Regulation of c-myc Gene by Nitric Oxide via Inactivating NF-κB Complex in P19 Mouse Embryonal Carcinoma Cells
2003; Elsevier BV; Volume: 278; Issue: 32 Linguagem: Inglês
10.1074/jbc.m303306200
ISSN1083-351X
Autores Tópico(s)Peroxisome Proliferator-Activated Receptors
ResumoNitric oxide (NO) may regulate gene expression by directly modifying redox state-sensitive residues of transcription factors. Here we show that the NO donor, sodium nitroprusside (SNP), rapidly represses c-myc gene transcription in a protein synthesis-independent manner in P19 embryonal carcinoma cells by inactivation of NF-κB. SNP treatment reduces the DNA binding ability of the constitutively active NF-κB heterodimer, p65/p50, and its consequent transactivation of the c-myc promoter. Repression can be blocked by the peroxynitrite scavenger, deferoxamine, but not by dithiothreitol, which triggers reduction of S-nitrosylated residues. In HEK293 cells, where tumor necrosis factor-α can activate NF-κB, SNP likewise suppresses the binding of the active NF-κB complex, restoring the binding of the repressive p50/p50 homodimer complex. This effect of SNP in HEK293 cells is also blocked by deferoxamine. Chromatin immunoprecipitation analysis of SNP-treated P19 cells reveals reduced association of p65, but not of p50, with the promoter region of the endogenous c-myc gene. SNP-induced p65 dissociation was associated with the recruitment of histone deacetylase 1 and 2 to the endogenous c-myc gene promoter and the subsequent deacetylation of its chromatin histone. This study is the first to demonstrate that NO modulates the transcriptional activity of the c-myc gene promoter by dissociating the active form of NF-κB and replacing it with a repressive NF-κB complex, correlated with the recruitment of gene-silencing histone deacetylases. In light of findings that NF-κB stimulates Myc oncoprotein expression in cancers, our findings suggest that NO should be investigated as a prospective therapeutic cancer agent. Nitric oxide (NO) may regulate gene expression by directly modifying redox state-sensitive residues of transcription factors. Here we show that the NO donor, sodium nitroprusside (SNP), rapidly represses c-myc gene transcription in a protein synthesis-independent manner in P19 embryonal carcinoma cells by inactivation of NF-κB. SNP treatment reduces the DNA binding ability of the constitutively active NF-κB heterodimer, p65/p50, and its consequent transactivation of the c-myc promoter. Repression can be blocked by the peroxynitrite scavenger, deferoxamine, but not by dithiothreitol, which triggers reduction of S-nitrosylated residues. In HEK293 cells, where tumor necrosis factor-α can activate NF-κB, SNP likewise suppresses the binding of the active NF-κB complex, restoring the binding of the repressive p50/p50 homodimer complex. This effect of SNP in HEK293 cells is also blocked by deferoxamine. Chromatin immunoprecipitation analysis of SNP-treated P19 cells reveals reduced association of p65, but not of p50, with the promoter region of the endogenous c-myc gene. SNP-induced p65 dissociation was associated with the recruitment of histone deacetylase 1 and 2 to the endogenous c-myc gene promoter and the subsequent deacetylation of its chromatin histone. This study is the first to demonstrate that NO modulates the transcriptional activity of the c-myc gene promoter by dissociating the active form of NF-κB and replacing it with a repressive NF-κB complex, correlated with the recruitment of gene-silencing histone deacetylases. In light of findings that NF-κB stimulates Myc oncoprotein expression in cancers, our findings suggest that NO should be investigated as a prospective therapeutic cancer agent. Nitric oxide (NO) 1The abbreviations used are: NO, nitric