Portal de Periódicos da CAPES

The Large Subunit of Replication Factor C Interacts with the Histone Deacetylase, HDAC1

2002; Elsevier BV; Volume: 277; Issue: 33 Linguagem: Inglês

10.1074/jbc.m200513200

1083-351X

Lisa A. Anderson, Neil D. Perkins,

DNA Repair Mechanisms

Replication factor C (RFC) is a pentameric complex of five distinct subunits that functions as a clamp loader, facilitating the loading of proliferating cell nuclear antigen (PCNA) onto DNA during replication and repair. More recently the large subunit of RFC, RFC (p140), has been found to interact with the retinoblastoma (Rb) tumor suppressor and the CCAAT/enhancer-binding protein α (C/EBPα) transcription factor. We now report that RFC (p140) associates with histone deacetylase activity and interacts with histone deacetylase 1 (HDAC1). This complex is functional and when targeted to promoters as a Gal4 fusion, RFC (p140) is a strong, deacetylase-dependent repressor of transcription. Further analysis revealed that RFC (p140) contains two distinct transcriptional repression domains. Moreover, both of these domains interact separately with HDAC1. Replication factor C (RFC) is a pentameric complex of five distinct subunits that functions as a clamp loader, facilitating the loading of proliferating cell nuclear antigen (PCNA) onto DNA during replication and repair. More recently the large subunit of RFC, RFC (p140), has been found to interact with the retinoblastoma (Rb) tumor suppressor and the CCAAT/enhancer-binding protein α (C/EBPα) transcription factor. We now report that RFC (p140) associates with histone deacetylase activity and interacts with histone deacetylase 1 (HDAC1). This complex is functional and when targeted to promoters as a Gal4 fusion, RFC (p140) is a strong, deacetylase-dependent repressor of transcription. Further analysis revealed that RFC (p140) contains two distinct transcriptional repression domains. Moreover, both of these domains interact separately with HDAC1. histone acetyltransferase replication factor C CCAAT/enhancer-binding protein α histone deacetylase 1 proliferating cell nuclear antigen retinoblastoma protein breast cancer susceptibility gene 1 glutathioneS-transferase: TSA, trichostatin A thymidine kinase chloramphenicol acetyltransferase cAMP-response element-binding protein human embryonic kidney hemagglutinin A common problem faced during the assembly of protein complexes on eukaryotic chromosomal DNA is the inhibitory effect of chromatin structure. For example, the assembly of a gene into chromatin represses transcription by limiting access of the transcriptional machinery to the DNA template (1Tyler J.K. Kadonaga J.T. Cell. 1999; 99: 443-446Abstract Full Text Full Text PDF PubMed Scopus (204) Google Scholar). Transcriptional regulatory proteins therefore recruit chromatin remodeling activities to either facilitate or inhibit access of DNA-binding proteins to their target sequences (2Brown C.E. Lechner T. Howe L. Workman J.L. Trends Biochem. Sci. 2000; 25: 15-19Abstract Full Text Full Text PDF PubMed Scopus (303) Google Scholar, 3Vignali M. Hassan A.H. Neely K.E. Workman J.L. Mol. Cell. Biol. 2000; 20: 1899-1910Crossref PubMed Scopus (588) Google Scholar). It can be predicted, therefore, that proteins associated with DNA replication and repair will also have to be able to regulate chromatin structure. Consistent with this, the largest subunit of the human origin recognition complex, ORC1, and the replication factor minichromosome maintenance protein 2 (MCM2) have both been found to interact with factors containing histone acetyltransferase (HAT)1 activity (4Iizuka M. Stillman B. J. Biol. Chem. 1999; 274: 23027-23034Abstract Full Text Full Text PDF PubMed Scopus (263) Google Scholar, 5Burke T.W. Cook J.G. Asano M. Nevins J.R. J. Biol. Chem. 