity of TRIP12 HECT domain was subse- of the Sox6 protein is potentially a vital mechanism to quently demonstrated [18]. The TRIP12 substrates known regulate development of both cardiac and skeletal to date include APP1-BP1 (amyloid beta precursor muscle. protein-binding protein 1) [32], the SWI/SNF chromatin To identify Sox6 interacting proteins, we performed remodeling complex subunit BAF57 [31], the tumor sup- yeast two-hybrid screening and identified TRIP12, an E3 pressor protein ARF [30], and the RING finger E3 ligase ubiquitin ligase, as a SOX6-interacting protein. In adult RNF168 [47]. Our current report adds the transcription mouse tissues, the Trip12 protein is expressed highest in factor Sox6 as a TRIP12 substrate polyubiquitinated in testis and moderately in both skeletal muscle and the muscle cells. Since the TRIP12 clone identified by yeast heart (Figure 2A). Endogenous Trip12 and Sox6 proteins two-hybrid screening contained only the HECT domain in C2C12 cells directly interact (Figure 2B), and Trip12 (Figure 1B) and the TRIP12 HECT domain alone could polyubiquitinates Sox6 both in vitro and in vivo (Figure 3). interact with the SOX6 coiled-coil domain (Figure 1D), we Ubiquitination is a complex and multifaceted regulatory speculate that the HECT domain is the minimal necessary system involving the modification of a target protein by requirement for TRIP12 to recognize SOX6 as a substrate. addition of either one or multiple ubiquitin molecules In fact, the dual role of the TRIP12 HECT domain (E3 leading to many different outcomes depending on the ligase catalytic activity and the substrate recognition) has number of ubiquitins added or the topology of the ubiqui- been previously demonstrated [18]. Park and colleagues tin chain [14,45]. The addition of one or more ubiquitin reported that the HECT domain of TRIP12, but not molecules to a protein is a multistep process involving at UBE3A (E6-AP), could recognize and ubiquitinate an E3 least three ubiquitin enzymes, E1, E2 and E3. In this art- ligase substrate on its own [32]. Our current result indi- cates that this observation holds for the recognition of icle, we focused on Trip 12, an E3 ubiquitin ligase, which polyubiquitinates Sox6. The E3 ligases’ role in the SOX6 as a substrate by TRIP12. The ability of the HECT ubiquitination chain comes at the end of a multistep domain to function as a ligase catalytic domain as well as a substrate recognition module could be a unique trait for process, where it functions to attach one or more ubiqui- tin molecules to a specific target protein [13,45]. the TRIP12 HECT domain, which will be tested in the Currently, it is estimated that more than 1,000 distinct E3 future as more information becomes available for the other HECT domain E3 ligases. ligase genes are encoded in the human genome [13]. The majority of these are classified into the RING family of E3 The addition of one or more ubiquitin molecules to a ligases with a much smaller family of genes belonging to target protein lends a tremendous flexibility to the cellu- the HECT family, of which Trip12 is a member [28,46]. lar fate of that protein. Depending on the ubiquitin code The HECT domain consists of approximately 350 amino (for review, see Komander et al. [45]), the activity, loca- acids encoding the E3 ligase catalytic domain. It is located tion, partners and even the protein’s survival can be An et al. Skeletal Muscle 2013, 3:11 Page 12 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 altered. We found that Sox6 is polyubiquitinated by outcomes on differentiation of various cell types. Although TRIP12 and is degraded by the 26S proteasome. In con- Sox6 mRNA is transcribed in multiple tissues at moderate trast to other critical myogenic regulatory factors such as to high levels [17,34], its role