The retinoblastoma tumor suppressor, RB, is thought to inhibit cell cycle progression through transcriptional repression. E2F-regulated genes have been viewed as presumptive targets of RB-mediated repression. However, we found that specific E2F targets were not regulated in a consistent manner by the action of a RB allele that is refractory to cyclin-dependent kinase/cyclin-mediated phosphorylation (PSM-RB) when compared with E2F2 overproduction. Therefore, we used Affymetrix GeneChips as an unbiased approach to identify RB targets. We found that expression of PSM-RB significantly attenuates >200 targets, the majority of which are involved in cell cycle control (DNA replication or G2-M), DNA repair, or transcription/chromatin structure. The observed repression was due to the action of RB and not merely a manifestation of altered cell cycle distribution. Additionally, the majority of RB repression targets were confirmed through the blockade of endogenous RB phosphorylation via p16ink4a overexpression. Thus, these results have utility in assigning RB pathway activation in more complex systems of cell cycle inhibition (e.g., mitogen withdrawal, senescence, or DNA damage checkpoint). As expected, a significant fraction of RB-repressed genes have promoters that are bound/regulated by E2F family members. However, targets were identified that are distinct from genes known to be stimulated by overexpression of specific E2F proteins. Moreover, the relative action of RB versus E2F2 overexpression on specific genes demonstrates that a simple opposition model does not explain the relative contribution of RB to gene regulation. Thus, this study provides the first unbiased description of RB-repressed genes, thereby delineating new aspects of RB-mediated transcriptional control and novel targets involved in diverse cellular processes.

The retinoblastoma tumor suppressor protein, RB, is a potent inhibitor of cell cycle progression. During mitogen-induced proliferation, RB is transiently inactivated by phosphorylation catalyzed by CDK3/cyclin complexes (1, 2, 3). Specifically, CDK4 complexes initiate RB phosphorylation, and subsequent phosphorylation by CDK2 complexes leads to the complete inactivation of RB (3, 4). Phosphorylation disrupts RB function, an event that is critically important for cell cycle progression, because blockade of RB phosphorylation represents a mechanism of cell cycle arrest in both G1 and S phase (2, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14).

Underscoring its potency as an antiproliferative molecule, RB is targeted for functional inactivation at high frequency in human cancer (2, 15, 16, 17, 18, 19, 20, 21). Inactivation of RB occurs through a variety of means, including the bialleleic inactivation of the RB gene and sequestration via the oncoproteins of DNA tumor viruses (2, 15, 16, 17, 18, 19, 20, 21, 22). Alternatively, amplification of CDK4/cyclin D1 activity or loss of p16ink4a in cancer cells leads to deregulated RB phosphorylation/inactivation (2, 15, 16, 17, 18, 19, 20, 21). A common theme of these diverse mechanisms of inactivation is that they disrupt the ability of RB to assemble critical protein complexes.

RB interacts with numerous cellular proteins to assemble a transcriptional repressor module at promoters (2, 8, 23, 24, 25). For example, RB associates with the E2F family of transcriptional regulators (23, 26). The E2F proteins (E2F1–6) bind to DNA as heterodimers with DP proteins (DP1 and DP2), and these complexes bind to similar DNA sequences. The majority of E2F family members (E2F1–5) have transactivation domains and can activate the expression of target genes when bound to DNA (25, 26). In general, E2F4 and E2F5 are thought to function largely in quiescence, whereas E2F1, E2F2, and E2F3 function during G1-S progression (25, 26). The binding of RB to these E2F complexes not only antagonizes the function of the E2F proteins in activating transcription but may also convert E2F binding activity to a repressor element on the promoter of specific genes (5, 27, 28, 29). However, the extent to which repression can occur on E2F-regulated genes has not been systematically documented. The mode of RB-mediated transcriptional repression appears to be promoter specific and is dependent on a variety of proteins that are recruited by RB to promoters (24, 30, 31, 32, 33, 34, 35, 36, 37). The significance of this repressive activity of RB was demonstrated in studies wherein specific disruption of RB-mediated transcriptional repression bypassed the antiproliferative action of RB (8, 31, 32, 35, 38, 39).

Despite the link between RB and transcriptional repression, the targets of RB action remain largely undescribed. Prior studies have identified a number of genes that are specifically up-regulated upon the ectopic expression of E2F1, E2F2, and E2F3. Initially, this was determined through sequence analysis of the promoter elements of known genes and elucidation of the functional effect of E2F overproduction on transcription (40, 41, 42, 43). Subsequent microarray analyses have delineated a number of additional E2F-stimulated genes (44, 50). Whereas many of these targets are postulated to be influenced by RB, few genes have in fact been assessed for RB regulation (31, 32, 35, 40, 45, 46, 47). Thus, although transcriptional repression is assumed to play an important role in RB-mediated tumor suppression, relatively few transcriptional targets have been identified.

Cell Culture.

