A Novel RNA Polymerase II C-terminal Domain Phosphatase That Preferentially Dephosphorylates Serine 5
2003; Elsevier BV; Volume: 278; Issue: 28 Linguagem: Inglês
10.1074/jbc.m301791200
ISSN1083-351X
AutoresMichele Yeo, Patrick S. Lin, Michael Dahmus, Gordon N. Gill,
Tópico(s)RNA modifications and cancer
ResumoThe transcription and processing of pre-mRNA in eukaryotic cells are regulated in part by reversible phosphorylation of the C-terminal domain of the largest RNA polymerase (RNAP) II subunit. The CTD phosphatase, FCP1, catalyzes the dephosphorylation of RNAP II and is thought to play a major role in polymerase recycling. This study describes a family of small CTD phosphatases (SCPs) that preferentially catalyze the dephosphorylation of Ser5 within the consensus repeat. The preferred substrate for SCP1 is RNAP II phosphorylated by TFIIH. Like FCP1, the activity of SCP1 is enhanced by the RAP74 subunit of TFIIF. Expression of SCP1 inhibits activated transcription from a number of promoters, whereas a phosphatase-inactive mutant of SCP1 enhances transcription. Accordingly, SCP1 may play a role in the regulation of gene expression, possibly by controlling the transition from initiation/capping to processive transcript elongation. The transcription and processing of pre-mRNA in eukaryotic cells are regulated in part by reversible phosphorylation of the C-terminal domain of the largest RNA polymerase (RNAP) II subunit. The CTD phosphatase, FCP1, catalyzes the dephosphorylation of RNAP II and is thought to play a major role in polymerase recycling. This study describes a family of small CTD phosphatases (SCPs) that preferentially catalyze the dephosphorylation of Ser5 within the consensus repeat. The preferred substrate for SCP1 is RNAP II phosphorylated by TFIIH. Like FCP1, the activity of SCP1 is enhanced by the RAP74 subunit of TFIIF. Expression of SCP1 inhibits activated transcription from a number of promoters, whereas a phosphatase-inactive mutant of SCP1 enhances transcription. Accordingly, SCP1 may play a role in the regulation of gene expression, possibly by controlling the transition from initiation/capping to processive transcript elongation. The largest subunit of RNA polymerase (RNAP) 1The abbreviations used are: RNAP, RNA polymerase; CKII, casein kinase II; CTD, C-terminal domain of RNA polymerase II; FCP1, TFIIF-associated CTD phosphatase; PNθP, para-nitrophenylphosphate; P-TEFb, positive transcription elongation factor b; RAP74, RNA polymerase II-associated protein of 74 kDa (larger subunit of TFIIF); SCP, small CTD phosphatase; GST, glutathione S-transferase; MAPK, mitogen-activated protein kinase; ERK, extracellular signal-regulated kinase; DTT, dithiothreitol; PPARγ, peroxisome proliferator-activated receptor γ; rCTDo, recombinant CTDo; TK, thymidine kinase. II contains a C-terminal domain (CTD) composed of multiple repeats of the consensus sequence Tyr1-Ser2-Pro3-Thr4-Ser5-Pro6-Ser7. The progression of RNAP II through the transcription cycle is regulated by both the state of CTD phosphorylation and the specific site of phosphorylation within the consensus repeat (1Dahmus M.E. J. Biol. Chem. 1996; 271: 19009-19012Abstract Full Text Full Text PDF PubMed Scopus (464) Google Scholar, 2Majello B. Napolitano G. Front. Biosci. 2001; 6: 1358-1368Crossref PubMed Google Scholar). The emerging overview of this process is that unphosphorylated RNAP II, designated RNAP IIA, enters the preinitiation complex, where phosphorylation of Ser5 is catalyzed by TFIIH (which contains Cdk7/cyclin H subunits) concomitant with transcript initiation (3Lu H. Zawel L. Fisher L. Egly J.M. Reinberg D. Nature. 