- Email: [email protected]
Transcriptional Repression Activity of N-MYCProtein Requires Phosphorylation by MAP Kinase
JOBNAME: BBRC 219#3 PAGE: 1 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.
219, 813–823 (1996)
0316
Transcriptional Repression Activity of N-MYC Protein Requires Phosphorylation by MAP Kinase Akira Manabe,*,† Sanae M. M. Iguchi-Ariga,‡ Hajime Iizuka,† and Hiroyoshi Ariga*,1 *Faculty of Pharmaceutical Sciences and ‡College of Medical Technology, Hokkaido University, Kita-ku, Sapporo, 060; and †Department of Dermatology, Asahikawa Medical College, Nishikagura, Asahikawa 078, Japan Received January 24, 1996 The transrepressing function of the N-myc protein is due to the distinct domains located at the N-terminus. In this report we introduced various point mutations around the myc boxes of the N-myc protein to examine whether the phosphorylation of the protein affected its transrepressing function. Serine (Ser) residues located at amino acid numbers 12, 31, 51, and 65 were changed to leucine or arginine, and the expression vectors of the mutant proteins were transfected to HeLa cells together with the luciferase gene linked to the MHC class I gene. Among the mutants, only the N(51)-myc carrying mutation at Ser51, a target for mitogen-activated protein kinase (MAP kinase), lost the repression activity, while the other mutant proteins preserved it. Formation in vitro of the specific nucleoprotein complexes on the H2TF1/NFkB element, a major target for transrepression by N-myc protein, was interfered by the wild-type N-myc protein, but not by the Ser51-mutated protein. The results suggest that the phosphorylation of the N-myc protein at Ser51 by MAP kinase is required for the transcriptional repression activity of the protein. © 1996 Academic Press, Inc.
The N-myc gene, a myc family oncogene, encodes a nuclear phosphoprotein. The N-myc protein is composed of several characteristic domains including two Myc boxes (the regions highly conserved among myc family proteins), an acidic region (a region rich in acidic amino acids), and a bHLH-Zip (a basic, helix-loop-helix, leucine zipper region) (1, 2) (Fig. 1). These domains are suggested to be important for the transrepression activity of the protein against the genes including the major histocompatibility (MHC) class I antigen (3, 4), the neural cell adhesion molecule (NCAM) (5), and the N-myc (6). Down-regulation of the MHC class I expression by N-myc and c-myc was reported in neuroblastoma and melanoma cells, respectively, and the regulation by myc proteins was at transcriptional level (see review, 7). N-myc was shown to repress the transcription from all the MHC class I loci, while the regulation by c-myc is predominantly restricted to the HLA-b locus (7). The expression of the MHC class I gene is dependent on the two major enhancer elements, enhancer A and enhancer B, located between −190 and −138 from, and immediately upstream from, the transcriptional start site (+1), respectively. Of the two enhancers, enhancer A shows a transcriptional activity much stronger than enhancer B, and three copies of the TGGGGATTCCCA sequence or its variants near the 39-terminus of enhancer A efficiently promotes the transcription of the reporter gene linked to the elements (4). Transfection experiments of the N-myc gene to cells indicated that TGGGGATTCCCA in enhancer A was a target sequence of transrepression of the MHC class I gene by N-myc (4). Several transcription factors, including KBF-1 of 48 KDa (8, 9), KBF2 of 58 KDa (10), H2TF1 of 110 KDa (11) and NFkB (11), have been identified to recognize the sequence. All of the factors belong to rel family proteins. An elevated level of the N-myc expression in neuroblastoma or transfected cells was correlated with a reduced binding of H2TF1 to enhancer A (11). The fact suggested that the N-myc protein directly or indirectly interferes with H2TF1 or NFkB family proteins in inactivating enhancer A. Precise molecular mechanisms of the transrepression by N-myc, however, remain unclarified.
1
To whom correspondence should be addressed. Fax: +81 11 (706) 4988. 813 0006-291X/96 $18.00 Copyright © 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
JOBNAME: BBRC 219#3 PAGE: 2 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Phosphorylation is also considered to be important for the functions of myc proteins (see reviews, 4, 12, 13 and references therein). The c-myc protein has been reported to be phosphorylated by casein kinase II (ckII) at the central (amino acid residues 240–262) or C-terminal region (residues 342–357), but roles of the phosphorylation are still unclarified (12–14): Deletion of the regions affected little the activities of the c-myc protein in transformation and transcriptional regulation (15–18), although the ckII phosphorylation was correlated with transformation of chicken haematopoietic cells by the v-myc protein (19). Mitogen-activated protein kinase (MAP kinase) has also been reported to phosphorylate a site near myc boxes of the c-myc protein in vitro (14, 20–24). The site, located in the N-terminal region important for both transactivation and transformation activities (14–16, 18, 25), was often found to be mutated in the c-myc proteins recovered from various cancer cells (26). As for phosphorylation of the N-myc protein, only the sites within the acidic region have been reported so far to be phosphorylated by ckII (27, 28), although a putative target for MAP kinase exists near the myc boxes of the protein. Here we introduced point mutations into putative phosphorylation sites around the myc boxes of the N-myc protein, and examined the mutant proteins for transrepressing activity on the regulatory sequences of the MHC class I gene. The results showed that the phosphorylation at Ser51 by MAP kinase is required for the transrepression activity of the N-myc protein. Formation in vitro of the specific nucleoprotein complexes on the H2TF1/NFkB element was interfered by the wild type N-myc protein, but not by the Ser51-mutated protein. MATERIALS AND METHODS Plasmids and introduction of mutations into N-myc protein. pGBV-Luc, a plasmid containing the luciferase gene (Wako Pure Chemicals Co. Ltd., Kyoto, Japan), was digested with HindIII, treated with Klenow fragment, and then self-ligated to destroy HindIII site (named pGBVH-Luc). The ScaI-ScaI fragment containing multicloning sites from pUC18 was inserted to the ScaI-SacI sites of pGBVH-Luc, to obtain pGBV(18). pH-2Kbm1-CAT was kindly provided by Dr. Kazushige Yokoyama, Riken, Tsukuba, Japan. pH-2Kbm1-CAT, which contains the transcriptional regulatory sequences of the mouse MHC class I gene linked to the bacterial chloramphenicol acetyltransferase gene (28), was digested with HindIII and NruI, and the fragment containing the MHC class I promoter was inserted between the HindIII-SmaI sites of pGBV(18). The construct was named pMHC-Luc. As for pNFk-TATA-Luc construction, oligonucleotides as follows were synthesized: 59-tcgagTGGGGATTCCCCATCTCCACa-39 (upper strand) and 59-agcttGTGGAGATGGGGAATCCCCAc-39 (lower strand), in which small letter indicates XhoI and HindIII restriction recognition sequences. The oligonucleotides were annealed and inserted first into the XhoI-HindIII sites of pBluescript SK(-). The SmaI-SacI fragment of the resultant plasmid containing the insertion was then cloned into the SmaI-SacI sites of pHS-TATA-Luc which contains TATA box of the human heat shock protein 70 gene linked to the luciferase gene in pGBV. pMHC-Luc and pNFk-TATA-Luc were used as reporter genes in co-transfection experiments. Mutations into the N-myc protein were introduced by polymerase chain reactions (PCR). Briefly, the fragment from EcoRI to SacI (from nucleotide number 1 to 970) of the human N-myc cDNA was cloned to the EcoRI-SacI sites of pUC19, to obtain pHN(E-S). The first PCR reaction was carried out on a pHE(E-S) template using mutation primers shown in Fig. 1 and conventional M4 or RV primer to amplify the left or right half segment, respectively. The amplified left and right fragments were mixed, and the second PCR reaction was carried out using M4 and RV primers. The conditions for both the reactions were 1 cycle of 94°C for 5 min, and 30 cycles of 94°C for 1 min, 55°C for 2 min, and 72°C for 3 min. Nucleotide sequences of upper (U) and lower (L) oligonucleotides used as primers are the following: N-myc(12); U:59GAGTTTGACTtGCTACAG39, L: 59CTGTAGCaAGTCAAACTC39, N-myc(31); U:59CGGCCCCGACTtGACCCC39, L: GGGGTCaAGTCGGGGCCG39, N-myc(51), U:59CCGCTGTtGCCCAGCCGT39, L: 59ACGGCTGGGCaACAGCGG39, N-myc(65); U: 59 CCCCCGAGaTGGGTCCCGGAGATG39, L: 59CATCTCCGGGACCCAtCTCGGGGG39. Small letters indicate the mutated nucleotides. The resulting fragments were digested with EcoRI and SacI, and cloned between the EcoRI/BgIII sites of pSRa296, together with the SacI-BgIII fragment of the C-terminal fragment of the N-myc cDNA. Plasmid DNA containing desired mutation was confirmed by the sequencing analysis of DNA. Cell culture and transfection. Human HeLa cells were cultured in Dulbecco’s modified Eagle medium supplemented with 10% calf serum. Various amounts of N-myc expression vectors were transfected to HeLa cells about 50% confluent in a 6 cm dish by the calcium phosphate precipitation method (29), together with 2 mg of pMHC-Luc and 1 mg of pCMV-b-gal (an expression vector of the b-galactosidase gene linked to the cytomegalovirus promoter). At 4 hr after transfection, the cells were boosted with 10% glycerol and cultured. At 48 hr after transfection, the cells were harvested and cell extracts were prepared by adding the Triton X-100-containing solution from Pica Gene kit (Wako Pure Chemicals Co. Ltd., Kyoto, Japan). One tenth volume of the extracts was used to examine the b-galactosidase activity due to the pCMV-b-gal and then 814
JOBNAME: BBRC 219#3 PAGE: 3 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
assayed for the luciferase activity using the Pica Gene kit after standardization of transfection efficiency by b-galactosidase activity (Note, N-myc protein does not affect the activity of CMV promoter). For transfection of an anti-MAP kinase polyclonal antibody (ERK 2 (C-14) raised against a carboxyl-terminal epitope, Santa Cruz Biotechnology, Inc., California), plasmid DNAs, or both DNA and antibody, a commercial reagent of transfection based on liposome, “transfection reagent” (Boehringer Mannheim GmbH, Biochemica, Germany) was used. In transfection of either DNA or antibody alone, 18 ml of solution A containing DNA or antibody was gently mixed with 18 ml of solution B containing 11 ml of transfection reagent. The mixture was stood for 15 min at room temperature and added to the cells in 6 cm dish in 1.5 ml of medium without serum. At 18 hr after transfection, final 10% of serum was added back to medium and the cells were cultured for another 30 hr. In transfection of both DNA and antibody, 18 ml of solution A containing antibody was gently mixed with 18 ml of solution B containing 11 ml of transfection reagent. The mixture was stood for 15 min at room temperature and added to the cells as above. At 4 hr after transfection, solution C containing DNA and 11 ml of the transfection reagent was added to cells. At 14 hr after transfection, final 10% of serum was added back to medium and the cells were cultured for another 30 hr. One mg of the anti-MAP kinase antibody or non-specific rabbit anti-human IgG was added to the cells. Assays of the enzyme activities were carried out as described below. All the experiments were repeated more than three times. Bandshift assay. Oligonucleotides corresponding to the H2TF1/NFkB element (59-GAT CCT GGG GAT TCC CCA G-39; 59-TCG ACT GGG GAA TCC CCA G-39) were chemically synthesized, annealed and end-labelled using Klenow enzyme and [a-32P]dCTP. The H2TF1 probe (10,000 cpm) was incubated in 15 ml with 2 mg of poly(dI-dC) and 30 mg of the nuclear extracts prepared from transfected or untransfected HeLa cells according to Dignam et al. (30) with minor modifications (31) in a buffer containing 50 mM KCl, 1 mM EDTA, 1 mM dithiothreitol, 4 mg/ml bovine serum albumin and 4% Ficoll 400. After 15 min at room temperature, the DNA/proteins mixtures were separated in a 4% polyacrylamide gel containing 0.25 × TBE (1 × TBE contains 90 mM Tris (pH 8.3), 90 mM boric acid, and 2.5 mM EDTA) and autoradiographed.