as a transcription factor is MyoD or myogenin, the half-life of Sox6 is surprisingly quite distinct in different cell types [22]. This is because, long (~24 h). Half-lives of MyoD, Myf5, and myogenin in part, cell type-specific Sox6 functions are dictated by its have been reported as 0.8-1.0 h, <1.0 h, and ~1.0 h, regulatory co-factors [22,24]. To achieve developmental respectively [39,48,49]. The stable nature of Sox6 may stage-specific functions, however, close monitoring of the reflect its function in muscle development; Sox6 plays a Sox6 protein expression level is also critical. In light of role in specification of myotube phenotype by regulating this, the ubiquitin-proteasome system mediated degrad- fiber type differentiation and is expressed at high levels in ation of the Sox6 protein likely plays a crucial role in tem- myotubes [10]. Therefore, the relative stability of Sox6 poral regulation of the Sox6 protein level during may be necessary for stably maintaining the fiber pheno- development. It is possible that a different set of E3 ligases type of myotubes. The ubiquitously expressed nuclear catalyzes ubiquitination of the Sox6 protein depending on protein Tbp is also stable, supporting the correlation of different cellular environments (e.g.. different cell types) or protein stability with maintenance of gene expression. different signals. Therefore, further uncovering of the In muscle, the overall stability of a multigene regulator ubiquitin-proteasome regulation of transcription factors such as Sox6 lends stability to a phenotype such as will enrich our understanding of the multi-layered mecha- muscle fiber type, but this same stability could quickly nisms of transcriptional regulation during development. become a hindrance when external signals demand a fiber type shift. Thus, the polyubiquitination of Sox6 by Conclusions the Trip12 E3 ligase leading to its degradation would We have shown here that Trip12, a HECT domain E3 allow for a swift response in the face of changing work- ubiquitin ligase, targets Sox6, a suppressor for slow loads of the muscle. Therefore, Trip12 likely functions as fiber-specific genes. In addition, we showed that Trip12 a pivot point in fiber type transition, altering the balance is involved in regulation of the fast MyHC isoform gene between slow and fast fiber gene expression. To date, expression. Based on our current data, we propose that two other mechanisms regulating Sox6 protein activities in skeletal muscle, E3 ligases have a significant role in within the cells have been reported: regulation by regulating fiber type-specific gene expression, expanding microRNA and sumoylation. The Sox6 mRNA has a their functional importance in muscle beyond their well- long (~5 kb) 3’-UTR sequence, which contains multiple established role in atrophy. microRNA seed sequences [22]. MicroRNAs are expressed in a cell type-specific manner and are known Additional files to regulate the protein levels of cell type-specific genes [50]. miR-499 is known to target the Sox6 mRNA and Additional file 1: Table S1. siRNAs, PrimeTime qPCR Assays, and suppresses Sox6 protein expression in skeletal muscle TaqMan Gene Expression Assay used in the experiments. [51-54] and in differentiating cardiomyocytes [44]. In the Additional file 2: Figure S1. Temporal changes in Sox6 expression central nervous system, it has been shown that miR-219 level in differentiating C2C12 cells. (A) RNA was prepared every 24 h from C2C12 cells grown in DM at the indicated time. RT-qPCR for Sox6 targets Sox6 in oligodendrocytes and induces the ter- was performed using TaqMan Gene Expression Assays (Applied minal differentiation of oligodendrocytes [55,56]. Post- Biosystems). Gapdh expression level was used to normalize data. Fold translational modification can also control