The A5-1, A2-4, and 10-5 cell lines were derived and cultured as described previously (54). Twenty-four h before harvest, cells were washed extensively and then cultured in the presence or absence of Dox. RNA was harvested for array analysis using Trizol (Life Technologies, Inc.) according to the manufacturer’s suggested protocol. Concurrent cultures were used to prepare protein lysates to confirm the induction of RB proteins and cell cycle arrest. Cells were synchronized in S phase as described previously (54). U2OS cells were cultured in DMEM supplemented with 10% heat-inactivated fetal bovine serum plus glutamine and penicillin/streptomycin.

Adenoviral Infections, Flow Cytometry, and Reporter Assays.

Cells were infected at a calculated multiplicity of infection of 50–100 (actual infection efficiency was 95–100% as determined by GFP immunofluorescence). Infected cells were harvested at 24 (A2-4 cells) or 36 (U2OS cells) h postinfection. Flow cytometry was carried out as described previously (6). Reporter assays were carried out as described previously (12, 54).

Array Analysis.

Total RNA was isolated from the cells and subjected to reverse transcription using random hexamers. Samples were then biotinylated and hybridized to the Affymetrix GeneChips RatU34A, U34B and U34C using the Affymetrix-recommended protocol (63, 64). Affymetrix Micro Array Suite version 4.0 was used to scan and quantitate GeneChips using default scan settings. Intensity data were collected from each chip, and the results were analyzed using both Micro Array Suite and GeneSpring 4.0 (Silicon Genetics, Inc., Redwood City, CA).

RNA Sample Validation.

Before submission of RNA samples for analysis, protein extracts prepared from replicate plates of the corresponding cell cultures were analyzed for expected induction of RB and reduction of cyclin A using Western blots. RNA integrity was also verified for lack of degradation by formaldehyde gel electrophoresis and a Agilent Bioanalyzer 2100 (with typical 28S:18S ratios = 2 ± 0.1).

Gene Annotation.

ESTs presented on the Rat genome U34A, B, and C GeneChips by Affymetrix were based on the Rat UniGene Build #34 assembly and were subjected to reannotation using the National Center for Biotechnology Information BLAST and UniGene Build #93 database resources (November 2001), Known genes were based on those present in GenBank (Oct 12, 2001). Using these resources, the ESTs that were present in the 341 total implicated targets were subsequently collapsed into nonredundant genes.

Immunoblotting.

Equal protein was loaded in each lane, as verified by naphthal black staining and by immunoblotting against CDK4 or vimentin. Western blots were performed using the following commercial antibodies: (a) cyclin E, cyclin B1, RNR-M2, p16ink4a, p55CDC, cyclin A, CDK2, CDC2, CDK4, PLK1, MCM5, E2F2, MCM6, MCM7, PTTG, and PCNA (all from Santa Cruz Biotechnology); (b) Fen1, DNA polymerase δ, and DHFR (all from Transduction Laboratories); (c) HMG2 and EGR1 (both from Geneka); and (d) topoisomerase IIα (Topogen). The following antibodies were gifts from the individuals indicated: (a) RB (Dr. Jean Wang); (b) vimentin (Dr. Wally Ip); (c) DNMT1 (Dr. Keith Robertson); and (d) thymidylate synthase (Dr. Masakazu Fukushima).

RB-mediated Target Attenuation Is Not Synonymous with E2F Activation.

To investigate the role of RB in signaling to downstream targets, a number of cell lines were produced. These cell lines are based in a Rat-1-derived cell line (Rat-16) that carries the elements for tetracycline-regulated gene expression. Due to the highly efficient phosphorylation/inactivation of RB in most cells, we used a mutant of RB that is refractory to CDK/cyclin-mediated phosphorylation (PSM-RB), and we have previously documented that such proteins are potent inhibitors of cell cycle progression (10, 11, 48). For the present study, we used two independent cell lines (A2-4 and A5-1) that express PSM-RB upon withdrawal of tetracycline/Dox from the media. We also used a cell line that expresses the minimal growth suppression region of RB, WT, which can be phosphorylated by endogenous CDK/cyclin activity (10-5 cell line) (49). Before analysis, we confirmed the expression and activity of RB in each sample harvested. In the presence of Dox, expression of the RB proteins was not detected (Fig. 1,A, top panel, Lanes 1, 3, 5, and 7). However, when Dox was removed from the media, the ectopically expressed RB proteins were readily detectable (Fig. 1,A, top panel, Lanes 4, 6, and 8). In the case of 10-5 cells, the expressed WT protein was efficiently phosphorylated (Fig. 1,A, top panel, Lane 4). As such, it had minimal effect on the known downstream target, cyclin A (Fig. 1,A, bottom panel, compare Lanes 3 and 4). Additionally, induction of WT in 10-5 cells had limited influence on cell cycle distribution (Fig. 1,B, compare top and bottom panels). In fact, 10-5 cells readily proliferate in the absence of Dox (data not shown), consistent with prior observations that WT is a poor inhibitor of cell cycle progression. In contrast, A2-4 and A5-1 cells expressed PSM-RB upon Dox withdrawal (Fig. 1,A, top panel, Lanes 6 and 8), leading to the strong inhibition of cyclin A expression (Fig. 1,A, top panel, compare Lanes 5 and 6 with Lanes 7 and 8) and concomitant G1-S arrest (Fig. 1,B, compare top and bottom panels for each line). As expected, parental Rat-16 cells did not ectopically express RB (Fig. 1,A, top panel, Lanes 1 and 2), modify cyclin A expression (Fig. 1,A, bottom panel, Lanes 1 and 2), or change cell cycle profile (Fig. 1 B) in the absence of Dox. Therefore, whereas Rat-16 and 10-5 cells represent cell types with largely inactive RB, A5-1 and A2-4 cells serve as potent inducers of RB activity.

