• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of molcellbPermissionsJournals.ASM.orgJournalMCB ArticleJournal InfoAuthorsReviewers
Mol Cell Biol. Oct 1999; 19(10): 6632–6641.
PMCID: PMC84642

RBP1 Recruits Both Histone Deacetylase-Dependent and -Independent Repression Activities to Retinoblastoma Family Proteins

Abstract

Retinoblastoma (RB) tumor suppressor family proteins block cell proliferation in part by repressing certain E2F-specific promoters. Both histone deacetylase (HDAC)-dependent and -independent repression activities are associated with the RB “pocket.” The mechanism by which these two repression functions occupy the pocket is unknown. A known RB-binding protein, RBP1, was previously found by our group to be an active corepressor which, if overexpressed, represses E2F-mediated transcription via its association with the pocket. We show here that RBP1 contains two repression domains, one of which binds all three known HDACs and represses them in an HDAC-dependent manner while the other domain functions independently of the HDACs. Thus, RB family members repress transcription by recruiting RBP1 to the pocket. RBP1, in turn, serves as a bridging molecule to recruit HDACs and, in addition, provides a second HDAC-independent repression function.

The retinoblastoma (RB) tumor suppressor protein pRB plays a critical role in the control of cell proliferation. In addition, loss of the Rb gene is known to contribute to the establishment of a variety of cancers. pRB and the related proteins p130 and p107 control cell cycle progression through interactions with the E2F family of transcription factors (reviewed in reference 11). Such interactions regulate transcription by mechanisms requiring the “pocket” domain of RB family members (1, 3, 5, 18, 32, 39, 40, 46). One role of the pocket is to interact with and mask the transcriptional activation domain of E2F; however, this mechanism does not explain the repression of E2F-dependent promoters in which E2F binding sites act as negative elements that, if deleted, result in relief from repression (7, 11, 20, 28). The pocket can interact simultaneously with both E2F and certain cellular and viral proteins that bind by utilizing a conserved Leu-X-Cys-X-Glu (LXCXE) sequence (10, 14, 25, 38, 41). Many of these cellular LXCXE-containing proteins have been shown to be transcriptional repressors, including RBP1 (24), HBP1 (37), RIZ (4), RBP2 (unpublished data), and histone deacetylase 1 (HDAC1) and HDAC2 (19, 42, 43). Recently, it has been proposed that the pockets of the pRB-E2F, p130-E2F, and p107-E2F complexes actively repress E2F-dependent transcription by two mechanisms, one involving the recruitment of HDAC1 (and possibly HDAC2) (2, 16, 26, 27) and the other independent of HDACs (26, 30). HDACs are multiprotein complexes in which the human homologues of the yeast RPD3 protein, HDAC1, HDAC2, and HDAC3, constitute catalytic subunits (8, 13, 35, 42, 43). It is widely believed that histone deacetylation condenses chromatin structure, thereby shutting down transcription (reviewed in references 17, 19, 31, and 34). The recruitment of HDAC1 and HDAC2 to the pocket may therefore account for active repression by RB family members through deacetylation at the promoter level, and in fact, enzymatically active forms of HDACs have been detected both in vitro and in vivo in association with both the pocket and E2F (2, 26, 27). Magnaghi-Jaulin et al. (27) and Ferreira et al. (16) have shown that a region within HDAC1 containing an IXCXE motif is important for interactions with RB family members because its deletion resulted in a reduction in binding. Thus, it was proposed that a direct physical interaction between the degenerate IXCXE motif of HDAC1 and the pocket of RB family members occurs in a manner analogous to interactions with LXCXE-containing viral transforming proteins. One problem with this interpretation concerns the fact that in addition to the IXCXE motif, the deletion mutants used in these studies eliminated additional HDAC1-coding sequences. Second, the in vitro binding assays used by these groups utilized HDAC1 translated in vitro with reticulocyte lysates. Thus, it is possible that other factors might function in the interaction, perhaps even serving as linkers for HDAC1. We present evidence herein that a known RB pocket-binding protein, RBP1, links all three HDACs to RB proteins and, in addition, provides a second HDAC-independent repression function.

MATERIALS AND METHODS

Cell culture and transfection.

Human lung carcinoma H1299 cells were grown in Dulbecco’s modified Eagle medium containing 10% fetal calf serum. 293 cells, 293T cells (a variant of 293 cells that expresses simian virus 40 [SV40] large T antigen), and Chinese hamster ovary (CHO) cells (ATCC CCL-61) were grown in α-minimal essential medium supplemented with 10% fetal calf serum. Transfections for binding studies were carried out with Lipofectamine reagents (NEN Life Science), and transfections for chloramphenicol acetyltransferase (CAT) assays were done by the calcium phosphate precipitation method with the pGEM plasmid as the carrier DNA, as described previously (24).

Antibodies.