the function increase in mRNA levels (0 h in DM = 1) was calculated at each time point. Each data set represents four independent RT-qPCR experiments of a protein. In the case of Sox6, it has been shown that (mean ± SD). (B) Crude protein extract prepared from differentiating SUMO (small ubiquitin-related modifier) reduces the C2C12 cells (0, 24, 48, and 72 h in DM) was separated on a 7% SDS-PAGE binding affinity of Sox6 to DNA, thus changing its tran- gel (100 μg per well), and Western blotting was performed using Sox6 antibody. A ~90 kDa Sox6 band was detected in all C2C12 cell extracts. scriptional activity [57]. The highest Sox6 protein expression was observed at 24 h in DM and In the current report, we have demonstrated that then rapidly dropped close to the 0 h level at 48 h in DM. β-actin was Trip12 ubiquitinates the Sox6 protein, Sox6 is degraded used as a loading control. by the 26S proteasome, and in both these instances, the Additional file 3: Figure S2. MG132 treatment shifted the intensity of the two Sox6 bands in differentiating C2C12 cells. (A) A procedure for expression of Sox6 target genes are affected. Since skel- MG132 experiments. C2C12 cells were seeded in six-well plates at a etal muscle is highly plastic in nature and in transitions 5 density of 3 × 10 cells/well and incubated in GM for 24 h. Cells were between the slow and fast fiber types [58], the speedy re- rinsed with PBS once and incubated in DM. After 48 h, 100 μg/ml CHX and 1 μM MG132 or the same volume of DMSO (with which MG132 moval of relevant transcription factors (e.g., Sox6) via stock solution was prepared) were added to DM, and cells were the ubiquitin-proteasome system could allow for a rapid incubated for another 6 h before Western blotting. (B) Western blots of response to external cues. MG132-treated cells. Three independent samples were prepared for each treatment. Tbp was used as a loading control. (C) Phosphatase treatment Taken together, regulation of Sox6 protein activity, both of C2C12 nuclear protein; 5 μg of nuclear protein prepared from post-transcriptional and post-translational, has critical An et al. Skeletal Muscle 2013, 3:11 Page 13 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 11. Hagiwara N, Ma B, Ly A: Slow and fast fiber isoform gene expression is differentiating C2C12 cells without phosphatase inhibitor cocktail was systematically altered in skeletal muscle of the Sox6 mutant, p100H. treated with 10 units of calf intestinal alkaline phosphatase (CIP) in the Dev Dyn 2005, 234:301–311. absence or presence of phosphatase inhibitor cocktail (PIC) for 10 min at 12. Quiat D, Voelker KA, Pei J, Grishin NV, Grange RW, Bassel-Duby R, Olson EN: 37°C and analyzed by Western blotting using a 7.5% gel. Positions of two Concerted regulation of myofiber-specific gene expression and muscle Sox6 bands, which are always seen on 7.5% gels but not on 4-15% performance by the transcriptional repressor Sox6. Proc Natl Acad Sci USA gradient gels, are indicated. CIP treatment shifted the position of Sox6 2011, 108:10196–10201. band from upper to lower size, indicating that the upper band consists 13. Varshavsky A: The ubiquitin system, an immense realm. Annu Rev Biochem of a phosphorylated form of Sox6. 2012, 81:167–176. Additional file 4: Figure S3. Relative mRNA levels of fiber-type-specific 14. Geng F, Wenzel S, Tansey WP: Ubiquitin and proteasomes in transcription. genes shown in Figure 6. Relative mRNA levels against reference genes Annu Rev Biochem 2012, 81:177–201. -ΔCt (Huwe1 and Tbp) were calculated using the formula 2 and 15. Glass DJ: Signaling pathways perturbing muscle mass. Curr Opin Clin Nutr represented as mean ± SD (n=3). Metab Care 2010, 13:225–229. 