To better understand how RB signals to mediate cell cycle inhibition, the effect of PSM-RB on specific E2F-induced gene products (recently described in microarray studies) was investigated (44, 50, 51). Initially, the behavior of genes previously shown to be induced by E2F overproduction (cyclin E, Cdc25a, DHFR, and PCNA) was examined in response to PSM-RB (41, 44, 50, 51). Interestingly, we failed to observe a significant attenuation of these targets after the induction of PSM-RB expression, whereas cyclin A was readily attenuated (Fig. 1,C, compare Lanes 1 and 2). As a control for the regulation of these proteins in this cell line, the cells were infected with recombinant adenovirus encoding E2F2 (Fig. 1 C, Lane 3). As expected, ectopic expression of E2F2 led to the induction of both cyclin E and Cdc25a. In contrast, ectopic E2F2 expression had little effect on the levels of PCNA, DHFR, and cyclin A. Together, these results indicate that E2F2 targets are not necessarily consistent with those of RB-mediated attenuation; therefore, we undertook an unbiased approach to identify targets of the unphosphorylated actively repressing form of RB.

Target Genes of RB-mediated Arrest.

To identify additional targets of RB-mediated transcriptional control, Affymetrix GeneChip arrays were used that contain probe sets recognizing approximately 25,000 rat genes or UniGene-clustered ESTs. Each cell line (Rat-16, 10-5, A2-4, and A5-1) was cultured in the presence or absence of Dox. For each condition, two independent cultures were used, and each one was subjected to independent RNA isolation, labeling, and GeneChip hybridization. GeneChips were scanned and quantified using the algorithms implemented within Affymetrix Micro Array Suite 4.0 software. GeneSpring software was then used to impose a series of normalization and filtering criteria to identify a list of genes reproducibly responsive to PSM-RB. These analyses yielded 341 targets from the replicated experiments, which were reproducibly repressed at least 1.7-fold in both cell lines expressing PSM-RB.

The graphic in Fig. 2,A shows the overall behavior of these targets (fully annotated data set is available online4). The basal state of each gene and EST is depicted as a yellow bar. The repression of each target is visualized by the change in color from yellow toward blue, whereas activation is shown as a change from yellow toward red (see color bar). For the entire set of 341 targets, the average effect of WT (1.3-fold repression) was considerably more modest than that observed for PSM-RB (2.9-fold repression). Using clustering algorithms, the most prominent repression targets of RB (in all conditions studied) fall into two mathematically defined clusters containing a total of 251 genes/ESTs, which exhibit very similar behavior and are depicted together (Fig. 2,B). In the graph shown (Fig. 2,B), each line represents an individual gene or EST, with the change in RNA levels represented by the slope of the line. For these genes, the average repression achieved by the expression of WT (10-5 cells, +Dox versus −Dox) was 1.2-fold, whereas the average repression achieved through the expression of PSM-RB was 3.4-fold (A2-4 and A5-1 cells, +Dox versus −Dox). Analysis of the genes in these clusters showed a large fraction of targets involved in DNA replication, DNA repair, chromatin structure/transcription, and G2-M progression (Table 1). A number of these targets have recently been shown to be bound by endogenous E2F proteins (52, 53) or induced by E2F overexpression (Refs. 44, 50, and 51; genes/ESTs are summarized in Table 1). However, this list represents the first unbiased analysis of RB-mediated repression and clearly defines a large number of new targets not previously linked to E2F.

Cell Cycle Dependence and Endogenous RB in Transcriptional Repression.