2001; 276: 15397-15408Abstract Full Text Full Text PDF PubMed Scopus (163) Google Scholar). This observation suggested that ORC1 might fulfill an important function regulating chromatin structure and facilitating the binding of other components of the DNA replication machinery. Acetylation of the amino-terminal tails of core histones alters nucleosome structure and facilitates both the recruitment of DNA-binding proteins and other chromatin remodeling activities (2Brown C.E. Lechner T. Howe L. Workman J.L. Trends Biochem. Sci. 2000; 25: 15-19Abstract Full Text Full Text PDF PubMed Scopus (303) Google Scholar, 6Winston F. Allis C.D. Nat. Struct. Biol. 1999; 6: 601-604Crossref PubMed Scopus (225) Google Scholar, 7Strahl B.D. Allis C.D. Nature. 2000; 403: 41-45Crossref PubMed Scopus (6584) Google Scholar). In the context of transcriptional regulation, HAT activity is therefore generally associated with the stimulation of gene expression, and it can be assumed that it will be similarly associated with the positive regulation of DNA replication and repair (8Chen H.W. Tini M. Evans R.M. Curr. Opin. Cell Biol. 2001; 13: 218-224Crossref PubMed Scopus (151) Google Scholar). Proteins with HAT activity, however, do not always acetylate histones. Many non-histone substrates have been identified, including DNA-binding proteins such as the tumor suppressor p53, components of the basal transcription apparatus, and the architectural transcription factor high mobility group protein I(Y) (HMG I(Y)) (2Brown C.E. Lechner T. Howe L. Workman J.L. Trends Biochem. Sci. 2000; 25: 15-19Abstract Full Text Full Text PDF PubMed Scopus (303) Google Scholar, 9Gu W. Roeder R.G. Cell. 1997; 90: 595-606Abstract Full Text Full Text PDF PubMed Scopus (2168) Google Scholar, 10Imhof A. Yang X.J. Ogryzko V.V. Nakatani Y. Wolffe A.P. Ge H. Curr. Biol. 1997; 7: 689-692Abstract Full Text Full Text PDF PubMed Scopus (533) Google Scholar, 11Munshi N. Agalioti T. Lomvardas S. Merika M. Chen G.Y. Thanos D. Science. 2001; 293: 1133-1136Crossref PubMed Scopus (182) Google Scholar). Interestingly, in this latter case, acetylation of HMG I(Y) on lysine 71 by GCN5 stimulates the ability of the protein to promote transcription from the β-interferon promoter. In contrast, acetylation on lysine 65 by CREB-binding protein (CBP) inhibits HMG I(Y) function and is associated with the termination of β-interferon transcription (11Munshi N. Agalioti T. Lomvardas S. Merika M. Chen G.Y. Thanos D. Science. 2001; 293: 1133-1136Crossref PubMed Scopus (182) Google Scholar). Thus the association of HAT activity with a protein should not automatically be interpreted as an indication of an effect on chromatin structure. In contrast to proteins with HAT activity, histone deacetylases are often associated with transcriptional repression (12Kuo M.H. Allis C.D. Bioessays. 1998; 20: 615-626Crossref PubMed Scopus (1067) Google Scholar, 13Kouzarides T. Curr. Opin. Genet. Dev. 1999; 9: 40-48Crossref PubMed Scopus (587) Google Scholar). However, proteins that participate in DNA replication and repair have also been described associating with histone deacetylase activity. For example, the ataxia telangiectasia-mutated (ATM) kinase and breast cancer susceptibility gene 1 (BRCA1) have been shown to interact with histone deacetylases (14Yarden R.I. Brody L.C. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 4983-4988Crossref PubMed Scopus (285) Google Scholar, 15Kim G.D. Choi Y.H. Dimtchev A. Jeong S.J. Dritschilo A. Jung M. J. Biol. Chem. 1999; 274: 31127-31130Abstract Full Text Full Text PDF PubMed Scopus (105) Google Scholar). Furthermore, the RbAp48 subunit of chromatin assembly factor 1 (CAF1), a protein complex that facilitates the assembly of nucleosomes onto newly replicated DNA (16Verreault A. Kaufman P.D. Kobayashi R. Stillman B. Cell. 1996; 87: 95-104Abstract Full Text Full Text PDF PubMed Scopus (524) Google Scholar, 17Gaillard P.H.L. Martini E.M.D. Kaufman P.D. Stillman B. Moustacchi E. Almouzni G. Cell. 1996; 86: 887-896Abstract Full Text Full Text PDF PubMed Scopus (272) Google Scholar), has also been identified as a protein known to interact with histone deacetylases and associate with the tumor suppressor protein Rb (16Verreault A. Kaufman P.D. Kobayashi R. Stillman B. Cell. 1996; 87: 95-104Abstract Full Text Full Text PDF PubMed Scopus (524) Google Scholar,18Nicolas E. Morales V. Magnaghi-Jaulin L. Harel-Bellan A. Richard-Foy H. Trouche D. J. Biol. Chem. 2000; 275: 9797-9804Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar, 19Tyler J.K. Bulger M. Kamakaka R.T. Kobayashi R. Kadonaga J.T. Mol. Cell. Biol. 1996; 16: 6149-6159Crossref PubMed Scopus (119) Google Scholar). In this study we have investigated the interaction of RFC (p140) with histone deacetylase activity. RFC (p140) is a component of the pentameric replication factor C (RFC) complex (20Mossi R. Hubscher U. Eur. J. Biochem. 1998; 254: 209-216PubMed Google Scholar). RFC functions as a clamp loader and facilitates both the loading and unloading of PCNA at sites of DNA synthesis (20Mossi R. Hubscher U. Eur. J. Biochem. 1998; 254: 209-216PubMed Google Scholar). PCNA functions both to recruit DNA polymerase and CAF1, giving it a role both in stimulating DNA synthesis and in subsequent chromatin assembly (20Mossi R. Hubscher U. Eur. J. Biochem. 1998; 254: 209-216PubMed Google Scholar, 21Shibahara K. Stillman B. Cell. 1999; 96: 575-585Abstract Full Text Full Text PDF PubMed Scopus (539) Google Scholar). It is apparent, however, that RFC (p140) has the potential to regulate other cellular processes and participate as a component of many protein complexes. RFC (p140) has been described in a large protein complex containing both BRCA1 and ATM, termed the BRCA1-associated genome surveillance complex (BASC), which has been postulated to act as a sensor for DNA damage (22Wang Y. Cortez D. Yazdi P. Neff N. Elledge S.J. Qin J. Genes Dev. 2000; 14: 927-939Crossref PubMed Scopus (95) Google Scholar). Moreover, RFC (p140) contains an LXCXE motif and can bind directly to Rb (23Pennaneach V. Salles-Passador I. Munshi A. Brickner H. Regazzoni K. Dick F. Dyson N. Chen T.T. Wang J.Y.J. Fotedar R. Fotedar A. Mol. Cell. 2001; 7: 715-727Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar). Interestingly, and consistent with it having a dynamic regulatory function, RFC (p140) binding to Rb was shown to have a prosurvival function following ultraviolet light stimulation (23Pennaneach V. Salles-Passador I. Munshi A. Brickner H. Regazzoni K. Dick F. Dyson N. Chen T.T. Wang J.Y.J. Fotedar R. Fotedar A. Mol. Cell. 2001; 7: 715-727Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar). Both Rb and BRCA1 have important functions as transcriptional regulators (24Harbour J.W. Dean D.C. Genes Dev. 2000; 14: 2393-2409Crossref PubMed Scopus (957) Google Scholar, 25Chen Y.M. Lee W.H. Chew H.K. J. Cell. Physiol. 1999; 181: 385-392Crossref PubMed Scopus (77) Google Scholar). This raises the possibility that RFC (p140) might also have a dual role as a regulator of gene expression. Consistent with this, RFC (p140) was recently described as a transcriptional coactivator for the bZIP transcription factor C/EBPα (26Hong S.H. Park S.J. Kong H.J. Shuman J.D. Cheong J.H. J. Biol. Chem. 2001; 276: 28098-28105Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar). In this study, we report that RFC (p140) is associated with histone deacetylase activity and can directly bind HDAC1. Moreover, we demonstrate that when directly targeted to a promoter, RFC (p140) functions as a potent, histone deacetylase-dependent repressor