oxide; SNP, sodium nitroprusside; DTT, dithiothreitol; DFO, deferoxamine; HDAC, histone deacetylase; TNF-α, tumor necrosis factor-α; URE, upstream responsive element; IRE, internal responsive element; ActD, actinomycin D; RT-PCR, reverse transcriptase-coupled polymerase chain reaction; ChIP, chromatin immunoprecipitation; UTR, untranslated region; KOR, kappa opioid receptor; EMSA, electrophoretic mobility shift assay. is a signaling molecule involved in a wide spectrum of pathophysiological processes such as inflammation, apoptosis, regulation of enzyme activity, and gene expression. NO may modulate the cellular redox state by acting as an oxidant, thereby activating or inhibiting various enzymes involved in a number of signal transduction pathways. It is known that NO can regulate gene expression by modulating transcription factors, the translation or stability of mRNA (1Bouton C. Demple B. J. Biol. Chem. 2000; 275: 32688-32693Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar), and the modification of proteins (2Shindler H. Bogdan C. Int. J. Immunopharmacol. 2001; 1: 1443-1455Crossref Scopus (46) Google Scholar, 3Marshall H.E. Merchant K. Stamler J.S. FASEB J. 2000; 14: 1889-1900Crossref PubMed Scopus (373) Google Scholar, 4Bogdan C. Trends Cell Biol. 2001; 11: 66-75Abstract Full Text Full Text PDF PubMed Scopus (457) Google Scholar). It is also known that these effects of NO are differentially elicited by various concentrations of NO in the microenvironment. For instance, a low level of NO activates the cGMP second messenger system, namely the cGMP-dependent protein kinase pathway, in vascular smooth muscle and others (5Ignarro L.J. Buga G.M. Wood K.S. Byrns R.E. Chaudhuri G. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 9265-9269Crossref PubMed Scopus (4347) Google Scholar, 6Gu M. Lynch J. Brecher P. J. Biol. Chem. 2000; 275: 11389-11396Abstract Full Text Full Text PDF PubMed Scopus (75) Google Scholar, 7Lincoln T.M. Dey N. Sellak H. J. Appl. Physiol. 2001; 91: 1421-1430Crossref PubMed Scopus (417) Google Scholar). However, a high level of NO, such as that generated by inducible NO synthase or administered exogenously, will directly modify target proteins through S-nitrosylation of cysteine residues or nitration of tyrosine (Tyr) residues (2Shindler H. Bogdan C. Int. J. Immunopharmacol. 2001; 1: 1443-1455Crossref Scopus (46) Google Scholar, 3Marshall H.E. Merchant K. Stamler J.S. FASEB J. 2000; 14: 1889-1900Crossref PubMed Scopus (373) Google Scholar, 4Bogdan C. Trends Cell Biol. 2001; 11: 66-75Abstract Full Text Full Text PDF PubMed Scopus (457) Google Scholar, 8Hausladen A. Privalle C.T. Keng T. DeAngelo J. Stamler J.S. Cell. 1996; 86: 719-729Abstract Full Text Full Text PDF PubMed Scopus (312) Google Scholar, 9Haqqani A.S. Kelly J.F. Birnboim H.C. J. Biol. Chem. 2002; 277: 3614-3621Abstract Full Text Full Text PDF PubMed Scopus (70) Google Scholar). We have previously identified a novel inhibitory role for NO in the transcription of the mouse kappa opioid receptor (KOR) gene in P19 embryonal carcinoma cells, a role mediated primarily through reduced expression of c-Myc that activates KOR gene transcription (10Park S.W. Li J. Loh H.H. Wei L.N. J. Neurosci. 2002; 22: 7941-7947Crossref PubMed Google Scholar). c-Myc is a transcription factor of the basic helix-loop-helix leucine zipper family. It forms a heterodimer with Max on the E-box of target gene promoters to activate transcription. In contrast, a heterodimer of Max and Mad binds to the E-box to suppress transcription. A functional role for c-Myc has been established in a wide variety of cellular processes including proliferation, metabolism, apoptosis, differentiation, genomic stability (11Dang C.V. Mol. Cell. Biol. 1999; 19: 1-11Crossref PubMed Scopus (1384) Google Scholar, 12Eisenman R.N. Genes Dev. 2001; 15: 2023-2030Crossref PubMed Scopus (322) Google Scholar, 13Pelengaris S. Littlewood T. Khan M. Elia G. Evan G. Mol. Cell. 1999; 3: 565-577Abstract Full Text Full Text PDF PubMed Scopus (401) Google Scholar, 14Pelengaris S. Khan M. Evan G. Nat. Rev. Cancer. 