This Article Abstract Full Text (PDF) Supplemental material Other Versions of this Article: MCB.02020-06v1 MCB.02020-06v2 27/12/4374    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wang, H. Articles by Guttridge, D. C. Search for Related Content PubMed PubMed Citation Articles by Wang, H. Articles by Guttridge, D. C.

 Previous Article  |  Next Article 

Molecular and Cellular Biology, June 2007, p. 4374-4387, Vol. 27, No. 12
0270-7306/07/$08.00+0     doi:10.1128/MCB.02020-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

NF-B Regulation of YY1 Inhibits Skeletal Myogenesis through Transcriptional Silencing of Myofibrillar Genes ^,

Human Cancer Genetics Program, Department of Molecular Virology, Immunology and Medical Genetics,¹ Integrated Biomedical Graduate Program,² Molecular, Cellular, and Developmental Biology Graduate Program,³ The Arthur G. James Comprehensive Cancer Center, The Ohio State University, Columbus, Ohio 43210⁴

Received 27 October 2006/ Returned for modification 11 December 2006/ Accepted 20 March 2007

	ABSTRACT

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
NF-B signaling is implicated as an important regulator of skeletal muscle homeostasis, but the mechanisms by which this transcription factor contributes to muscle maturation and turnover remain unclear. To gain insight into these mechanisms, gene expression profiling was examined in C2C12 myoblasts devoid of NF-B activity. Interestingly, even in proliferating myoblasts, the absence of NF-B caused the pronounced induction of several myofibrillar genes, suggesting that NF-B functions as a negative regulator of late-stage muscle differentiation. Although several myofibrillar promoters contain predicted NF-B binding sites, functional analysis using the troponin-I2 gene as a model revealed that NF-B-mediated repression does not occur through direct DNA binding. In the search for an indirect mediator, the transcriptional repressor YinYang1 (YY1) was identified. While inducers of NF-B stimulated YY1 expression in multiple cell types, genetic ablation of the RelA/p65 subunit of NF-B in both cultured cells and adult skeletal muscle correlated with reduced YY1 transcripts and protein. NF-B regulation of YY1 occurred at the transcriptional level, mediated by direct binding of the p50/p65 heterodimer complex to the YY1 promoter. Furthermore, YY1 was found associated with multiple myofibrillar promoters in C2C12 myoblasts containing NF-B activity. Based on these results, we propose that NF-B regulation of YY1 and transcriptional silencing of myofibrillar genes represent a new mechanism by which NF-B functions in myoblasts to modulate skeletal muscle differentiation.

	INTRODUCTION

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
NF-B belongs to the Rel family of transcription factors, which regulates an exceptionally large number of genes, particularly those involved in immune and inflammatory responses (28, 35). Five mammalian Rel proteins, RelA/p65, c-Rel, RelB, p50 (NF-B1), and p52 (NF-B2), have been identified (22, 55). All of these proteins share a highly conserved 300-amino-acid Rel homology domain in the amino-terminal half of the protein. NF-B subunits are able to homo- or heterodimerize to form transcription factor complexes with a range of DNA binding and activation potentials. Different NF-B dimers exhibit different binding affinities for B sites bearing the consensus sequence GGGRNNYYCC, where R is purine, Y is pyrimidine, and N is any base (43). Although all Rel members bind DNA, only RelA/p65 (from here on referred to as p65), c-Rel, and RelB have an extended carboxy terminus harboring a transactivation domain. The most widely studied form of NF-B is a heterodimer composed of p50 and p65 subunits containing a potent transactivator function.

In most cells, NF-B is found mainly sequestered in the cytoplasm, bound in an inactive complex with an IB inhibitory protein family member. This family includes IB, IBß, IB, Bcl-3, p100, and p105 (8, 9, 22, 61). These proteins function as inhibitors through ankyrin repeats that bind to the DNA binding domain of NF-B and mask the nuclear localization signal, thus maintaining the transcription factor as an inactive complex. Upon stimulation, a rapid and transient activation of IB kinase (IKK) occurs, leading to the phosphorylation of IB (66). This phosphorylation event targets IB for ubiquitin-dependent degradation by the 26S proteasome complex, resulting in the liberation and nuclear translocation of NF-B (11, 35). Classical NF-B activation is induced by a wide variety of different stimuli, including inflammatory cytokines, such as tumor necrosis factor alpha (TNF-) and interleukin 1 (IL-1), bacterial lipopolysaccharide (LPS), and viruses that signal through the ß catalytic and regulatory subunits of IKK (16, 28, 52).

In addition to its well-established role in activating the transcription of genes involved in immunological responses, NF-B also functions in the regulation of multiple cellular processes related to proliferation, adhesion, migration, and viability (7). A role for NF-B in skeletal myogenesis also has emerged in recent years. Studies performed with cultured myoblasts demonstrate that NF-B DNA binding and transcriptional activities decrease during differentiation, and this activity of NF-B functions to maintain cells in an undifferentiated state (26, 40). This occurs through the ability of NF-B to stimulate cell cycle progression concomitant with its regulation of cyclin D1. Consistent with these findings, the upstream activator of NF-B, RIP2, has also been found to negatively regulate myogenesis (46). Furthermore, activation of NF-B in response to proinflammatory cytokines, such as TNF-, is required to block muscle differentiation by suppressing the synthesis of the skeletal muscle-specific transcription factor, MyoD (27, 37). In C2C12 myotubes (27, 36, 49) or in intact muscles (13, 31), NF-B activity has also been shown to regulate muscle wasting in various forms of atrophy. The mechanisms by which NF-B promotes muscle wasting remain unclear but may be related to its ability to inhibit myogenesis (27) or to induce protein turnover through the regulation of the ATP-dependent ubiquitin proteasome pathway (13, 42, 49).