of transcription. These results suggest that the cellular function of RFC (p140) is complex, and that in addition to its role as a clamp loader it also has the potential to regulate chromatin structure and repress transcription. The full-length RFC (p140) cDNA was isolated from human foreskin fibroblast cell RNA by reverse transcription polymerase chain reaction (PCR) and inserted into the KpnI site of pCDNA3 or pCGN, which inserted an HA tag at its amino terminus and the NcoI/NotI sites of pVR1012Gal4 (27Snowden A.W. Anderson L.A. Webster G.A. Perkins N.D. Mol. Cell. Biol. 2000; 20: 2676-2686Crossref PubMed Scopus (112) Google Scholar). Fragments of RFC (p140) were isolated by PCR from the original clone and inserted into the NcoI/XhoI sites of pGEX CD1 (to make glutathione S-transferase (GST) fusion proteins), the KpnI site of pCGN (to make HA epitope-tagged proteins), or the SalI/XhoI sites of pVR1012Gal4. RFC (p140) F1Δ1–36 was generated by PCR using the following oligonucleotide primer: CTGCGGCCGCCAGCAAAGAAAGGAATAAAGGAAATC. RFC (p140) F3 LXGXK was generated by overlap extension PCR using the following primers to create the mutation: ctggtgggtcagaagttgggatacagctacgtggaactg and cttctgacccaccagggaagctgtggtggttttgccaacacc. GST fusions were created by subcloning into the NcoI and XhoI sites of PGEX CD1, similar to their wild type equivalents. Gal4 fusions were created by subcloning into the NotI/XbaI sites of pVR1012Gal4. Since these sites differed from the wild type versions of Gal4-RFC (p140) F1 and F3 described above, the wild type versions were remade using the same sites. These latter plasmids were used only in Fig. 3, D and E and were functionally identical to the original Gal4 fusion proteins. All PCR products and mutations were confirmed by sequencing. The pCDNA3 HA-tagged HDAC1 and pING 14A-HDAC1 plasmids were supplied by Dr. Andy Bannister and Professor Tony Kouzarides (University of Cambridge). The reporter plasmids G5 TK-CAT and G0 TK-CAT were supplied by Dr. Stefan Roberts. The polyclonal RFC (p140) antibody was generated using a purified, His-tagged fragment of the protein (amino acids 1–369) expressed in Escherichia coli. The antibody was raised in sheep by the Scottish Antibody Production Unit (SAPU). The HA tag antibody was obtained from Dr. Barbara Spruce (University of Dundee). The Gal4 antibody was supplied by Santa Cruz Biotechnology (cat. no. sc-510). The HDAC1 antibody was supplied by CN Biosciences (cat. no. PC544T). Nuclear extracts, Western blots, and immunoprecipitations were prepared and performed essentially as described previously (28Webster G.A. Perkins N.D. Mol. Cell. Biol. 1999; 19: 3485-3495Crossref PubMed Google Scholar). HeLa cell nuclear protein extracts were obtained from the Computer Cell Culture Center (Belgium). HEK293 cell nuclear extracts were prepared essentially as described (29Dignam J.D. Martin P.L. Shastry B.S. Roeder R.G. Methods Enzymol. 1983; 101: 582-598Crossref PubMed Scopus (745) Google Scholar). Where indicated, addition of ethidium bromide occurred at the point of addition of RFC (p140) antibody to the HeLa nuclear protein extract. GST and GST·RFC (p140) domains F1–F4 were expressed in E. coli and purified using glutathione-agarose as described previously (30Chapman N.R. Perkins N.D. J. Biol. Chem. 2000; 275: 4719-4725Abstract Full Text Full Text PDF PubMed Scopus (105) Google Scholar). GST pull-down assays were also performed as previously described (30Chapman N.R. Perkins N.D. J. Biol. Chem. 2000; 275: 4719-4725Abstract Full Text Full Text PDF PubMed Scopus (105) Google Scholar). To enhance solubility, protein extracts for both wild type and mutant versions of GST·RFC (p140) F3 were made in the presence of 2% sarkosyl. Subsequent purification was then performed in the presence of 1.5% Triton as described previously (31Frangioni J.V. Neel B.G. Anal. Biochem. 