RESULTS Effect of the Mutations Introduced into the N-terminal Region on Transrepression Activity of the N-myc Protein We have previously identified several domains of N-myc protein responsible for its transrepression function (6). The domains include the N-terminal region containing the myc boxes, the acidic region, and the bHLH-Zip region (see Fig. 1, upper panel). The N-terminal and the acidic regions contain putative sites to be phosphorylated: The former region contains a target for MAP kinase, which overlaps a recognition consensus for cdc2 kinase, as in the corresponding region of the c-myc protein. We introduced mutations in the N-myc cDNA by a PCR-mediated method to substitute leucine (Leu) or arginine (Arg) for putative phosphorylation sites, Ser12, Ser31, Ser51 and Ser 65 among the 10 serine residues in the N-terminal portion of the N-myc protein (see Fig. 1, lower panel). The wild type and mutant N-myc cDNAs were linked to the SRa promoter (32) in order to express them efficiently in cells. The constructs were transfected to human HeLa cells and the levels of their expression were examined by Western blotting using a polyclonal anti-N-myc antibody. The anti-N-myc antibody used here did not recognize c-myc protein. As previously described (33), HeLa cells did not express, or at most expressed a trace amount of, the N-myc protein endogenously. The expression of the N-myc proteins due to transfection of the wild type (pN-myc) and mutant (pN-(12)myc, pN(31)-myc, pN(51)-myc, and pN(65)-myc) expression vectors were of the same or similar level (data not shown). The mutations we introduced thus did not affect the expression efficiency of the N-myc protein in transfected cells. Various amounts of the wild type N-myc expression vector, pN-myc, were transfected to HeLa cells together with a constant amount of pMHC-Luc, a reporter plasmid carrying the luciferase gene linked to the MHC Class I promoter region. The luciferase activity was transrepressed by pN-myc in a dose-dependent manner, while a vector pSRa296 (pSRa) alone affected little the activity (data not shown). In several independent experiments, 30–50% repression was observed when 2 mg of pN-myc were co-transfected. The inhibition of transcription by N-myc proteins was not so strong as expected, probably because there exist a number of transcriptional regulatory sequences other than the N-myc target element in the reporter plasmid, pMHC-Luc. We thereby transfected 2 mg of expression vectors for various mutant proteins to examine their transrepression activity. The 815
JOBNAME: BBRC 219#3 PAGE: 4 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Structure of the human N-myc protein and the amino acid and nucleotide sequences around the mutations introduced in the N-terminal region. Upper panel schematically represents characteristic domains of the human N-myc protein. Numbers in the figure correspond to the amino acid numbers of the protein. In lower panel, the nucleotide and amino acid sequences around the mutations introduced in the protein are shown. Mutated nucleotides are written in small letters, and the amino acids thereby mutated are squared.
expression vector for the wild type (pN-myc) or a mutant protein (pN-(12)myc, pN(31)-myc, pN(51)-myc, or pN(65)-myc) or the vector alone (pSRa296) was co-transfected to HeLa cells with pMHC-Luc, and the luciferase activity was assayed (Fig. 2). The wild type and the 3 mutants at Ser12, Ser31, and Ser65 suppressed the luciferase activity to similar extents. No repression was observed, however, by the Ser51 mutant protein: Alteration of Ser51 into Leu abolished the transrepression activity of the N-myc protein. Since the Ser51 is a putative target for MAP kinase, phosphorylation of the N-myc protein at Ser51 by MAP kinase was suggested to be required for the transrepression activity of the protein. Effect of Anti-MAP Kinase Antibody on Transrepression Activity of the N-myc Protein To verify the suggestion, an anti-MAP kinase antibody was introduced to the cells transfected with the wild type (pN-myc) or Ser51 mutant (pN(51)-myc) expression vector. Introduction of the antibody was carried out using “transfection reagent”, a commercial kit of liposome-mediated transfection (Boehringer Mannheim GmbH, Biochemica, Germany). Indirect immunofluorescence staining of the cells revealed not only that the anti-MAP kinase antibody and non-specific IgG introduced in the cells were stable at 48 hr after transfection and distributed both in nuclei and cytoplasm (data not shown). The transfection efficiency of the antibodies by the liposome-mediated method was comparable to that of the expression vector plasmids by the standard calcium phosphate precipitation method as described. Plasmid DNAs were also transfected to HeLa cells by the liposome-mediated method as efficiently as by the calcium phosphate procedure: the luciferase 816
JOBNAME: BBRC 219#3 PAGE: 5 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Effect of the point of mutations introduced in the N-myc protein on its transrepression activity. Two µg of expression vectors for the wild-type or mutated N-myc protein were cotransfected to HeLa cells together with 1 µg of pCMV-b-gal and 2 µg of pMHC-Luc, and the luciferase activity was assayed. Relative luciferase activities to the value of a reporter alone (pMHC-Luc alone was set as 100) are shown.