To confirm that the defined targets were indeed responsive to RB and not merely an indirect effect of cell cycle arrest, the DNA polymerase inhibitor APH was used to hold both induced and uninduced cells in S phase. As depicted in Fig. 3,A, A5-1 cells were synchronized in S phase in the presence of Dox by the addition of APH to the media. In the continued presence of APH, Dox was then removed from half of the cultures to induce PSM-RB (Fig. 3,A). We have previously used this approach to investigate signaling from RB in S-phase cells (54). Under these conditions, PSM-RB mediated transcriptional repression of an E2F reporter (Fig. 3,B, left panel) and led to the attenuation of cyclin A (Fig. 3,B, right panel, compare Lanes 1 and 2), verifying its activity in S phase. Interestingly, a small subset of the defined repression targets were dependent on cell cycle arrest (Fig. 3,D). This subset was characterized by histone genes (histones H1 and H3) and the growth factor-inducible genes c-Jun and EGR-1. However, in the S-phase-synchronized cultures (A5-1 APH +Dox/–Dox), PSM-RB mediated repression of 94% of those targets identified as repressed in asynchronous cultures (Fig. 3,C, compare A5-1 +Dox/−Dox versus A5-1 APH +Dox/−Dox). Additionally, the average fold repression in S-phase-synchronized cells was similar to that observed in asynchronous cultures (Fig. 3 C). Thus, the majority of the RB-mediated repression response was intrinsic to the action of RB and not merely a consequence of changes in cell cycle distribution.

Whereas it is clear that PSM-RB alleles are important tools that specifically elicit downstream signaling pathways from RB, it is critical to compare this with the effects exerted by physiological activation of endogenous RB. To activate endogenous rat RB, we used ectopic expression of p16ink4a. It is known that p16ink4a prevents phosphorylation of RB through the direct inhibition of CDK4 and CDK2 activity (1). The observation that RB is a critical target of p16ink4a is supported by results showing that Rb−/ cells are refractory to the effects of p16ink4a on cell cycle progression (55, 56). As expected, ectopic expression of p16ink4a from recombinant adenovirus mediated an accumulation of cells in G1 as determined by flow cytometry, when compared with infection with a GFP-encoding adenovirus (Fig. 3,E, left panel). This arrest was accompanied by dephosphorylation of endogenous RB (data not shown) and attenuation of cyclin A expression (Fig. 3,E, right panel, compare Lanes 1 and 2). Therefore, p16ink4a expression is a viable means to prevent the phosphorylation of endogenous RB and monitor subsequent changes in gene expression. Infection with GFP-encoding adenovirus did not result in significant changes in gene expression, when compared with uninfected controls. As compared with cells expressing GFP-control, p16ink4a expression resulted in significant transcriptional repression (Fig. 3,F). The repression targets of p16ink4a included 87% of those genes/ESTs repressed by PSM-RB (Fig. 3 F, compare A5-1 +Dox/−Dox versus A5-1 APH +Dox/−Dox). Thus, the repression targets elucidated by use of PSM-RB are indicative of the gene spectrum targeted through the activation of the endogenous RB.

Validation of RB Repression Targets.

Specific targets identified in these microarray screens were verified through analysis of protein levels, cell cycle position, and alternative cell systems. First, although the data described above have extensive utility in profiling RB-mediated transcriptional repression, the effect on cell cycle control downstream from RB would likely be represented by changes in protein levels as opposed to RNA. Therefore, we assessed the action of RB on a number of strong [cyclin A, ribonucleotide reductase M2 subunit (RRM2), MCM6, MCM7, DNA polymerase δ, topoisomerase IIα, PCNA, CDK2, p55CDC, cyclin B1, CDC2, Egr1, and HMG2] and relatively weak (Fen-1, DHFR, and cyclin E) repression targets at the level of protein expression. Cells cultured in the presence or absence of Dox for 24 h were harvested, and the levels of proteins were determined by immunoblotting. In Rat-16 cells, there were no significant changes in protein levels after the removal of Dox (Fig. 4,A, compare Lanes 1 and 2). Similarly, the expression of WT had little effect on the expression of target proteins (data not shown). However, for the majority of targets, activation of PSM-RB in the A5-1 cells resulted in a substantial reduction in their protein levels (Fig. 4 A, compare Lanes 3 and 4). Similar protein attenuation was observed in the A2-4 cell line (data not shown). Interestingly, proteins such as PCNA and CDK2 do not significantly change at the protein level, suggesting that the failure to change could reflect a relatively long half-life. Consistent with this idea, expression of PSM-RB for longer intervals (48–72 h) resulted in significant attenuation of both PCNA and CDK2 (data not shown). Cyclin E and DHFR, which exhibit a relatively weak response at the level of RNA (approximately 2-fold), showed no attenuation at the protein level. Together, these data support the observation that the diminution of specific target RNAs leads to meaningful changes in proteins.

Second, to confirm that the down-regulation of proteins was not merely a consequence of cell cycle phase, we used S-phase-synchronized cells. A5-1 cells were synchronized in S-phase with APH, and PSM-RB was induced as described above (Fig. 3,A). Protein levels from the cells were then examined by immunoblotting (Fig. 4 B, Lanes 1 and 2). Consistent with the effects of RB on transcription in S phase, the attenuation of proteins was similar to that observed in asynchronous cells. These analyses confirmed that the target proteins were down-regulated by RB in a manner that was independent from changes in cell cycle phase.