1992; 358: 641-645Crossref PubMed Scopus (330) Google Scholar, 4Hengartner C.J. Myer V.E. Liao S.M. Wilson C.J. Koh S.S. Young R.A. Mol. Cell. 1998; 2: 43-53Abstract Full Text Full Text PDF PubMed Scopus (336) Google Scholar). This generates the phosphorylated form of RNAP II, designated RNAP IIO. Ser5 phosphorylation facilitates the recruitment of the 7-methyl G capping enzyme complex (5Cho E.-J. Takagi T. Moore C.R. Buratowski S. Genes Dev. 1997; 11: 3319-3326Crossref PubMed Scopus (373) Google Scholar, 6McCracken S. Fong N. Rosonina E. Yankulov K. Brothers G. Siderovski D. Hessel A. Foster S. Shuman S. Bentley D.L. Genes Dev. 1997; 11: 3306-3318Crossref PubMed Scopus (433) Google Scholar, 7Yue Z. Maldonado E. Pillutla R. Cho H. Reinberg D. Shatkin A.J. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 12898-12903Crossref PubMed Scopus (196) Google Scholar, 8Ho C.K. Shuman S. Mol. Cell. 1999; 3: 405-411Abstract Full Text Full Text PDF PubMed Scopus (267) Google Scholar, 9Rodriguez C.R. Cho E.-J. Keogh M.C. Moore C.L. Greenleaf A.L. Buratowski S. Mol. Cell. Biol. 2000; 20: 104-112Crossref PubMed Scopus (161) Google Scholar). Phosphorylation of Ser2 is catalyzed by the cyclin-dependent kinase P-TEFb (which contains Cdk9/cyclin T subunits) (10Zhou M. Halanski M.A. Radonovich M.F. Kashanchi F. Peng J. Price D.H. Brady J.N. Mol. Cell. Biol. 2000; 20: 5077-5086Crossref PubMed Scopus (220) Google Scholar, 11Shim E.Y. Walker A.K. Shi Y. Blackwell T.K. Genes Dev. 2002; 16: 2135-2146Crossref PubMed Scopus (165) Google Scholar). Although Ser5 phosphorylation precedes Ser2 phosphorylation (12Komarnitsky P. Cho E.-J. Buratowski S. Genes Dev. 2000; 14: 2452-2460Crossref PubMed Scopus (802) Google Scholar), it is not clear if the dephosphorylation of Ser5 precedes Ser2 phosphorylation. During transcript elongation in yeast, there is extensive turnover of Ser2 phosphates mediated by FCP1 and Ctk1, the putative P-TEFb homolog (13Cho E.-J. Kobor M.S. Kim M. Greenblatt J. Buratowski S. Genes Dev. 2001; 15: 3319-3329Crossref PubMed Scopus (340) Google Scholar). Finally, dephosphorylation of Ser2 by the FCP1 phosphatase regenerates RNAP IIA, thereby completing the cycle (14Cho H. Kim T.-K. Mancebo H. Lane W.S. Flores O. Reinberg D. Genes Dev. 1999; 13: 1540-1552Crossref PubMed Scopus (172) Google Scholar). FCP1 is a class C phosphatase containing a BRCT domain that is required for interaction with RNAP II and dephosphorylation of the CTD (15Archambault J. Chambers R.S. Kobor M.S. Ho Y. Cartier M. Bolotin D. Andrews B. Kane C.M. Greenblatt J. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 14300-14305Crossref PubMed Scopus (129) Google Scholar, 16Kobor M.S. Simon L.D. Omichinski J. Zhong G. Archambault J. Greenblatt J. Mol. Cell. Biol. 2000; 20: 7438-7449Crossref PubMed Scopus (63) Google Scholar). FCP1 interacts with and is stimulated by RAP74, the larger subunit of TFIIF (16Kobor M.S. Simon L.D. Omichinski J. Zhong G. Archambault J. Greenblatt J. Mol. Cell. Biol. 2000; 20: 7438-7449Crossref PubMed Scopus (63) Google Scholar, 17Chambers R.S. Wang B.Q. Burton Z.F. Dahmus M.E. J. Biol. Chem. 1995; 270: 14962-14969Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar). Class C phosphatases are resistant to inhibitors that block other classes of Ser/Thr phosphatases and bind Mg2+ or Mn2+ in the binuclear metal center of the catalytic site (18Das A.K. Helps N.R. Cohen P.T.W. Barford D. EMBO J. 1996; 15: 6798-6809Crossref PubMed Scopus (391) Google Scholar, 19Barford D. Das A.K. Egloff M.-P. Annu. Rev. Biophys. Biomol. Struct. 1998; 27: 133-164Crossref PubMed Scopus (571) Google Scholar). The ψψψDXDX(T/V)ψψ motif (where ψ represents a hydrophobic residue) present in the FCP1 homology domain characterizes a subfamily of class C phosphatases, with both Asp residues being essential for activity (20Collet J.