activities of the same levels as those in Fig. 2 were recovered from the cells transfected with the same plasmids by the liposome-mediated method (data not shown). We then carried out the co-transfection experiments of the N-myc expression vectors and the reporter pMHC-Luc with the antibodies (Fig. 3). Whole cell extracts were prepared from the cells. The results of western blot analyses indicate that the amounts of the wild type and mutant N-myc proteins in the extracts were comparable (data not shown). The repression of the luciferase activity
FIG. 3. Effect of an anti-MAP kinase antibody on the repression activity of N-myc protein by using pMHC-Luc as a reporter gene. Two µg of the expression vectors for the wild-type or mutated N-myc proteins, or pSRa vector, were cotransfected to HeLa cells together with 2 < gkm>g of pMHC-Luc, 1 µg of pCMV-b-gal, and 0.2 µg of an anti-MAP kinase antibody or nonspecific IgG by a liposome-mediated procedure as described in Materials and methods. The luciferase activity was assayed and relative luciferase activities to the value of a reporter alone (pMHC-Luc alone was set as 100) are shown. 817
JOBNAME: BBRC 219#3 PAGE: 6 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
due to pMHC-Luc by the expression vector for the wild type N-myc protein (pN-myc) was abrogated by the anti-MAP kinase antibody co-transfected to the cells, while non-specific IgG hardly affected the transrepression. Similar results as that of wild type N-myc protein were also obtained in the experiments using the mutant N-myc proteins of N−(21), N−(31), and N−(65) (data not shown). The expression of the Ser51-mutated protein due to pN(51)-myc did not repress the luciferase activity in the absence or presence of either the anti-MAP kinase antibody or nonspecific IgG. These results indicate that phosphorylation of the N-myc protein at Ser51 residue by MAP kinase is required for the transrepression activity of the protein. Interference of the Nucleoprotein Complex Formation on the H2TF1/NFkB Element by the N-myc Protein In this report, we examined the transrepression activity of the N-myc protein to the promoter region of MHC class I gene of about 2,000 bp, in which the major target for the N-myc protein is the H2TF1/NFkB element (4). To verify the target sequence for the N-myc protein in the MHC class I promoter, the H2TF1/NFkB element was ligated to TATA box of the heat shock protein 70 (hsp70) gene linked to the luciferase gene, and the construct pNFk-TATA-Luc was used as a reporter plasmid for the N-myc-induced transrepression experiments in HeLa cells (Fig. 4). A greater transrepression by co-transfected pN-myc was observed for pNFk-TATA-Luc than for pMHC-Luc, probably because pMHC-Luc contains various transcriptional elements, in addition to the H2TF1/NFkB element, which may be involved in the MHC expression. The expression vector for the Ser51-mutated N-myc protein hardly affected the transcription due to pNFk-TATA-Luc as well as the transcription due to pMHC-Luc. Co-introduction of the anti-MAP kinase antibody abrogated the transrepression activity of the N -myc protein, while IgG little affected. The results implicated that the N-myc protein represses transcription via the H2TF1/NFkB element only when the protein is phosphorylated by MAP kinase. Repression of transcription due to a specific DNA element is usually accompanied with a
FIG. 4. Effect of an anti-MAP kinase antibody on the repression activity of N-myc protein using pNFk -TATA-Luc as a reporter gene. The transfection and luciferase assay experiments were carried out as described in the legend of Fig. 3 except for pNFk-TATA-Luc used as a reporter gene. Two µg of the expression vector for the wild-type or mutated N-myc protein, or pSRa vector, were cotransfected to HeLa cells together with 2 µg of pNFk-TATA-Luc, 1 µg of pCMV-b-gal, and 0.2 µg of an anti-MAP kinase antibody or nonspecific IgG by the liposome-mediated procedure as described in Materials and methods. The luciferase activity was assayed and relative luciferase activities to the value of a reporter alone (pNFk-TATA-Luc alone was set as 100) are shown. 818
JOBNAME: BBRC 219#3 PAGE: 7 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
binding of negative factors to or release of positive factors from the element. We hence examined whether the N-myc protein affect the nucleoprotein complex formation on the element. Oligonucleotides corresponding to the H2TF1/NFkB element were labelled and incubated with the nuclear extract prepared from HeLa cells before or after transfection of the expression vector for the wild type (pN-myc) or mutant (pN(51)-myc) N-myc protein. Several bands of complexes were detected with the extract from untransfected cells (Fig. 5A, lane 2). Competition experiments using nonlabelled homo- or hetero-oligonucleotides including two points mutation revealed that three bands indicated as I, II and III were due to proteins specifically bound to the H2TF1/NFkB element (data not shown). As for the extracts prepared from the cells transfected with the wild type pN-myc, the specific bands I, II and III disappeared with the dose of transfected pN-myc (Fig. 5A, lanes 3–5). The expression of the Ser51-mutated protein due to pN(51)-myc, on the other hand, did not alter the pattern of the nucleoprotein complexes (Fig. 5A, lanes 6–8). Since the transrepression was observed with pN-myc, but not with pN(51)-myc (Fig. 2), the complexes I, II and III are suggested to be of positive factors for transcription due to the H2TF1 element, and the release of the complexes from the element by the N-myc protein may result in the transrepression. This possibility was further suggested by the mixing experiments between the extracts from the cells transfected with either pN-myc or pN(51)-myc. Extract from pN-myc transfected cells abolished the specific bands derived from non-transfected cells, while extract from pN(51)-myc transfected cells did not. Addition of non-transfected cell extract to that of pN-myc transfected cells, on the other hand, restored the specific bands in dose dependent manner (data not shown). Furthermore, the complexes I, II and III with the untransfected cell extract, which disappeared by the wild type N-myc protein due to transfected pN-myc, were restored by the co-transfected anti-MAP kinase antibody, but not by non-specific IgG (Fig. 5B, lanes 4 and 5). The results suggest that the N-myc protein
FIG. 5. Inhibition by N-myc protein of nucleoprotein complexes formation on the H2TF1 element. (A) Various amounts of an expression vector for the wild-type or mutant N-myc protein (pN-myc or pN-(51)myc) were transfected to HeLa cells by the calcium phosphate precipitation method. Nuclear extracts were prepared from the cells as well as from nontransfected cells, incubated with 32P-labelled oligonucleotides corresponding to the H2TF1 element and separated on a 4% polyacrylamide gel containing 0.25× TBE. Three bands due to the nucleoprotein complexes specifically interfered by the wild-type N-myc protein are indicated by arrows, I, II and III. Arrow F indicates the position of free probe. (B) Two µg of pN-myc were transfected to HeLa cells together with or without 2 µg of an anti-MAP kinase antibody or nonspecific IgG by the liposome-mediated procedure. Nuclear extracts were prepared from the cells as well as nontransfected cells, incubated with 32 P-labelled H2TF1 probe and analyzed as in A. Arrows I, II and III correspond to the same nucleoprotein complexes indicated in A. Arrows I, II and III correspond to the same nucleoprotein complexes indicated in A. Arrow F indicates the position of free probe. 819
JOBNAME: BBRC 219#3 PAGE: 8 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS 51
phosphorylated at Ser by MAP kinase interferes the formation of specific nucleoprotein complexes, probably of positive factors, on the H2TF1/NFkB element and that the protein thereby represses the transcription due to the element. Phosphorylation of N-myc Protein by MAP Kinase The expression vector pN-myc or pN(51)-myc was then co-transfected to HeLa cells together with or without the anti-MAP kinase antibody or non-specific IgG. Whole cell extracts were prepared from the cells, incubated in the presence of [g-32P]ATP, immunoprecipitated with the anti-N-myc antibody and electrophoresed through an SDS-containing 10% polyacrylamide gel. The results of western blot analyses indicate that the amounts of the wild type and mutant N-myc proteins in the extracts were comparable (Fig. 6A, lower panel). In the extract from the cells transfected with pN-myc, a clear band of a phosphorylated protein was detected at the size expected for the N-myc protein (Fig. 6A upper part, lane 4). The band was missing in the extracts from the cells co-transfected with pN-myc and the anti-MAP kinase antibody (Fig. 6A upper part, lane 6). Non-specific IgG co-transfected with pN-myc to the cells did not alter the electrophoretic pattern of phosphorylated proteins (Fig. 6A, upper panel, lane 5 compared with lane 4). The extracts from the cells transfected with pN(51)-myc carrying the mutation at Ser51 yielded no band at the size for N-myc protein regardless to the co-transfection of the antibodies (Fig. 6A upper panel, lanes 1 - 3). The results indicate that the N-myc protein expressed due to pN-myc in HeLa cells was mainly phosphorylated at Ser51 by MAP kinase. To examine whether the N-myc protein was phosphorylated by MAP kinase in cytoplasm or in nuclei, phosphorylation experiments were carried out using transfected cells. HeLa cells were transfected with the N-myc expression vector (pN-myc), either alone or together with the anti-MAP kinase antibody or non-specific IgG, and labelled in vivo with 32P-orthophosphate. Cytoplasmic fraction and nuclear extract were prepared from the cells and analyzed by immunoprecipitation using the anti-N-myc antibody. As for the cells transfected with the N-myc expression vector alone or together with non-specific IgG, the phosphorylated N-myc protein was detected only in the nuclear extract, but not in the cytoplasmic fraction (Fig. 6B, lanes 1, 3, 4 and 6). Neither in the nuclear extract nor in the cytoplasmic fraction prepared from the cells co-transfected with pN-myc and the anti-MAP kinase antibody, on the contrary, the 32P-labelled N-myc protein was identified (Fig. 6B, lanes 2 and 5). Phosphorylated N-myc protein was hardly detected in either nuclei or cytoplasm of the cells transfected with pN(51)-myc (data not shown). DISCUSSION In this report, we show the involvement of MAP kinase-dependent phosphorylation in the transcriptional activity of N-myc protein. The consensus sequence for a MAP kinase target is Pro-Leu-Ser/Thr-Pro-Pro, and the sequence is present between two myc boxes both in the c-myc (amino acid number from 60 to 63) and the N-myc (amino acid number from 49 to 52) proteins. A recent report tells that MAP kinase directly phosphorylated the Ser62 of the c-myc protein in vivo (34). The corresponding Ser51 of the N-myc protein is thus suggested to be phosphorylated directly by MAP kinase. The Ser62 of the c-myc protein was also reported to be important for transactivation by the protein (34), which is reminiscent of the results reported here: phosphorylation of the Ser51 of the N-myc protein is required for the transrepression activity of the protein. MAP kinase has been widely reported to regulate functions of a variety of proteins including cell surface proteins, cytoskeletal proteins, and the nuclear proteins including c-jun, c-myc, NF-IL6, ATF-2, TAL1, and p62TCF/ELK-1, involved in signal transduction (see reviews 35, 36, and references therein). As for the phosphorylation of nuclear proteins, translocation of MAP kinase from cytoplasm to nucleus has been considered (23, 33, 37-39): MAP kinase may phosphorylate the proteins in nucleus. Alternatively, the proteins may translocate to nucleus after being phosphory820
JOBNAME: BBRC 219#3 PAGE: 9 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 6. Inhibition of phosphorylation of the N-myc protein by anti-MAP kinase antibody. (A) Two µg of an expression vector for the wild-type (W) or mutant (M, mutated at Ser51) N-myc protein were transfected to HeLa cells together with or without 2 µg of an anti-MAP kinase antibody or nonspecific IgG by the liposome-mediated procedure. Whole cell extracts were prepared from the cells and incubated with [g-32P]ATP. The N-myc protein was immunoprecipitated with an anti-N-myc antibody, separated in an SDS-containing 10% polyacrylamide gel and autoradiographed (upper panel, kinase reaction). An aliquot of immunoprecipitates were blotted to nitrocellulose filter, reacted with anti-N-myc antibody, and visualized by ECL-detection system (lower panel, Western). Arrows (N-MYC) indicates the positions of the band due to N-myc protein. (B) Two µg of pN-myc, an expression vector for the wild-type N-myc protein, were transfected to HeLa cells together with or without 2 µg of an anti-MAP kinase antibody or nonspecific IgG by the liposome-mediated procedure. The transfected cells were labelled in vivo with 32P-orthophosphate for 4 hr prior to harvest. Nuclear extracts as well as cytoplasmic fractions (S100 fractions) were prepared from the cells. The N-myc protein was immunoprecipitated with an anti-N-myc antibody, separated in an SDS-containing 10% polyacrylamide gel and autoradiographed. Arrow N-MYC indicates the position of the band due to the N-myc protein.
