JUMONJI, a Critical Factor for Cardiac Development, Functions as a Transcriptional Repressor
2003; Elsevier BV; Volume: 278; Issue: 43 Linguagem: Inglês
10.1074/jbc.m307386200
ISSN1083-351X
AutoresTae‐Gyun Kim, Jonathan C. Kraus, Junqin Chen, Youngsook Lee,
Tópico(s)Genomics and Chromatin Dynamics
ResumoJUMONJI (JMJ) is a nuclear factor that is critical for normal cardiovascular development, evidenced by the analysis of jmj homozygous mutant mice. However, the molecular function of JMJ remains to be elucidated. In the present study, we investigated whether JMJ is a transcriptional modulator. Reporter gene assays using the GAL4-DNA binding domain fused to JMJ and a reporter gene consisting of the GAL4 binding sites upstream of a luciferase reporter gene indicated that JMJ functions as a powerful transcriptional repressor. The DNA binding motif of JMJ was determined using CASTing experiments by incubating a random oligonucleotide library with the GST-JMJ fusion protein coupled to agarose beads. Among the selected binding oligonucleotides, the high affinity DNA binding sequences were identified by gel retardation assays. JMJ repressed expression of the reporter genes containing the high affinity JMJ binding sequences, indicating that JMJ is a DNA-binding transcriptional repressor. The domains for transcriptional repression, DNA binding, and nuclear localization signal were mapped by mutational analyses using reporter gene assays, gel retardation assays, and immunostaining experiments, respectively. The present data demonstrate for the first time that JMJ functions as a DNA-binding transcriptional repressor. Therefore, JMJ may play a critical role in transcription factor cascade to regulate expression of heart-specific genes and normal cardiac development. JUMONJI (JMJ) is a nuclear factor that is critical for normal cardiovascular development, evidenced by the analysis of jmj homozygous mutant mice. However, the molecular function of JMJ remains to be elucidated. In the present study, we investigated whether JMJ is a transcriptional modulator. Reporter gene assays using the GAL4-DNA binding domain fused to JMJ and a reporter gene consisting of the GAL4 binding sites upstream of a luciferase reporter gene indicated that JMJ functions as a powerful transcriptional repressor. The DNA binding motif of JMJ was determined using CASTing experiments by incubating a random oligonucleotide library with the GST-JMJ fusion protein coupled to agarose beads. Among the selected binding oligonucleotides, the high affinity DNA binding sequences were identified by gel retardation assays. JMJ repressed expression of the reporter genes containing the high affinity JMJ binding sequences, indicating that JMJ is a DNA-binding transcriptional repressor. The domains for transcriptional repression, DNA binding, and nuclear localization signal were mapped by mutational analyses using reporter gene assays, gel retardation assays, and immunostaining experiments, respectively. The present data demonstrate for the first time that JMJ functions as a DNA-binding transcriptional repressor. Therefore, JMJ may play a critical role in transcription factor cascade to regulate expression of heart-specific genes and normal cardiac development. We have previously reported that JUMONJI (JMJ) 1The abbreviations used are: JMJ, JUMONJI; DBD, DNA binding domain; ARID, AT-rich interaction domain; NLS, nuclear localization signal; GMSA, gel mobility shift assay; CASTing, cyclic amplification and selection of target; SAAB, site selection and amplification binding; HDAC, histone deacetylase; TSA, trichostatin A; CtBP, C-terminal binding protein; GST, glutathione S-transferase; aa, amino acid(s). is a nuclear protein that is critical for normal