1993; 210: 179-187Crossref PubMed Scopus (831) Google Scholar). Where indicated, addition of ethidium bromide occurred at the point GST fusion proteins were mixed with reticulocyte lysate extracts. HEK293 cells were cultured in Dulbecco's Modified Eagles Medium (Sigma) supplemented with 10% fetal bovine serum (Invitrogen), l-glutamine (2 mm), penicillin (100 units/ml), and streptomycin (0.1 mg/ml). Calcium phosphate transfections of HEK293 cells have been described previously (28Webster G.A. Perkins N.D. Mol. Cell. Biol. 1999; 19: 3485-3495Crossref PubMed Google Scholar). HEK293 cells were split 2 h before transfection and harvested after 48 h. Inhibition of enzyme activity by trichostatin A (TSA) (Upstate Biotechnology) was performed by incubation of samples with 300 nm TSA for 24 h before harvesting cells. All CAT assays shown are representative of at least three separate experiments. 3H-acetylated H4 peptide (corresponding to the amino terminal 24 amino acids of bovine histone H4) was supplied by Prof. Tony Kouzarides and has been described previously (32Taunton J. Hassig C.A. Schreiber S.L. Science. 1996; 272: 408-411Crossref PubMed Scopus (1530) Google Scholar). HeLa cell nuclear protein extract (Computer Cell Culture Center, Belgium) was immunoprecipitated with either an anti-RFC (p140) antibody or an IgG control antibody. Immunoprecipitates were incubated with H4 peptide for 3 h at 37 °C. HDAC activity was determined by scintillation counting of the ethyl acetate-soluble [3H]acetic acid. To determine whether RFC (p140) associates with histone deacetylase activity, the endogenous protein was immunoprecipitated from either HeLa or HEK293 cell nuclear protein extracts. Analysis of the immunoprecipitated complex in a histone deacetylase assay demonstrated the presence of deacetylase activity associated with the endogenous RFC (p140) complex (Fig. 1, A and B). No significant deacetylase activity was observed with preimmune serum. To confirm this interaction, RFC (p140) was immunoprecipitated from nuclear extracts prepared from HEK293 cells transfected with both RFC (p140) and (HA) epitope-tagged HDAC1 expression plasmids. Western blot analysis of the immunoprecipitated complex with anti-HA antibody revealed that RFC interacted with HDAC1 (Fig. 1C). To extend this observation, endogenous RFC (p140) was again immunoprecipitated from either HeLa or HEK293 cell nuclear protein extracts, but here the complex was Western blotted with an anti HDAC1 antibody. In both cases it was found that endogenous HDAC1 associated with endogenous RFC (p140) (Fig.1D). Some co-immunoprecipitations have been reported to result from binding to contaminating DNA fragments (33Lai J. Herr W. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 6958-6962Crossref PubMed Scopus (397) Google Scholar). To eliminate this possibly artifactual explanation for these results, this experiment was repeated using HeLa cell nuclear protein extracts but with the inclusion of ethidium bromide, which has been shown to disrupt such nonspecific DNA-protein interactions (33Lai J. Herr W. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 6958-6962Crossref PubMed Scopus (397) Google Scholar). Demonstrating that this is a real and specific interaction, endogenous HDAC1 still co-immunoprecipitated with RFC (p140) to the same extent as the ethidium bromide-free sample (Fig. 1E). We next wished to determine if this interaction was functional in vivo. Recruitment of histone deacetylase activity to a promoter should result in transcriptional repression. It was decided, therefore, through the creation of a Gal4 fusion protein to investigate whether DNA-targeted RFC (p140) could repress transcription. Cotransfection of