lated by MAP kinase in cytoplasm. A recent report showed that MAP kinase directly binds to the N-terminal portion of the c myc protein covering Ser62 to phosphorylate the protein (40). Since the c-myc protein is always localized in cell nuclei, binding and phosphorylation of the protein by MAP kinase should occur in nuclei. The data described in this report suggest that the N-myc protein is phosphorylated by MAP kinase in nuclei. The H2TF1 element has been identified as the target sequence in the MHC class I promoter for transrepression by the N-myc protein (4). We described here the results that the N-myc protein efficiently represses the transcription due to pNFk-TATA-Luc which carries the H2TF1 element linked to TATA box of the hsp70 gene. As reported, c-myc protein also repressed MHC class I promoter activity (7). Target sequence of c-myc protein on the MHC class I gene, however, is the 821
JOBNAME: BBRC 219#3 PAGE: 10 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
region containing TATA box (7). The transrepression by N-myc protein was smaller on pMHCLuc, which contains the region of about 2 kb upstream from the MHC class I promoter, than on pNFk-TATA-Luc, probably because various transcriptional regulatory elements other than the H2TF1 element also contributed to the expression of pMHC-Luc. The formation of specific nucleoprotein complexes on the H2TF1 element was inhibited by the N-myc protein, and the N-myc protein thereby repressed the transcription due to the element. Introduction of the mutations into H2TF1 element lost the transrepression activity of N-myc protein in the luciferase construct as well as nucleoprotein complex formation (data not shown). The abrogation of the nucleoprotein complex on the H2TF1 element was observed with the N-myc protein phosphorylated at Ser51 by MAP kinase which repressed the transcription due to the element, but not affected by the mutant unphosphorylated protein which lost the transrepressing activity. The results suggest that the release of the positive regulatory factors of transcription from the H2TF1 element by the N-myc protein may result in the transrepression. Since no bands due to different nucleoprotein complexes on the same probe alternatively appeared, the N-myc protein may not directly compete with the factors in binding to the element. The N-myc protein may rather interact with certain member(s) involved in the complexes, including NFkB and other associated proteins, and thus interfere the binding of the original complexes. The interaction between the N-myc protein and such factors probably requires the phosphorylation of the N-myc protein at Ser51 by MAP kinase. A previous report consistently describes that the overexpression of the N-myc in a neuroblastoma cell line may impair the function of the MHC class I gene enhancer by altering the levels of the specific binding proteins to the H2TF1 element and thereby result in the suppression of the MHC class I gene expression in the cells (4). ACKNOWLEDGMENT We thank Kiyomi Takaya for expert experimental assistance. This work was supported by the grants from the Ministry of Education, Culture and Science in Japan and the Sagawa Cancer Research Foundation.
REFERENCES 1. Kohl, N. E., Kanda, N., Schreck, R. R., Burn, s G., Latt, S. A., Gilbert, F., and Alt, F. W., (1983). Cell 35, 359–367. 2. Schwab, M., Alitalo, K., Klempnauer, K.-H., Varmus, H. E., Bishop, J. M., Gilbert, F., Brodeur, G., Goldstein, M., and Traut, J., (1983). Nature 305, 245–248. 3. Bernards, R., Dessain, S. K., and Weinberg, R. A., (1986). Cell 47, 667–674. 4. Lenardo, M., Rustgi, A. K., Schievella, A. R., and Bernards, R., (1989). EMBO. J. 8, 3351–3355. 5. Akeson, R., and Bernards, R., (1990). Mol. Cell. Biol. 10, 2012–2016. 6. Sato, M., Imamura, Y., Iguchi-Ariga, S.M. M., and Ariga, H., (1992). Int. N. J. Onc. 1, 539–545. 7. Schrier, P. I., and Peltenburg, T.T. C., (1993). Ad. Cancer. Res. 60, 181–246. 8. Kieran, M., Blank, V., Logeat, F., Vandekerckhove, J., Lottspeich, F., Le Bail, O., Urban, M. B., Kourilsky, P., Bacuerle, P. A., and Israe, I. A., (1990). Cell 62, 1007–18. 9. Yano, O., Kanellopolous, J., Kiernan, M., LeBail, O., Israel, A., and Kourilsky, P., (1987). EMBO. J. 6, 3317–3324. 10. Israel, A., Yano, O., Logcat, F., Kieran, M., and Kourilsky, P., (1989). Nucleic Acids Res. 17, 5245–57. 11. Baldwin, A. S., and Sharp, P. A., (1988). Proc. Natl. Acad. Sci. USA 85, 723–727. 12. Lüscher, B., and Eisenman, R. N., (1991). Genes Dev. 4, 2025–2035. 13. Wang, J.Y. J., (1992). Biochim. Biophys. Acta 1114, 179–192. 14. Seth, A., Alvarez, E., Gupta, S., and Davis, R. J., (1991). J. Biol. Chem. 266, 23521–23524. 15. Amin, C., Wagner, A. J., and Hay, N., (1993). Mol. Cell. Biol. 13, 383–390. 16. Kato, G. J., Barrett, J., Villa-Garcia, M., and Dang, C. V., (1990). Mol. Cell. Biol. 10, 5914–5920. 17. Kitagawa, M., Okabe, T., Ogino, H., Matsumoto, H., Suzuki-Takahashi, I., Kokubo, T., Higashi, H., Saitoh, S., Taya, Y., Yasuda, H., Ohba, Y., Nishimura, S., Tanaka, N., and Okuyama, A., (1993). Oncogene 8, 2425–2432. 18. Penn, L.J. Z., Brooks, M. W., Laufer, E. M., Littlewood, T. D., Morgenstern, J. P., Evan, G. I., Lee, W.M. F., and Land, H., (1990). Mol. Cell. Biol. 10, 4961–4966. 19. Bister, K., Trachmann, C., Jansen, H. W., Schoeer, B., and Patschinsky, T., (1986). Oncogene 1, 97–109. 20. Alvarez, E., Northwood, I. C., Gonzalez, F. A., Latour, D. A., Seth, A., Abate, C., Curran, T., and Davis, R. J., (1991). J. Biol. Chem. 266, 15277–15285. 21. Gonzalez, F., Randen, D. L., and Davis, R. J., (1991). J. Biol. Chem. 266, 22159–22163. 822
JOBNAME: BBRC 219#3 PAGE: 11 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40.
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Henriksson, M., Bakardjiev, A., Klein, G., and Lüscher, B., (1993). Oncogene 8, 3199–3201. Seth, A., Gonzalez, P. A., Gupta, S., Randen, D. L., and Davis, R. J., (1992). J. Biol. Chem. 267, 24796–24804. Seth, A., Gupta, S., and Davis, R. J., (1993). Mol. Cell. Biol. 13, 4125–4136. Kitaura, H., Galli, I., Taira, T., Iguchi-Ariga, S.M. M., and Ariga, H., (1991). FEBS Lett. 290, 147–152. Albert, T., Urlbauer, B., Kohlhuber, F., Hammersen, B., and Eick, D., (1994). Oncogene 9, 759–763. Hagiwara, T., Nakaya, K., Nakamura, Y., Nakajima, H., Nishimura, S., and Taya, Y., (1992). Eur. J. Biochem. 209, 945–950. Hamann, U., Wenzel, A., Frank, R., and Schwab, M., (1991). Oncogene 6, 1745–1751. Graham, F. L., and van der Eb, A. J., (1973). Virology 52, 456–467. Dignam, J. D., Lebovitz, R. M., and Roeder, R. G., (1983). Nucleic Acids Res. 11, 1475–1489. Westin, G., Gerster, T., Müller, M. M., Schaffner, G., and Schaffner, W., (1987). Nucleic Acids Res. 15, 6787–6797. Takebe, Y., Seiki, M., Fujisawa, J., Hoy, P., Yokota, K., Arai, K., Yoshida, M., and Arai, N., (1988). Mol. Cell. Biol. 8, 466–472. Kubota, Y., Kim, S.-H., Iguchi-Ariga, S.M. M., and Ariga, H., (1989). Biochem. Biophys. Res. Comm. 162, 991–997. Gupta, S., Seth, A., and Davis, R. J., (1993). Proc. Natl. Acad. Sci. USA. 90, 3216–3220. Blenis, J., (1993). Proc. Natl. Acad. Sci. USA 90, 5889–5692. Davis, R., (1993). J. Biol. Chem. 268, 14553–14556. Chen, R.-H., Sarnecki, C., and Blenis, J., (1992). Mol. Cell. Biol. 12, 915–927. Gonzalez, F. A., Seth, A., Randen, D. L., Bowman, D. S., Fay, F. S., and Davis, R. J., (1993). J. Cell. Biol. 122, 1089–1101. Lenormand, P., Sardet, C., Pagès, G., Allemain, G. L., Brunet, A., and Pouysségur, J., (1993). J. Cell. Biol. 122, 1079–1088. Gupta, S., and Davis, R. J., (1994). FEBS Lett. 353, 281–285.