In order to enhance our understanding of NF-B function in skeletal muscle, we performed gene expression profiling with C2C12 myoblasts containing or lacking NF-B activity. Interestingly, even under proliferating conditions, myoblasts devoid of NF-B function expressed uncharacteristically high levels of myofibrillar genes, suggesting that the previously described basal activity of NF-B in proliferating myoblasts might function to suppress muscle genes involved in late-stage differentiation. Although troponin promoters and enhancer elements were found to contain several potential NF-B binding sites, our evidence indicates that NF-B regulation of troponin-I2 expression does not occur through direct DNA binding. The search for an indirect mediator revealed the myogenic transcriptional repressor YinYang1 (YY1). This ubiquitous zinc finger transcription factor is highly conserved and targets a variety of cellular and viral genes (21, 24, 60), recognizing a core 5'-CCATNTT-3' CCAT box sequence flanked by flexible nucleotides (32). Depending on the cell type and promoter context, YY1 can either activate or repress transcription, and in doing so has been found to physically associate with histone acetyltransferases CBP and p300, transcription factors such as Sp1 and p53, the histone deacetylases HDAC1 to -3, and the Arg-specific methytransferase PRMT1 (4, 50, 54, 58, 65).

In skeletal muscle, YY1 is considered to play a negative role in myogenesis by directly repressing the synthesis of late-stage differentiation genes, including -skeletal actin (39), muscle creatine kinase (MCK), and myosin heavy chain IIb (MyHCIIb) (15, 62). Recent findings demonstrate that transcriptional silencing of the MCK enhancer and the MyHCIIb promoter in proliferating myoblasts is regulated by YY1 binding and the subsequent recruitment of the Polycomb suppressor complex containing the Ezh2 methyltransferase in association with HDAC1 (15). During skeletal myogenesis, the YY1/Ezh2/HDAC repressive complex is removed from MCK and MyHC DNA and replaced by activators serum response factor and MyoD along with associated acetyltransferases CBP and p300/CBP-associated factor (PCAF), which are necessary for the expression of the late-stage differentiation genes.

In this report, we identify YY1 as a direct target gene of NF-B and demonstrate that, analogous to the MyHCIIb gene, YY1 binding to the troponin-I2 enhancer functions to recruit a chromatin silencing complex that maintains transcriptional suppression of myofibrillar genes in undifferentiated myoblasts. These results highlight a new function of NF-B in skeletal myogenesis, a finding that may prove useful for better understanding the role of this transcription factor in both physiological and pathophysiological states of skeletal muscle.

	MATERIALS AND METHODS

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Cell culture and transfections. C2C12, 10T1/2, and 293T cells were obtained from ATCC. C2C12 myoblasts were cultured and differentiated as previously described (26), while 10T1/2 and 293T cells were grown in Dulbecco's modified Eagle medium containing 5% fetal bovine serum. Transient transfections of C2C12 and 10T1/2 cells were performed using 12-well plates with SuperFect reagent (QIAGEN). Cell extracts were prepared and luciferase activity was monitored as previously described (26). Transient transfection into 293T cells was performed by using a 6-cm dish with ProFection mammalian transfection system-calcium phosphate reagent (Promega). Forty-eight hours posttransfection, RNA and protein extracts were prepared for analysis. Stable transfection into C2C12 cells was performed with SuperFect reagent (QIAGEN) followed by G418 selection as described earlier (26, 40). Small interfering RNA (siRNA) transfection was performed using Lipofectamine reagent as suggested by the manufacturer (Invitrogen).

Plasmids. To construct a YY1-luciferase reporter plasmid, an amplified fragment (–774 to +69) of the mouse YY1 promoter was cut with Xho1 and HindIII and subcloned into the XhoI and HindIII sites of the pGL3 basic vector (Promega). Tnni2 luciferase reporter (TnI-Luc) plasmid was used as described previously (26, 40). A mutant version of this reporter plasmid containing a mutant YY1 binding site (TnI-Luc-Mut) was generated using a QuikChange XL mutagenesis kit (Stratagene). The YY1 expression plasmid was a gift from Y. Shi (Harvard University). siRNA sequences targeting p65 were self-designed, and annealed oligonucleotides were cloned into the pSuper retrovirus vector (OligoEngine). siRNA oligomers targeting YY1 and MyoD were obtained from Santa Cruz Biotechnology.

EMSA, immunoblotting, and immunostaining. Electrophoretic mobility shift assays (EMSAs) were prepared as previously described (26). Briefly, 5 µg of nuclear extract was incubated with 1 mM phenylmethylsulfonyl fluoride and 1 µg of poly(dI-dC)-poly(dI-dC) (Amersham Biosciences) for 10 min at room temperature. To this mixture, 2 x 10⁴ cpm of a ³²P-labeled oligonucleotide probe was added in a buffer consisting of 10 mM Tris-HCl (pH 7.7), 50 mM NaCl, 0.5 mM EDTA, 1 mM dithiothreitol, and 10% glycerol. Complexes were resolved on a nondenaturing 5% polyacrylamide gel and subsequently exposed on X-OMAT film (Kodak) or on a Typhoon phosphorimager (Amersham). For supershift assays, antibodies raised against p65 (Rockland), p50 (NLS; Santa Cruz Biotechnology), and YY1 (Santa Cruz Biotechnology) were preincubated with nuclear extracts for 10 min at room temperature before the addition of phenylmethylsulfonyl fluoride and poly(dI-dC)-poly(dI-dC). Immunoblotting was performed as previously described (29). Immunofluorescence assays of muscle sections were performed as described previously (1, 29), using an anti-YY1 monoclonal antibody (Santa Cruz Biotechnology) at 1:100 dilutions. Immunofluorescence assays of embryonic sections were performed as described previously (29), using an anti-MyHC monoclonal antibody (Sigma) at 1:100 dilutions and an anti-YY1 polyclonal antibody (Santa Cruz Biotechnology) at 1:100 dilutions.