Gal4-RFC (p140) with Gal4-E1B-CAT, a reporter plasmid with a low basal level of activity, demonstrated that RFC (p140) did not activate transcription (data not shown). Gal4-RFC (p140) was next analyzed using a G5 TK-CAT reporter plasmid, which has five Gal4 sites upstream of the thymidine kinase (TK) promoter. The TK promoter has a significant level of activity that facilitates the analysis of transcriptional repression by Gal4 fusion proteins. Interestingly, strong repression of CAT activity by Gal4-RFC (p140) was observed in HEK293 cells (Fig.2A), whereas Gal4-RFC (p140) did not repress the G0-TK-CAT reporter plasmid, which lacks Gal4 sites (Fig 2B). Many histone deacetylases including HDAC1 are inhibited by TSA (34Yoshida M. Kijima M. Akita M. Beppu T. J. Biol. Chem. 1990; 265: 17174-17179Abstract Full Text PDF PubMed Google Scholar). To confirm that the transcriptional repression observed with RFC (p140) resulted from its association with histone deacetylase activity, the experiment performed above was repeated in the presence of TSA. Significantly, TSA virtually abolished transcriptional repression by RFC (p140) (Fig. 2C). These experiments confirmed that the physical interaction between RFC (p140) and HDAC1 has functional consequences in vivo. We next wished to determine which domain of RFC (p140) interacts with HDAC1. To facilitate this, a series of Gal4 and GST fusions with different fragments of RFC (p140) were constructed (Fig.3A). RFC (p140) F1 (amino acids 1–369) encodes the amino terminus of the protein. RFC (p140) F2 (amino acids 367–493) encodes a domain with homology to DNA ligases and has a BRCT domain also found in BRCA1 and other proteins involved in DNA repair (20Mossi R. Hubscher U. Eur. J. Biochem. 1998; 254: 209-216PubMed Google Scholar). RFC (p140) F3 (amino acids 480–882) is responsible for binding to PCNA and also contains the domain homologous to other RFC subunits, an LXCXE motif required for binding to Rb and a caspase-3 cleavage site (20Mossi R. Hubscher U. Eur. J. Biochem. 1998; 254: 209-216PubMed Google Scholar, 23Pennaneach V. Salles-Passador I. Munshi A. Brickner H. Regazzoni K. Dick F. Dyson N. Chen T.T. Wang J.Y.J. Fotedar R. Fotedar A. Mol. Cell. 2001; 7: 715-727Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar, 35Rheaume E. Cohen L.Y. Uhlmann F. Lazure C. Alam A. Hurwitz J. Sekaly P.P. Denis F. EMBO J. 1997; 16: 6346-6354Crossref PubMed Scopus (78) Google Scholar). RFC (p140) F4 (amino acids 728–1148) contains a domain required for association with other RFC subunits and also has two caspase-3 cleavage sites (20Mossi R. Hubscher U. Eur. J. Biochem. 1998; 254: 209-216PubMed Google Scholar, 35Rheaume E. Cohen L.Y. Uhlmann F. Lazure C. Alam A. Hurwitz J. Sekaly P.P. Denis F. EMBO J. 1997; 16: 6346-6354Crossref PubMed Scopus (78) Google Scholar, 36Montecucco A. Rossi R. Levin D.S. Gary R. Park M.S. Motycka T.A. Ciarrocchi G. Villa A. Biamonti G. Tomkinson A.E. EMBO J. 1998; 17: 3786-3795Crossref PubMed Scopus (172) Google Scholar). Western blot analysis demonstrated that Gal4-RFC (p140) F1–F4 were all expressed close to their expected sizes (57, 30, 60, and 62 kDa, respectively), although Gal4-RFC (p140) F1 ran with an apparent molecular mass slightly higher than expected. Sequencing confirmed that the insert was correct, suggesting that this slightly altered mobility results from a structural feature of the protein fragment, possibly reflecting its high basic content. Gal4-RFC (p140) F1 was also expressed at a slightly higher level relative to the other fragments, and some protein degradation or premature translational termination was observed (Fig. 3B). Initially, we investigated the ability of these RFC (140) fragments to repress transcription using G5 TK-CAT. Strong repression of CAT activity, relative