823
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.
219, 813–823 (1996)
0316
Transcriptional Repression Activity of N-MYC Protein Requires Phosphorylation by MAP Kinase Akira Manabe,*,† Sanae M. M. Iguchi-Ariga,‡ Hajime Iizuka,† and Hiroyoshi Ariga*,1 *Faculty of Pharmaceutical Sciences and ‡College of Medical Technology, Hokkaido University, Kita-ku, Sapporo, 060; and †Department of Dermatology, Asahikawa Medical College, Nishikagura, Asahikawa 078, Japan Received January 24, 1996 The transrepressing function of the N-myc protein is due to the distinct domains located at the N-terminus. In this report we introduced various point mutations around the myc boxes of the N-myc protein to examine whether the phosphorylation of the protein affected its transrepressing function. Serine (Ser) residues located at amino acid numbers 12, 31, 51, and 65 were changed to leucine or arginine, and the expression vectors of the mutant proteins were transfected to HeLa cells together with the luciferase gene linked to the MHC class I gene. Among the mutants, only the N(51)-myc carrying mutation at Ser51, a target for mitogen-activated protein kinase (MAP kinase), lost the repression activity, while the other mutant proteins preserved it. Formation in vitro of the specific nucleoprotein complexes on the H2TF1/NFkB element, a major target for transrepression by N-myc protein, was interfered by the wild-type N-myc protein, but not by the Ser51-mutated protein. The results suggest that the phosphorylation of the N-myc protein at Ser51 by MAP kinase is required for the transcriptional repression activity of the protein. © 1996 Academic Press, Inc.
The N-myc gene, a myc family oncogene, encodes a nuclear phosphoprotein. The N-myc protein is composed of several characteristic domains including two Myc boxes (the regions highly conserved among myc family proteins), an acidic region (a region rich in acidic amino acids), and a bHLH-Zip (a basic, helix-loop-helix, leucine zipper region) (1, 2) (Fig. 1). These domains are suggested to be important for the transrepression activity of the protein against the genes including the major histocompatibility (MHC) class I antigen (3, 4), the neural cell adhesion molecule (NCAM) (5), and the N-myc (6). Down-regulation of the MHC class I expression by N-myc and c-myc was reported in neuroblastoma and melanoma cells, respectively, and the regulation by myc proteins was at transcriptional level (see review, 7). N-myc was shown to repress the transcription from all the MHC class I loci, while the regulation by c-myc is predominantly restricted to the HLA-b locus (7). The expression of the MHC class I gene is dependent on the two major enhancer elements, enhancer A and enhancer B, located between −190 and −138 from, and immediately upstream from, the transcriptional start site (+1), respectively. Of the two enhancers, enhancer A shows a transcriptional activity much stronger than enhancer B, and three copies of the TGGGGATTCCCA sequence or its variants near the 39-terminus of enhancer A efficiently promotes the transcription of the reporter gene linked to the elements (4). Transfection experiments of the N-myc gene to cells indicated that TGGGGATTCCCA in enhancer A was a target sequence of transrepression of the MHC class I gene by N-myc (4). Several transcription factors, including KBF-1 of 48 KDa (8, 9), KBF2 of 58 KDa (10), H2TF1 of 110 KDa (11) and NFkB (11), have been identified to recognize the sequence. All of the factors belong to rel family proteins. An elevated level of the N-myc expression in neuroblastoma or transfected cells was correlated with a reduced binding of H2TF1 to enhancer A (11). The fact suggested that the N-myc protein directly or indirectly interferes with H2TF1 or NFkB family proteins in inactivating enhancer A. Precise molecular mechanisms of the transrepression by N-myc, however, remain unclarified.
1
To whom correspondence should be addressed. Fax: +81 11 (706) 4988. 813 0006-291X/96 $18.00 Copyright © 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
JOBNAME: BBRC 219#3 PAGE: 2 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Phosphorylation is also considered to be important for the functions of myc proteins (see reviews, 4, 12, 13 and references therein). The c-myc protein has been reported to be phosphorylated by casein kinase II (ckII) at the central (amino acid residues 240–262) or C-terminal region (residues 342–357), but roles of the phosphorylation are still unclarified (12–14): Deletion of the regions affected little the activities of the c-myc protein in transformation and transcriptional regulation (15–18), although the ckII phosphorylation was correlated with transformation of chicken haematopoietic cells by the v-myc protein (19). Mitogen-activated protein kinase (MAP kinase) has also been reported to phosphorylate a site near myc boxes of the c-myc protein in vitro (14, 20–24). The site, located in the N-terminal region important for both transactivation and transformation activities (14–16, 18, 25), was often found to be mutated in the c-myc proteins recovered from various cancer cells (26). As for phosphorylation of the N-myc protein, only the sites within the acidic region have been reported so far to be phosphorylated by ckII (27, 28), although a putative target for MAP kinase exists near the myc boxes of the protein. Here we introduced point mutations into putative phosphorylation sites around the myc boxes of the N-myc protein, and examined the mutant proteins for transrepressing activity on the regulatory sequences of the MHC class I gene. The results showed that the phosphorylation at Ser51 by MAP kinase is required for the transrepression activity of the N-myc protein. Formation in vitro of the specific nucleoprotein complexes on the H2TF1/NFkB element was interfered by the wild type N-myc protein, but not by the Ser51-mutated protein. MATERIALS AND METHODS Plasmids and introduction of mutations into N-myc protein. pGBV-Luc, a plasmid containing the luciferase gene (Wako Pure Chemicals Co. Ltd., Kyoto, Japan), was digested with HindIII, treated with Klenow fragment, and then self-ligated to destroy HindIII site (named pGBVH-Luc). The ScaI-ScaI fragment containing multicloning sites from pUC18 was inserted to the ScaI-SacI sites of pGBVH-Luc, to obtain pGBV(18). pH-2Kbm1-CAT was kindly provided by Dr. Kazushige Yokoyama, Riken, Tsukuba, Japan. pH-2Kbm1-CAT, which contains the transcriptional regulatory sequences of the mouse MHC class I gene linked to the bacterial chloramphenicol acetyltransferase gene (28), was digested with HindIII and NruI, and the fragment containing the MHC class I promoter was inserted between the HindIII-SmaI sites of pGBV(18). The construct was named pMHC-Luc. As for pNFk-TATA-Luc construction, oligonucleotides as follows were synthesized: 59-tcgagTGGGGATTCCCCATCTCCACa-39 (upper strand) and 59-agcttGTGGAGATGGGGAATCCCCAc-39 (lower strand), in which small letter indicates XhoI and HindIII restriction recognition sequences. The oligonucleotides were annealed and inserted first into the XhoI-HindIII sites of pBluescript SK(-). The SmaI-SacI fragment of the resultant plasmid containing the insertion was then cloned into the SmaI-SacI sites of pHS-TATA-Luc which contains TATA box of the human heat shock protein 70 gene linked to the luciferase gene in pGBV. pMHC-Luc and pNFk-TATA-Luc were used as reporter genes in co-transfection experiments. Mutations into the N-myc protein were introduced by polymerase chain reactions (PCR). Briefly, the fragment from EcoRI to SacI (from nucleotide number 1 to 970) of the human N-myc cDNA was cloned to the EcoRI-SacI sites of pUC19, to obtain pHN(E-S). The first PCR reaction was carried out on a pHE(E-S) template using mutation primers shown in Fig. 1 and conventional M4 or RV primer to amplify the left or right half segment, respectively. The amplified left and right fragments were mixed, and the second PCR reaction was carried out using M4 and RV primers. The conditions for both the reactions were 1 cycle of 94°C for 5 min, and 30 cycles of 94°C for 1 min, 55°C for 2 min, and 72°C for 3 min. Nucleotide sequences of upper (U) and lower (L) oligonucleotides used as primers are the following: N-myc(12); U:59GAGTTTGACTtGCTACAG39, L: 59CTGTAGCaAGTCAAACTC39, N-myc(31); U:59CGGCCCCGACTtGACCCC39, L: GGGGTCaAGTCGGGGCCG39, N-myc(51), U:59CCGCTGTtGCCCAGCCGT39, L: 59ACGGCTGGGCaACAGCGG39, N-myc(65); U: 59 CCCCCGAGaTGGGTCCCGGAGATG39, L: 59CATCTCCGGGACCCAtCTCGGGGG39. Small letters indicate the mutated nucleotides. The resulting fragments were digested with EcoRI and SacI, and cloned between the EcoRI/BgIII sites of pSRa296, together with the SacI-BgIII fragment of the C-terminal fragment of the N-myc cDNA. Plasmid DNA containing desired mutation was confirmed by the sequencing analysis of DNA. Cell culture and transfection. Human HeLa cells were cultured in Dulbecco’s modified Eagle medium supplemented with 10% calf serum. Various amounts of N-myc expression vectors were transfected to HeLa cells about 50% confluent in a 6 cm dish by the calcium phosphate precipitation method (29), together with 2 mg of pMHC-Luc and 1 mg of pCMV-b-gal (an expression vector of the b-galactosidase gene linked to the cytomegalovirus promoter). At 4 hr after transfection, the cells were boosted with 10% glycerol and cultured. At 48 hr after transfection, the cells were harvested and cell extracts were prepared by adding the Triton X-100-containing solution from Pica Gene kit (Wako Pure Chemicals Co. Ltd., Kyoto, Japan). One tenth volume of the extracts was used to examine the b-galactosidase activity due to the pCMV-b-gal and then 814
JOBNAME: BBRC 219#3 PAGE: 3 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
assayed for the luciferase activity using the Pica Gene kit after standardization of transfection efficiency by b-galactosidase activity (Note, N-myc protein does not affect the activity of CMV promoter). For transfection of an anti-MAP kinase polyclonal antibody (ERK 2 (C-14) raised against a carboxyl-terminal epitope, Santa Cruz Biotechnology, Inc., California), plasmid DNAs, or both DNA and antibody, a commercial reagent of transfection based on liposome, “transfection reagent” (Boehringer Mannheim GmbH, Biochemica, Germany) was used. In transfection of either DNA or antibody alone, 18 ml of solution A containing DNA or antibody was gently mixed with 18 ml of solution B containing 11 ml of transfection reagent. The mixture was stood for 15 min at room temperature and added to the cells in 6 cm dish in 1.5 ml of medium without serum. At 18 hr after transfection, final 10% of serum was added back to medium and the cells were cultured for another 30 hr. In transfection of both DNA and antibody, 18 ml of solution A containing antibody was gently mixed with 18 ml of solution B containing 11 ml of transfection reagent. The mixture was stood for 15 min at room temperature and added to the cells as above. At 4 hr after transfection, solution C containing DNA and 11 ml of the transfection reagent was added to cells. At 14 hr after transfection, final 10% of serum was added back to medium and the cells were cultured for another 30 hr. One mg of the anti-MAP kinase antibody or non-specific rabbit anti-human IgG was added to the cells. Assays of the enzyme activities were carried out as described below. All the experiments were repeated more than three times. Bandshift assay. Oligonucleotides corresponding to the H2TF1/NFkB element (59-GAT CCT GGG GAT TCC CCA G-39; 59-TCG ACT GGG GAA TCC CCA G-39) were chemically synthesized, annealed and end-labelled using Klenow enzyme and [a-32P]dCTP. The H2TF1 probe (10,000 cpm) was incubated in 15 ml with 2 mg of poly(dI-dC) and 30 mg of the nuclear extracts prepared from transfected or untransfected HeLa cells according to Dignam et al. (30) with minor modifications (31) in a buffer containing 50 mM KCl, 1 mM EDTA, 1 mM dithiothreitol, 4 mg/ml bovine serum albumin and 4% Ficoll 400. After 15 min at room temperature, the DNA/proteins mixtures were separated in a 4% polyacrylamide gel containing 0.25 × TBE (1 × TBE contains 90 mM Tris (pH 8.3), 90 mM boric acid, and 2.5 mM EDTA) and autoradiographed.