2002; 2: 764-776Crossref PubMed Scopus (924) Google Scholar), and cell cycle progression from quiescent to synthetic phases (15Eilers M. Schirm S. Bishop J. EMBO J. 1991; 10: 133-141Crossref PubMed Scopus (530) Google Scholar, 16Mateyak M.K. Obaya A.J. Adachi S. Cell Growth & Differ. 1997; 8: 1039-1048PubMed Google Scholar). Our recent study of P19 stem cells presented evidence that c-Myc functions as an activator for the mouse KOR gene by binding to its promoter. We also showed that NO suppressed KOR expression by rapidly down-regulating c-Myc protein synthesis in P19 cells (10Park S.W. Li J. Loh H.H. Wei L.N. J. Neurosci. 2002; 22: 7941-7947Crossref PubMed Google Scholar). This result prompted us to investigate the mechanisms underlying rapid down-regulation of c-Myc by NO in P19 cells. Regulation of c-myc gene expression has been shown to occur at multiple levels, including gene transcription, premature termination of translation (17Krumm A. Meulia T. Brunvand M. Groudine M. Genes Dev. 1992; 6: 2201-2213Crossref PubMed Scopus (221) Google Scholar, 18Perez-Juste G. Garcia-Silva S. Aranda A. J. Biol. Chem. 2000; 275: 1307-1314Abstract Full Text Full Text PDF PubMed Scopus (63) Google Scholar), and translocation (19Kanda K. Hu H.M. Zhang L. Grandchamps J. Boxer L.M. J. Biol. Chem. 2000; 275: 32338-32346Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar, 20Ratsch A. Joos S. Kioschis P. Lichter P. Exp. Cell Res. 2002; 273: 12-20Crossref PubMed Scopus (25) Google Scholar). In Burkitt's lymphoma the c-myc gene is translocated to the immunoglobulin heavy chain gene loci on either chromosome 2, 14, or 22. Transcription of the mouse gene can be initiated from two promoters, P1 and P2, separated by ∼160 bp. P2 is the major promoter where several transcription factors and regulatory DNA elements have been identified. For instance, three cis-acting elements for ME1a2, E2F, and ME1a1 have been found in P2 (21Moberg K.H. Tyndall W.A. Hall D.J. J. Cell. Biochem. 1992; 49: 208-215Crossref PubMed Scopus (48) Google Scholar, 22Rhee K. Ma T. Thompson E.A. J. Biol. Chem. 1994; 269: 17035-17042Abstract Full Text PDF PubMed Google Scholar). The pituitary tumor-transforming gene product can bind to a region near the P2 initiation site as a protein complex containing the upstream stimulatory factor-1 (23Pei L. J. Biol. Chem. 2001; 276: 8484-8491Abstract Full Text Full Text PDF PubMed Scopus (131) Google Scholar). A Smad-responsive element has been shown to be responsible for transforming growth factor-α mediated suppression of c-myc transcription (24Yagi K. Furuhashi M. Aoki H. Goto D. Kuwano H. Sugamura K. Miyazono K. Kato M. J. Biol. Chem. 2002; 277: 854-861Abstract Full Text Full Text PDF PubMed Scopus (209) Google Scholar). An intron IX box is known to be involved in leukemia HL-60 differentiation (25Pan Q. Simpson R.U. J. Biol. Chem. 1999; 274: 8437-8444Abstract Full Text Full Text PDF PubMed Scopus (56) Google Scholar, 26Chen L. Smith L. Johnson M.R. Wang K. Diasio R.B. Smith J.B. J. Biol. Chem. 2000; 275: 32227-32233Abstract Full Text Full Text PDF PubMed Scopus (36) Google Scholar). In addition, two functional NF-κB elements have been identified in both the human and the murine c-myc genes (27Kessler D.J. Duyao M.P. Spicer D.B. Sonenshein G.E. J. Exp. Med. 1992; 176: 787-792Crossref PubMed Scopus (67) Google Scholar, 28La Rosa F.A. Pierce J.W. Sonenshein G.E. Mol. Cell. Biol. 1994; 14: 1039-1044Crossref