cardiovascular development by characterizing JMJ knockout mice (1Lee Y. Song A.J. Baker R. Micales B. Conway S.J. Lyons G.E. Circ. Res. 2000; 86: 932-938Crossref PubMed Scopus (125) Google Scholar). The jmj mutant embryos showed heart malformations including ventricular septal defect, noncompaction of the ventricular wall, double outlet right ventricle, and dilated atria. All homozygous jmj mutants died soon after birth. Cardiac marker analysis by in situ hybridization suggested that cardiomyocytes were differentiated but developmental regulation of chamber-specific genes was defective in late stage embryos. The jmj gene was first identified and described as a developmentally important gene in the nervous system and subsequently in liver, spleen, and thymus development (2Takeuchi T. Yamazaki Y. Katoh-Fukui Y. Tsuchiya R. Kondo S. Motoyama J. Higashinakagawa T. Genes Dev. 1995; 9: 1211-1222Crossref PubMed Scopus (261) Google Scholar, 3Motoyama J. Kitajima K. Kojima M. Kondo S. Takeuchi T. Mech. Dev. 1997; 66: 27-37Crossref PubMed Scopus (65) Google Scholar), when knockout mice were generated in genetic backgrounds that are different from ours (1Lee Y. Song A.J. Baker R. Micales B. Conway S.J. Lyons G.E. Circ. Res. 2000; 86: 932-938Crossref PubMed Scopus (125) Google Scholar). During mouse embryonic development, JMJ is widely expressed, including in the developing heart. In adult mice, it is expressed at a higher level in heart, skeletal muscle, brain, and thymus (1Lee Y. Song A.J. Baker R. Micales B. Conway S.J. Lyons G.E. Circ. Res. 2000; 86: 932-938Crossref PubMed Scopus (125) Google Scholar). Continuous expression of jmj in the heart suggests that JMJ plays an important role in both the developing and adult heart. Although JMJ may be involved in cell growth when JMJ is overexpressed (4Toyoda M. Kojima M. Takeuchi T. Biochem. Biophys. Res. Commun. 2000; 274: 332-336Crossref PubMed Scopus (45) Google Scholar), the molecular function of JMJ remains unknown. The deduced amino acid sequence of JMJ reveals a putative DNA binding domain (DBD) homologous to the DBD of a DNA-binding protein family, AT-rich interaction domain (ARID) (5Gregory S.L. Kortschak R.D. Kalionis B. Saint R. Mol. Cell Biol. 1996; 16: 792-799Crossref PubMed Scopus (138) Google Scholar), suggesting that JMJ is a transcription factor. However, the homology of this domain with the DBDs of other ARID family members is low, with only about 30% amino acid identity in the putative DBD. There is an increasing number of factors that belong to an ARID family, which show diverse functions in vertebrates, plants, and fungi (6Balciunas D. Ronne H. Trends Biochem. Sci. 2000; 25: 274-276Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar). Cardiac development is a complex biological process requiring the integration of cell specification, differentiation, and morphogenesis. Many factors have been implicated in this process on the basis of their spatial and temporal expression patterns or their phenotypic effects when they are functionally inactivated in flies or mice (for reviews, see Refs. 7Cripps R.M. Olson E.N. Dev. Biol. 2002; 246: 14-28Crossref PubMed Scopus (272) Google Scholar, 8Zaffran S. Frasch M. Circ. Res. 2002; 91: 457-469Crossref PubMed Scopus (253) Google Scholar, 9Lyons G.E. Curr. Opin. Genet. Dev. 1996; 6: 454-460Crossref PubMed Scopus (74) Google Scholar). Recently, significant advances have been made in understanding the role of transcription factors during heart development. Transcription factors playing critical roles in early cardiac morphogenesis include the homeodomain protein, Nkx2.5 (10Lyons I. Parsons L.M. Hartley L. Li R. Andrews J.E. Robb L. Harvey R.P. Genes Dev. 1995; 9: 1654-1666Crossref PubMed Scopus (969) Google Scholar), MEF2C, a MADS box factor (11Lin Q. Schwarz J. Bucana C. Olson E.N. Science. 