to Gal4 alone, was observed with Gal4-RFC (p140) F1 and F3 (Fig. 3C). In contrast, RFC (p140) F2 and F4 had only a weak inhibitory effect. RFC (p140) therefore contains two domains capable of strongly repressing transcription. The RFC (p140) F1 fragment has been shown to possess a PCNA binding site between amino acids 1 and 24 (36). Deletion of this domain, which had no effect, demonstrated that transcriptional repression was not mediated by PCNA binding to this region (Fig. 3D). Recently, RFC (p140) has been shown to interact with the Rb tumor suppressor through an LXCXE motif present within the F3 region (23Pennaneach V. Salles-Passador I. Munshi A. Brickner H. Regazzoni K. Dick F. Dyson N. Chen T.T. Wang J.Y.J. Fotedar R. Fotedar A. Mol. Cell. 2001; 7: 715-727Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar). Since Rb is a strong repressor of transcription that has also been shown to interact indirectly with histone deacetylases (18Nicolas E. Morales V. Magnaghi-Jaulin L. Harel-Bellan A. Richard-Foy H. Trouche D. J. Biol. Chem. 2000; 275: 9797-9804Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar, 37Laj A. Lee J.M. Yang W.M. DeCaprio J.A. Kaelin W.G. Seto E. Branton P.E. Mol. Cell. Biol. 1999; 19: 6632-6641Crossref PubMed Scopus (142) Google Scholar) it was important that its involvement in repression by the RFC (p140) F3 region be determined. To facilitate this, a mutant form of RFC (p140) in which the LXCXE motif was mutated to LXGXK was generated. This mutation has previously been found to disrupt the interaction between Rb and RFC (p140) (23Pennaneach V. Salles-Passador I. Munshi A. Brickner H. Regazzoni K. Dick F. Dyson N. Chen T.T. Wang J.Y.J. Fotedar R. Fotedar A. Mol. Cell. 2001; 7: 715-727Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar). As a Gal4 fusion however, RFC (p140) F3 LXGXK still repressed transcription to the same extent as wild type RFC (p140) F3, suggesting that association with Rb does not account for the effects seen here (Fig. 3E). We next investigated whether these domains could interact with HDAC1. Consistent with their ability to repress transcription, both Gal4-RFC (p140) F1 and F3 co-immunoprecipitated with HDAC1, whereas no association was seen with Gal4-RFC (p140) F2 and F4 (Fig.4A). Finally, we determined whether this interaction could be seen in vitro in a GST pull-down assay. Bacterially expressed GST·RFC (p140) F1 and F3 bound reticulocyte lysate-translated HDAC1 (Fig.4B). No significant interaction was seen with GST alone or GST·RFC (p140) F2 and F4. None of these proteins interacted with a luciferase control protein. Consistent with the results seen as Gal4 fusion proteins (Fig. 3, D and E) both RFC (p140) F1Δ1–36 and RFC (p140) F3 LXGXK still interacted with HDAC1in vitro (Fig. 4, C and D). Similar to the result seen in Fig. 1E, the inclusion of ethidium bromide failed to disrupt the interaction between RFC (p140) F1 or F3 with HDAC1 in vitro (Fig. 4E). Similar results were seen with DNase I (data not shown). Association between these proteins does not, therefore, result from fortuitous colocalization on DNA present in the assay. Interestingly, the inclusion of ethidium bromide actually stimulated the association between these proteins, suggesting that DNA might have an inhibitory effect on their association. In this report we have demonstrated that the large subunit of replication factor C, RFC (p140), can interact with the histone deacetylase HDAC1 through two distinct domains. It cannot be ruled out that RFC (p140) might interact with other cellular deacetylase activities, because these were not investigated. Although the in vitro association between these proteins suggests a direct interaction (Fig. 4), it cannot be ruled out at this time that an intermediary protein exists that