PubMed Google Scholar, 29Ji L. Arcinas M. Boxer L.M. Mol. Cell. Biol. 1994; 14: 7967-7974Crossref PubMed Google Scholar), one located from —1261 to —1251 in the upstream region of P2 and the other located from +280 to +289. Importantly, transcription of the translocated c-myc gene in Burkitt's lymphoma is stimulated by constitutively expressed NF-κB (19Kanda K. Hu H.M. Zhang L. Grandchamps J. Boxer L.M. J. Biol. Chem. 2000; 275: 32338-32346Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar). Our finding of rapid repression of c-Myc by NO in P19 cells (10Park S.W. Li J. Loh H.H. Wei L.N. J. Neurosci. 2002; 22: 7941-7947Crossref PubMed Google Scholar) and the identification of two NF-κB binding sites in the c-myc promoter prompted us to examine the potential role of NF-κB in mediating NO regulation of c-myc in P19 cells. NF-κB is an inducible transcription factor. Five members of NF-κB are known, including p50 (NF-κB1), p52 (NF-κB2), c-Rel, RelB, and p65 (RelA). The major active form of this transcription factor is the p65/p50 heterodimer (30Chen F.E. Huang D.B. Chen Y.Q. Ghosh G. Nature. 1998; 391: 410-413Crossref PubMed Scopus (338) Google Scholar, 31Karin M. Cao Y. Greten F.R. Li Z.W. Nat. Rev. Cancer. 2002; 2: 301-310Crossref PubMed Scopus (2258) Google Scholar, 32Li Q. Verma I.M. Nat. Rev. Immunol. 2002; 2: 725-734Crossref PubMed Scopus (3328) Google Scholar). In most cells, NF-κB complexes are predominantly cytoplasmic and are usually associated with an inhibitory protein, the IκB family. Upon stimulation by pro-inflammatory cytokines such as tumor necrosis factor-α (TNF-α) or interleukin-1, NF-κB dimer dissociates from IκBα and is translocated into nuclei to act on the target gene (33Ghosh S. Karin M. Cell. 2002; 109: S81-S96Abstract Full Text Full Text PDF PubMed Scopus (3294) Google Scholar, 34Chen G. Goeddel D.V. Science. 2002; 296: 1634-1635Crossref PubMed Scopus (1498) Google Scholar). Translocation into nuclei is considered to be a general, underlying mechanism for the activation of NF-κB in gene regulation. It has been shown that NO can inhibit NF-κB by modifying the p50 subunit via redox-based S-nitrosylation of a cysteine residue in its N-terminal region. This effect of NO is sensitive to dithiothreitol (DTT) (4Bogdan C. Trends Cell Biol. 2001; 11: 66-75Abstract Full Text Full Text PDF PubMed Scopus (457) Google Scholar, 35Matthews J.R. Botting C.H. Panico M. Morris H.R. Hay R.T. Nucleic Acids Res. 1996; 24: 2236-2242Crossref PubMed Scopus (461) Google Scholar, 36DelaTorre A. Schroeder R.A. Punzalan C. Kuo P.C. J. Immunol. 1999; 162: 4101-4108PubMed Google Scholar). In our previous study, the effect of NO on c-Myc in P19 cells was not blocked by DTT (10Park S.W. Li J. Loh H.H. Wei L.N. J. Neurosci. 2002; 22: 7941-7947Crossref PubMed Google Scholar), indicating the involvement of other NO-triggered mechanisms in the regulation of NF-κB activity. In this study we found that a NO donor sodium nitroprusside (SNP) repressed the DNA binding and activating function of the constitutively active p65/p50 form of NF-κB in P19 cells. Although not affected by DTT, it was blocked by deferoxamine (DFO), a peroxynitrite scavenger. Similarly, the NO donor also repressed TNF-α-activated NF-κB in cells where NF-κB normally existed as an inactive complex. The effect of NO in these cells was likewise blocked by DFO. We then confirmed the effect of NO on NF-κB and the c-myc promoter by examining the dynamic behavior of the NF-κB complex and histone deacetylases (HDACs), and chromatin acetylation on the endogenous c-myc promoter. This study demonstrated, for the first time, that NO modulates the transcription of the c-myc gene by inducing the dissociation of p65/p50 NF-κB from the c-myc promoter region and the