Lastly, we verified that the effects of PSM-RB on targets observed in the Rat-1 cell systems were recapitulated by activation of endogenous RB in human cells. To activate the RB pathway in a human cell line, U2OS cells were either mock-infected or infected with p16ink4a-encoding adenovirus to inhibit the phosphorylation of RB. As predicted, p16ink4a overexpression led to the activation of endogenous RB (data not shown) and mediated cell cycle arrest, as observed via flow cytometry (Fig. 4,C, left panels). Under these conditions, attenuation of target proteins was observed, similar to the rat cells with PSM-RB induced (Fig. 4 C, right panel). Interestingly, some targets were more strikingly attenuated in the U2OS cells, such as PCNA. Together, these data confirm the targets of RB identified through the microarray analysis and verify their conservation in both human and rodent models.

Effect of E2F2 Overexpression on RB Repression Targets.

Because E2F complexes are believed to function as the principal transcriptional targeting system for RB, we assessed the effect of E2F2 overproduction on the behavior of RB targets. First, we investigated the extent to which ectopic E2F2 expression is able to reverse RB-mediated repression using a synthetic 3XE2F reporter construct (Fig. 4,D). A5-1 cells harboring inducible expression of PSM-RB were either mock-infected or infected with GFP-encoding adenovirus or adenovirus encoding for E2F2 in the presence or absence of Dox. Infection with GFP adenovirus did not significantly influence E2F activity (Fig. 4,D, compare Mock +Dox and GFP +Dox). The PSM-RB-mediated repression was readily apparent in GFP-infected cells (Fig. 4 D, compare GFP +Dox and GFP −Dox bars). In the absence of PSM-RB, ectopic expression of E2F2 weakly stimulated 3XE2F promoter activity (compare GFP +Dox and E2F2 +Dox). E2F2 expression blocked the repression elicited by PSM-RB (compare GFP −Dox and E2F2 −Dox). Thus, the ectopic expression of E2F2 can quantitatively antagonize PSM-RB-mediated repression.

To examine the effect of E2F2 overexpression on PSM-RB-mediated cell cycle inhibition, we used flow cytometry. Ectopic expression of E2F2 had little effect on cell cycle distribution in the presence of Dox (Fig. 4,E, left panels, Mock +Dox versus E2F2 +Dox). In the absence of Dox, E2F2 expression overcomes the RB-mediated cell cycle block, allowing an accumulation of cells in S phase (Fig. 4 E, left panels, Mock −Dox versus E2F2 −Dox).

Analysis of RB repression targets under conditions of mock infection showed that cyclin A, topoisomerase IIα, p55CDC, RRM2, thymidylate synthase, and cyclin B1 were all down-regulated by expression of PSM-RB (Fig. 4,E, right panel, compare Lanes 1 and 2). However, these same targets were not significantly stimulated by the overexpression of E2F2 (Fig. 4,E, right panel, compare Lanes 1 and 3). In contrast, expression of E2F2 did block the RB-mediated attenuation of these proteins (Fig. 4,E, right panel, compare Lanes 1 and 2 versus Lanes 3 and 4). Thus, E2F2 overproduction serves to disrupt RB-mediated repression, but only in the case of specific targets (e.g., cyclin E, see Fig. 1 C) does it induce their expression.

RB-mediated Transcriptional Repression: Targets of RB.

Although RB is believed to inhibit proliferation through regulation of transcription, targets of RB action were speculative. In this report we identify a comprehensive list of genes that respond to the hypophosphorylated form of RB capable of mediating transcriptional repression. Using inducible cell lines that either express an allele of RB that is readily phosphorylated and inactivated (WT) or an allele of RB that is inert to phosphorylation (PSM-RB), unbiased microarray analyses were performed. From approximately 25,000 targets, 341 were identified that were reproducibly repressed by PSM-RB in two independent cell lines. These repression targets were highly enriched for cell cycle genes, particularly those involved in DNA replication and G2-M control. It could be argued that a significant number of the repression targets of PSM-RB are a result of cell cycle repositioning during the RB-mediated arrest. However, we were able to dissociate cell cycle from transcriptional repression by arresting all cells in S phase with aphidicolin and then inducing active RB. Some genes were identified that failed to be responsive to RB in S phase. These included histones, whose transcription is coupled to ongoing DNA replication (57).

Typically, the action of RB is elicited in response to physiological stresses (2). The PSM-RB alleles used provide an important tool for specifically focusing on targets of RB in isolation; however, it is equally important to determine whether they reflect the action of the endogenous RB. To assess the role of endogenous RB, we used p16ink4a, which, by virtue of inhibiting CDK activity, causes the accumulation of dephosphorylated RB (1). Analyses of genes regulated by p16ink4a were largely consistent with those of genes regulated by the action of PSM-RB. These results lend further support to the idea that the transcriptional repression program elicited by p16ink4a is mediated through RB. Additionally, they suggest that these same targets will be repressed in response to environmental signals (e.g., DNA damage that elicits RB activation).