1997; 276: 1404-1407Crossref PubMed Scopus (790) Google Scholar), GATA-4, a zinc finger domain protein (12Molkentin J.D. Lin Q. Duncan S.A. Olson E.N. Genes Dev. 1997; 11: 1061-1072Crossref PubMed Scopus (968) Google Scholar, 13Kuo C.T. Morrisey E.E. Anandappa R. Sigrist K. Lu M.M. Parmacek M.S. Soudais C. Leiden J.M. Genes Dev. 1997; 11: 1048-1060Crossref PubMed Scopus (878) Google Scholar), and the basic helix-loop-helix factors, eHAND and dHAND (14Srivastava D. Cserjesi P. Olson E.N. Science. 1995; 270: 1995-1999Crossref PubMed Scopus (456) Google Scholar). Mutations in other nuclear factors, such as RXR (15Sucov H.M. Dyson E. Gumeringer C.L. Price J. Chien K.R. Evans R.M. Genes Dev. 1994; 8: 1007-1018Crossref PubMed Scopus (543) Google Scholar), RAR (16Mendelsohn C. Lohnes D. Decimo D. Lufkin T. LeMeur M. Chambon P. Mark M. Development. 1994; 120: 2749-2771Crossref PubMed Google Scholar), TEF-1 (17Chen Z. Friedrich G.A. Soriano P. Genes Dev. 1994; 8: 2293-2301Crossref PubMed Scopus (269) Google Scholar), Tbx5 (18Bruneau B.G. Nemer G. Schmitt J.P. Charron F. Robitaille L. Caron S. Conner D.A. Gessler M. Nemer M. Seidman C.E. Seidman J.G. Cell. 2001; 106: 70-721Abstract Full Text Full Text PDF Scopus (872) Google Scholar), and Pitx2 (19Yoshioka H. Meno C. Koshiba K. Sugihara M. Itoh H. Ishimaru Y. Inoue T. Ohuchi H. Semina E.V. Murray J.C. Hamada H. Noji S. Cell. 1998; 94: 299-305Abstract Full Text Full Text PDF PubMed Scopus (330) Google Scholar), resulted in various abnormal cardiac developments. Identification of molecular pathways involved in normal heart development led to the discovery of the genetic basis for human congenital heart disease. For example, congenital heart disease characterized by septal defects and abnormal atrioventricular conduction is caused by mutations in the transcription factor Nkx2.5 (20Schott J.J. Benson D.W. Basson C.T. Pease W. Silberbach G.M. Moak J.P. Maron B.J. Seidman C.E. Seidman J.G. Science. 1998; 281: 108-111Crossref PubMed Scopus (1123) Google Scholar). Another group of human cardiac defects referred to as CATCH-22 (cardiac defects, abnormal facies, thymic hypoplasia, cleft palate, and hypocalcemia, associated with chromosome 22 microdeletion) may be caused by mutation in a downstream gene of dHAND (21Yamagishi H. Garg V. Matsuoka R. Thomas T. Srivastava D. Science. 1999; 283: 1158-1161Crossref PubMed Scopus (241) Google Scholar). Holt-Oram syndrome, characterized by upper limb malformations and cardiac septation defects, is caused by mutations in the tbx5 gene (18Bruneau B.G. Nemer G. Schmitt J.P. Charron F. Robitaille L. Caron S. Conner D.A. Gessler M. Nemer M. Seidman C.E. Seidman J.G. Cell. 2001; 106: 70-721Abstract Full Text Full Text PDF Scopus (872) Google Scholar, 22Hiroi Y. Kudoh S. Monzen K. Ikeda Y. Yazaki Y. Nagai R. Komuro I. Nat. Genet. 2001; 28: 276-280Crossref PubMed Scopus (489) Google Scholar, 23Basson C.T. Bachinsky D.R. Lin R.C. Levi T. Elkins J.A. Soults J. Grayzel D. Kroumpouzou E. Traill T.A. Leblanc-Straceski J. Renault B. Kucherlapati R. Seidman J.G. Seidman C.E. Nat. Genet. 1997; 15: 30-35Crossref PubMed Scopus (921) Google Scholar). Therefore, it is critical to characterize the molecular function of JMJ. We have used in vitro mutagenesis to functionally characterize the JMJ protein and to better understand the structural/functional relationship of JMJ. The present study indicates that JMJ is a DNA-binding nuclear factor that contains a strong transcriptional repressor domain. JMJ may regulate the expression of cardiac genes, and therefore this study will form a basis to further identify molecular mechanism of cardiac development. Plasmid Construction—Cloning