recruitment of the repressive p50 homodimer complex, in correlation with increased recruitment of HDACs and decreased histone acetylation on the endogenous c-myc promoter. Antibodies and Plasmids—Antibodies against c-Myc (N-262), p65 (H-286), HDAC1 (C-19), and HDAC2 (H-54) were from Santa Cruz Biotechnology Inc. (Santa Cruz, CA); anti-p50 (06-886) and anti-acetylated histone H4 (06-866) were from Upstate Biotechnology (Lake Placid, NY). The expression vectors, pMT2T-p65 and pMT2T-p50, were kind gifts from Dr. U. Siebenlist (NIH) (37Franzoso G. Bours V. Park S. Tomita-Yamaguchi M. Kelly K. Siebenlist U. Nature. 1992; 359: 339-342Crossref PubMed Scopus (265) Google Scholar), pcDNA-His-p65 was a kind gift from Dr. A. Carter (University of Iowa) (38Carter A.B. Knudtson K.L. Monick M.M. Hunninghake G.W. J. Biol. Chem. 1999; 274: 30858-30863Abstract Full Text Full Text PDF PubMed Scopus (419) Google Scholar), and NF-κB specific reporter construct, p(Igk)4-LUC, that contains four tandem repeats of NF-κB binding sites was a gift from Dr. N. Mackman (Scripps Institute) (39O'Connell M.A. Bennett B.L. Mercurio F. Manning A.M. Mackman N. J. Biol. Chem. 1998; 273: 30410-30414Abstract Full Text Full Text PDF PubMed Scopus (164) Google Scholar). Cell Culture and Transient Transfection—P19 embryonal carcinoma cell and COS-1 cell lines were maintained as described previously (10Park S.W. Li J. Loh H.H. Wei L.N. J. Neurosci. 2002; 22: 7941-7947Crossref PubMed Google Scholar). HEK293 cells were grown in Dulbecco's modified Eagle's medium (Invitrogen) supplemented with 10% fetal bovine serum and 1% nonessential amino acids. Transient transfection was performed in P19 and COS-1 cells using the calcium chloride method, and reporter assays were conducted as described previously (40Hu X. Bi J. Loh H.H. Wei L.N. J. Biol. Chem. 2001; 276: 4597-4603Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar). Analyses of c-myc mRNA—The endogenous c-myc mRNA was analyzed by a reverse transcriptase-coupled polymerase chain reaction (RT-PCR) as described (10Park S.W. Li J. Loh H.H. Wei L.N. J. Neurosci. 2002; 22: 7941-7947Crossref PubMed Google Scholar). Total RNA was isolated (by using a Trizol® kit (Invitrogen)) from P19 cells which were pretreated with 2 μg/ml actinomycin D (ActD) or 2 μg/ml cycloheximide. Five mm DTT or 0.1 mm DFO was added for 30 min prior to the addition of 0.5 mm SNP for 6 h. To determine c-myc mRNA stability, P19 cells were treated with 2 μg/ml ActD for different durations in the presence of 1 mm SNP for 6 h. Primers specific to c-myc mRNA for the amplification procedure are 5′-CCATATGCCCCTCAACGTGAAC-3′ and 5′-GGGATCCTTATGCACCAGAGTTT-3′, which span a 1350-bp coding region. Electrophoretic Mobility Shift Assay (EMSA) and Western Blot— Nuclear extracts and whole cell lysates were isolated from P19 cells treated as above. HEK293 cells were treated with 10 ng/ml TNF-α for 8 h to activate NF-κB, followed by the addition of SNP for 6 h. EMSA was performed as described (41Li J. Park S.W. Loh H.H. Wei L.N. J. Biol. Chem. 2002; 277: 39967-39972Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar). Four types of double-stranded oligonucleotides (upstream responsive element (URE), internal responsive element (IRE), mt-URE, and nonspecific KOR gene promoter sequences) (see Fig. 2) were each used as probes that were labeled with [α-32P]dCTP or used as competitors without labeling. Whole cell lysates were analyzed on a Western blot by using anti-c-Myc antibody as described (10Park S.W. Li J. Loh H.H. Wei L.N. J. Neurosci. 2002; 22: 7941-7947Crossref PubMed Google Scholar). Chromatin Immunoprecipitation (ChIP) Assays—P19 cells were treated with SNP or SNP plus DFO for 6 h (or in a time-dependent manner), followed by cross-linking with 1% formaldehyde. ChIP assay was as described previously (10Park S.W. Li J. Loh H.H. Wei L.N. J. Neurosci. 