ChIP assays. Chromatin immunoprecipitation (ChIP) assays were performed as recommended by the manufacturer (Upstate), using 2 µg of antibodies against p65 (Upstate), YY1 (Santa Cruz Biotechnology), Ezh2 (Zymed), HDAC1 (Santa Cruz Biotechnology), trimethyl-histone H3-K27 (Upstate), PCAF (Santa Cruz Biotechnology), or acetyl-histone H4-K9 (Upstate), or with isotype immunoglobulin G (IgG; Sigma) as the negative control. Genomic DNA pellets were resuspended in 20 µl of water. PCR was performed with 2 µl of immunoprecipitated material, and products were analyzed by using agarose gel and visualized with a GelDoc documentation system (Bio-Rad Laboratories).

Reverse transcriptase (RT) PCR and real-time PCR. Total RNA from cells was extracted using TRIzol reagent (Invitrogen). cDNA was synthesized using a SuperScript double-stranded cDNA synthesis system (Invitrogen), and 2 µl of cDNA was PCR amplified. Real-time PCR was performed using an iCycler (Bio-Rad Laboratories) with a SyberGreen MasterMix (Bio-Rad, Hercules, CA). One microliter of cDNA was used as the template in a total reaction volume of 25 µl containing final concentrations of 1x SYBR Green Super mix and 0.5 µM (each) for the forward and reverse primers. A list of oligonucleotides sequences used for various assays throughout the study is available (see Table S1 in the supplemental material).

Mice and genotyping. Animals were housed in the animal facility at the Ohio State University Comprehensive Cancer Center under sterile conditions, with temperature and humidity kept constant, and were fed a standard diet. Treatment of the mice was in accordance to the institutional guidelines of the Animal Care and Use Committee. Genotyping of mice was performed by PCR analysis from prepared tail DNA. For cardiotoxin (CTX) studies, approximately 4-week-old mice were anesthetized with isofluorane, and 50 µl of CTX at 10 µg/ml was injected into the tibialis anterior muscle. Muscles were harvested, and total protein was extracted for Western blot analysis or frozen for immunostaining procedures.

Microarray and bioinformatic analysis. RNA was extracted using TRIzol reagent (Invitrogen) and purified using RNeasy columns (QIAGEN). Eight micrograms of total RNA was reverse transcribed to cDNA, which was subsequently used as a template to prepare biotinylated cRNA. Mu74Av2 arrays from Affymetrix were hybridized at The Ohio State University Comprehensive Cancer Center microarray core facility. To obtain change values (n-fold), analysis was performed using MicroArray Suite (MAS) software, version 4.0. Complete gene expression array data are available upon request. Promoter sequences were retrieved from the previously described Mammalian Promoter Database (59). Putative NF-B binding sites were predicted using MATCH software.

	RESULTS

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
NF-B negatively regulates myofibrillar gene expression in proliferating myoblasts. The ability of NF-B to repress myogenesis by promoting the cell cycle (26) and inhibiting MyoD synthesis (27, 36, 56) suggests that this transcription factor functions in muscle differentiation through multiple mechanisms. To address this point, microarray analysis was performed with both proliferating C2C12 vector control myoblasts and myoblasts stably expressing the IB super repressor (IB-SR) mutant, in which both basal and activated forms of NF-B are strongly inhibited (Fig. 1A). Results revealed that several myofibrillar genes were highly induced in myoblasts lacking NF-B activity, including several isoforms of the troponin genes, Tnnc, Tnnt, and Tnni, as well as myosin light and heavy chains and -actin (Table 1). This was a surprising finding, given that myofibrillar genes are normally transcriptionally silent in proliferating undifferentiated muscle cells, and it supported the notion that such genes may be suppressive target genes of NF-B.

View larger version (18K): [in this window] [in a new window]   FIG. 1. Myofibrillar gene expression is increased in IB-SR-expressing C2C12 myoblasts. (A) EMSA for NF-B binding activity in TNF--treated vector and IB-SR-expressing C2C12 myoblasts. (B and C) Confirmation of microarray results (Table 1) was performed with semiquantitative RT-PCR (B) and real-time PCR (C) analyses of myofibrillar genes expressed in vector and IB-SR-expressing C2C12 myoblasts. Real-time PCR data were normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA. (D) C2C12 vector and IB-SR cells were cultured in 12-well plates and transiently transfected in triplicate with a TnI-Luc reporter plasmid along with LacZ-expressing plasmids. Cell extracts were prepared 48 h posttransfection, and relative luciferase units were determined by normalizing to ß-galactosidase (ß-Gal) activity. (E and F) Proposed models for how NF-B functions to repress myofibrillar gene expression in proliferating skeletal myoblasts, either through direct DNA binding (E) or indirectly through the regulation of an unknown downstream transcriptional repressor (F).

View this table: [in this window] [in a new window]   TABLE 1. Representative genes expressed in C2C12 IB-SR myoblasts

NF-B represses the Tnni2 promoter by a mechanism independent of direct DNA binding. NF-B-mediated transcriptional repression of myofibrillar genes could be envisioned to occur through several mechanisms. One possibility is that NF-B directly binds to the promoter of these genes to inhibit their transcription, as modeled in Fig. 1E. Although NF-B is not generally considered to be a direct transcriptional repressor, evidence exists that in response to UV radiation or chemotherapy, DNA-bound NF-B undergoes posttranslational modifications that switch the transcription factor from an transcriptional activator to a transcriptional suppressor (14). In addition, NF-B has been found to participate in a silencing complex present on the 3'-flanking sequence of the -globin gene (64). In contrast to these conditions, we reasoned that a second possible way by which NF-B mediates suppression of myofibrillar genes is indirectly through the regulation of one or more downstream mediators (Fig. 1F).