A rabbit polyclonal antibody raised against HDAC2 was described previously (23). Antibodies raised by E. Seto were prepared against peptides corresponding to the unique carboxy termini of HDAC1 (EEKPEAKGVKEEVKLA) and HDAC3 (NEFYDGDHDNDKESDVEI), which had been coupled to keyhole limpet hemocyanin and injected separately into New Zealand White rabbits. The resulting antibodies were immunoaffinity purified on peptide columns. LY11 and LY32 monoclonal antibodies raised by J. A. DeCaprio and W. G. Kaelin specifically against RBP1 were described previously (24). Anti-pRB antibody G3-245 was purchased from Pharmingen. Antibodies against p130 (C-20), p107 (C-18), and the Gal4 DNA-binding domain (Gal4DBD) (RK5C1) were purchased from Santa Cruz. Antihemagglutinin (anti-HA) antibody HA.11 was purchased from Babco, and anti-Flag antibody M2 was from Sigma.

Plasmids.

A mammalian expression plasmid expressing the small pocket of pRB as a fusion product with Gal4DBD, termed Gal4-pRB(pocket), was provided by Tony Kouzarides (2). Flag-HDAC1, -HDAC2, and -HDAC3 and Gal4-HDAC1, -HDAC2, and -HDAC3 constructs have been described elsewhere (42, 43). Gal4-VP16 was provided by Arnie Berk (44). Constructs expressing Gal4-RBP1, RBP1-HA, Gal4-RBP1dl-LXCXE, and RBP1dl-LXCXE-HA mutants which lack the LXCXE pocket-binding motif were described previously (24). The G5TKCAT reporter construct has been described elsewhere (36, 45). The G5MLPCAT reporter was provided by Doug Dean (26). The E2F1-luc reporter construct has been described elsewhere (24). Mutants of Gal4DBD-RBP1 were generated as follows. Gal4-dlR1 mutants were generated with two specific primers close to the 3′ end of the region corresponding to R1 and the end of the RBP1-coding sequence. PCR was done to generate fragments with unique restriction sites (Bsp1107I and HindIII) which were subcloned into digested RBP1 constructs, in which all the RBP1-coding sequences corresponding to between the beginning of R1 and the end of the protein had been removed with the same restriction enzymes. All carboxy-terminal-deletion mutants of Gal4-dlR1 were generated from fragments produced by restriction enzyme digestion of the Gal4-dlR1 plasmid DNA, and then these were subcloned into a modified pcDNA3 (Invitrogen) construct containing stop codons inserted 3′ of the multicloning cassette. Gal4-R2 truncation mutants were constructed by subcloning the restriction enzyme-digested fragments of the Gal4-RBP1-coding region that correspond to residues 1311 to 1404, 1314 to 1404, or 1263 to 1404 into pSG424. Mammalian expression plasmids encoding the pocket of pRB and the inactive pRB pocket mutant mRB(C706F) were described previously (22).

CAT, β-galactosidase, and luciferase assays.

Transcriptional analyses involving CAT, β-galactosidase, and luciferase assays were performed as described previously (24, 36).

Binding assays.

One microgram each of cDNAs encoding Gal4-pRB(pocket), RBP1, or RBP1 mutants was introduced with Lipofectamine (NEN Life Science) along with 1 μg each of those encoding Flag-tagged HDAC1, HDAC2, or HDAC3 into H1299 or 293T cells. In some experiments, 1 μg each of cDNAs encoding HA-tagged RBP1 or RBP1 mutants was introduced with Lipofectamine along with 1 μg each of those encoding Gal4DBD-fused HDAC1, HDAC2, or HDAC3 into H1299 or 293T cells. Cells were harvested 40 h posttransfection and lysed with low-stringency buffer (24). Cell extracts were diluted to 150 mM KCl in a 1-ml volume and precleared with protein G-Sepharose (Pharmacia) for 2 h. Precleared extracts were incubated with 1 μg of Gal4DBD antibody RK5C1 (Santa Cruz) and 30 μl of a 50% slurry of protein G-Sepharose for at least 12 h. Immunoprecipitated Gal4-tagged protein complexes were washed six times with lysis buffer and eluted by boiling in 2× sample buffer. Eluted proteins were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) with a 10% polyacrylamide gel, and proteins were then transferred to polyvinylidene difluoride membranes (Millipore) and were probed with either anti-HA (HA.11 [Babco]) or anti-Flag (M2; Sigma) monoclonal antibody and then with horseradish peroxidase-conjugated goat anti-mouse (λ light chain-specific) secondary antibody (PharMingen). Binding was detected by Enhanced Luminol Reagent (NEN Life Science).

RESULTS

Endogenous interactions between RB family members and HDACs.

Coimmunoprecipitation experiments were conducted to determine the ability of pRB to interact with various HDAC enzymes in vivo. Rabbit polyclonal antibodies that specifically recognize HDAC2 (23) or HDAC1 and HDAC3 (see Materials and Methods) were used under low-stringency conditions to immunoprecipitate endogenous HDAC species from extracts of H1299 cells. In each case, significant amounts of corresponding HDAC proteins were immunoprecipitated (data not shown). The presence of pRB in these complexes was determined by Western blot analysis with the G3-245 monoclonal antibody, which recognizes both hyperphosphorylated and hypophosphorylated forms of pRB. Figure Figure1A1A shows that the hypophosphorylated form of pRB coprecipitated with HDAC1, as reported by others (26). Interestingly, significant amounts of hypophosphorylated pRB were also detected in association with HDAC2 and HDAC3. These interactions are specific, as pRB was not detected in immunoprecipitates prepared from the same cell extracts with antibody to CREB-binding protein, a histone acetyltransferase (data not shown). These results suggested that the active form of pRB is able to recruit all three forms of the HDAC.