2002; 22: 7941-7947Crossref PubMed Google Scholar) by using 2 μg of anti-acetylated histone H4, anti-p65, anti-p50, anti-HDAC1, or anti-HDAC2 antibodies. Precipitated DNA was amplified with primers specific to the URE flanking sequences in the endogenous c-myc promoter and to its 3′-untranslated region (UTR), followed by Southern blot analyses of amplified endogenous c-myc sequences. Transcriptional Repression of c-myc Gene by NO—We have previously shown that several NO donors suppressed KOR gene expression in P19 cells through repressing c-Myc protein expression and its binding to the KOR promoter (10Park S.W. Li J. Loh H.H. Wei L.N. J. Neurosci. 2002; 22: 7941-7947Crossref PubMed Google Scholar). To determine whether the repression of c-Myc by NO in P19 cells occurred at transcriptional or post-transcriptional levels, P19 cells were treated with SNP, a NO donor, in the absence or presence of ActD or cycloheximide followed by an analysis of the c-myc mRNA level with an established RT-PCR (Fig. 1A). NO dramatically reduced the constitutively expressed c-myc mRNA level (lane 2), and ActD effectively blocked c-myc transcription in the absence (lane 3) or presence (lane 4) of the NO donor SNP. Interestingly, although cycloheximide slightly increased c-myc expression (lane 5), it had little effect on the repression elicited by NO (lane 6), indicating that NO directly repressed c-myc transcription and that protein synthesis was probably not involved in this repression. To determine if c-myc mRNA stability was affected by NO, an experiment was carried out to determine c-myc mRNA half-life by treating P19 cells with ActD in the presence of SNP for various periods of time. As shown in Fig. 1B, the c-myc mRNA half-life remained at ∼20 min in both control cells (lanes 1–5) and cells exposed to NO (lanes 6–10). Therefore, c-myc mRNA stability is not affected by NO, whereas its gene transcription is directly repressed by NO signal. Repression of c-myc Gene Transcription by NO through Inactivation of NF-κB Activity—To define the target of NO-mediated repression of c-myc gene transcription, the regulatory regions of c-myc gene were carefully examined. This gene utilizes two promoters located ∼160 bp apart, with the second promoter residing in exon 1. Interestingly, two putative NF-κB binding sites, an URE (GGGTTTCCCC) located at —1261 to —1251 bp in the 5′-flanking region of promoter 1 and an IRE (GGGAATTTTT) located at +280 to +289 relative to the promoter P2 (27Kessler D.J. Duyao M.P. Spicer D.B. Sonenshein G.E. J. Exp. Med. 1992; 176: 787-792Crossref PubMed Scopus (67) Google Scholar, 28La Rosa F.A. Pierce J.W. Sonenshein G.E. Mol. Cell. Biol. 1994; 14: 1039-1044Crossref PubMed Google Scholar), were found to be relevant (Fig. 2A). NF-κB has been shown to be a target protein of NO signal (3Marshall H.E. Merchant K. Stamler J.S. FASEB J. 2000; 14: 1889-1900Crossref PubMed Scopus (373) Google Scholar, 42Colasanti M. Persichini T. Brain Res. Bull. 2000; 52: 155-161Crossref PubMed Scopus (76) Google Scholar). We examined how the endogenous NF-κB activity in P19 cells might be affected by NO and augment c-myc gene expression. P19 cells were treated with SNP, followed by analyzing its nuclear extract for DNA binding ability to the κB binding sites of the c-myc promoter in EMSA. As shown in Fig. 2B, the 32P-labeled URE probe revealed a major retarded band and several minor bands (lane 1), which could be competed out specifically by the wild type cold probe (lane 2) but not by the mutated cold probe (mt-URE, lane 3) or nonspecific oligonucleotides (lane 4). The anti-p65 antibody was able to block the major binding species that