-F. Stroobant V. Pirard M. Delpierre G. Van Schaftingen E. J. Biol. Chem. 1998; 273: 14107-14112Abstract Full Text Full Text PDF PubMed Scopus (227) Google Scholar, 21Kobor M.S. Archambault J. Lester W. Holstege F.C.P. Gileadi O. Jansma D.B. Jennings E.G. Kouyoumdjian F. Davidson A.R. Young R.A. Greenblatt J. Mol. Cell. 1999; 4: 55-62Abstract Full Text Full Text PDF PubMed Scopus (169) Google Scholar). Synthetic lethality is observed between mutant FCP1 and reduced levels of RNAP II in Saccharomyces cerevisiae and Schizosaccharomyces pombe, indicating that FCP1 is an essential gene (21Kobor M.S. Archambault J. Lester W. Holstege F.C.P. Gileadi O. Jansma D.B. Jennings E.G. Kouyoumdjian F. Davidson A.R. Young R.A. Greenblatt J. Mol. Cell. 1999; 4: 55-62Abstract Full Text Full Text PDF PubMed Scopus (169) Google Scholar, 22Kimura M. Suzuki H. Ishihama A. Mol. Cell. Biol. 2002; 22: 1577-1588Crossref PubMed Scopus (84) Google Scholar). It remains uncertain whether the activity of yeast FCP1 accounts for the dephosphorylation in vivo of both Ser5 and Ser2 and whether this is the sole activity that catalyzes CTD dephosphorylation. Mutations in FCP1 lead to increased phosphorylation of Ser2, suggesting that it functions in vivo in the dephosphorylation of Ser2 (13Cho E.-J. Kobor M.S. Kim M. Greenblatt J. Buratowski S. Genes Dev. 2001; 15: 3319-3329Crossref PubMed Scopus (340) Google Scholar). Yeast FCP1 appears more specific for Ser2 phosphate when synthetic peptides are used as substrate (23Hausmann S. Shuman S. J. Biol. Chem. 2002; 277: 21213-21220Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar). However, mammalian FCP1 dephosphorylates both Ser2 and Ser5 in vitro in the context of native RNAP II (24Lin P.S. Dubois M.-F. Dahmus M.E. J. Biol. Chem. 2002; 277: 45949-45956Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar). Furthermore, modifications such as that catalyzed by the Ess1/Pin1 peptidyl prolyl isomerase may alter the activity and specificity of FCP1 (25Morris D.P. Phatnani H.P. Greenleaf A.L. J. Biol. Chem. 1999; 274: 31583-31587Abstract Full Text Full Text PDF PubMed Scopus (129) Google Scholar, 26Wu X. Wilcox C.B. Devasahayam G. Hackett R.L. Arévalo-Rodríguez M. Cardenas M.E. Heitman J. Hanes S.D. EMBO J. 2000; 19: 3727-3738Crossref PubMed Scopus (142) Google Scholar, 27Kops O. Zhou Z. Lu K.P. FEBS Lett. 2002; 513: 305-311Crossref PubMed Scopus (53) Google Scholar). Given the importance of CTD phosphorylation in gene expression (28Hirose Y. Manley J.L. Genes Dev. 2000; 14: 1415-1429Crossref PubMed Google Scholar), it is essential to know whether additional CTD phosphatases exist and, if so, their specificity and the mechanisms by which they are targeted to RNAP II at discrete stages of the transcription cycle. Although FCP1 is the only reported CTD phosphatase, examination of the data bases reveals additional genes that consist principally of a domain with homology to the CTD phosphatase domain of FCP1. Three closely related human genes encoding small proteins with CTD phosphatase domain homology, but lacking a BRCT domain, have been identified. In the present study, we show that a gene located on chromosome 2 encodes a nuclear CTD phosphatase. This protein preferentially dephosphorylates Ser5 within the CTD of RNAP II and is stimulated by RAP74. Expression of this small CTD phosphatase (SCP1) inhibits activated transcription from a variety of promoter-reporter gene constructs, whereas expression of a mutant lacking phosphatase activity enhances transcription. This newly identified small CTD phosphatase appears to play an important role in the regulation of RNAP II transcription. Materials—SCP1 and SCP2 were obtained as expressed sequence tag clones from Invitrogen. The full-length cDNA for SCP1 (261 aa; AL520011), the