of the cDNA encoding jmj was previously described (1Lee Y. Song A.J. Baker R. Micales B. Conway S.J. Lyons G.E. Circ. Res. 2000; 86: 932-938Crossref PubMed Scopus (125) Google Scholar). For the GAL4 DBD-JMJ chimera, the wild type and various deletion mutants were constructed as follows. The various regions of JMJ were PCR-amplified with primers containing appropriate restriction digestion sites. The PCR products were digested and subcloned in frame to the GAL4 DBD in the pcDNA3 vector (24Fry C.J. Slansky J.E. Farnham P.J. Mol. Cell Biol. 1997; 17: 1966-1976Crossref PubMed Scopus (55) Google Scholar) and also subcloned in frame to an Xpress tag in pcDNA3.1/HisB (Invitrogen). The reporter gene consisting of 5× GAL4 binding site upstream of the adenovirus major late promoter TATA box linked to luciferase (pG5Ti-Luc, obtained from Dr. P. Farnham) (24Fry C.J. Slansky J.E. Farnham P.J. Mol. Cell Biol. 1997; 17: 1966-1976Crossref PubMed Scopus (55) Google Scholar) exhibited low basal activity. We constructed the reporter gene by subcloning 5× GAL4 binding sites into the pGL3 vector (Promega) that consisted of the SV40 promoter linked to a luciferase reporter gene, 5×G-pGL3. This reporter gene exhibited high basal reporter gene activity, which is suitable to study transcriptional repression activity. The reporter genes containing selected oligonucleotides by CASTing were constructed by subcloning the high affinity binding site of JMJ upstream of the SV40 promoter linked to luciferase in pGL3. All new constructs were confirmed by restriction enzyme digestion and sequencing. Preparation of the JMJ Fusion Protein and Anti-JMJ Antibody—The bacterially produced JMJ fusion protein was prepared for a gel mobility shift assay (GMSA), cyclic amplification and selection of target (CASTing), and production of antibodies. The regions containing a putative DNA-binding domain of JMJ (amino acids (aa) 529-798 and 529-1198) or N-terminal part (aa 1-517) were PCR-amplified using primers with appropriate linkers and were ligated in frame to the vector containing a cDNA for glutathione S-transferase (GST) (pGEX-2T; Amersham Biosciences). The GST-JMJ fusion proteins were prepared using glutathione-agarose beads (Sigma) as described previously (25Lee Y. Mahdavi V. J. Biol. Chem. 1993; 268: 2021-2028Abstract Full Text PDF PubMed Google Scholar, 26Lee Y. Nadal-Ginard B. Mahdavi V. Izumo S. Mol. Cell Biol. 1997; 17: 2745-2755Crossref PubMed Google Scholar, 27Lee Y. Shioi T. Kasahara H. Jobe S.M. Wiese R.J. Markham B.E. Izumo S. Mol. Cell Biol. 1998; 18: 3120-3129Crossref PubMed Scopus (245) Google Scholar). To generate anti-JMJ antibody against the JMJ fusion protein, GST-JMJ 529-798 was loaded onto SDS-PAGE, and a protein band visualized by Coomassie Brilliant Blue was used to inoculate a rabbit (service provided by Covance). The anti-JMJ antibodies were characterized by immunostaining, immunoprecipitation, and Western blot analysis as described previously (1Lee Y. Song A.J. Baker R. Micales B. Conway S.J. Lyons G.E. Circ. Res. 2000; 86: 932-938Crossref PubMed Scopus (125) Google Scholar). Briefly, for immunoprecipitation, in vitro translated and 35S-labeled JMJ (TNT kit; Promega) was incubated with anti-JMJ antibody in NETN buffer (100 mm NaCl, 1 mm EDTA, 20 mm Tris, pH 8.0, 0.5% Nonidet P-40) for 2 h at 4 °C followed by incubation with protein A-agarose beads for 1 h. The precipitate was washed and loaded onto SDS-PAGE and autoradiographed. For Western blot analysis, COS cells were transfected with a plasmid encoding JMJ. Whole cell lysate was prepared as described elsewhere (4Toyoda M. Kojima M. Takeuchi T. Biochem. Biophys. Res. Commun. 