RB-mediated transcriptional repression is complicated, with multiple factors cooperating for transcriptional repression on specific promoters. For example, repression of cyclin A is dependent on SWI/SNF chromatin remodeling, whereas other forms of transcriptional repression are dependent on histone deacetylases or polycomb repressor components (31, 32, 35, 36). The relatively small list of RB transcriptional targets has stymied efforts to analyze the contributions of different corepressors. Thus, by elucidating the targets, we are now in the position to delineate those activities (i.e., SWI/SNF, HDAC, and so forth) responsible for the observed transcriptional repression effects. Additionally, because loss of SWI/SNF disrupts RB-mediated cell cycle inhibition components (31, 32), we can begin to decipher the functional significance of targets by analyzing the effect of this lesion on specific targets.

Whether RB interacts directly with the promoters of all of the targets identified in this study remains undetermined. Specific promoters have been shown to be occupied by RB during p16ink4a-mediated cell cycle arrest (35), and these were identified in our screen. Moreover, members of a subset of targets identified herein have been shown through chromatin immunoprecipitation analysis to harbor E2F complexes on their promoters (Refs. 52 and 53; summarized in Table 1). Given the documented interaction between E2F and RB, it is highly likely that RB is capable of directly binding the regulatory regions of these E2F-associated genes. Analysis of the remaining RB target promoter regions is the focus of current study. Regardless of the nature of transcriptional repression (i.e., direct or indirect), these data provide powerful insight into the transcriptional and thus biological consequence of RB activation.

E2F as a Target of RB.

RB assembles repressor complexes to actively repress promoters containing E2F sites (8, 24, 28, 30, 31, 32, 36, 37). The importance of this activity of RB has been demonstrated in several ways. First, ectopic overexpression of E2F proteins was shown to overcome RB-mediated cell cycle arrest (39, 58, 59). Additionally, using mutants of E2F that specifically displace the endogenous proteins from promoters, it was shown that disruption of repression was sufficient to overcome RB-dependent arrest (8, 38). These results have led to the hypotheses that genes activated via the ectopic expression of E2F proteins are universally repressed by RB and that RB repression targets are activated by E2F. The results shown here argue against simple antagonism in relation to RB/E2F in several ways. First, some of the genes up-regulated by the overexpression of E2F proteins are not appreciably attenuated by PSM-RB. For example, ectopic expression of E2F1 or E2F2 stimulates the expression of cyclin E RNA by 14.4- and 22.3-fold, respectively (44), whereas PSM-RB only represses cyclin E by 2.1-fold. This dichotomy is also observed at the level of protein, where ectopic expression of E2F2 results in significant augmentation of cyclin E protein levels, whereas PSM-RB does not influence the level of cyclin E protein (Fig. 1 C). Second, there are a large number of RB targets that are only weakly activated by the ectopic expression of E2F proteins. For example, Ki-67, cyclin A2, Cdc2, and MCM-7 are repressed efficiently by PSM-RB (12.4-, 7.5-, 4.2-, and 6.3-fold, respectively) but stimulated weakly by E2F1 (2.2-, 2.2-, 1.9-, and 1.9-fold, respectively) or E2F2 (2.7-, 2.7-, 2.2-, and 2.4-fold, respectively) overexpression (44). Interestingly, in studies where ectopic expression of E2F1 and E2F2 (44) or E2F1 and E2F3 (51) was studied side by side, the magnitude of target gene activation was similar between the different E2F-family members used. Thus, although induction by E2F family members tends to be similar in degree, there are differences between the relative induction by E2F proteins and repression by RB.

To further delineate the role of E2F in RB-mediated transcriptional repression, ectopic expression of E2F2 was used to antagonize RB signaling. Interestingly, E2F2 overexpression stimulated the expression of cdc25A and cyclin E protein but had little influence on stimulating the accumulation of other proteins (cyclin A, p55CDC, RNR-M2, cyclin B1, and topoisomerase IIα) in the absence of PSM-RB. These results suggest that under normal growth conditions, there is potentially little effect of E2F2 overproduction, and perhaps only under conditions where pocket proteins are active (e.g., during quiescence) are these targets clearly apparent. This idea would explain the relatively disparate sets of target genes identified between E2F1, E2F2, or E2F3 overexpression in proliferating cells (50)versus those that are stimulated with E2F from quiescence (44, 51). In general, the targets of RB identified in our screen fall into this latter category, suggesting that much of the observed stimulation due to E2F overexpression is due to derepression. Consistent with this view, we show that E2F2 overproduction can uncouple RB-mediated target attenuation (Fig. 4 E). Together, these results show that there are inherent differences of specific targets for the relative role of RB repression versus E2F activation.

RB Targets and Cell Cycle Control.

Once activated/dephosphorylated, RB can inhibit G1 and S-phase progression. The finding that RB functions to down-regulate a large number of DNA replication factors is consistent with its role in the inhibition of DNA replication. Because the encoded proteins are important participants in cell cycle progression, we sought to determine whether the activation of RB actually attenuated the target proteins. Immunoblot analysis of several of the replication protein targets confirmed that they were down-regulated at the level of protein (Fig. 4 A). Furthermore, attenuation of these proteins occurred in concert with cell cycle inhibition. Thus, loss of these proteins could participate in the observed blockage of replication. Ongoing studies are determining the functional role of these targets in RB-mediated replication control.