FIG. 1
Endogenous interactions of pRB (A) and p107 (B) with HDACs. (A) Immunoprecipitations were done in lysates from either H1299 or 293T cells. Plates (100 mm) of both cell types were lysed in low-stringency buffer. Cell extracts were incubated with 1 μg ...

Previous studies showed that HDAC1 also interacts with the pRB-related family members p107 and p130 (16, 21, 33). Thus, immunoprecipitates from H1299 cells containing various HDACs were immunoblotted with C-18 polyclonal antibody, which recognizes up to three different phosphorylated forms of p107 in various cell lines. Figure Figure1B1B shows that p107, its underphosphorylated forms in particular, also coprecipitated with all the endogenous HDAC species, especially HDAC1. We attempted a similar experiment with p130, but as H1299 cells die upon serum starvation and p130 is expressed largely at growth arrest, we were unable to detect sufficient quantities. Significant amounts of p130 were present in asynchronized H1299 cells, but these p130 species were mostly hyperphosphorylated and did not appear to associate with HDACs at significant levels (data not shown).

Adenovirus E1A protein and SV40 large T antigen associate with the pRB pocket via LXCXE-binding motifs (12) and have been found to disrupt interactions between pRB and HDAC1 (2, 26, 27). We therefore conducted a parallel series of binding studies in 293T cells, which express high levels of both adenovirus type 5 (Ad5) E1A proteins and SV40 large T antigen. Figure Figure1A1A shows that no interactions were apparent between pRB and any of the HDAC enzymes, including HDAC3. Figure Figure1B1B shows that a similar effect was apparent with p107. Thus, in both cases, binding of all three HDAC enzymes seemed to require the pocket region targeted by DNA tumor virus proteins.

The small pocket interacts with different HDACs.

It has been proposed that HDAC1 utilizes a degenerate IXCXE motif to interact with the small pocket (residues 379 to 792) of pRB (16, 27). To directly determine if this region is involved in the binding of all of the HDACs, studies were carried out with extracts from H1299 cells cotransfected with plasmid DNAs expressing Gal4-pRB(pocket) and Flag-tagged versions of HDAC1, HDAC2, or HDAC3. Following immunoprecipitation with an antibody against the Gal4DBD, precipitates were resolved by SDS-PAGE, and after transfer, the presence of HDAC1 to HDAC3 was analyzed by Western blotting with anti-Flag antibody. Figure Figure2A2A shows that all three HDACs, which were detected at similar levels in whole-cell extracts, associated with the small pocket of pRB in vivo. Anti-Flag antibody recognized three Flag-HDAC1 species, but only the slowest-migrating form was evident in Gal4-pRB(pocket) precipitates, suggesting that the faster-migrating species may be degradation products. Flag-HDAC3 was detected as two closely migrating species that were both associated with the Gal4-pRB(pocket). These interactions were highly specific; a cDNA expressing the Gal4DBD linked to VP16, a known transcriptional activator, did not associate with any of the HDACs (Fig. (Fig.2A)2A) even though Gal4-pRB(pocket) and Gal4-VP16 were expressed at comparable levels (data not shown). We believe that these results, as well as those described above for antibody against CREB-binding protein, demonstrate clearly the specificity of interactions involving HDACs and rule out the possibility that the interactions we observed were nonspecific.

FIG. 2FIG. 2
In vivo interaction of the small pocket of pRB with HDACs. (A) cDNAs encoding the Gal4DBD fused with the pocket of pRB were overexpressed with those encoding Flag-tagged HDAC1, HDAC2, or HDAC3 in H1299 cells. RK5C1 (anti-Gal4DBD) antibody (Santa Cruz) ...

Figure Figure2B2B shows that in a similar experiment with 293T cells, none of the HDAC proteins were present in immunoprecipitates containing Gal4-pRB(pocket). Identical results were also obtained with 293 cells, which express only Ad5 E1A proteins (Fig. (Fig.2C).2C). Figure Figure2D2D shows that addition prior to immunoprecipitation of increasing amounts of in vitro-synthesized E1A protein fused to glutathione S-transferase (GST-E1A) (6) to extracts from H1299 cells expressing Gal4-pRB(pocket) and Flag-HDAC1 caused a decrease in the binding of HDAC1 to the small pocket of pRB. Figure Figure2D2D also demonstrates that pRB coimmunoprecipitated with E1A protein but not with GST. Similar results were also obtained with Flag-HDAC2 and Flag-HDAC3 (data not shown). Thus, the small pocket of pRB is important for interactions not only with HDAC1 but also with HDAC2 and HDAC3. Disruption of such binding by the E1A protein or T antigen also implied that the associations of HDACs may be mediated by LXCXE-like interactions. Although HDAC1 and HDAC2 contain IXCXE motifs, the mechanism of binding of HDAC3 was uncertain, as HDAC3 lacks such sequences. Of further interest, no interaction between the pocket of pRB and HDAC1 (2) or HDAC2 (32a) was observed in studies involving the yeast two-hybrid method. As all three human HDACs are highly homologous, it is possible that the interactions of all three with the pocket may be indirect and may involve an additional LXCXE-containing protein as a linker.