RESULTS Effect of the Mutations Introduced into the N-terminal Region on Transrepression Activity of the N-myc Protein We have previously identified several domains of N-myc protein responsible for its transrepression function (6). The domains include the N-terminal region containing the myc boxes, the acidic region, and the bHLH-Zip region (see Fig. 1, upper panel). The N-terminal and the acidic regions contain putative sites to be phosphorylated: The former region contains a target for MAP kinase, which overlaps a recognition consensus for cdc2 kinase, as in the corresponding region of the c-myc protein. We introduced mutations in the N-myc cDNA by a PCR-mediated method to substitute leucine (Leu) or arginine (Arg) for putative phosphorylation sites, Ser12, Ser31, Ser51 and Ser 65 among the 10 serine residues in the N-terminal portion of the N-myc protein (see Fig. 1, lower panel). The wild type and mutant N-myc cDNAs were linked to the SRa promoter (32) in order to express them efficiently in cells. The constructs were transfected to human HeLa cells and the levels of their expression were examined by Western blotting using a polyclonal anti-N-myc antibody. The anti-N-myc antibody used here did not recognize c-myc protein. As previously described (33), HeLa cells did not express, or at most expressed a trace amount of, the N-myc protein endogenously. The expression of the N-myc proteins due to transfection of the wild type (pN-myc) and mutant (pN-(12)myc, pN(31)-myc, pN(51)-myc, and pN(65)-myc) expression vectors were of the same or similar level (data not shown). The mutations we introduced thus did not affect the expression efficiency of the N-myc protein in transfected cells. Various amounts of the wild type N-myc expression vector, pN-myc, were transfected to HeLa cells together with a constant amount of pMHC-Luc, a reporter plasmid carrying the luciferase gene linked to the MHC Class I promoter region. The luciferase activity was transrepressed by pN-myc in a dose-dependent manner, while a vector pSRa296 (pSRa) alone affected little the activity (data not shown). In several independent experiments, 30–50% repression was observed when 2 mg of pN-myc were co-transfected. The inhibition of transcription by N-myc proteins was not so strong as expected, probably because there exist a number of transcriptional regulatory sequences other than the N-myc target element in the reporter plasmid, pMHC-Luc. We thereby transfected 2 mg of expression vectors for various mutant proteins to examine their transrepression activity. The 815
JOBNAME: BBRC 219#3 PAGE: 4 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Structure of the human N-myc protein and the amino acid and nucleotide sequences around the mutations introduced in the N-terminal region. Upper panel schematically represents characteristic domains of the human N-myc protein. Numbers in the figure correspond to the amino acid numbers of the protein. In lower panel, the nucleotide and amino acid sequences around the mutations introduced in the protein are shown. Mutated nucleotides are written in small letters, and the amino acids thereby mutated are squared.
expression vector for the wild type (pN-myc) or a mutant protein (pN-(12)myc, pN(31)-myc, pN(51)-myc, or pN(65)-myc) or the vector alone (pSRa296) was co-transfected to HeLa cells with pMHC-Luc, and the luciferase activity was assayed (Fig. 2). The wild type and the 3 mutants at Ser12, Ser31, and Ser65 suppressed the luciferase activity to similar extents. No repression was observed, however, by the Ser51 mutant protein: Alteration of Ser51 into Leu abolished the transrepression activity of the N-myc protein. Since the Ser51 is a putative target for MAP kinase, phosphorylation of the N-myc protein at Ser51 by MAP kinase was suggested to be required for the transrepression activity of the protein. Effect of Anti-MAP Kinase Antibody on Transrepression Activity of the N-myc Protein To verify the suggestion, an anti-MAP kinase antibody was introduced to the cells transfected with the wild type (pN-myc) or Ser51 mutant (pN(51)-myc) expression vector. Introduction of the antibody was carried out using “transfection reagent”, a commercial kit of liposome-mediated transfection (Boehringer Mannheim GmbH, Biochemica, Germany). Indirect immunofluorescence staining of the cells revealed not only that the anti-MAP kinase antibody and non-specific IgG introduced in the cells were stable at 48 hr after transfection and distributed both in nuclei and cytoplasm (data not shown). The transfection efficiency of the antibodies by the liposome-mediated method was comparable to that of the expression vector plasmids by the standard calcium phosphate precipitation method as described. Plasmid DNAs were also transfected to HeLa cells by the liposome-mediated method as efficiently as by the calcium phosphate procedure: the luciferase 816
JOBNAME: BBRC 219#3 PAGE: 5 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Effect of the point of mutations introduced in the N-myc protein on its transrepression activity. Two µg of expression vectors for the wild-type or mutated N-myc protein were cotransfected to HeLa cells together with 1 µg of pCMV-b-gal and 2 µg of pMHC-Luc, and the luciferase activity was assayed. Relative luciferase activities to the value of a reporter alone (pMHC-Luc alone was set as 100) are shown.