The targeting of proteins involved in DNA repair and G2-M control also suggests that RB could participate in the inhibition of those processes. This would be of interest given the role of RB in DNA damage checkpoint regulation (12, 60, 61, 62). Consistent with the role of RB in this form of G2-M control, it has been reported that RB is required for the maintenance of the DNA damage-induced G2-M checkpoint that is facilitated by down-regulation of cyclin B1 (a target identified in our screen; Ref. 62).

In summary, the data presented here provide a clear outline of RB action in the cessation of proliferation. Our unbiased screen revealed transcriptional targets of RB that were independent of cell cycle position and validated to the level of protein in both rat and human cells. Whereas a subset of targets is shared with known E2F-regulated genes, many novel targets were identified linking RB to additional layers of cell cycle and checkpoint control. Moreover, our results draw significant distinctions between RB repression and E2F activation, illustrating that the relative contributions to transcriptional control cannot be attributed to simple antagonism. Together, these data reveal the complex nature of RB/E2F signaling and identify functional targets that contribute to RB biological activity.

Fig. 1.

RB-mediated arrest elicits specific target response. A, the indicated cell lines were cultured in the presence (Lanes 1, 3, 5, and 7) or absence of Dox (Lanes 2, 4, 6, and 8) for 24 h. Cells were harvested, protein lysates were prepared, and equal protein was resolved by SDS-PAGE. The RB and cyclin A proteins were then detected by immunoblotting. B, the cell lines cultured as described in A were harvested, fixed with ethanol, and stained with propidium iodide. The cell cycle distribution of stained cells was determined by flow cytometry. The percentage of cells in G0-G1, S, and G2-M was determined using ModFit software. C, A2-4 cells were either mock-infected (Lanes 1 and 2) or infected with E2F2-encoding adenovirus (Lane 3) in the presence (Lanes 1 and 3) or absence of Dox (Lane 2). Cells were harvested 24 h postinfection, and equal total protein was resolved by SDS-PAGE. The indicated proteins were detected by immunoblotting.

Fig. 1.

RB-mediated arrest elicits specific target response. A, the indicated cell lines were cultured in the presence (Lanes 1, 3, 5, and 7) or absence of Dox (Lanes 2, 4, 6, and 8) for 24 h. Cells were harvested, protein lysates were prepared, and equal protein was resolved by SDS-PAGE. The RB and cyclin A proteins were then detected by immunoblotting. B, the cell lines cultured as described in A were harvested, fixed with ethanol, and stained with propidium iodide. The cell cycle distribution of stained cells was determined by flow cytometry. The percentage of cells in G0-G1, S, and G2-M was determined using ModFit software. C, A2-4 cells were either mock-infected (Lanes 1 and 2) or infected with E2F2-encoding adenovirus (Lane 3) in the presence (Lanes 1 and 3) or absence of Dox (Lane 2). Cells were harvested 24 h postinfection, and equal total protein was resolved by SDS-PAGE. The indicated proteins were detected by immunoblotting.

Close modal
Fig. 2.

Microarray analysis of RB-repression targets. A, hierarchical tree depiction of the 341 genes/ESTs that exhibit replicate-reproducible repression after the induction of PSM-RB in the A2-4 and A5-1 cell lines. Each horizontal block represents a single gene/EST. Basal expression of each gene or EST in the individual cell lines is represented in yellow (Rat-16, 10-5, A2-4, and A5-1 +Dox). A shift in color toward blue signifies a decrease in RNA levels (repression), whereas a shift toward red indicates an increase in RNA levels (activation), when the cell line is cultured in the absence of Dox (Rat-16, 10-5, A2-4, and A5-1 +Dox). The color bar depicts the relationship between gene/EST level and color. B, graph of the behavior of 251 clustered repression targets of PSM-RB. In this graphical format, the behavior of each gene or EST is represented as a single line. The basal state (+Dox) of each gene or EST is normalized to 1, and the relative change in RNA level is graphed (−Dox). Data are shown for each of the cell lines.

Fig. 2.

Microarray analysis of RB-repression targets. A, hierarchical tree depiction of the 341 genes/ESTs that exhibit replicate-reproducible repression after the induction of PSM-RB in the A2-4 and A5-1 cell lines. Each horizontal block represents a single gene/EST. Basal expression of each gene or EST in the individual cell lines is represented in yellow (Rat-16, 10-5, A2-4, and A5-1 +Dox). A shift in color toward blue signifies a decrease in RNA levels (repression), whereas a shift toward red indicates an increase in RNA levels (activation), when the cell line is cultured in the absence of Dox (Rat-16, 10-5, A2-4, and A5-1 +Dox). The color bar depicts the relationship between gene/EST level and color. B, graph of the behavior of 251 clustered repression targets of PSM-RB. In this graphical format, the behavior of each gene or EST is represented as a single line. The basal state (+Dox) of each gene or EST is normalized to 1, and the relative change in RNA level is graphed (−Dox). Data are shown for each of the cell lines.