One possible candidate for a linker is RBP1, a known nuclear pRB-binding phosphoprotein that interacts with the pocket via an LXCXE motif (9, 15, 22). In previous studies, it was shown that RBP1 associates with both pRB-E2F and p130-E2F complexes following serum starvation and, if overexpressed, induces both growth arrest and repression of E2F-dependent transcription (24). In addition, in studies involving RBP1 fused to the Gal4DBD, RBP1 was shown to be an active repressor containing an isolable repression domain (24), termed R1.

RBP1 interacts with HDACs in vivo.

The ability of RBP1 to associate with HDACs in vivo was tested. A binding experiment similar to that described above for Gal4-pRB(pocket) was performed with RBP1 fused to the Gal4DBD. Figure Figure3A3A shows that when expressed in H1299 cells, Gal4-RBP1, but not Gal4-VP16, interacted with all three HDACs. Again, only the slowest-migrating Flag-HDAC1 species was evident, and Flag-HDAC3 was present as a doublet in Gal4-RBP1 precipitates.

FIG. 3
In vivo interaction of RBP1 with HDACs. Binding studies similar to those described for Fig. Fig.2A2A and B were performed with H1299 cells (A) and 293T cells (B) by using Gal4-RBP1 instead of Gal4-pRB. (C) Binding studies similar to those described ...

Figure Figure3B3B shows that in 293T cells expressing E1A products and SV40 large T antigen, both HDAC1 and HDAC3 associated with Gal4-RBP1, indicating that these interactions were unaffected by high levels of these viral pocket-binding proteins. The binding of HDAC2 was not highly evident under these conditions. Whereas Flag-HDAC1 and Flag-HDAC3 expression relied on the cytomegalovirus promoter (32a) that yielded high levels of these products, the expression of Flag-HDAC2 was achieved in these experiments with a hybrid SV40-human T-cell leukemia virus type 1 promoter that produced smaller amounts of product in 293T and 293 cells (Fig. (Fig.2C).2C). To reexamine RBP1-HDAC binding under conditions in which all three HDACs were expressed at high levels, constructs in which all three HDACs were expressed in 293T cells along with HA-tagged RBP1 as Gal4DBD-HDAC fusion products under the control of the SV40 early promoter were prepared. Figure Figure3C3C shows that all the HDACs were expressed at comparable levels, as determined by the immunoblotting of whole-cell extracts with anti-Gal4DBD antibody. Similarly, all cells expressed comparable amounts of RBP1, as detected with anti-HA antibody. Figure Figure3C3C also shows that interactions between all three HDACs and HA-RBP1 were evident following the immunoblotting of individual HDAC immunoprecipitates with anti-HA antibody. No such binding was observed with Gal4 alone or with Gal4-VP16. Figure Figure3E3E shows that interactions between RBP1 and endogenous HDAC enzymes could also be detected in H1299 cells. With immunoprecipitates prepared with the same polyclonal antibodies against individual HDACs as employed for Fig. Fig.1,1, RBP1 was found to be in association with all three HDACs, as detected by immunoblotting with LY32 monoclonal antibody, raised against human RBP1 and found previously to interact specifically with this polypeptide (8a). Interactions with HDAC1 clearly occurred at much higher levels than those with HDAC2 and HDAC3, implying that RBP1-HDAC1 complexes are the predominant species in H1299 cells. This observation may suggest that RBP1 could be a component of the HDAC1 core complex. If this is the case, much of the endogenous RBP1-HDAC1 complex could be targeted to sites other than RB family members, as the levels of RBP1 detected in association with these proteins were much lower than those observed with HDAC1. Figure Figure3E3E also shows that RBP1 associates with all members of the RB family, and examination of pertinent lanes in Fig. Fig.11 indicated that this association occurred preferentially with hypophosphorylated forms of pRB (Fig. (Fig.1A)1A) and p107 (Fig. (Fig.1B).1B). In addition, Fig. Fig.3E3E shows that such interactions of RBP1 with pRB, p107, and p130 were disrupted in 293T cells, as expected. Taken together, these results suggest that interactions between RBP1 and RB family members are truly pocket dependent, as they were disrupted by T antigen and E1A pocket-binding proteins. On the other hand, interactions between RBP1 and HDACs are not at all sensitive to disruption by these viral pocket-binding proteins, suggesting that HDACs cannot be recruited to RBP1 via endogenous RB family members present in 293T cells. Instead, RBP1 could mediate the association between RB family members and all three HDAC enzymes.

It is unlikely that the interactions of HDACs with RBP1 occur indirectly through the recruitment of RB family members by RBP1. In 293T and 293 cells, we failed to observe interactions either between the pocket of pRB and these enzymes (Fig. (Fig.2B2B and C) or between RBP1 and RB family members (Fig. (Fig.11 and and3E).3E). In addition, we studied interactions between the HDACs and an RBP1 mutant, RBP1dl-LXCXE, that fails to interact with pRB because of the removal of the conserved LXCXE pocket-binding motif by an internal deletion (9, 24). Figure Figure3D3D shows that in 293T cells, both Gal4-RBP1dl-LXCXE and RBP1dl-LXCXE-HA mutants interacted with Flag-tagged or Gal4DBD-tagged HDACs. Similar results were also obtained with human H1299 cells (data not shown). Thus, HDACs appear to interact with RBP1 in a region apart from the pocket-binding motif, further supporting a role for RBP1 in bridging interactions between HDACs and pRB-E2F complexes.