2000; 274: 332-336Crossref PubMed Scopus (45) Google Scholar). The lysates were sonicated and centrifuged to remove the cell debris. Cell extracts were loaded onto 7.5% SDS-PAGE and transferred onto polyvinylidene difluoride membrane. The membrane was incubated with anti-JMJ antibody (1000-fold dilution), and the protein band was detected with the ECL kit (Amersham Biosciences). GMSA—To examine whether JMJ binds to DNA in vitro, GMSAs were performed as described previously (25Lee Y. Mahdavi V. J. Biol. Chem. 1993; 268: 2021-2028Abstract Full Text PDF PubMed Google Scholar). Briefly, the purified GST-JMJ protein eluted from the glutathione-agarose beads was incubated with 1 μg of poly(dI-dC) in the reaction buffer (10 mm Tris, pH 8.0, 1 mm dithiothreitol, 1 mm NaH2PO4, 5% glycerol, 50 mm NaCl) for 10 min on ice. After 10 fmol of 32P-labeled double-stranded oligonucleotide (about 30,000 cpm/lane) were added for 20 min at room temperature, the reaction mixture was loaded onto a 5% nondenaturing polyacrylamide gel. The gel was run in 1× protein buffer (25 mm Tris and 192 mm glycine) at 200 V for 2 h and autoradiographed. The Tx2 oligonucleotide sequence, a duplicated binding site of the Bright protein (28Herrscher R.F. Kaplan M.H. Lelsz D.L. Das C. Scheuermann R. Tucker P.W. Genes Dev. 1995; 9: 3067-3082Crossref PubMed Scopus (233) Google Scholar), is 5′-GC(GTTAAATCACAATAAAATATTG)2. The NP3 oligonucleotide, a trimer of the consensus Engrailed binding sequence that is also a binding site of the dri gene product (5Gregory S.L. Kortschak R.D. Kalionis B. Saint R. Mol. Cell Biol. 1996; 16: 792-799Crossref PubMed Scopus (138) Google Scholar), is 5′-GC(TCAATTAAATGA)3. The sense and antisense oligonucleotides were synthesized and annealed, followed by 32P-end-labeling by T4 kinase. CASTing—To select the consensus DNA-binding motif of JMJ in vitro, CASTing was performed as described previously (29Wilson D. Sheng G. Lecuit T. Dostatni N. Desplan C. Genes Dev. 1993; 7: 2120-2134Crossref PubMed Scopus (332) Google Scholar, 30Funk W.D. Wright W.E. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 9484-9488Crossref PubMed Scopus (85) Google Scholar). The GST-JMJ aa 529-798 fusion protein coupled to glutathione-coated agarose beads was prepared. The random sequence library was prepared by synthesizing oligonucleotide pools comprising a random 30-bp sequence flanked by 18 bases of nonrandom sequence 5′-GCGAAGCTTTGACTGAGGN30TTGATGCCGAGGATCCCG-3′ (Biotechnology Center at the University Of Wisconsin). BamHI and HindIII sites are underlined. The PCR primers annealing to the 18 bases of each top and bottom strand were synthesized. The 3′ primer was used for a primer extension reaction to make double-stranded oligonucleotides, which was purified by PAGE. The selection procedure was initiated by mixing 10 μg of the purified double-stranded random oligonucleotides with 50 ng of GST-JMJ coupled to glutathione-agarose beads in buffer A (0.5 ml of 20 mm Tris, pH 8.0, 50 mm KCl, 0.5 mm EDTA, 10% glycerol, 20 μg/ml bovine serum albumin, and 2 μg of poly(dI-dC)). The mixture was rotated at 4 °C for 1 h and then centrifuged. The pellet was washed twice with buffer A, resuspended in 95 μl of PCR buffer, boiled for 3 min, and centrifuged. 90 μl of this supernatant was used as template for a 100-μl PCR. The product after 10, 15, and 20 PCR cycles was examined by running 6.0 μl of the PCR on an agarose gel. Then the PCR fraction that was barely visible on the agarose gel was used for the second round of selection. The second generation began when 20 μl of the above PCR was mixed with the GST-JMJ fusion protein exactly as in the first generation. This step was repeated seven times. After the last PCR amplification step, a fraction of the PCR product was cut with BamHI and HindIII and cloned into the pBluescript