16. Cohen-Barak O, Yi Z, Hagiwara N, Monzen K, Komuro I, Brilliant MH: Sox6 regulation of cardiac myocyte development. Nucleic Acids Res 2003, Abbreviations 31:5941–5948. CC: Coiled-coil; Co-IP: Co-immunoprecipitation; CHX: Cycloheximide; 17. Cohen-Barak O, Hagiwara N, Arlt MF, Horton JP, Brilliant MH: Cloning, DM: Differentiation medium; GM: Growth medium; HECT: Homologous to characterization and chromosome mapping of the human SOX6 gene. the E6-AP carboxyl terminus; IP: Immunoprecipitation; MyHC: Myosin heavy Gene 2001, 265:157–164. chain; qPCR: Quantitative PCR; RT-qPCR: Reverse transcription-quantitative 18. Park Y, Yoon SK, Yoon JB: The HECT domain of TRIP12 ubiquitinates PCR; siRNA: Small interfering RNA; WB: Western blot. substrates of the ubiquitin fusion degradation pathway. J Biol Chem 2009, 284:1540–1549. Competing interests 19. Niikura Y, Nonaka T, Imajoh-Ohmi S: Monitoring of caspase-8/FLICE The authors declare no competing interests. processing and activation upon Fas stimulation with novel antibodies directed against a cleavage site for caspase-8 and its substrate, FLICE- Authors’ contributions like inhibitory protein (FLIP). J Biochem 2002, 132:53–62. CIA performed Co-IP, in vitro and in vivo ubiquitination assays, siRNA 20. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, experiments, cycloheximide and MG132 experiments, RT-qPCR, Western Speleman F: Accurate normalization of real-time quantitative RT-PCR blotting, and densitometric analysis and contributed to writing the data by geometric averaging of multiple internal control genes. manuscript. EG carried out yeast two-hybrid screening to identify TRIP12 and Genome Biol 2002, 3:RESEARCH0034. vector construction. NH conceived, designed, and supervised the study and 21. Andersen CL, Jensen JL, Orntoft TF: Normalization of real-time contributed to writing the manuscript. All authors read and approved the quantitative reverse transcription-PCR data: a model-based variance final manuscript. estimation approach to identify genes suited for normalization, applied to bladder and colon cancer data sets. Cancer Res 2004, 64:5245–5250. Acknowledgements 22. Hagiwara N: Sox6, jack of all trades: a versatile regulatory protein in The authors thank Dr. Aldrin V. Gomes for both technical support and vertebrate development. Dev Dyn 2011, 240:1311–1321. discussion and Mr. Adam Jenkins for helpful discussion. This work was 23. Iguchi H, Urashima Y, Inagaki Y, Ikeda Y, Okamura M, Tanaka T, Uchida A, supported by grants from the National Institutes of Health (AR055209) and Yamamoto TT, Kodama T, Sakai J: SOX6 suppresses cyclin D1 promoter American Heart Association (0635070N) to N.H. activity by interacting with beta-catenin and histone deacetylase 1, and its down-regulation induces pancreatic beta-cell proliferation. Received: 25 October 2012 Accepted: 8 April 2013 J Biol Chem 2007, 282:19052–19061. Published: 10 May 2013 24. Kamachi Y, Uchikawa M, Kondoh H: Pairing SOX off: with partners in the regulation of embryonic development. Trends Genet 2000, 16:182–187. 25. Lefebvre V, Li P, de Crombrugghe B: A new long form of Sox5 (L-Sox5), References Sox6 and Sox9 are coexpressed in chondrogenesis and cooperatively 1. Baylor SM, Hollingworth S: Intracellular calcium movements during activate the type II collagen gene. EMBO J 1998, 17:5718–5733. excitation-contraction coupling in mammalian slow-twitch and fast- 26. Ohe K, Tamai KT, Parvinen M, Sassone-Corsi P: DAX-1 and SOX6 molecular twitch muscle fibers. J Gen Physiol 2012, 139:261–272. interplay results in an antagonistic effect in pre-mRNA splicing. 2. Schiaffino S, Reggiani C: Fiber types in mammalian skeletal muscles. Dev Dyn 2009, 238:1595–1604. Physiol Rev 2011, 91:1447–1531. 