is the heterodimeric p65/p50 NF-κB complex (lane 5), and the anti-p50 antibody was able to supershift the minor species, the p50 complex (lane 6). The control, a nonspecific antibody such as the anti-Sp1, had no effect (lane 7). Therefore, the URE consensus sequences of the c-myc gene indeed can be bound by p65/p50 heterodimer and p50 homodimer of NF-κB, but the major complex in P19 cells is the transcriptionally active p65/p50 heterodimer. The second putative NF-κB binding site, IRE, was also tested in EMSA as shown in Fig. 2C, which appeared to share a very similar binding pattern as that of the URE. Furthermore, SNP decreased the intensity of the p65/p50 NF-κB band but increased the p50 complex for both NF-κB sites (lanes 1–4), revealing a decreased activator p65/p50-DNA interaction of P19 nuclear extract as a result of SNP treatment. To determine whether the transcriptional activity of endogenous NF-κB was affected by NO, transient transfection assays were conducted in P19 stem cells and fully differentiated COS-1 cells by using a specific NF-κB reporter, p(Igk)4-LUC (39O'Connell M.A. Bennett B.L. Mercurio F. Manning A.M. Mackman N. J. Biol. Chem. 1998; 273: 30410-30414Abstract Full Text Full Text PDF PubMed Scopus (164) Google Scholar), without introducing any other expression vectors. As shown in Fig. 3A, SNP effectively suppressed the specific reporter activity in a dose-dependent manner in P19 cells but not in COS-1 cells, which is consistent with our previous conclusion that NO suppressed KOR gene transcription only in P19 cells but not in COS-1 cells (10Park S.W. Li J. Loh H.H. Wei L.N. J. Neurosci. 2002; 22: 7941-7947Crossref PubMed Google Scholar). This suppressive effect of NO on NF-κB reporter plasmid in P19 cells agreed with the reduced DNA binding activity of NF-κB from SNP-treated P19 cells (Fig. 2C). The resistance of the same reporter activity to SNP in COS-1 cells was because of the absence of p65/p50 species binding to the URE in COS-1 cells that contained primarily the p50 species binding to the URE site (data not shown). To further confirm that the effect of SNP was indeed mediated by p65/p50 NF-κB, the specific reporter and a pMT2T-p65 expression vector (37Franzoso G. Bours V. Park S. Tomita-Yamaguchi M. Kelly K. Siebenlist U. Nature. 1992; 359: 339-342Crossref PubMed Scopus (265) Google Scholar) were simultaneously introduced into COS-1 cells (Fig. 3A). Indeed, ectopic expression of p65 increased the specific reporter activity in COS-1 cells, which was again suppressed by SNP in a dose-dependent manner (Fig. 3B). Thus, NO not only suppresses the DNA binding activity of p65/p50 NF-κB, but also efficiently reduces NF-κB activity in transcriptional control. The effect is primarily at the p65 subunit. Repression of c-myc Gene Transcription by NO through Peroxynitrite—Previously we have suggested that S-nitrosylation was not involved in c-myc repression by SNP, a NO donor, in P19 cells because this suppressive effect could not be reversed by DTT, an agent that blocks S-nitrosylation of Cys residues by NO (10Park S.W. Li J. Loh H.H. Wei L.N. J. Neurosci. 2002; 22: 7941-7947Crossref PubMed Google Scholar). Although it is accepted that NF-κB activity can be inhibited by NO through S-nitrosylation on a Cys residue of its p50 subunit (35Matthews J.R. Botting C.H. Panico M. Morris H.R. Hay R.T. Nucleic Acids Res. 1996; 24: 2236-2242Crossref PubMed Scopus (461) Google Scholar, 36DelaTorre A. Schroeder R.A. Punzalan C. Kuo P.C. J. Immunol. 