RBP1 contains two independent transcriptional repression domains.

Previous studies showed the existence of both pRB-E2F-RBP1 and p130-E2F-RBP1 complexes in growth-arrested cells that correlated with the ability of RB family members to actively repress E2F-dependent transcription (24). These studies also used Gal4-RBP1 fusion products to map a repression domain, R1, between residues 388 and 599 of RBP1, comprising an ARID sequence and a region predicted to have an α-helical structure. Figure Figure4A4A shows that Gal4-RBP1 is able to repress the expression of CAT under the control of the Gal4 minimal herpesvirus thymidine kinase promoter. Similar repression was also seen with Gal4-R1 that contains only the R1 repression domain of RBP1. Curiously, when R1 was deleted (Gal4-RBP1dl-R1), this mutant RBP1 product still repressed CAT expression at high levels (Fig. (Fig.4A),4A), suggesting that a second repression domain may exist towards the carboxy terminus of RBP1. We therefore generated a series of in-frame carboxy-terminal-deletion mutants that also lacked R1 (Fig. (Fig.4B)4B) and found that all, including Gal4-dl-R1-93C, which lacked only 93 residues at the carboxy terminus, failed to repress CAT expression (Fig. (Fig.4A).4A). All mutants used in the experiments illustrated in Fig. Fig.44 were shown to be expressed at similarly high levels (data not shown). As these data suggested that the second repression domain likely lies at the most carboxy-terminal portion, three additional constructs that contained only various amounts of the carboxy terminus of RBP1 linked to the Gal4DBD were generated. Figure Figure4A4A shows that all three constructs, Gal4-R2(1311-C), Gal4-R2(1314-C), and Gal4-R2(1263-C), repressed CAT expression. Thus, a second RBP1 repression domain, R2, exists between residues 1314 and 1404. Furthermore, repression by neither R1 nor R2 relies on the LXCXE pocket-binding motif when RBP1 is tethered to DNA by a heterologous DNA-binding domain, like the Gal4DBD.

FIG. 4FIG. 4
Mapping of transcriptional repression domains in RBP1 and effect of TSA on RBP1 repression activity. (A) Repression by RBP1 mutants. CAT assays were performed with CHO cells, as described previously (24), with G5TKCAT as the reporter. (B) Illustration ...

Transcriptional repression by RBP1 is both dependent and independent of HDAC activity.

Previous studies by Luo et al. (26) suggested that the ability of the pocket to actively repress transcription relies on both HDAC-dependent and -independent mechanisms. These authors also showed that only a subset of promoters repressed by the pocket of pRB was sensitive to the specific HDAC inhibitor trichostatin A (TSA). Among these, the only Gal4-dependent promoter/reporter construct found to be repressed by the pocket and to be sensitive to TSA is G5MLPCAT, which contains the adenovirus major late promoter. Figure Figure4C4C shows that both Gal4-RBP1 and Gal4-pRB(pocket) repressed the G5MLPCAT reporter, as did both Gal4-R1 and Gal4-R2, which contain only R1 and R2, respectively. Figure Figure4C4C also shows that following the treatment of transfected cells with 330 nM TSA for 24 h prior to harvesting, repression by both Gal4-pRB(pocket) and Gal4-RBP1 was partially relieved; however, whereas drug treatment completely abolished repression by Gal4-R2, it had no effect on that by Gal4-R1. Thus, it appears that repression by R2 depends on HDACs, whereas that by R1 does not. We also noted that whereas Gal4-HDAC1 repression activity is completely relieved by TSA (Fig. (Fig.4C),4C), repression by Gal4-Ad5-E1B-55K, an adenoviral repressor known to block p53-dependent transactivation (36, 45), at this promoter is not affected by TSA (data not shown), suggesting that the adenovirus major late promoter can be subject to both HDAC-dependent and -independent repression. These results therefore strengthen the possibility that RBP1 plays an important role in repression by pRB, as TSA only partially relieves repression by the pRB pocket. Interestingly, TSA had little effect on repression by Gal4-RBP1 or Gal4-R2 with either the G5TKCAT or G5SV40CAT promoter (data not shown), as determined previously with Gal4-pRB(pocket) by Luo et al. (26). Studies were extended from the synthetic G5MLPCAT construct to the E2F-dependent promoter regulating E2F-1 expression (E2F1-luc). Figure Figure4D4D shows that both RBP1 and the pocket of pRB repressed the expression of luciferase from the E2F1-luc reporter, whereas RBP1 lacking the LXCXE pocket-binding motif or a pRB point mutant [mRB(C706F)] did not. Members of our group had demonstrated previously that mutation of the E2F binding site in this reporter ablated repression by RBP1 (24). Addition of TSA partially relieved repression by both RBP1 and the pRB pocket. These results indicated that both the RBP1 HDAC-dependent and -independent repression activities can repress this E2F-dependent promoter via interactions with RB family members, and thus interactions with RBP1 could provide both types of repression activities attributed to the pocket of RB family members (26).

Only one repression domain of RBP1 interacts with HDACs.