To test each of these conditions, we first used computational analysis to screen for putative NF-B consensus binding sites. Results revealed a relatively high number of sites in the 5'-end regions of numerous myofibrillar genes (data not shown). To refine our search, we modeled myofibrillar genes after troponin promoters, since these were the genes that were highly responsive to the absence of NF-B activity in C2C12 myoblasts. A scanning of the Tnni2 proximal promoter (45) revealed five sequence motifs with similarity to NF-B consensus binding sites, GGGRNNYYCC (referred to as sites A to E) (Fig. 2A). To test whether sites A to E were competent for NF-B binding, radiolabeled probes were generated for EMSA analysis with nuclear extracts prepared from C2C12 myoblasts either untreated or treated with TNF-. An oligonucleotide derived from the major histocompatability complex class I (MHC-I) promoter was used as a positive control. Results showed that of the five elements, only sites A and C produced binding complexes whose migrations were comparable to that of the MHC-I control probe (Fig. 2B). TNF- treatment increased NF-B DNA activity on site A but not on site C. Furthermore, a supershift analysis revealed that unlike the complexes formed with MHC-I, which contained both p65 and p50 subunits, sites A and C from the Tnni2 promoter contained neither of these proteins (Fig. 2C).

View larger version (37K): [in this window] [in a new window]   FIG. 2. NF-B does not directly bind to the Tnni2 promoter. (A) Schematic illustration of the tnni2 promoter. Five putative NF-B binding sites, A to E, are indicated as black boxes upstream of the transcriptional start site, and sequences are shown. (B) Nuclear extracts were prepared from C2C12 myoblasts either untreated (–) or treated (+) with TNF- for 15 min, and EMSAs were performed with radiolabeled probes corresponding to putative NF-B sites A to E or to the MHC-I probe containing a bona fide NF-B binding site. (C) Supershift EMSAs were performed with TNF-treated C2C12 nuclear extracts preincubated with IgG (lanes 1, 5, and 9) or antisera specific for p65 (lanes 2, 6, and 10) or p50 subunits (lanes 3, 7, and 11). For competition EMSAs, extracts were preincubated with a 100-fold molar excess of unlabeled oligonucleotides containing MHC-I NF-B binding sites (lanes 4, 8, and 12). Arrows denote p65- and p50-containing complexes and supershift complexes.

Identification of YY1 as a novel transcriptional target of NF-B. The data above indicated that NF-B inhibition of Tnni2 gene expression is likely to follow the model illustrated in Fig. 1E. Interestingly, a second microarray analysis that was performed with TNF--treated mouse embryonic fibroblasts (MEFs) identified YY1 as a potential NF-B target gene (Table 2). Given that YY1 functions as a transcriptional repressor of -actin and MyHCIIb genes in undifferentiated myoblasts (15) and that its expression is under control of the NF-B-inducing cytokine, IL-1ß (47, 48), this result suggested that NF-B-mediated repression of myofibrillar genes might occur through YY1. To investigate this possibility, microarray results were confirmed by examining YY1 expression levels in TNF--treated C2C12 myoblasts. Results showed that TNF- induction of YY1 occurred at levels similar to those of IB (Fig. 3A), which was used as a positive control NF-B-regulated gene. Real-time PCR results for YY1 and IB were found to be quite consistent with induction levels obtained from microarray analysis with TNF--treated MEFs. YY1 regulation by TNF- was also observed for murine 10T1/2 fibroblasts (Fig. 3B) and human 293T cells (data not shown), demonstrating that NF-B regulation of YY1 is not cell type or species specific. In addition, classical activation of NF-B by IL-1ß and LPS stimulation also led to YY1 induction (Fig. 3C and D). Furthermore, consistent with our previous findings that NF-B activity decreases during C2C12 myogenesis (26, 40), we found that this regulation was, in turn, associated with a reduction in YY1 expression (Fig. 3E).

View this table: [in this window] [in a new window]   TABLE 2. Representative genes induced by TNF- in MEFs

View larger version (20K): [in this window] [in a new window]   FIG. 3. YY1 expression is regulated by NF-B. C2C12 (A) or 10T1/2 (B) cells were plated in 10-cm plates and treated with 10 ng/ml of TNF-. Total RNA was prepared at the indicated times, and the amount of YY1 mRNA was quantified by real-time PCR and normalized to GAPDH. For comparison, real-time PCR was repeated while probing for a known NF-B target gene, IB. Induction changes were calculated by setting YY1 or IB levels in untreated cells to a value of 1. 10T1/2 cells were also treated with 10 ng/ml of IL-1ß (C) or 50 ng/ml LPS (D). The amounts of YY1 mRNA were quantified by real-time PCR. (E) Nuclear extracts were prepared from C2C12 at the indicated times in differentiating myoblasts (DM), and EMSAs were performed with radiolabeled probes corresponding to the MHC-I probe. The amounts of YY1 mRNA at the indicated times were quantified by real-time PCR.

View larger version (15K): [in this window] [in a new window]   FIG. 4. YY1 regulation is p65 dependent. (A) Total RNA and protein were prepared from C2C12 myoblasts stably expressing vector or p65, and YY1 RNA was measured by real-time PCR. The inset shows the expression of Flag-tagged p65 normalized to tubulin. (B) Total RNA and protein were prepared from vector control (V) and IB-SR (SR)-expressing C2C12 myoblasts, and YY1 RNA and protein were measured by real-time PCR and Western blot analysis. (C) 293T cells were plated in 10-cm plates and transfected with either vector control, p65 shRNA, or a negative control shRNA construct (Cont. shRNA). Total RNA was prepared from the same transfected plate, and RT-PCR was performed to measure the expression of YY1 normalized to GAPDH mRNA. Total protein was also extracted, and Western blots were performed to detect the expression of p65. The blot was reprobed for -tubulin as a loading control. (D and E) p65+/+ and p65–/– or p50+/+ and p50–/– MEF cells were treated with TNF- for 1 or 2 h and quantified for YY1 mRNA by real-time PCR.