activities of the same levels as those in Fig. 2 were recovered from the cells transfected with the same plasmids by the liposome-mediated method (data not shown). We then carried out the co-transfection experiments of the N-myc expression vectors and the reporter pMHC-Luc with the antibodies (Fig. 3). Whole cell extracts were prepared from the cells. The results of western blot analyses indicate that the amounts of the wild type and mutant N-myc proteins in the extracts were comparable (data not shown). The repression of the luciferase activity
FIG. 3. Effect of an anti-MAP kinase antibody on the repression activity of N-myc protein by using pMHC-Luc as a reporter gene. Two µg of the expression vectors for the wild-type or mutated N-myc proteins, or pSRa vector, were cotransfected to HeLa cells together with 2 < gkm>g of pMHC-Luc, 1 µg of pCMV-b-gal, and 0.2 µg of an anti-MAP kinase antibody or nonspecific IgG by a liposome-mediated procedure as described in Materials and methods. The luciferase activity was assayed and relative luciferase activities to the value of a reporter alone (pMHC-Luc alone was set as 100) are shown. 817
JOBNAME: BBRC 219#3 PAGE: 6 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
due to pMHC-Luc by the expression vector for the wild type N-myc protein (pN-myc) was abrogated by the anti-MAP kinase antibody co-transfected to the cells, while non-specific IgG hardly affected the transrepression. Similar results as that of wild type N-myc protein were also obtained in the experiments using the mutant N-myc proteins of N−(21), N−(31), and N−(65) (data not shown). The expression of the Ser51-mutated protein due to pN(51)-myc did not repress the luciferase activity in the absence or presence of either the anti-MAP kinase antibody or nonspecific IgG. These results indicate that phosphorylation of the N-myc protein at Ser51 residue by MAP kinase is required for the transrepression activity of the protein. Interference of the Nucleoprotein Complex Formation on the H2TF1/NFkB Element by the N-myc Protein In this report, we examined the transrepression activity of the N-myc protein to the promoter region of MHC class I gene of about 2,000 bp, in which the major target for the N-myc protein is the H2TF1/NFkB element (4). To verify the target sequence for the N-myc protein in the MHC class I promoter, the H2TF1/NFkB element was ligated to TATA box of the heat shock protein 70 (hsp70) gene linked to the luciferase gene, and the construct pNFk-TATA-Luc was used as a reporter plasmid for the N-myc-induced transrepression experiments in HeLa cells (Fig. 4). A greater transrepression by co-transfected pN-myc was observed for pNFk-TATA-Luc than for pMHC-Luc, probably because pMHC-Luc contains various transcriptional elements, in addition to the H2TF1/NFkB element, which may be involved in the MHC expression. The expression vector for the Ser51-mutated N-myc protein hardly affected the transcription due to pNFk-TATA-Luc as well as the transcription due to pMHC-Luc. Co-introduction of the anti-MAP kinase antibody abrogated the transrepression activity of the N -myc protein, while IgG little affected. The results implicated that the N-myc protein represses transcription via the H2TF1/NFkB element only when the protein is phosphorylated by MAP kinase. Repression of transcription due to a specific DNA element is usually accompanied with a
FIG. 4. Effect of an anti-MAP kinase antibody on the repression activity of N-myc protein using pNFk -TATA-Luc as a reporter gene. The transfection and luciferase assay experiments were carried out as described in the legend of Fig. 3 except for pNFk-TATA-Luc used as a reporter gene. Two µg of the expression vector for the wild-type or mutated N-myc protein, or pSRa vector, were cotransfected to HeLa cells together with 2 µg of pNFk-TATA-Luc, 1 µg of pCMV-b-gal, and 0.2 µg of an anti-MAP kinase antibody or nonspecific IgG by the liposome-mediated procedure as described in Materials and methods. The luciferase activity was assayed and relative luciferase activities to the value of a reporter alone (pNFk-TATA-Luc alone was set as 100) are shown. 818
JOBNAME: BBRC 219#3 PAGE: 7 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
binding of negative factors to or release of positive factors from the element. We hence examined whether the N-myc protein affect the nucleoprotein complex formation on the element. Oligonucleotides corresponding to the H2TF1/NFkB element were labelled and incubated with the nuclear extract prepared from HeLa cells before or after transfection of the expression vector for the wild type (pN-myc) or mutant (pN(51)-myc) N-myc protein. Several bands of complexes were detected with the extract from untransfected cells (Fig. 5A, lane 2). Competition experiments using nonlabelled homo- or hetero-oligonucleotides including two points mutation revealed that three bands indicated as I, II and III were due to proteins specifically bound to the H2TF1/NFkB element (data not shown). As for the extracts prepared from the cells transfected with the wild type pN-myc, the specific bands I, II and III disappeared with the dose of transfected pN-myc (Fig. 5A, lanes 3–5). The expression of the Ser51-mutated protein due to pN(51)-myc, on the other hand, did not alter the pattern of the nucleoprotein complexes (Fig. 5A, lanes 6–8). Since the transrepression was observed with pN-myc, but not with pN(51)-myc (Fig. 2), the complexes I, II and III are suggested to be of positive factors for transcription due to the H2TF1 element, and the release of the complexes from the element by the N-myc protein may result in the transrepression. This possibility was further suggested by the mixing experiments between the extracts from the cells transfected with either pN-myc or pN(51)-myc. Extract from pN-myc transfected cells abolished the specific bands derived from non-transfected cells, while extract from pN(51)-myc transfected cells did not. Addition of non-transfected cell extract to that of pN-myc transfected cells, on the other hand, restored the specific bands in dose dependent manner (data not shown). Furthermore, the complexes I, II and III with the untransfected cell extract, which disappeared by the wild type N-myc protein due to transfected pN-myc, were restored by the co-transfected anti-MAP kinase antibody, but not by non-specific IgG (Fig. 5B, lanes 4 and 5). The results suggest that the N-myc protein
FIG. 5. Inhibition by N-myc protein of nucleoprotein complexes formation on the H2TF1 element. (A) Various amounts of an expression vector for the wild-type or mutant N-myc protein (pN-myc or pN-(51)myc) were transfected to HeLa cells by the calcium phosphate precipitation method. Nuclear extracts were prepared from the cells as well as from nontransfected cells, incubated with 32P-labelled oligonucleotides corresponding to the H2TF1 element and separated on a 4% polyacrylamide gel containing 0.25× TBE. Three bands due to the nucleoprotein complexes specifically interfered by the wild-type N-myc protein are indicated by arrows, I, II and III. Arrow F indicates the position of free probe. (B) Two µg of pN-myc were transfected to HeLa cells together with or without 2 µg of an anti-MAP kinase antibody or nonspecific IgG by the liposome-mediated procedure. Nuclear extracts were prepared from the cells as well as nontransfected cells, incubated with 32 P-labelled H2TF1 probe and analyzed as in A. Arrows I, II and III correspond to the same nucleoprotein complexes indicated in A. Arrows I, II and III correspond to the same nucleoprotein complexes indicated in A. Arrow F indicates the position of free probe. 819
JOBNAME: BBRC 219#3 PAGE: 8 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS 51
phosphorylated at Ser by MAP kinase interferes the formation of specific nucleoprotein complexes, probably of positive factors, on the H2TF1/NFkB element and that the protein thereby represses the transcription due to the element. Phosphorylation of N-myc Protein by MAP Kinase The expression vector pN-myc or pN(51)-myc was then co-transfected to HeLa cells together with or without the anti-MAP kinase antibody or non-specific IgG. Whole cell extracts were prepared from the cells, incubated in the presence of [g-32P]ATP, immunoprecipitated with the anti-N-myc antibody and electrophoresed through an SDS-containing 10% polyacrylamide gel. The results of western blot analyses indicate that the amounts of the wild type and mutant N-myc proteins in the extracts were comparable (Fig. 6A, lower panel). In the extract from the cells transfected with pN-myc, a clear band of a phosphorylated protein was detected at the size expected for the N-myc protein (Fig. 6A upper part, lane 4). The band was missing in the extracts from the cells co-transfected with pN-myc and the anti-MAP kinase antibody (Fig. 6A upper part, lane 6). Non-specific IgG co-transfected with pN-myc to the cells did not alter the electrophoretic pattern of phosphorylated proteins (Fig. 6A, upper panel, lane 5 compared with lane 4). The extracts from the cells transfected with pN(51)-myc carrying the mutation at Ser51 yielded no band at the size for N-myc protein regardless to the co-transfection of the antibodies (Fig. 6A upper panel, lanes 1 - 3). The results indicate that the N-myc protein expressed due to pN-myc in HeLa cells was mainly phosphorylated at Ser51 by MAP kinase. To examine whether the N-myc protein was phosphorylated by MAP kinase in cytoplasm or in nuclei, phosphorylation experiments were carried out using transfected cells. HeLa cells were transfected with the N-myc expression vector (pN-myc), either alone or together with the anti-MAP kinase antibody or non-specific IgG, and labelled in vivo with 32P-orthophosphate. Cytoplasmic fraction and nuclear extract were prepared from the cells and analyzed by immunoprecipitation using the anti-N-myc antibody. As for the cells transfected with the N-myc expression vector alone or together with non-specific IgG, the phosphorylated N-myc protein was detected only in the nuclear extract, but not in the cytoplasmic fraction (Fig. 6B, lanes 1, 3, 4 and 6). Neither in the nuclear extract nor in the cytoplasmic fraction prepared from the cells co-transfected with pN-myc and the anti-MAP kinase antibody, on the contrary, the 32P-labelled N-myc protein was identified (Fig. 6B, lanes 2 and 5). Phosphorylated N-myc protein was hardly detected in either nuclei or cytoplasm of the cells transfected with pN(51)-myc (data not shown). DISCUSSION In this report, we show the involvement of MAP kinase-dependent phosphorylation in the transcriptional activity of N-myc protein. The consensus sequence for a MAP kinase target is Pro-Leu-Ser/Thr-Pro-Pro, and the sequence is present between two myc boxes both in the c-myc (amino acid number from 60 to 63) and the N-myc (amino acid number from 49 to 52) proteins. A recent report tells that MAP kinase directly phosphorylated the Ser62 of the c-myc protein in vivo (34). The corresponding Ser51 of the N-myc protein is thus suggested to be phosphorylated directly by MAP kinase. The Ser62 of the c-myc protein was also reported to be important for transactivation by the protein (34), which is reminiscent of the results reported here: phosphorylation of the Ser51 of the N-myc protein is required for the transrepression activity of the protein. MAP kinase has been widely reported to regulate functions of a variety of proteins including cell surface proteins, cytoskeletal proteins, and the nuclear proteins including c-jun, c-myc, NF-IL6, ATF-2, TAL1, and p62TCF/ELK-1, involved in signal transduction (see reviews 35, 36, and references therein). As for the phosphorylation of nuclear proteins, translocation of MAP kinase from cytoplasm to nucleus has been considered (23, 33, 37-39): MAP kinase may phosphorylate the proteins in nucleus. Alternatively, the proteins may translocate to nucleus after being phosphory820
JOBNAME: BBRC 219#3 PAGE: 9 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 6. Inhibition of phosphorylation of the N-myc protein by anti-MAP kinase antibody. (A) Two µg of an expression vector for the wild-type (W) or mutant (M, mutated at Ser51) N-myc protein were transfected to HeLa cells together with or without 2 µg of an anti-MAP kinase antibody or nonspecific IgG by the liposome-mediated procedure. Whole cell extracts were prepared from the cells and incubated with [g-32P]ATP. The N-myc protein was immunoprecipitated with an anti-N-myc antibody, separated in an SDS-containing 10% polyacrylamide gel and autoradiographed (upper panel, kinase reaction). An aliquot of immunoprecipitates were blotted to nitrocellulose filter, reacted with anti-N-myc antibody, and visualized by ECL-detection system (lower panel, Western). Arrows (N-MYC) indicates the positions of the band due to N-myc protein. (B) Two µg of pN-myc, an expression vector for the wild-type N-myc protein, were transfected to HeLa cells together with or without 2 µg of an anti-MAP kinase antibody or nonspecific IgG by the liposome-mediated procedure. The transfected cells were labelled in vivo with 32P-orthophosphate for 4 hr prior to harvest. Nuclear extracts as well as cytoplasmic fractions (S100 fractions) were prepared from the cells. The N-myc protein was immunoprecipitated with an anti-N-myc antibody, separated in an SDS-containing 10% polyacrylamide gel and autoradiographed. Arrow N-MYC indicates the position of the band due to the N-myc protein.