vector (Stratagene) and sequenced (Biotechnology Center, University of Wisconsin, Madison, WI). Binding to JMJ was confirmed by GMSA as described above. Site Selection and Amplification Binding (SAAB)—To identify the DNA-binding consensus sequences, the SAAB method was also carried out as described previously (31Chen C.Y. Schwartz R.J. J. Biol. Chem. 1995; 270: 15628-15633Abstract Full Text Full Text PDF PubMed Scopus (272) Google Scholar, 32Amendt B.A. Sutherland L.B. Russo A.F. J. Biol. Chem. 1999; 274: 11635-11642Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar) with minor modifications. 1 μg of GST-JMJ eluted from the beads was incubated with 1.6 × 106 cpm of 32P-labeled random double-stranded probe. This probe was generated by PCR using the primer sets as described above. The mixture was electrophoresed on native polyacrylamide gels. After the shifted band was excised, the DNA was eluted from the gel into 0.5 mm ammonium acetate, 1 mm EDTA, and 0.1% SDS for 5 h at 37 °C. The recovered DNA was amplified by 15 cycles of PCR with the 5′ and 3′ primers that were used in the CASTing experiments. The selection procedure was repeated for four rounds. After the last round of SAAB, PCR-amplified products were directly subcloned into the pGEMT vector (Promega) and sequenced. These oligonucleotides were then subjected to GMSA to confirm their binding to JMJ, followed by the reporter gene assays as described above. Transfection Assay and Immunostaining—Transient transfection assays were performed using the calcium phosphate precipitation method as previously described (27Lee Y. Shioi T. Kasahara H. Jobe S.M. Wiese R.J. Markham B.E. Izumo S. Mol. Cell Biol. 1998; 18: 3120-3129Crossref PubMed Scopus (245) Google Scholar) or LipofectAMINE 2000 (Invitrogen). Mouse fibroblast 10T1/2 cells in 60-mm plates were transfected with 1-2 μg of reporter gene and various amounts of JMJ in mammalian expression vectors and 1 μg of CMV-βgal. Two days after glycerol shock, cell lysates were assayed for luciferase activity according to the manufacturer's recommendation (Promega) using a luminometer. Luciferase activity was normalized to β-galactosidase activity to correct for variations in transfection efficiency. Indirect immunostaining experiments were performed as described (1Lee Y. Song A.J. Baker R. Micales B. Conway S.J. Lyons G.E. Circ. Res. 2000; 86: 932-938Crossref PubMed Scopus (125) Google Scholar). COS cells transfected with JMJ mutants were fixed in 4% formaldehyde for 15 min and incubated with the indicated antibodies followed by incubation with secondary antibodies coupled to fluorescein isothiocyanate (Sigma) or Texas Red (Amersham Biosciences) in phosphate-buffered saline containing 0.1% Nonidet P-40. Fluorescence was visualized using a Zeiss Axiophot microscope and a confocal laser microscope. JMJ Represses the GAL4 Reporter Gene—Although JMJ has been previously identified as a nuclear factor that is developmentally important (1Lee Y. Song A.J. Baker R. Micales B. Conway S.J. Lyons G.E. Circ. Res. 2000; 86: 932-938Crossref PubMed Scopus (125) Google Scholar, 2Takeuchi T. Yamazaki Y. Katoh-Fukui Y. Tsuchiya R. Kondo S. Motoyama J. Higashinakagawa T. Genes Dev. 1995; 9: 1211-1222Crossref PubMed Scopus (261) Google Scholar), the molecular function of JMJ remains largely unknown. The deduced amino acid sequence of jmj revealed a homologous region to the DBD of an ARID factor family. In addition, the jmj homozygous mutant embryos showed defective regulation of heart-specific gene expression (1Lee Y. Song A.J. Baker R. Micales B. Conway S.J. Lyons G.E. Circ. Res. 2000; 86: 932-938Crossref PubMed Scopus (125) Google Scholar), suggesting that JMJ is a transcriptional regulator. To investigate whether JMJ functions as a transcriptional regulator, transient transfection assays were performed using a reporter gene containing five tandem copies of the GAL4-DNA binding sites linked to the heterologous SV40 promoter upstream of luciferase in the pGL3 vector (5×G-pGL3). The reporter gene was cotransfected into 10T1/2 cells with the expression plasmid encoding the GAL4-DNA binding domain fused to the full-length JMJ consisting of 1234 amino acids (G-J1-1234). The reporter gene will be activated or repressed if JMJ contains a transactivation or repressor domain, respectively. Interestingly, G-J1-1234 markedly repressed the reporter gene activity up to 80% in a dose-dependent manner (Fig. 1), indicating that JMJ contains the repressor domain. The N terminus of JMJ consisting of aa 1-528 (G-J1-528) repressed the reporter gene as efficiently as the full-length JMJ in a dose-dependent manner. In contrast, the C terminus containing JMJ chimera (G-J529-1234) neither repressed nor activated the reporter gene, suggesting the N terminus contains a repressor domain. This repression is specific to JMJ, because GAL4-DBD alone did not have any effect on the reporter gene activity. In addition, the positive control plasmid, GAL4-DBD, fused to a VP16 activation domain (G-VP16), activated the reporter gene about 1000-fold as compared with the reporter gene alone. Cotransfection assays using the reporter gene containing a different promoter, pG5Ti-Luc, 5× GAL4-DNA binding site linked to the adenovirus major late promoter, yielded the same results as that of 5×G-pGL3 (data not shown), confirming that repression by JMJ is not dependent on the promoter. Determination of a Transcriptional Repression Domain in JMJ—Since the JMJ N terminus (aa 1-528) mediated repression of the target gene expression, serial deletion mutants were examined for their transcriptional activities to map the repressor domain (Fig. 2) in association with their intracellular location (Fig. 3, A and B). Both G-J1-377 and G-J1-222 markedly repressed the reporter gene as efficiently as the full-length chimera G-J1-1234. In contrast, G-J1-130 completely lost repressor activity, indicating that the repressor domain is located between aa 131 and 222. When this repressor domain was deleted from the full-length JMJ, this mutant (G-J1-130/225-1234) lost most of its repression function, exhibiting about 80% activity as compared with the reporter gene alone. These results indicated that the region between aa 131-222 is necessary. This region between aa 131 and 222, together with the nuclear localization domain (aa 1-130; see Fig. 3A), is sufficient to mediate major transcriptional repressor activity as indicated by maximum repression by G-J1-222. Therefore, this region (aa 131-222) was designated as a transcriptional repressor domain. All of these N terminus-containing chimeras were localized in the nucleus (Fig. 3, A and B).Fig. 3JMJ contains the nuclear localization signal. A, the immunostaining experiments with JMJ antibody. COS cells were transfected with various JMJ mutants in pcDNA3.1/HisB (a-f) or G-JMJ chimeras (g-i). Fixed cells were incubated with anti-JMJ antibodies or anti-Xpress antibodies followed by anti-rabbit IgG conjugated to fluorescein isothiocyanate or anti-mouse IgG-Texas Red, respectively. Subcellular localization of JMJ mutants was examined under epifluorescence or confocal laser microscope. B, the same field of 4′,6-diamidino-2-phenylindole staining, which stains the nuclei. C, characterization of the anti-JMJ antibody. Polyclonal antibody against GST-JMJ fusion protein was raised in rabbit and characterized by immunoprecipitation (lanes 1-4) and Western blot analysis (lanes 5 and 6). For immunoprecipitation, control reticulocyte lysate and in