We tested the ability of individual R1 and R2 transcriptional repression domains to interact with HDACs. Figure Figure55 shows results of coimmunoprecipitation experiments using 293T cells expression Gal4-RBP1 constructs and Flag-HDAC3, in which RBP1 was immunoprecipitated with anti-Gal4DBD antibodies and HDAC3 binding was detected by immunoblotting with anti-Flag antibody. No HDAC3 binding was observed with either all of R1 or just the ARID portion of R1, but such binding was clearly evident with both R2 constructs and with RBP1 lacking R1. Binding was eliminated or greatly reduced with RBP1 lacking R2. Similar results were also obtained with Flag-HDAC1 and Flag-HDAC2 (data not shown).

FIG. 5
Mapping of specific binding of HDAC3 to RBP1. Binding studies similar to those described for Fig. Fig.3B3B were done with 293T cells by using either Gal4-R1, Gal4-R2(1314-C), Gal4-R2(1263-C), Gal4 alone, Gal4-RBP1 (wild type [WT]), ...

DISCUSSION

The present study reports for the first time that the pocket of pRB, and possibly other RB family members, interacts with all three cloned human HDACs. Furthermore, the RB-binding protein RBP1 also binds these enzymes, but unlike pRB, the pocket-binding E1A protein and large T antigen do not affect interactions between RBP1 and any of the HDACs. It is currently believed that HDAC1 or HDAC2 interacts directly with the pocket via degenerate IXCXE motifs, but this model does not account for the interactions of HDAC3, which lacks such sequences. Our data indicated that the binding of HDAC3 to RB family members also requires the pocket, suggesting that, in this case at least, such interactions may be indirect and rely on a pocket-binding protein, such as RBP1. The previous observation that RBP1 is present in pRB-E2F and p130-E2F complexes following serum starvation and the effects of its overexpression on inducing growth arrest and the repression of E2F-dependent transcription (24) suggest that RBP1 may be functionally important in the repression of E2F-dependent transcription by linking HDACs to the pockets of RB family proteins. Figure Figure66 shows a model in which the repression of E2F-dependent transcription by pRB and p130 at growth arrest or by p107 in G1 results from the binding of RBP1 at the pocket, thus introducing not only HDACs via interactions with the R2 domain but also a second R1 repression domain that functions by another mechanism.

FIG. 6
Model of repression of E2F-dependent promoters by RB family members and RBP1.

At present, it is not known if HDACs bind directly to the RBP1 or if they interact indirectly via an additional R2-binding protein. Previous work with anti-RBP1 LY11 antibodies indicated the presence of several proteins that coprecipitate with RBP1, including pRB and p130 (24). The most prominent species was a protein of about 48 kDa. HDAC1 copurifies with RBAP48, which, along with RBAP46, plays a role in targeting HDAC1 to histones (47). It is possible that both RBP1 and either RBAP48 or RBAP46 could coexist in a single HDAC complex. RBAP48 binds to the so-called extended pocket, including a carboxy-terminal portion of pRB, and thus might play a role as a linker for HDAC1; however, it lacks the LXCXE-binding motif and thus is not targeted to the small pocket. In addition, no interactions between the pRB pocket and HDAC1 (2) or HDAC2 (32a) have been detected by the yeast two-hybrid system, even though yeast cells contain high levels of MSI1, a protein that is highly homologous to RBAP48 and RBAP46 (47). The present results strongly suggest that RBP1 is responsible for bridging the pocket of RB family members to HDAC complexes to repress a diversity of E2F-dependent promoters. RBP1 therefore appears to represent a major component of the growth-regulatory machinery controlled by RB family members.

ACKNOWLEDGMENTS

We thank Tony Kouzarides for Gal4-pRB(pocket) and for helpful discussions; Xiang-Jiao Yang and Brian Kennedy for critical review of the manuscript; Arnie Berk for Gal4-VP16, pSG424, and G5TKCAT; and Doug Dean and Don Ayer for G5MLPCAT and G5SV40CAT. We also thank Dennis Paquette for the construction of the Gal4-RBP1dl-R1 mutant.

This work was supported through grants from the National Cancer Institute of Canada and the Medical Research Council of Canada.

REFERENCES

1. Adnane J, Shao Z, Robbins P D. The retinoblastoma susceptibility gene product represses transcription when directly bound to the promoter. J Biol Chem. 1995;270:8837–8843. [PubMed]
2. Brehm A, Miska E A, McCance D J, Reid J L, Bannister A J, Kouzarides T. Retinoblastoma protein recruits histone deacetylase to repress transcription. Nature. 1998;391:597–601. [PubMed]
3. Bremner R, Cohen B L, Sopta M, Hamel P A, Ingles C J, Gallie B L, Phillips R A. Direct transcriptional repression by pRB and its reversal by specific cyclins. Mol Cell Biol. 1995;15:3256–3265. [PMC free article] [PubMed]
4. Buyse I M, Shao G, Huang S. The retinoblastoma protein binds to RIZ, a zinc-finger protein that shares an epitope with the adenovirus E1A protein. Proc Natl Acad Sci USA. 1995;92:4467–4471. [PMC free article] [PubMed]
5. Chow K N B, Dean D C. Domains A and B in the Rb pocket interact to form a transcriptional repressor motif. Mol Cell Biol. 1996;16:4862–4868. [PMC free article] [PubMed]
6. Corbeil H B, Branton P E. Functional importance of complex formation between the retinoblastoma tumor suppressor family and adenovirus E1A proteins as determined by mutational analysis of E1A conserved region 2. J Virol. 1994;68:6697–6709. [PMC free article] [PubMed]
7. Dalton S. Cell cycle regulation of the human cdc2 gene. EMBO J. 1992;11:1797–1804. [PMC free article] [PubMed]
8. Dangond F, Hafler D A, Tong J K, Randall J, Kojima R, Utku N, Gullans S R. Differential display cloning of a novel human histone deacetylase (HDAC3) cDNA from PHA-activated immune cells. Biochem Biophys Res Commun. 1998;242:648–652. [PubMed]
8a. DeCaprio, J., and W. Kaelin. Unpublished results.
9. Defeo-Jones D, Huang P S, Jones R E, Haskell K M, Vuocolo G A, Hanobik M G, Huber H E, Oliff A. Cloning of cDNAs for cellular proteins that bind to the retinoblastoma gene product. Nature. 1991;352:251–254. [PubMed]
10. Dunaief J L, Strober B E, Guha S, Khavari P A, Alin K, Luban J, Begemann M, Crabtree G R, Goff S P. The retinoblastoma protein and BRG1 form a complex and cooperate to induce cell cycle arrest. Cell. 1994;79:119–130. [PubMed]
11. Dyson N. The regulation of E2F by pRB-family proteins. Genes Dev. 1998;12:2245–2262. [PubMed]
12. Dyson N, Guida P, McCall C, Harlow E. Adenovirus E1A makes two distinct contacts with the retinoblastoma protein. J Virol. 1992;66:4606–4611. [PMC free article] [PubMed]
13. Emiliani S, Fischle W, Van Lint C, Al-Abed Y, Verdin E. Characterization of a human RPD3 ortholog, HDAC3. Proc Natl Acad Sci USA. 1998;95:2795–2800. [PMC free article] [PubMed]
14. Fattaey A R, Harlow E, Helin K. Independent regions of adenovirus E1A are required for binding to and dissociation of E2F-protein complexes. Mol Cell Biol. 1993;13:7267–7277. [PMC free article] [PubMed]
15. Fattaey A R, Helin K, Dembski M S, Dyson N, Harlow E, Vuocolo G A, Hanobik M G, Haskell K M, Oliff A, Defeo-Jones D, et al. Characterization of the retinoblastoma binding proteins RBP1 and RBP2. Oncogene. 1993;8:3149–3156. [PubMed]
16. Ferreira R, Magnaghi-Jaulin L, Robin P, Harel-Bellan A, Trouche D. The three members of the pocket proteins family share the ability to repress E2F activity through recruitment of a histone deacetylase. Proc Natl Acad Sci USA. 1998;95:10493–10498. [PMC free article] [PubMed]
17. Grunstein M. Histone acetylation in chromatin structure and transcription. Nature. 1997;389:349–352. [PubMed]
18. Hamel P A, Gill R M, Phillips R A, Gallie B L. Transcriptional repression of the E2-containing promoters EIIaE, c-myc, and RB1 by the product of the RB1 gene. Mol Cell Biol. 1992;12:3431–3438. [PMC free article] [PubMed]
19. Hassig C A, Tong J K, Fleischer T C, Owa T, Grable P G, Ayer D E, Schreiber S L. A role for histone deacetylase activity in HDAC1-mediated transcriptional repression. Proc Natl Acad Sci USA. 1998;95:3519–3524. [PMC free article] [PubMed]
20. Hsiao K M, McMahon S L, Farnham P J. Multiple DNA elements are required for the growth regulation of the mouse E2F1 promoter. Genes Dev. 1994;8:1526–1537. [PubMed]
21. Iavarone A, Massagué J. E2F and histone deacetylase mediate transforming growth factor β repression of cdc25A during keratinocyte cell cycle arrest. Mol Cell Biol. 1999;19:916–922. [PMC free article] [PubMed]
22. Kaelin W G, Jr, Krek W, Sellers W R, DeCaprio J A, Ajchenbaum F, Fuchs C S, Chittenden T, Li Y, Farnham P J, Blanar M A, et al. Expression cloning of a cDNA encoding a retinoblastoma-binding protein with E2F-like properties. Cell. 1992;70:351–364. [PubMed]
23. Laherty C D, Yang W M, Sun J M, Davie J R, Seto E, Eisenman R N. Histone deacetylases associated with the mSin3 corepressor mediate mad transcriptional repression. Cell. 1997;89:349–356. [PubMed]
24. Lai A, Marcellus R C, Corbeil H B, Branton P E. RBP1 induces growth arrest by repression of E2F-dependent transcription. Oncogene. 1999;18:2091–2100. [PubMed]