cDNA encoding the variant of SCP1 (214 aa; BE300370), and SCP2 (AL520463) were subcloned into EcoRI-XhoI sites of pGEX4T-1 and pcDNA3FLAG vectors by PCR. All clones were sequenced, and the corrected sequences have been deposited in the NCBI data base. The D96E, D98N mutant of SCP1 261 and the corresponding mutant of SCP1 214, D48E, D50N, were generated by QuikChange (Stratagene). All constructs were verified by sequencing. GST fusions were purified by glutathione-Sepharose chromatography, and SCP1 261 was generated by cleavage at the thrombin site encoded in the vector. Recombinant FCP1 was expressed and purified as described previously (24Lin P.S. Dubois M.-F. Dahmus M.E. J. Biol. Chem. 2002; 277: 45949-45956Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar). Human recombinant casein kinase II (CKII) and mouse recombinant MAPK2/ERK2 were obtained from Upstate Biotechnology, Inc. (Lake Placid, NY). Human CTDK1/CTDK2 were purified as described by Payne and Dahmus (29Payne J.M. Dahmus M.E. J. Biol. Chem. 1993; 268: 80-87Abstract Full Text PDF PubMed Google Scholar). Human TFIIH was generously provided by Jean-Marc Egly (30Gerard M. Fischer L. Monocollin V. Chipoulet J.M. Chambon P. Egly J.M. J. Biol. Chem. 1991; 266: 20940-20945Abstract Full Text PDF PubMed Google Scholar). Human P-TEFb was partially purified from HeLa S-100 extract by chromatography on heparin-Sepharose (Amersham Biosciences), DEAE 15HR (Millipore Corp.), and HiTrap S and phenyl-Superose (both from Amersham Biosciences). P-TEFb was dialyzed against 25 mm Hepes, pH 7.9, 20% glycerol, 25 mm KCl, 0.1 mm EDTA, 1 mm DTT, 1 mm phenylmethylsulfonyl fluoride. Human recombinant Cdc2 kinase was purchased from New England Biolabs. Rabbits were immunized with GST-SCP1 214. Anti-GST antibodies were adsorbed to GST-agarose and rabbit anti-SCP1 IgG (antibody 6703) was prepared by ammonium sulfate fractionation and protein G-Sepharose chromatography. RNAP II antibodies (8WG16, H5, and H14) were obtained from Covance. Preparation and Purification of 32P-Labeled RNAP IIO Isozymes and [32P]-GST-CTDo—Calf thymus RNAP IIA was purified by the method of Hodo and Blatti (31Hodo H.G. Blatti S.P. Biochemistry. 1977; 16: 2334-2343Crossref PubMed Scopus (129) Google Scholar) with modifications as described by Kang and Dahmus (32Kang M.E. Dahmus M.E. J. Biol. Chem. 1993; 268: 25033-25040Abstract Full Text PDF PubMed Google Scholar). Specific isozymes of 32P-labeled RNAP IIO were prepared by phosphorylation at the most C-terminal serine (CKII site) in the largest subunit of purified RNAP IIA with recombinant CKII and [γ-32P]ATP (17Chambers R.S. Wang B.Q. Burton Z.F. Dahmus M.E. J. Biol. Chem. 1995; 270: 14962-14969Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar, 33Chesnut J.D. Stephens J.H. Dahmus M.E. J. Biol. Chem. 1992; 267: 10500-10506Abstract Full Text PDF PubMed Google Scholar), followed by CTD phosphorylation in the presence of 2 mm ATP with either purified CTDK1/CTDK2, TFIIH, P-TEFb, recombinant MAPK2/ERK2, or recombinant Cdc2 kinase. The RNAP IIO isozymes were individually purified over a DE53 column with a step elution of 500 mm KCl (29Payne J.M. Dahmus M.E. J. Biol. Chem. 1993; 268: 80-87Abstract Full Text PDF PubMed Google Scholar). Because only the most C-terminal serine is labeled with 32P and lies outside the consensus repeat, dephosphorylation by CTD phosphatase results in an electrophoretic mobility shift in SDS-PAGE of subunit IIo to the position of subunit IIa without the loss of label. GST-CTD was prepared and purified as previously described (34Kang M.E. Dahmus M.E. J. Biol. Chem. 1995; 270: 23390-23397Abstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar). 