facilitates the association of HDAC1 with RFC (p140). These interactions are functional in vivobecause RFC (p140) can repress transcription in a TSA-sensitive manner when targeted to a promoter. The actual function of this RFC (p140)-associated deacetylase complex in vivo is unclear, however. RFC (p140) has been previously implicated as a transcriptional regulator. For example, it has recently been shown to interact with the bZIP transcription factor C/EBPα and was reported to stimulate transcription (26Hong S.H. Park S.J. Kong H.J. Shuman J.D. Cheong J.H. J. Biol. Chem. 2001; 276: 28098-28105Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar). This observation is not obviously consistent with an interaction between RFC (p140) and histone deacetylases, suggesting that such effects might be context specific or influenced by the cell type in which the study is performed. Significantly, as a Gal4 fusion we did not find that RFC (p140) stimulated gene expression (data not shown), suggesting transcriptional activation is not an intrinsic property of RFC (p140). More consistent with our observations is the report that RFC (p140) can interact with the tumor suppressor Rb (23Pennaneach V. Salles-Passador I. Munshi A. Brickner H. Regazzoni K. Dick F. Dyson N. Chen T.T. Wang J.Y.J. Fotedar R. Fotedar A. Mol. Cell. 2001; 7: 715-727Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar). Rb is an important repressor of E2F transcriptional activity. Moreover, it accomplishes this in part through interacting with proteins that recruit histone deacetylase activity (18Nicolas E. Morales V. Magnaghi-Jaulin L. Harel-Bellan A. Richard-Foy H. Trouche D. J. Biol. Chem. 2000; 275: 9797-9804Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar, 37Laj A. Lee J.M. Yang W.M. DeCaprio J.A. Kaelin W.G. Seto E. Branton P.E. Mol. Cell. Biol. 1999; 19: 6632-6641Crossref PubMed Scopus (142) Google Scholar). It is possible, therefore, that RFC (p140) might participate in this process. Although we show that the RFC (p140) histone deacetylase complex has the potential to repress transcription as a Gal4 fusion, it is perfectly possible that this activity could be involved in a process other than the regulation of gene expression. Replication factor C also participates in the removal of PCNA following replication (20Mossi R. Hubscher U. Eur. J. Biochem. 1998; 254: 209-216PubMed Google Scholar). As such, it might have a function in regulating newly assembled nucleosomes and re-establishing chromatin structure following DNA synthesis. Alternatively, this deacetylase activity might regulate the function of an RFC (p140)-associated protein. The Gal4 fusion experiments were performed principally to allow the activity of RFC (p140)-associated histone deacetylase activity to be analyzed in vivo, and it would be an overinterpretation to assume that it must therefore have a role in transcriptional regulation. In conclusion, there are a number of possible roles that the RFC (p140)-associated deacetylase activity might perform, and these will require further experimentation to fully understand. Nonetheless, this study adds to a growing body of evidence that the function of RFC (p140) is more complex and dynamic that originally thought. It also highlights the similarities and possible dual functions of proteins associated with replication and transcription. Given the complexity of chromatin-remodeling activities associated with transcriptional regulators, it can be anticipated that many similar interactions remain to be discovered. We thank Donna Bumpass, Neil Chapman, David Gregory, Sonia Rocha, Kevin Roche, Alison Sleigh, Niall McTavish, Barbara Spruce, Tom Owen-Hughes, Stefan Roberts, Joost Zomerdijk, Julian Blow, Andy Bannister, and Tony Kouzarides for help during this project.