View larger version (30K): [in this window] [in a new window]   FIG. 5. YY1 is regulated by p65 in vivo. (A) Tibialis anterior muscles were isolated from 4-week-old TNF–/–; p65+/+ (p65+/+) and TNF–/–; p65–/– (p65–/–) mice, and YY1 was measured by RT-PCR (upper panel) and Western blot analysis (lower panel). (B) YY1 measurement from quadriceps or gastrocnemius muscles from p65+/+ and p65–/– mice. (C) YY1 RNA expression from nonmuscle tissues (spleen, thymus, heart, and lung) isolated from 4-week-old p65+/+ and p65–/– mice. (D) Tibialis anterior muscles from control mice were injected with CTX and, at indicated days, Western blot analysis was performed with probes for YY1 and -tubulin. (E) Analysis of YY1 from tibialis anterior muscles from p65+/+ and p65–/– after 6 days of CTX injections by Western blotting. (F) Uninjected (control) or 6-day CTX-injected tibialis anterior muscles from p65+/+ and p65–/– mice were sectioned and immunostained for YY1 (red) and nuclei (Hoechst; blue) and photographed at x20 magnification. Arrows indicate the muscle fibers with YY1 expression localized in the nuclei. Scale bar = 50 µM. (G) Percentage of central nucleation (%CLN) was determined for 6-day CTX-treated tibialis anterior muscles from control and p65-null littermate pairs (n = 2) by counting the number of central nuclei in randomly chosen fields from a minimum of 1,500 fibers; P < 0.05, Student's t test.

Next, we asked whether such regulation was pertinent in postnatal regenerating muscle. Tibialis anterior muscles from 4-week-old mice were therefore injected with CTX to induce muscle injury and subsequent regeneration. Results showed an evident induction of YY1 protein (Fig. 5D). Further immunostaining analysis with wild-type mice revealed that induced YY1 expression was localized to both the nuclei (see arrows) and cytoplasm of muscle fibers (Fig. 5F). Significantly, similar analysis performed with p65^–/– mice showed a dramatic impairment of YY1 induction (Fig. 5E and F), indicating that YY1 regulation in postnatal myogenesis is NF-B dependent. Furthermore, to determine whether the absence of p65 could modulate myogenesis in vivo, CTX-treated muscle sections were analyzed for central nucleation, which is a hallmark feature of regenerative muscle. In two littermate pairs of isolated progenies, muscle regeneration was significantly enhanced in p65^–/– mice (Fig. 5G), a finding that is in line with a recent report demonstrating that muscle regeneration is accelerated in CTX-injured mouse muscles lacking NF-B signaling (44) and that further supports our own results showing that NF-B functions as a negative regulator of myogenesis in dystrophic muscle (2).

NF-B regulation of YY1 occurs through promoter binding. To determine whether YY1 is a direct transcriptional target of NF-B, sequence analysis of the 5' flanking element of the YY1 gene was performed. Although the characterization of the YY1 promoter has been described (51), little is known regarding the cis- and trans-acting regulatory factors. Scanning the YY1 proximal promoter revealed two motifs with similarities to an NF-B consensus site, which we referred to as sites A (GGGGGCCCCC) and B (GGAGGACCCT), located at positions –170 and –153 relative to the start site of transcription (Fig. 6A). In an analogous fashion to our examination of putative NF-B binding sites in the Tnni2 promoter, radiolabeled oligonucleotides containing these sequences corresponding to sites A and B were used in EMSA analyses with nuclear extracts prepared from TNF--treated myoblasts. Results identified a complex formed at site A but not at site B, and this complex could be successfully supershifted with antibodies against both p65 and p50 (Fig. 6B). In addition, binding to site A was abolished either when site A was mutated to a nonconforming NF-B binding sequence, TTGGGCCCAA, or when an excess of unlabeled site A probe was used (Fig. 6B and data not shown). To validate NF-B binding, ChIP analysis was performed to assess the in vivo association of NF-B to the YY1 promoter. Chromatins derived from C2C12 vector or IB-SR myoblasts were immunoprecipitated with an antibody against p65, and a fragment encompassing the putative NF-B site in the YY1 promoter was amplified with specific primers. As shown in Fig. 6C, p65 was found associated with the YY1 promoter in the vector control, but not in IB-SR-expressing cells, in a manner similar to that seen with p65 binding to the IB gene. Together, these results indicate that the YY1 promoter contains a functional NF-B binding site.

View larger version (35K): [in this window] [in a new window]   FIG. 6. NF-B binds and transcriptionally regulates the YY1 promoter. (A) Schematic illustration of the murine YY1 promoter containing two putative NF-B consensus binding sequences (diamonds) and a Sp1 binding site (oval) upstream of the transcription start site. (B) C2C12 myoblasts were either untreated (–) or treated (+) with TNF- for 15 min, and EMSA was performed with radiolabeled probes containing putative YY1-A, YY1-B, or YY1-A-mutant NF-B binding sites. Complexes formed were identified by supershift EMSA with antisera specific to p65 (lanes 3, 8, and 13), p50 (lanes 4, 9, and 14) or IgG (lanes 5, 10, and 15). Arrows denote supershift complexes. (C) ChIP analysis with p65 or control IgG was performed with chromatin derived from either vector control (V) or IB-SR (SR)-expressing C2C12 myoblasts. The precipitated DNA fragments were amplified with specific oligonucleotides containing the putative NF-B binding site on the YY1 promoter (upper panel). As a control, ChIP assays were repeated with the IB promoter (bottom panel). Total inputs are indicated. (D) 10T1/2 cells were plated in triplicate in 12-well plates; the next day, transient transfections were performed with DNA consisting of 0.2 µg of YY1-Luc reporter plasmids along with 0.2 µg of a LacZ expression plasmid and 0.01 µg of a p65 or p50 expression plasmid. At 48 h posttransfection, cell extracts were prepared and relative luciferase units were determined by normalizing to ß-Gal protein. (E) C2C12 myoblast cell lines were generated to stably express luciferase reporters under the control of the YY1 promoter, either wild type (WT) or mutated (Mut) in the NF-B binding site. Luciferase reporter assays were then performed in cells untreated or treated with TNF- (10 ng/ml) for 6 h.