32P-Labeled GST-CTDo was subsequently prepared from GST-CTDa by CKII followed by MAPK2/ERK2. GST-CTDo was purified over a glutathione-agarose column with a step elution of 15 mm glutathione. Phosphatase Assays—PNθP reaction mixtures (200 μl) containing 50 mm Tris acetate, pH 5.5, 10 mm MgCl2, 0.5 mm DTT, 10% glycerol, 20 mm PNθP, and recombinant proteins were incubated at 30 °C for 1 h. The reactions were quenched by adding 800 μl of 0.25 n NaOH. Release of para-nitrophenyl was determined by measuring A 410. N-terminal biotinylated CTD phosphopeptides, composed of four tandem repeats YSPTSPS and containing phosphoserine at position 2 or 5, were synthesized (Alpha Diagnostics, San Antonio, TX). Phosphatase reaction mixtures (50 μl) containing 50 mm Tris acetate, pH 5.5, 10 mm MgCl2, 0.5 mm DTT, 10% glycerol, 25 μm phosphopeptide, and wild type or mutant SCP1 were incubated for 60 min at 37 °C. The reactions were quenched by adding 0.5 ml of malachite green (Biomol). Phosphate release was measured at A 620 and quantified relative to a phosphate standard curve. CTD phosphatase assays utilizing RNAP IIO and GST-CTDo as substrate were performed as described previously (35Chambers R.S. Dahmus M.E. J. Biol. Chem. 1994; 269: 26243-26248Abstract Full Text PDF PubMed Google Scholar) with minor modifications. Reactions were performed in 20 μl of CTD phosphatase buffer (50 mm Tris-HCl, pH 7.9, 10 mm MgCl2, 20% glycerol, 0.025% Tween 80, 0.1 mm EDTA, 5 mm DTT) in the presence of 20 mm KCl. Each reaction contained specified amounts of GST-CTDo and/or RNAP IIO and was carried out in the presence of 7 pmol of RAP74. Reactions were initiated by the addition of FCP1 or SCP1 and incubated at 30 °C for 30 min. Assays were terminated by the addition of 5× Laemmli buffer, and RNAP II subunits and GST-CTD were resolved on a 5% SDS-PAGE gel. The gel images were developed by autoradiography and scanned by an Amersham Biosciences Image Scanner Storm 860 in the phosphor screen mode. Data were quantitatively analyzed by ImageQuant software. Tissue Culture and Transfections—Human 293, COS-7, and CV1 cells were grown at 37 °C in Dulbecco's modified Eagle's medium supplemented with 10% normal calf serum (Invitrogen). Subconfluent cells were transfected in six-well tissue culture dishes using Effectene (Qiagen) according to the manufacturer's instructions. Reporter and activator plasmids (100 ng each) and FLAG SCP1 (80 ng) or its mutant were used per well. For T3 and PPARγ transfections, 20 ng of retinoid X receptor plasmid was also added. The amounts of ligands used were as follows: 100 nm T3, 1 μm PPARγ 609843 (Ligand Pharmaceuticals), and 100 nm dexamethasone. LMX1 and E47 expression plasmids (100 ng) were cotransfected with 100 ng of the rat insulin promoter-luciferase reporter construct with the indicated concentrations of SCP1 and phosphatase-deficient SCP1 expression plasmids. The total amounts of transfected DNA were kept constant by the addition of empty vector. Cells were harvested 48 h after transfections, and cellular extracts were assayed for luciferase activity using the luciferase assay system (Promega) according to the manufacturer's instructions. Immunofluorescence—Cells grown on coverslips were fixed in 2% paraformaldehyde, neutralized, and blocked using 2.5% fetal calf serum/phosphate-buffered saline. Rabbit polyclonal IgG 6703 was used at 1:100 dilution, followed by goat anti-rabbit IgG H+L chains conjugated to Alexa Fluor 488 (1:250) (Molecular Probes, Inc., Eugene, OR). Mouse anti EEA1 was used at 1:1000, followed by goat anti-mouse IgG conjugated to Alexa Fluor 594 (1:250). Omission of primary antibodies was used as a negative control. The coverslips were viewed using the Zeiss Axiophot, which is equipped with a Hamamatsu Orca ER firewire camera that runs on Improvision Openlab 