lated by MAP kinase in cytoplasm. A recent report showed that MAP kinase directly binds to the N-terminal portion of the c myc protein covering Ser62 to phosphorylate the protein (40). Since the c-myc protein is always localized in cell nuclei, binding and phosphorylation of the protein by MAP kinase should occur in nuclei. The data described in this report suggest that the N-myc protein is phosphorylated by MAP kinase in nuclei. The H2TF1 element has been identified as the target sequence in the MHC class I promoter for transrepression by the N-myc protein (4). We described here the results that the N-myc protein efficiently represses the transcription due to pNFk-TATA-Luc which carries the H2TF1 element linked to TATA box of the hsp70 gene. As reported, c-myc protein also repressed MHC class I promoter activity (7). Target sequence of c-myc protein on the MHC class I gene, however, is the 821
JOBNAME: BBRC 219#3 PAGE: 10 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
region containing TATA box (7). The transrepression by N-myc protein was smaller on pMHCLuc, which contains the region of about 2 kb upstream from the MHC class I promoter, than on pNFk-TATA-Luc, probably because various transcriptional regulatory elements other than the H2TF1 element also contributed to the expression of pMHC-Luc. The formation of specific nucleoprotein complexes on the H2TF1 element was inhibited by the N-myc protein, and the N-myc protein thereby repressed the transcription due to the element. Introduction of the mutations into H2TF1 element lost the transrepression activity of N-myc protein in the luciferase construct as well as nucleoprotein complex formation (data not shown). The abrogation of the nucleoprotein complex on the H2TF1 element was observed with the N-myc protein phosphorylated at Ser51 by MAP kinase which repressed the transcription due to the element, but not affected by the mutant unphosphorylated protein which lost the transrepressing activity. The results suggest that the release of the positive regulatory factors of transcription from the H2TF1 element by the N-myc protein may result in the transrepression. Since no bands due to different nucleoprotein complexes on the same probe alternatively appeared, the N-myc protein may not directly compete with the factors in binding to the element. The N-myc protein may rather interact with certain member(s) involved in the complexes, including NFkB and other associated proteins, and thus interfere the binding of the original complexes. The interaction between the N-myc protein and such factors probably requires the phosphorylation of the N-myc protein at Ser51 by MAP kinase. A previous report consistently describes that the overexpression of the N-myc in a neuroblastoma cell line may impair the function of the MHC class I gene enhancer by altering the levels of the specific binding proteins to the H2TF1 element and thereby result in the suppression of the MHC class I gene expression in the cells (4). ACKNOWLEDGMENT We thank Kiyomi Takaya for expert experimental assistance. This work was supported by the grants from the Ministry of Education, Culture and Science in Japan and the Sagawa Cancer Research Foundation.
REFERENCES 1. Kohl, N. E., Kanda, N., Schreck, R. R., Burn, s G., Latt, S. A., Gilbert, F., and Alt, F. W., (1983). Cell 35, 359–367. 2. Schwab, M., Alitalo, K., Klempnauer, K.-H., Varmus, H. E., Bishop, J. M., Gilbert, F., Brodeur, G., Goldstein, M., and Traut, J., (1983). Nature 305, 245–248. 3. Bernards, R., Dessain, S. K., and Weinberg, R. A., (1986). Cell 47, 667–674. 4. Lenardo, M., Rustgi, A. K., Schievella, A. R., and Bernards, R., (1989). EMBO. J. 8, 3351–3355. 5. Akeson, R., and Bernards, R., (1990). Mol. Cell. Biol. 10, 2012–2016. 6. Sato, M., Imamura, Y., Iguchi-Ariga, S.M. M., and Ariga, H., (1992). Int. N. J. Onc. 1, 539–545. 7. Schrier, P. I., and Peltenburg, T.T. C., (1993). Ad. Cancer. Res. 60, 181–246. 8. Kieran, M., Blank, V., Logeat, F., Vandekerckhove, J., Lottspeich, F., Le Bail, O., Urban, M. B., Kourilsky, P., Bacuerle, P. A., and Israe, I. A., (1990). Cell 62, 1007–18. 9. Yano, O., Kanellopolous, J., Kiernan, M., LeBail, O., Israel, A., and Kourilsky, P., (1987). EMBO. J. 6, 3317–3324. 10. Israel, A., Yano, O., Logcat, F., Kieran, M., and Kourilsky, P., (1989). Nucleic Acids Res. 17, 5245–57. 11. Baldwin, A. S., and Sharp, P. A., (1988). Proc. Natl. Acad. Sci. USA 85, 723–727. 12. Lüscher, B., and Eisenman, R. N., (1991). Genes Dev. 4, 2025–2035. 13. Wang, J.Y. J., (1992). Biochim. Biophys. Acta 1114, 179–192. 14. Seth, A., Alvarez, E., Gupta, S., and Davis, R. J., (1991). J. Biol. Chem. 266, 23521–23524. 15. Amin, C., Wagner, A. J., and Hay, N., (1993). Mol. Cell. Biol. 13, 383–390. 16. Kato, G. J., Barrett, J., Villa-Garcia, M., and Dang, C. V., (1990). Mol. Cell. Biol. 10, 5914–5920. 17. Kitagawa, M., Okabe, T., Ogino, H., Matsumoto, H., Suzuki-Takahashi, I., Kokubo, T., Higashi, H., Saitoh, S., Taya, Y., Yasuda, H., Ohba, Y., Nishimura, S., Tanaka, N., and Okuyama, A., (1993). Oncogene 8, 2425–2432. 18. Penn, L.J. Z., Brooks, M. W., Laufer, E. M., Littlewood, T. D., Morgenstern, J. P., Evan, G. I., Lee, W.M. F., and Land, H., (1990). Mol. Cell. Biol. 10, 4961–4966. 19. Bister, K., Trachmann, C., Jansen, H. W., Schoeer, B., and Patschinsky, T., (1986). Oncogene 1, 97–109. 20. Alvarez, E., Northwood, I. C., Gonzalez, F. A., Latour, D. A., Seth, A., Abate, C., Curran, T., and Davis, R. J., (1991). J. Biol. Chem. 266, 15277–15285. 21. Gonzalez, F., Randen, D. L., and Davis, R. J., (1991). J. Biol. Chem. 266, 22159–22163. 822
JOBNAME: BBRC 219#3 PAGE: 11 SESS: 10 OUTPUT: Tue Apr 2 13:44:42 1996 /xypage/worksmart/tsp000/68953f/48
Vol. 219, No. 3, 1996 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40.
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Henriksson, M., Bakardjiev, A., Klein, G., and Lüscher, B., (1993). Oncogene 8, 3199–3201. Seth, A., Gonzalez, P. A., Gupta, S., Randen, D. L., and Davis, R. J., (1992). J. Biol. Chem. 267, 24796–24804. Seth, A., Gupta, S., and Davis, R. J., (1993). Mol. Cell. Biol. 13, 4125–4136. Kitaura, H., Galli, I., Taira, T., Iguchi-Ariga, S.M. M., and Ariga, H., (1991). FEBS Lett. 290, 147–152. Albert, T., Urlbauer, B., Kohlhuber, F., Hammersen, B., and Eick, D., (1994). Oncogene 9, 759–763. Hagiwara, T., Nakaya, K., Nakamura, Y., Nakajima, H., Nishimura, S., and Taya, Y., (1992). Eur. J. Biochem. 209, 945–950. Hamann, U., Wenzel, A., Frank, R., and Schwab, M., (1991). Oncogene 6, 1745–1751. Graham, F. L., and van der Eb, A. J., (1973). Virology 52, 456–467. Dignam, J. D., Lebovitz, R. M., and Roeder, R. G., (1983). Nucleic Acids Res. 11, 1475–1489. Westin, G., Gerster, T., Müller, M. M., Schaffner, G., and Schaffner, W., (1987). Nucleic Acids Res. 15, 6787–6797. Takebe, Y., Seiki, M., Fujisawa, J., Hoy, P., Yokota, K., Arai, K., Yoshida, M., and Arai, N., (1988). Mol. Cell. Biol. 8, 466–472. Kubota, Y., Kim, S.-H., Iguchi-Ariga, S.M. M., and Ariga, H., (1989). Biochem. Biophys. Res. Comm. 162, 991–997. Gupta, S., Seth, A., and Davis, R. J., (1993). Proc. Natl. Acad. Sci. USA. 90, 3216–3220. Blenis, J., (1993). Proc. Natl. Acad. Sci. USA 90, 5889–5692. Davis, R., (1993). J. Biol. Chem. 268, 14553–14556. Chen, R.-H., Sarnecki, C., and Blenis, J., (1992). Mol. Cell. Biol. 12, 915–927. Gonzalez, F. A., Seth, A., Randen, D. L., Bowman, D. S., Fay, F. S., and Davis, R. J., (1993). J. Cell. Biol. 122, 1089–1101. Lenormand, P., Sardet, C., Pagès, G., Allemain, G. L., Brunet, A., and Pouysségur, J., (1993). J. Cell. Biol. 122, 1079–1088. Gupta, S., and Davis, R. J., (1994). FEBS Lett. 353, 281–285.
823