Myofibrillar genes are transcriptionally suppressed by YY1 in an NF-B-dependent fashion. Since our results suggested that YY1 was positively regulated by p65, we speculated that p65 inhibition of troponin, and possibly other myofibrillar genes, might occur through this transcriptional mediator. This was an appealing model since YY1 had already been shown to repress myofibrillar genes, such as -actin and MyHCIIb, which were also identified in Fig. 1 to be derepressed in the absence of NF-B. To test this hypothesis, we first determined whether YY1 was capable of binding to the Tnni2 regulatory region. YY1 has a loose consensus binding site, with the 5'-CCAT-3'core sequence being essential for binding (32). An inspection of the mouse Tnni2 promoter and enhancer regions revealed several essential regulatory elements, including MEF2, E-box, CCAC, and CAGG box elements, located in the first intron (Fig. 7A). In addition, three CCAT sites, referred to as A, B, and C, were identified in close proximity to these elements, with site A (GCCCATCTTC) exhibiting the highest consensus to an YY1 binding sequence, (C/G)(G/T/A)CCATNTTN. Oligonucleotides were therefore generated from site A, or from a known YY1 binding site from the MyHCIIb promoter (15), and EMSAs were subsequently performed with nuclear extracts prepared from either C2C12 myoblasts or myotubes. Consistent with previous findings (15), YY1 was found to bind to the MyHCIIb probe, which we could confirm by supershift analysis (Fig. 7B, lanes 1 and 2), and as expected, YY1 binding to MyHCIIb diminished in myotubes (Fig. 7B, lanes 3 and 4). EMSA results further revealed the formation of a complex bound to the putative YY1 site A of the Tnni2 enhancer (Fig. 7B, lane 5). This complex could also be supershifted with an YY1 antibody and, like the MyHCIIb promoter, was reduced in myotubes. In comparison, no complex was detected when the YY1 CCAT box was mutated to a GGAT sequence (Fig. 7B, lanes 9 to 12), supporting the functionality of this site.

View larger version (27K): [in this window] [in a new window]   FIG. 7. YY1 binds and represses the Tnni2 enhancer. (A) Schematic illustration of the murine Tnni2 first intron region containing a MEF2 binding site, two E-boxes, CAGG, CCAT regulatory elements, and three putative CCAT YY1 binding sites (diamonds; sites A, B, and C). (B) Nuclear extracts were isolated from C2C12 myoblasts (MB) or myotubes (MT), and EMSA was performed with probes containing either the known YY1 binding site in the MyHCIIb promoter or the putative wild-type YY1 site A (Tnni2) or mutant site A [Tnni2(Mut)] in the Tnni2 enhancer. Supershift EMSAs with no antibody (–) or with YY1 antibodies (+) were used to confirm YY1 binding complexes. The arrow denotes a supershift complex. (C) ChIP assays with either an YY1 antibody or control IgG were performed on chromatin isolated from C2C12 myoblasts. The precipitated DNA fragments were amplified with specific olionucleotides spanning regions A, B, or C of the Tnni2 enhancer. Total input is indicated. (D) C2C12 cells were plated in 12-well plates and transfected with 0.2 µg of MyHC-Luc, TnI-Luc, or TnI-Mut-Luc reporter plasmid along with 0.05 µg of MyoD and indicated amounts of a YY1 expression plasmid. Luciferase activities were determined 48 h posttransfection and normalized to ß-Gal protein. (E) C2C12 myoblasts were cotransfected with siRNA YY1 and a Tnni2 reporter plasmid, and reporter activity was determined by luciferase assays after 2 days in differentiation conditions. (F) ChIP assays were performed with C2C12 myoblasts (MB) or myotubes (MT) with antibodies against YY1, Ezh2, HDAC1, trimethyl-histone H3-k27, PCAF, or acetyl-histone H3-K9. Primers specific to Tnni2 intron region A were used for PCR amplification. Total inputs are indicated.

To assess the dependency of NF-B in YY1 repression of Tnni2, YY1 association with the Tnni2 enhancer was tested with TNF--treated C2C12 vector or IB-SR-expressing myoblasts. While YY1 binding to Tnni2 was readily detected in vector control cells by ChIP-PCR, this association was strongly reduced in myoblasts devoid of NF-B activity (Fig. 8A). Since increasing evidence has suggested that YY1 functions as an important regulator of skeletal muscle genes, we speculated that the NF-B regulation of YY1 could serve a general mechanism applying to other myofibrillar genes. To test this idea, YY1 association with several myofibrillar gene promoters was examined in C2C12 vector or IB-SR-expressing myoblasts. Results showed a clear association of YY1 on the promoters of two known target muscle genes, MyHCIIb and -actin, in vector control cells but not in SR cells (Fig. 8A). In addition, we were able to identify putative YY1 binding sites on promoter/enhancer regions of two other characterized myofibrillar genes, cardiac troponin T (cTnnt) and cardiac/slow troponin C (Tnnc) genes (18, 63), which were the same genes found highly expressed in C2C12 IB-SR myoblasts (Table 1). In a result similar to those for the other late differentiating genes, ChIP-PCR analysis detected elevated YY1 association with cTnnt and Tnnc promoters in vector control over that of IB-SR-containing cells (Fig. 8A). Together, results indicate that NF-B regulation of YY1 acts as a general mechanism in repressing myofibrillar gene expression in proliferating myoblasts.

View larger version (17K): [in this window] [in a new window]   FIG. 8. YY1-mediated repression of myofibrillar gene expression by NF-B. (A) ChIP assays were performed with chromatin from C2C12 vector (V) or IB-SR (SR) myoblasts with antibodies against YY1 or IgG. Primers specific to Tnni2, MyHCIIb, -skeletal actin, cTnnt, or Tnnc promoters/enhancers were used for PCR amplification. Total input DNA is indicated. (B) Total RNA was prepared from C2C12 IB-SR myoblasts transiently transfected with vector or YY1, and Tnni2 and MyHCIIb RNA were measured by real-time PCR. (C) Total RNA was prepared from C2C12 vector (V) or IB-SR (SR) myoblasts, and MyoD levels were measured by semiquantitative RT-PCR. (D) Western blot of MyoD C2C12 IB-SR myoblasts transfected with control or MyoD siRNA. (E) Cells were transfected as for panel D, and Tnni2 and MyHCIIb expression was measured by real-time PCR.