3.0.9 software. Immunoprecipitations—For immunoprecipitation experiments, 75% confluent COS-7 cells from a 10-cm dish were harvested in lysis buffer (phosphate-buffered saline containing 1% Nonidet P-40, 1 mm DTT, and protease inhibitors). Lysates were incubated with 20 μl of Sepharose-conjugated anti-SCP1 (6703) IgG at 4 °C for 6 h. Beads were washed with phosphate-buffered saline, and the complexes were evaluated by Western blotting using specific anti-RNAP II antibodies. Rabbit anti-SNX1 antibody was used as control IgG. SCP1 Is an RNAP II CTD Phosphatase—The alignment of three human proteins that are closely related to one another and have homology to the phosphatase domain of human FCP1 is shown in Fig. 1A. All contain the signature motif ψψψDXDX(T/V)ψψ. SCP1, located on chromosome 2q35, was initially designated nuclear LIM-interacting factor in the genome entry (36Marquet S. Lepage P. Hudson T.J. Musser J.M. Schurr E. Mamm. Genome. 2000; 11: 755-762Crossref PubMed Scopus (25) Google Scholar). The full-length 261-aa protein is encoded by seven exons; a shorter NH2-terminal splice version of 214 aa is present in expressed sequence tag data bases. SCP1 has ∼20% homology to human FCP1 in the phosphatase domain, whereas the three SCP proteins are >90% homologous in this region. SCP2/OS4 located on chromosome 12q13 was co-amplified with Cdk4 in sarcomas (37Su Y.A. Lee M.M. Hutter C.M. Meltzer P.S. Oncogene. 1997; 15: 1289-1294Crossref PubMed Scopus (42) Google Scholar), and SCP3/HYA22 located on chromosome 3q22 was part of a large chromosome deletion in a lung carcinoma cell line (38Ishikawa S. Kai M. Tamari M. Takei Y. Takeuchi K. Bandou H. Yamane Y. Ogawa M. Nakamura Y. DNA Res. 1997; 4: 35-43Crossref PubMed Scopus (48) Google Scholar). These represent a subset of proteins with putative CTD phosphatase-like catalytic domains found in plants, yeast, nematodes, and arthropods. The Drosophila and Anopheles genomes each contain a single highly conserved SCP ortholog. The SCP proteins lack the BRCT domain present in FCP1 (Fig. 1B). To determine whether SCP1 has phosphatase activity, the protein was expressed as a GST fusion and both SCP1 261 and SCP1 214 were assayed using PNθP as substrate. As reported for FCP1 from S. pombe (23Hausmann S. Shuman S. J. Biol. Chem. 2002; 277: 21213-21220Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar) utilizing PNθP as substrate, the pH optimum for SCP1 phosphatase activity is near 5 (Fig. 2A). Phosphatase activity was Mg2+-dependent and resistant to the phosphatase inhibitors okadaic acid and microcystin (Fig. 2B). Ca2+ could not substitute for Mg2+. Mutations of Asp96 to Glu (D96E) had little to no effect on phosphatase activity (data not shown), whereas mutating Asp98 to Asn (D98N) in conjunction with the D96E mutation completely abolished phosphatase activity (Fig. 2B). SCP1 is thus a class 2C phosphatase whose activity is dependent on acidic residues in the conserved DXD motif. SCP2 exhibited similar phosphatase activity (Fig. 2B). To determine whether GST-SCP1 214 has CTD phosphatase activity, GST-CTDo and RNAP IIO were utilized as substrates, and the activity of SCP1 was compared directly with that of FCP1. Recombinant CTDo (rCTDo) and RNAP IIO utilized as substrate in these experiments were prepared by the phosphorylation of purified GST-CTDa or RNAP IIA with casein kinase II (CKII) in the presence of [γ-32P]ATP, followed by phosphorylation with MAPK2/ERK2 in the presence of excess unlabeled ATP. MAPK2/ERK2 was used in these initial experiments, because it phosphorylates both GST-CTDa and RNAP IIA with comparable efficiency (24Lin P.S. Dubois M.