1999; 162: 4101-4108PubMed Google Scholar), the failure of DTT to reverse the repressive effects of SNP in P19 cells raised the question as to what other pathway could be involved in this novel suppressive effect of SNP in P19 cells. The alternative modification elicited by NO is Tyr nitration triggered by the highly reactive peroxynitrite (ONOO—) generated from NO and superoxide (4Bogdan C. Trends Cell Biol. 2001; 11: 66-75Abstract Full Text Full Text PDF PubMed Scopus (457) Google Scholar). To test this possibility, we treated P19 cells with DTT or DFO that is known to scavenge peroxynitrite, followed by examining the expression of c-myc mRNA as shown in Fig. 4. DTT failed to reverse the repressive effect of SNP on c-myc transcription (Fig. 4A, lane 3), whereas DFO efficiently blocked the repressive effect of SNP (lane 4). Consistently, c-Myc protein expression (Fig. 4B) was repressed by SNP (lane 2), which was effectively reversed by DFO (lane 4) but not by DTT (lane 3). The potent cGMP analog 8-bromo-cGMP and guanylate cyclase inhibitor 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one affected neither c-myc mRNA nor its protein level (data not shown). To confirm whether peroxynitrite was indeed responsible for inactivating NF-κB activity, transient transfection assays were performed using the specific reporter, p(Igk)4-LUC, in P19 cells. As shown in Fig. 4C, DTT failed to reverse the repressive effect of SNP on the NF-κB reporter, but DFO very efficiently reversed the repression to ∼75% of the control level. These results have ruled out the S-nitrosylation and cGMP-dependent pathways and suggested that peroxynitrite generated from NO could play a role in the repression of c-myc transcription in P19 cells, which may involve nitration of NF-κB. It was noted that changes had occurred in the DNA binding pattern of endogenous NF-κB and p50 homodimer in P19 cells following SNP treatment (Fig. 1C). To further confirm this point, we made use of HEK293 cells where the endogenous NF-κB existed primarily as the inactive complexes and could be activated by TNF-α. By testing DNA binding patterns of HEK293 extract as shown in Fig. 5, the portion of heterodimeric NF-κB increased in response to TNF-α stimulation as predicted (lane 2), which was blocked by SNP and replaced with the p50/p50 homodimer complex after SNP treatment (lane 3). Furthermore, the SNP-triggered blockage of NF-κB heterodimer formation was again reversed by DFO (lane 4). The identity of p65/p50 heterodimer and p50/p50 homodimer was again confirmed by using specific antibodies (lanes 5–8) that either blocked (for p65/p50 heterodimer) or supershifted (for p50/p50 homodimer) the complexes. This result confirms that indeed SNP triggers the dissociation of the p65/p50 complex from, and the association of p50 complex with, the target DNA, consistent with the effects of SNP and DFO in the regulation of c-myc gene, as well as its reporter, in P19 cells (Fig. 4). The results of the two cell lines have demonstrated that peroxynitrite indeed disrupts DNA binding of the activated NF-κB, i.e. the p65/p50 heterodimer, which can be blocked by DFO, a scavenger of peroxynitrite. Recruitment of HDACs 1 and 2 to the Endogenous c-myc Promoter by SNP Treatment—The activity of transcription factors can be modulated by altering the recruited coregulators in addition to changes in DNA binding ability. For the repressive activity of p50 homodimer, the recruitment of HDACs has been shown to play a major role (43Ashburner B.P. Westerheide S.D. Baldwin Jr., A.S. Mol. Cell. Biol. 2001; 21: 7065-7077Crossref PubMed Scopus (627) Google Scholar, 44Zhong H. May M.J. Jimi E. Ghosh S. Mol. Cell. 2002; 9: 625-636Abstract Full Text Full Text PDF PubMed Scopus (817) Google Scholar). To confirm the effect of SNP and thus replacement of p65/p50 with p50/p50 on its target gene, the role of HDAC activity was examined. P19 cells were transfected with the specific NF-κB reporter p(Igk)4-LUC and treated with SNP, trichostatin A (TSA), or
Referência(s)