	DISCUSSION

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Over the last few years, there has been increasing evidence supporting the involvement of the NF-B signaling pathway in regulating skeletal muscle homeostasis, findings potentially relevant to development, muscle repair, and disease (2, 13, 23, 25, 27, 33). In our own studies, we described the low constitutive NF-B activity contained by proliferating myoblasts, which functions to inhibit myogenesis by stimulating cell cycle progression (26) and by suppressing the synthesis of MyoD, especially in response to proinflammatory mediators (27, 36). However, other reports suggest that NF-B signaling may also possess promyogenic activity (5, 19, 34), which highlights the complexity of this signaling pathway in relation to muscle differentiation and argues that additional investigation into the role of this transcription factor is warranted. The current study was thus undertaken to gain insight into the mechanisms underlying NF-B regulation of myogenesis.

Identification of YY1 as an NF-B target gene. Microarray results, along with validating real-time PCR analysis, revealed to our surprise the pronounced induction of certain myofibrillar genes in proliferating myoblasts lacking NF-B activity. These genes, such as -actin, MyHC, and troponin, are typically transcriptionally silent in undifferentiated cells and do not become expressed until late in the myogenic program. The results therefore implied that NF-B participates in the transcriptional silencing of myofibrillar genes. Based on our sequence analysis that revealed several putative NF-B binding sites on the Tnni2 promoter, we considered the possibility that NF-B might mediate the general silencing of Tnni2 and other myofibrillar genes through direct DNA binding. However, repeated attempts using several experimental approaches failed to support this hypothesis. This does not imply that other myofibrillar promoters or enhancers that contain NF-B binding sites might not function by this type of mechanism, but since such examples for NF-B are rare, we considered the alternative model that negative regulation of myofibrillar genes by NF-B occurs through an indirect mechanism.

Our search for a secondary mediator uncovered the transcription factor YY1, which had appeared on a separate microarray analysis performed with TNF--treated fibroblasts. Several of the bona fide NF-B-regulated genes induced on our array list were present at levels that were comparable to the 4.4-fold increase detected with YY1; these included the IL-6 (3.8-fold), JunB (4.0-fold), selectin (4.9-fold), and IB (Nfbi; 7.9-fold) genes. YY1 was an appealing candidate as an NF-B secondary mediator regulating myogenesis since it had already been described as a repressor of skeletal muscle genes (24). Indeed, results from multiple assays showed that YY1 was regulated by NF-B and that regulation occurred through the binding of the p65/p50 subunits bound to the YY1 proximal promoter. Little is known regarding the transcriptional regulation of YY1 activity, and to date only Sp1 has been linked to this regulation (38). The NF-B site that we identified at position –170 lies in proximity to the Sp1 regulatory region at position –47, so it is possible that YY1 transcriptional activity is dependent on the cooperative function of both transcription factors.

Interestingly, YY1 and the RelB subunit of NF-B have been shown to physically interact in complex with the Oct-2 transcription factor on the IgH enhancer in B cells (53). Based on this evidence, we considered the possibility that NF-B regulation of YY1 through the p65 subunit may also occur by a posttranscriptional mechanism. Although we recently showed that p65 binding is critical in maintaining the stability of IBß expression (29), repeated attempts failed to demonstrate an association between p65 and YY1 (data not shown). We conclude from these data that p65 regulation of YY1 is predominantly at the transcriptional level.

Our in vivo studies utilizing p65 wild-type and p65-deficient mice confirmed that NF-B regulation of YY1 is present in muscle tissues, although the YY1 expression level was low in terminally differentiated adult muscles. However, a high YY1 expression level was detected in the regenerating muscle fibers of p65 wild-type mice, reinforcing the argument that YY1 is important during muscle differentiation. Increased YY1 expression was not observed in the regenerating muscle fibers of p65-null mice, further supporting the idea that YY1 expression is regulated by p65 in vivo during postnatal myogenesis. Furthermore, in response to CTX treatment, the absence of p65 caused a marked increase in the regenerative fibers, which is consistent with recent results with IKKß conditional mice in either a CTX (44) or mdx (2) model of muscle injury that NF-B functions as a negative regulator of muscle regeneration. Based on our in vitro data, we presume this inhibition on regeneration by NF-B is dependent on YY1, but this remains to be formally tested.

Given that NF-B regulation of YY1 was shown to occur in multiple cell types, these results also suggest that the potential relevance of this regulation may extend beyond skeletal muscle tissue. The proinflammatory cytokine IL-1ß which, like TNF-, is a potent activator of NF-B, has been shown to negatively regulate cardiac myogenesis through YY1 (47, 48), and YY1 has also been recently implicated in the reduction of cardiac myosin expression in failing hearts (57). Since chronic activation of NF-B has been shown to reduce skeletal muscle myosin expression (27, 36), it is tempting to speculate that NF-B regulation of YY1 may also contribute to the physiological and pathophysiological responses of cardiomyocytes. Moreover, tissue analysis from p65^+/+ and p65^–/– mice revealed that NF-B regulation of YY1 is not limited to skeletal muscle, suggesting the potential importance of this regulation in multiple tissues. Indeed, recent findings that were reported during the preparation of the manuscript demonstrate that in prostate cancer cell lines, YY1 inhibition of the proapoptotic gene Fas is under NF-B control (30).

YY1 binding negatively regulates myofibrillar gene expression. Another revealing aspect of this study was the identification of Tnni2 as a novel target gene for YY1. In the murine Tnni2 promoter, we identified several E boxes, a MEF2 regulatory region, and multiple YY1 consensus sites. EMSAs and ChIP analyses confirmed that only one of these YY1 sites (Fig. 7, site A) was competent for binding, and reporter assays supported the idea that YY1 binding led to transcriptional repression of Tnni2. The identification of Tnni2 adds one more target to a growing list of muscle genes regulated by YY1. In addition, our results with cTnnt and Tnnc enhancers using ChIP analysis lead us to believe these two genes are also YY1 targets. In fact, in performing genome-wide screenings, we have identified numerous YY1 binding sites on other myofibrillar promoters/enhancers, raising the possibilities that YY1 regulates a much larger group of myofibrillar genes. Such a notion is supported by a recent study showing that MEFs derived from YY1 hypomorphic mice have elevated levels of numerous muscle-specific transcripts, including various isoforms of MyHC and troponin genes (3). These data, in conjunction with our current study, underscore the significance of YY1 in regulating skeletal myogenesis during embryogenesis and muscle repair.