-F. Dahmus M.E. J. Biol. Chem. 2002; 277: 45949-45956Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar). As expected, FCP1 efficiently converts RNAP IIO to RNAP IIA in a processive manner (Fig. 2C, lanes 7–12) (24Lin P.S. Dubois M.-F. Dahmus M.E. J. Biol. Chem. 2002; 277: 45949-45956Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar, 35Chambers R.S. Dahmus M.E. J. Biol. Chem. 1994; 269: 26243-26248Abstract Full Text PDF PubMed Google Scholar). Even high concentrations of FCP1 did not result in measurable dephosphorylation of rCTDo (Fig. 2C, lanes 1–6) consistent with the idea that a docking site on RNAP II is required for activity (17Chambers R.S. Wang B.Q. Burton Z.F. Dahmus M.E. J. Biol. Chem. 1995; 270: 14962-14969Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar). GST-SCP1 214 catalyzed the dephosphorylation of both RNAP IIO and GST-CTDo with comparable efficiency (Fig. 2D). In contrast to FCP1, the SCP1-catalyzed dephosphorylation of RNAP IIO appears nonprocessive in that a number of phosphorylated intermediates are visible in SDS-PAGE. SCP1 is specific for dephosphorylation of the consensus repeat in that the phosphate at the CKII site is not removed. As shown below, the pattern of dephosphorylation varies depending on the CTD kinase used in the preparation of RNAP IIO. Mutant SCP1 (D96E,D98N) lacked activity on either substrate (data not shown). SCP1 is thus a CTD phosphatase that acts on both RNAP IIO and rCTDo. SCP2 exhibits comparable CTD phosphatase activity when RNAP IIO is utilized as substrate (see Fig. 4). SCP1 Preferentially Dephosphorylates Ser5 of the CTD Heptad Repeat—To determine the specificity of SCP1 with respect to its ability to dephosphorylate specific positions within the consensus repeat, RNAP IIO isozymes were prepared in vitro by the phosphorylation of RNAP IIA with CTD kinases of known specificity. TFIIH, P-TEFb, and MAPK2/ERK2 preferentially phosphorylate Ser5 when synthetic peptides serve as substrate (39Trigon S. Serizawa H. Conaway J.W. Conaway R.C. Jackson S.P. Morange M. J. Biol. Chem. 1998; 273: 6769-6775Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar, 40Ramanathan Y. Rajpara S.M. Reza S.M. Lees E. Shuman S. Mathews M.B. Pe'ery T. J. Biol. Chem. 2001; 276: 10913-10920Abstract Full Text Full Text PDF PubMed Scopus (108) Google Scholar), whereas Cdc2 kinase phosphorylates Ser2 and Ser5 (41Zhang J. Corden J.L. J. Biol. Chem. 1991; 266: 2290-2296Abstract Full Text PDF PubMed Google Scholar). Although the specificity appears relaxed when RNAP II serves as substrate, RNAP IIO prepared with Cdc2 kinase is clearly distinct from RNAP IIO generated by other CTD kinases (24Lin P.S. Dubois M.-F. Dahmus M.E. J. Biol. Chem. 2002; 277: 45949-45956Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar). Results presented in Fig. 3A indicate that RNAP IIO, prepared by the phosphorylation of RNAP IIA with distinct CTD kinases, exhibit a differential sensitivity to dephosphorylation with SCP1. SCP1 most efficiently dephosphorylates RNAP IIO generated by TFIIH and was unable to dephosphorylate RNAP IIO prepared with Cdc2 kinase. SCP1 was also unable to dephosphorylate RNAP IIO generated by Abl tyrosine kinase (data not shown). The dephosphorylation of RNAP IIO isozymes prepared with P-TEFb, MAPK2/ERK2, and CTDK1/CTDK2 occurred at a reduced rate relative to that of RNAP IIO prepared with TFIIH. Furthermore, whereas the dephosphorylation reaction appears processive for RNAP IIO prepared by TFIIH, it is clearly nonprocessive for RNAP IIO generated by MAPK2/ERK2. In contrast, FCP1 shows no preference for RNAP IIO generated by TFIIH and efficiently dephosphorylates RNAP IIO generated by Cdc2 kinase (24Lin P.S. Dubois M.