vitro translated and 35S-labeled JMJ and (2 μl each) were loaded onto 7.5% SDS-PAGE (lanes 1 and 2, respectively). JMJ (5 μl of programmed reticulocyte lysate) was incubated with the anti-JMJ antibody raised against GST-JMJ fusion protein or preimmune serum (lanes 3 and 4, respectively). The binding complex was then resolved by SDS-PAGE and autoradiographed. For Western blot analysis, whole COS cell extracts (50 μg/lane) either overexpressing the full-length JMJ 1-1234 (lane 2), or control extract (lane 1), were resolved by SDS-PAGE and transferred onto polyvinylidene difluoride membrane. The membrane was incubated with anti-JMJ antibody (1000-fold dilution), and the protein was visualized by ECL (Amersham Biosciences).View Large Image Figure ViewerDownload Hi-res image Download (PPT) None of the C terminus-containing JMJ chimeras, G-J529-1234 (Fig. 2), G-J529-806, and G-J807-1234 (data not shown), significantly repressed the reporter gene activity. However, these C-terminus-containing G-JMJ chimeras were not localized in the nucleus but rather in the cytoplasm, as visualized with immunostaining experiments (Fig. 3A). Therefore, the lack of transcriptional activity of these C-terminus-containing mutants could be attributed to their inability to localize in the nucleus. To determine whether there is an effector domain in this C terminus, the nuclear localization signal (NLS) domain (aa 1-130; see below) was fused to the C terminus. This NLS domain alone does not have any transcriptional regulatory activity, as indicated by lack of transcriptional activity of G-J1-130 (Fig. 2). These mutants, G-J1-130/529-1234 and G-J1-130/529-792, localized in the nucleus (Fig. 3A) but showed very weak repression if any (Fig. 2). These data indicate that a strong transcriptional repressor domain is located between aa 131 and 222, and perhaps a weak repressor domain exists between aa 529 and 798. The repressor domain deletion mutant G-J1-130/225-1234 was located in the nucleus but lost most of its repressor activity (Fig. 2). These data confirm that the strong repressor domain is situated between aa 131 and 222. Identification of a Nuclear Localization Signal in JMJ—Transcription factors should enter the nucleus where they regulate transcription of target genes. Many nuclear factors are actively transported into the nucleus, which requires an NLS (33Komeili A. O'Shea E.K. Annu. Rev. Genet. 2001; 35: 341-364Crossref PubMed Scopus (54) Google Scholar). When the NLS is deleted or blocked by steric hindrance, nuclear proteins will either accumulate in the cytoplasm or be rapidly degraded. Therefore, it is critical to examine intracellular location of various mutants. JMJ contains several putative NLS similar to that of the SV40 T-antigen at aa 104-110, 147-154, and 433-438. To examine intracellular location of various JMJ mutants, immunostaining experiments were performed on COS cells transfected with JMJ mutants in the mammalian expression vectors (Fig. 3, A and B). Depending on which mutant was used, fixed cells were incubated with either anti-JMJ polyclonal antibody or anti-Xpress antibody (Invitrogen). The JMJ antibodies raised against the peptide were described elsewhere (Fig. 3C) (1Lee Y. Song A.J. Baker R. Micales B. Conway S.J. Lyons G.E. Circ. Res. 2000; 86: 932-938Crossref PubMed Scopus (125) Google Scholar), and the GST-JMJ fusion proteins were characterized as described below. In Fig. 3, column A shows immunostaining of COS cells transfected with various JMJ mutants, and column B shows Hoechst nuclear staining of the same field. Nuclear staining of the wild type JMJ (panel Aa) was observed in COS cells transfected with a JMJ/pcDNA3.1/HisB under the epifluorescence and confocal microscope as previo
Referência(s)