27. Connor F, Wright E, Denny P, Koopman P, Ashworth A: The Sry-related 3. Zierath JR, Hawley JA: Skeletal muscle fiber type: influence on contractile HMG box-containing gene Sox6 is expressed in the adult testis and and metabolic properties. PLoS Biol 2004, 2:e348. developing nervous system of the mouse. Nucleic Acids Res 1995, 4. Chemello F, Bean C, Cancellara P, Laveder P, Reggiani C, Lanfranchi G: 23:3365–3372. Microgenomic analysis in skeletal muscle: expression signatures of 28. Rotin D, Kumar S: Physiological functions of the HECT family of ubiquitin individual fast and slow myofibers. PLoS One 2011, 6:e16807. ligases. Nat Rev Mol Cell Biol 2009, 10:398–409. 5. Seene T, Kaasik P, Umnova M: Structural rearrangements in contractile 29. Lee JW, Choi HS, Gyuris J, Brent R, Moore DD: Two classes of proteins apparatus and resulting skeletal muscle remodelling: effect of exercise dependent on either the presence or absence of thyroid hormone for training. J Sports Med Phys Fitness 2009, 49:410–423. interaction with the thyroid hormone receptor. Mol Endocrinol 1995, 6. Selsby JT, Morine KJ, Pendrak K, Barton ER, Sweeney HL: Rescue of 9:243–254. dystrophic skeletal muscle by PGC-1alpha involves a fast to slow fiber 30. Chen D, Shan J, Zhu WG, Qin J, Gu W: Transcription-independent ARF type shift in the mdx mouse. PLoS One 2012, 7:e30063. regulation in oncogenic stress-mediated p53 responses. Nature 2010, 7. Webster C, Silberstein L, Hays AP, Blau HM: Fast muscle fibers are 464:624–627. preferentially affected in Duchenne muscular dystrophy. Cell 1988, 31. Keppler BR, Archer TK: Ubiquitin-dependent and ubiquitin-independent 52:503–513. control of subunit stoichiometry in the SWI/SNF complex. J Biol Chem 8. Schakman O, Gilson H, Thissen JP: Mechanisms of glucocorticoid-induced 2010, 285:35665–35674. myopathy. J Endocrinol 2008, 197:1–10. 32. Park Y, Yoon SK, Yoon JB: TRIP12 functions as an E3 ubiquitin ligase of 9. An CI, Dong Y, Hagiwara N: Genome-wide mapping of Sox6 binding sites APP-BP1. Biochem Biophys Res Commun 2008, 374:294–298. in skeletal muscle reveals both direct and indirect regulation of muscle terminal differentiation by Sox6. BMC Dev Biol 2011, 11:59. 33. Kajiro M, Tsuchiya M, Kawabe Y, Furumai R, Iwasaki N, Hayashi Y, Katano M, 10. Hagiwara N, Yeh M, Liu A: Sox6 is required for normal fiber type Nakajima Y, Goto N, Watanabe T, et al: The E3 ubiquitin ligase activity of differentiation of fetal skeletal muscle in mice. Dev Dyn 2007, Trip12 is essential for mouse embryogenesis. PLoS One 2011, 236:2062–2076. 6:e25871. An et al. Skeletal Muscle 2013, 3:11 Page 14 of 14 http://www.skeletalmusclejournal.com/content/3/1/11 34. Hagiwara N, Klewer SE, Samson RA, Erickson DT, Lyon MF, Brilliant MH: Sox6 57. Fernandez-Lloris R, Osses N, Jaffray E, Shen LN, Vaughan OA, Girwood D, is a candidate gene for p100H myopathy, heart block, and sudden Bartrons R, Rosa JL, Hay RT, Ventura F: Repression of SOX6 transcriptional neonatal death. Proc Natl Acad Sci USA 2000, 97:4180–4185. activity by SUMO modification. FEBS Lett 2006, 580:1215–1221. 35. Hoppe T: Multiubiquitylation by E4 enzymes: 'one size' doesn't fit all. 58. Schiaffino S, Sandri M, Murgia M: Activity-dependent signaling pathways Trends Biochem Sci 2005, 30:183–187. controlling muscle diversity and plasticity. Physiology (Bethesda) 2007, 36. Zhou P: Determining protein half-lives. Methods Mol Biol 2004, 284:67–77. 22:269–278. 37. Boisvert FM, Ahmad Y, Gierlinski M, Charriere F, Lamont D, Scott M, Barton G, Lamond AI: A quantitative spatial proteomics analysis of proteome doi:10.1186/2044-5040-3-11 turnover in human cells. Mol Cell Proteomics 2012, 11:M111 011429. Cite this article as: An et al.: Trip12, a HECT domain E3 ubiquitin ligase, targets Sox6 for proteasomal degradation and affects fiber type-specific 38. Kisselev AF, Goldberg AL: Proteasome inhibitors: from research tools to gene expression in muscle cells. Skeletal Muscle 2013 3:11. drug candidates. Chem Biol 2001, 8:739–758. 