-F. Dahmus M.E. J. Biol. Chem. 2002; 277: 45949-45956Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar). These results suggest SCP1 differs from FCP1 in substrate specificity, showing relative preference for the dephosphorylation of Ser5 in the heptad repeat. To investigate the relative reactivity of SCP1 for Ser2 and Ser5, a synthetic 28-aa peptide containing four heptad repeats phosphorylated exclusively on Ser2 or on Ser5 was dephosphorylated in the presence of increasing amounts of SCP1. As shown in Fig. 3B, SCP1 preferentially dephosphorylates the Ser5 phosphopeptide compared with the Ser2 phosphopeptide. This substrate specificity contrasts to that reported for FCP1 from S. pombe, which preferentially dephosphorylate the Ser2 phosphopeptide (23Hausmann S. Shuman S. J. Biol. Chem. 2002; 277: 21213-21220Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar). Mammalian FCP1, within a comparable concentration range, did not act on either phosphopeptide (data not shown). These results using synthetic phosphopeptide substrates confirm that SCP1 preferentially dephosphorylates Ser5 phosphate of the CTD. Effect of RAP74 on the Activity of SCP1—The RAP74 subunit of TFIIF stimulates CTD phosphatase activity of FCP1 (17Chambers R.S. Wang B.Q. Burton Z.F. Dahmus M.E. J. Biol. Chem. 1995; 270: 14962-14969Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar). Furthermore, the domains of FCP1 that bind RAP74 are required for FCP1-dependent viability in S. cerevisiae (15Archambault J. Chambers R.S. Kobor M.S. Ho Y. Cartier M. Bolotin D. Andrews B. Kane C.M. Greenblatt J. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 14300-14305Crossref PubMed Scopus (129) Google Scholar). Therefore, it was of interest to determine whether RAP74 can also influence the activity of SCP. CTD phosphatase activity was measured at low enzyme concentrations to more readily detect stimulatory effects of RAP74. As shown in Fig. 4, RAP74 shifted the dose-response curve for SCP1-catalyzed dephosphorylation of RNAP IIO to an ∼10-fold lower concentration. The CTD phosphatase activity of the GST fusion forms of SCP1 261, SCP1 214, and SCP2 were also enhanced by RAP74. The presence of comparable concentrations of bovine serum albumin does not stimulate the CTD phosphatase activity of SCP1. In support of the conclusion that RAP74 stimulates the activity of SCPs, RAP74 bound directly to GST-SCP1 but not to GST (data not shown). The binding and stimulatory effects of RAP74 suggest that TFIIF is important for optimal CTD phosphatase activity for both FCP1 and SCP1. SCP1 Is Located in the Nucleus Associated with RNAP II— Although SCP1 lacks an obvious nuclear localization sequence, it is found in the nucleus. Immunofluorescence microscopy using a rabbit polyclonal anti-SCP1 antibody demonstrated nuclear localization of endogenous SCP1 in COS-7 cells (Fig. 5B). Co-staining with 4′,6-diamidino-2-phenylindole for nuclear identification and with the early endosomal marker EEA1 for cellular detail confirmed the specific localization of SCP1 in nuclei (Fig. 5, A and B). Co-immunoprecipitation was used to assess the association of SCP1 with RNAP II. Sepharose-immobilized anti-SCP1 IgG 6703 was used to immunoisolate SCP1 from COS-7 cells. Immunoisolates were resolved by SDS-PAGE and blotted with anti-RNAP II antibodies. As shown in Fig. 5C, RNAP II was present in SCP1 immunoprecipitates, indicating that SCP1 and RNAP II either interact directly or are in the same macromolecular complex. To determine whether SCP1 preferentially associated with either Ser2 or Ser5 phosphorylated RNAP IIO, lysates were prepared in the presence of EDTA, to inhibit phosphatase activity. SCP1 immunoprecipitates were then blotted with monoclonal antibodies specific for Ser2 phosphate (H5) and Ser5 phosphate (H14). Both forms of RNAP IIO were present in COS-7 cell lysates. Ser5 phosphate-enriched RNAP IIO appear
Referência(s)