25. Lee J O, Russo A A, Pavletich N P. Structure of the retinoblastoma tumour-suppressor pocket domain bound to a peptide from HPV E7. Nature. 1998;391:859–865. [PubMed]
26. Luo R X, Postigo A A, Dean D C. Rb interacts with histone deacetylase to repress transcription. Cell. 1998;92:463–473. [PubMed]
27. Magnaghi-Jaulin L, Groisman R, Naguibneva I, Robin P, Lorain S, Le Villain J P, Troalen F, Trouche D, Harel-Bellan A. Retinoblastoma protein represses transcription by recruiting a histone deacetylase. Nature. 1998;391:601–605. [PubMed]
28. Ohtani K, DeGregori J, Nevins J R. Regulation of the cyclin E gene by transcription factor E2F1. Proc Natl Acad Sci USA. 1995;92:12146–12150. [PMC free article] [PubMed]
29. Pazin M J, Kadonaga J T. What’s up and down with histone deacetylation and transcription? Cell. 1997;89:325–328. [PubMed]
30. Ross J F, Liu X, Dynlacht B D. Mechanism of transcriptional repression of E2F by the retinoblastoma tumor suppressor protein. Mol Cell. 1999;3:195–205. [PubMed]
31. Roth S Y, Allis C D. Histone acetylation and chromatin assembly: a single escort, multiple dances? Cell. 1996;87:5–8. [PubMed]
32. Sellers W R, Rodgers J W, Kaelin W G., Jr A potent transrepression domain in the retinoblastoma protein induces a cell cycle arrest when bound to E2F sites. Proc Natl Acad Sci USA. 1995;92:11544–11548. [PMC free article] [PubMed]
32a. Seto, E. Unpublished data.
33. Stiegler P, De Luca A, Bagella L, Giordano A. The COOH-terminal region of pRb2/p130 binds to histone deacetylase 1 (HDAC1), enhancing transcriptional repression of the E2F-dependent cyclin A promoter. Cancer Res. 1998;58:5049–5052. [PubMed]
34. Struhl K. Histone acetylation and transcriptional regulatory mechanisms. Genes Dev. 1998;12:599–606. [PubMed]
35. Taunton J, Hassig C A, Schreiber S L. A mammalian histone deacetylase related to the yeast transcriptional regulator Rpd3p. Science. 1996;272:408–411. [PubMed]
36. Teodoro J G, Branton P E. Regulation of p53-dependent apoptosis, transcriptional repression, and cell transformation by phosphorylation of the 55-kilodalton E1B protein of human adenovirus type 5. J Virol. 1997;71:3620–3627. [PMC free article] [PubMed]
37. Tevosian S G, Shih H H, Mendelson K G, Sheppard K A, Paulson K E, Yee A S. HBP1: a HMG box transcriptional repressor that is targeted by the retinoblastoma family. Genes Dev. 1997;11:383–396. [PubMed]
38. Trouche D, Le Chalony C, Muchardt C, Yaniv M, Kouzarides T. RB and hbrm cooperate to repress the activation functions of E2F1. Proc Natl Acad Sci USA. 1997;94:11268–11273. [PMC free article] [PubMed]
39. Weintraub S J, Chow K N, Luo R X, Zhang S H, He S, Dean D C. Mechanism of active transcriptional repression by the retinoblastoma protein. Nature. 1995;375:812–815. [PubMed]
40. Weintraub S J, Prater C A, Dean D C. Retinoblastoma protein switches the E2F site from positive to negative element. Nature. 1992;358:259–261. [PubMed]
41. Welch P J, Wang J Y. A C-terminal protein-binding domain in the retinoblastoma protein regulates nuclear c-Abl tyrosine kinase in the cell cycle. Cell. 1993;75:779–790. [PubMed]
42. Yang W M, Inouye C, Zeng Y, Bearss D, Seto E. Transcriptional repression by YY1 is mediated by interaction with a mammalian homolog of the yeast global regulator RPD3. Proc Natl Acad Sci USA. 1996;93:12845–12850. [PMC free article] [PubMed]
43. Yang W M, Yao Y L, Sun J M, Davie J R, Seto E. Isolation and characterization of cDNAs corresponding to an additional member of the human histone deacetylase gene family. J Biol Chem. 1997;272:28001–28007. [PubMed]
44. Yew P R, Berk A J. Inhibition of p53 transactivation required for transformation by adenovirus early 1B protein. Nature. 1992;357:82–85. [PubMed]
45. Yew P R, Liu X, Berk A J. Adenovirus E1B oncoprotein tethers a transcriptional repression domain to p53. Genes Dev. 1994;8:190–202. [PubMed]
46. Zacksenhaus E, Jiang Z, Phillips R A, Gallie B L. Dual mechanisms of repression of E2F1 activity by the retinoblastoma gene product. EMBO J. 1996;15:5917–5927. [PMC free article] [PubMed]
47. Zhang Y, Sun Z W, Iratni R, Erdjument-Bromage H, Tempst P, Hampsey M, Reinberg D. SAP30, a novel protein conserved between human and yeast, is a component of a histone deacetylase complex. Mol Cell. 1998;1:1021–1031. [PubMed]

Articles from Molecular and Cellular Biology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

  • Gene
    Gene
    Gene links
  • GEO Profiles
    GEO Profiles
    Related GEO records
  • HomoloGene
    HomoloGene
    HomoloGene links
  • MedGen
    MedGen
    Related information in MedGen
  • OMIM
    OMIM
    OMIM record citing PubMed
  • Pathways + GO
    Pathways + GO
    Pathways, annotations and biological systems (BioSystems) that cite the current article.
  • PubMed
    PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...