39. Vinals F, Ventura F: Myogenin protein stability is decreased by BMP-2 through a mechanism implicating Id1. J Biol Chem 2004, 279:45766–45772. 40. Silva JM, Li MZ, Chang K, Ge W, Golding MC, Rickles RJ, Siolas D, Hu G, Paddison PJ, Schlabach MR, et al: Second-generation shRNA libraries covering the mouse and human genomes. Nat Genet 2005, 37:1281–1288. 41. Hughes SM, Chi MM, Lowry OH, Gundersen K: Myogenin induces a shift of enzyme activity from glycolytic to oxidative metabolism in muscles of transgenic mice. J Cell Biol 1999, 145:633–642. 42. Brown DM, Parr T, Brameld JM: Myosin heavy chain mRNA isoforms are expressed in two distinct cohorts during C2C12 myogenesis. J Muscle Res Cell Motil 2012, 32:383–390. 43. von Hofsten J, Elworthy S, Gilchrist MJ, Smith JC, Wardle FC, Ingham PW: Prdm1- and Sox6-mediated transcriptional repression specifies muscle fibre type in the zebrafish embryo. EMBO Rep 2008, 9:683–689. 44. Sluijter JP, van Mil A, van Vliet P, Metz CH, Liu J, Doevendans PA, Goumans MJ: MicroRNA-1 and −499 regulate differentiation and proliferation in human-derived cardiomyocyte progenitor cells. Arterioscler Thromb Vasc Biol 2010, 30:859–868. 45. Komander D, Rape M: The ubiquitin code. Annu Rev Biochem 2012, 81:203–229. 46. Scheffner M, Staub O: HECT E3s and human disease. BMC Biochem 2007, 1(8 Suppl):S6. 47. Gudjonsson T, Altmeyer M, Savic V, Toledo L, Dinant C, Grofte M, Bartkova J, Poulsen M, Oka Y, Bekker-Jensen S, et al: TRIP12 and UBR5 suppress spreading of chromatin ubiquitylation at damaged chromosomes. Cell 2012, 150:697–709. 48. Lindon C, Albagli O, Domeyne P, Montarras D, Pinset C: Constitutive instability of muscle regulatory factor Myf5 is distinct from its mitosis- specific disappearance, which requires a D-box-like motif overlapping the basic domain. Mol Cell Biol 2000, 20:8923–8932. 49. Sun L, Trausch-Azar JS, Muglia LJ, Schwartz AL: Glucocorticoids differentially regulate degradation of MyoD and Id1 by N-terminal ubiquitination to promote muscle protein catabolism. Proc Natl Acad Sci USA 2008, 105:3339–3344. 50. Bartel DP: MicroRNAs: target recognition and regulatory functions. Cell 2009, 136:215–233. 51. Bell ML, Buvoli M, Leinwand LA: Uncoupling of expression of an intronic microRNA and its myosin host gene by exon skipping. Mol Cell Biol 2010, 30:1937–1945. 52. McCarthy JJ, Esser KA, Peterson CA, Dupont-Versteegden EE: Evidence of MyomiR network regulation of beta-myosin heavy chain gene expression during skeletal muscle atrophy. Physiol Genomics 2009, 39:219–226. 53. Rossi AC, Mammucari C, Argentini C, Reggiani C, Schiaffino S: Two novel/ ancient myosins in mammalian skeletal muscles: MYH14/7b and MYH15 are expressed in extraocular muscles and muscle spindles. J Physiol 2010, Submit your next manuscript to BioMed Central 588:353–364. and take full advantage of: 54. van Rooij E, Quiat D, Johnson BA, Sutherland LB, Qi X, Richardson JA, Kelm RJ Jr, Olson EN: A family of microRNAs encoded by myosin genes • Convenient online submission governs myosin expression and muscle performance. Dev Cell 2009, • Thorough peer review 17:662–673. 55. Dugas JC, Cuellar TL, Scholze A, Ason B, Ibrahim A, Emery B, Zamanian JL, • No space constraints or color figure charges Foo LC, McManus MT, Barres BA: Dicer1 and miR-219 Are required for • Immediate publication on acceptance normal oligodendrocyte differentiation and myelination. Neuron 2010, • Inclusion in PubMed, CAS, Scopus and Google Scholar 65:597–611. 56. Zhao X, He X, Han X, Yu Y, Ye F, Chen Y, Hoang T, Xu X, Mi QS, Xin M, et al: • Research which is freely available for redistribution MicroRNA-mediated control of oligodendrocyte differentiation. Neuron 2010, 65:612–626. Submit your manuscript at www.biomedcentral.com/submit

Journal

Skeletal MuscleSpringer Journals

Published: May 10, 2013

References