Regulation of YY1 represents a novel mechanism by which NF-B inhibits myogenesis. The identification of YY1 as a transcriptional target of NF-B provides further understanding of the function of NF-B in proliferating myoblasts. Collectively, our results support the model shown in Fig. 9, which predicts that basal nuclear activity of NF-B is necessary to regulate YY1 expression in undifferentiated myocytes. What controls basal activity of NF-B in these cells is currently not known. Although myostatin is produced from muscle cells, and similar to NF-B, functions to inhibit myogenesis through the inhibition of MyoD, our recent findings indicate that myostatin activity in myoblasts is independent of NF-B (6). Results have also shown that C2C12 myoblasts constitutively express TNF- (36), which was recently demonstrated to function in promoting myogenesis through the activation of p38 (17). It is therefore possible that NF-B activity in myoblasts is controlled through an autocrine TNF- signaling pathway. Since NF-B is known to regulate TNF-, these data suggest that constitutive activation of NF-B in myoblasts is controlled through a positive feedback loop.

View larger version (20K): [in this window] [in a new window]   FIG. 9. A model for how NF-B negatively regulates troponin and other myofibrillar genes. The model depicts basal NF-B activity regulating YY1 expression in undifferentiated myoblasts. YY1 expression, in turn, binds to the promoter or enhancer elements of various myofibrillar genes (troponin; Tnn is used as an example), thereby recruiting the HDAC1 corepressor and the Polycomb silencer complex to inhibit myogenic differentiation. At the onset of differentiation, YY1 is displaced from the chromatin and replaced by serum response factor (SRF), MyoD, and PCAF, which permits the initiation of transcription and subsequent myogenesis.

	ACKNOWLEDGMENTS

 
We thank members of the Guttridge laboratory for their support and insight throughout the course of this study, especially K. J. Ladner for technical assistance. We also thank T. Huang and A. Chang for assistance with ChIP assays and Y. Shi for the YY1 expression plasmid.

This work was supported by NIH grants CA97953 to D.C.G. and AR054244 to H.W.

	FOOTNOTES

 
* Corresponding author. Mailing address: Human Cancer Genetics Program, 420 W. 12th Avenue, The Ohio State University College of Medicine, Columbus, OH 43210. Phone: (614) 688-3137. Fax: (614) 688-4006. E-mail: denis.guttridge{at}osumc.edu

Published ahead of print on 16 April 2007.

Supplemental material for this article may be found at http://mcb.asm.org/.

	REFERENCES

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
1. Acharyya, S., K. J. Ladner, L. L. Nelsen, J. Damrauer, P. J. Reiser, S. Swoap, and D. C. Guttridge. 2004. Cancercachexia is regulated by selective targeting of skeletal muscle gene products. J. Clin. Investig. 114:370-378.[CrossRef][Medline]

2. Acharyya, S., S. A. Villalta, N. Bakkar, T. Bupha-Intr, P. M. Janssen, M. Carathers, Z. W. Li, A. A. Beg, S. Ghosh, Z. Sahenk, M. Weinstein, K. L. Gardner, J. A. Rafael-Fortney, M. Karin, J. G. Tidball, A. S. Baldwin, and D. C. Guttridge. 2007. Interplay of IKK/NF-kappaB signaling in macrophages and myofibers promotes muscle degeneration in Duchenne muscular dystrophy. J. Clin. Investig. 117:889-901.[CrossRef][Medline]

3. Affar, E. B., F. Gay, Y. Shi, H. Liu, M. Huarte, S. Wu, T. Collins, E. Li, and Y. Shi. 2006. Essential dosage-dependent functions of the transcription factor Yin Yang 1 in late embryonic development and cell cycle progression. Mol. Cell. Biol. 26:3565-3581.[Abstract/Free Full Text]

4. Austen, M., B. Luscher, and J. M. Luscher-Firzlaff. 1997. Characterization of the transcriptional regulator YY1. The bipartite transactivation domain is independent of interaction with the TATA box-binding protein, transcription factor IIB, TAFII55, or cAMP-responsive element-binding protein (CPB)-binding protein. J. Biol. Chem. 272:1709-1717.[Abstract/Free Full Text]

5. Baeza-Raja, B., and P. Munoz-Canoves. 2004. p38 MAPK-induced nuclear factor-kappaB activity is required for skeletal muscle differentiation: role of interleukin-6. Mol. Biol. Cell 15:2013-2026.[Abstract/Free Full Text]

6. Bakkar, N., H. Wackerhage, and D. C. Guttridge. 2005. Myostatin and NF-B regulate skeletal myogenesis through distinct signaling pathways. Signal Transduct. 5:202-210.[CrossRef]

7. Baldwin, A. S. 2001. Control of oncogenesis and cancer therapy resistance by the transcription factor NF-kappaB. J. Clin. Investig. 107:241-246.[CrossRef][Medline]

8. Baldwin, A. S., Jr. 1996. The NF-B and IB proteins: new discoveries and insights. Annu. Rev. Immunol. 14:649-683.[CrossRef][Medline]

9. Bates, P. W., and S. Miyamoto. 2004. Expanded nuclear roles for IkappaBs. Sci. STKE 254:pe48.

10. Beg, A. A., W. C. Sha, R. T. Bronson, S. Ghosh, and D. Baltimore. 1995. Embryonic lethality and liver degeneration in mice lacking the RelA component of NF-kappa B. Nature 376:167-1