Close modal
Fig. 3.

RB-mediated repression is not dependent on cell cycle position and can be recapitulated by active endogenous RB. A, A5-1 cells cultured in the presence of Dox were synchronized in S phase using aphidicolin (2 μg/ml). In the continued presence of aphidicolin, Dox was retained in half the cultures and removed from the other half to induce PSM-RB. The cells were then cultured for an additional 24 h. RB-mediated repression was then evaluated. B, PSM-RB represses transcription in S phase. Left panel, A5-1 cells were cotransfected with cytomegalovirus-β-galactosidase and 3xE2F-Luc reporter plasmids (12). These cells were synchronized in S phase and then cultured in the presence or absence of Dox for 24 h (as described in A). Lysates were prepared from the cells and used in analysis of β-galactosidase and luciferase activity. The luciferase activity was normalized to β-galactosidase, with +Dox set to 100. Data shown are from two independent experiments. Right panel, S-phase-synchronized A5-1 cells cultured in either the presence (Lane 1) or absence (Lane 2) of Dox for 24 h (as described in A) were harvested, and equal total protein was resolved by SDS-PAGE. RB and cyclin A were detected by immunoblotting. C, A5-1 cells were synchronized in S phase and then cultured in the presence or absence of Dox for 24 h (as described in A). These cells were harvested, and total RNA was used for microarray analysis. The graph depicts the behavior of the same 251 genes and ESTs shown in Fig. 2,B, in asynchronous Rat-16 and A5-1 cells (for comparison) and in S-phase-synchronized A5-1 cells (A5-1 APH +Dox/−Dox). D, a subset of genes that are dependent on cell cycle inhibition for RB-mediated repression. E, A5-1 cells cultured in the presence of Dox were infected with either GFP- or p16ink4a-encoding adenoviruses. Left panel, cells were fixed and stained with propidium iodide. The cell cycle distribution of these cells was then determined by flow cytometry. Right panel, cells were harvested, and the level of p16ink4a, GFP, and cyclin A proteins was determined by immunoblotting. F, A5-1 cells cultured in the presence of Dox were infected with GFP- or p16ink4a-encoding adenoviruses. Twenty-four h postinfection, cells were harvested, and total RNA was used for microarray analysis. The graph depicts the behavior of 251 target genes and ESTs (as in Fig. 2 B) in asynchronous Rat-16 and A5-1 cells (for comparison) and in adenovirus-infected A5-1 cells (A5-1 +Dox GFP versus p16).

Fig. 3.

RB-mediated repression is not dependent on cell cycle position and can be recapitulated by active endogenous RB. A, A5-1 cells cultured in the presence of Dox were synchronized in S phase using aphidicolin (2 μg/ml). In the continued presence of aphidicolin, Dox was retained in half the cultures and removed from the other half to induce PSM-RB. The cells were then cultured for an additional 24 h. RB-mediated repression was then evaluated. B, PSM-RB represses transcription in S phase. Left panel, A5-1 cells were cotransfected with cytomegalovirus-β-galactosidase and 3xE2F-Luc reporter plasmids (12). These cells were synchronized in S phase and then cultured in the presence or absence of Dox for 24 h (as described in A). Lysates were prepared from the cells and used in analysis of β-galactosidase and luciferase activity. The luciferase activity was normalized to β-galactosidase, with +Dox set to 100. Data shown are from two independent experiments. Right panel, S-phase-synchronized A5-1 cells cultured in either the presence (Lane 1) or absence (Lane 2) of Dox for 24 h (as described in A) were harvested, and equal total protein was resolved by SDS-PAGE. RB and cyclin A were detected by immunoblotting. C, A5-1 cells were synchronized in S phase and then cultured in the presence or absence of Dox for 24 h (as described in A). These cells were harvested, and total RNA was used for microarray analysis. The graph depicts the behavior of the same 251 genes and ESTs shown in Fig. 2,B, in asynchronous Rat-16 and A5-1 cells (for comparison) and in S-phase-synchronized A5-1 cells (A5-1 APH +Dox/−Dox). D, a subset of genes that are dependent on cell cycle inhibition for RB-mediated repression. E, A5-1 cells cultured in the presence of Dox were infected with either GFP- or p16ink4a-encoding adenoviruses. Left panel, cells were fixed and stained with propidium iodide. The cell cycle distribution of these cells was then determined by flow cytometry. Right panel, cells were harvested, and the level of p16ink4a, GFP, and cyclin A proteins was determined by immunoblotting. F, A5-1 cells cultured in the presence of Dox were infected with GFP- or p16ink4a-encoding adenoviruses. Twenty-four h postinfection, cells were harvested, and total RNA was used for microarray analysis. The graph depicts the behavior of 251 target genes and ESTs (as in Fig. 2 B) in asynchronous Rat-16 and A5-1 cells (for comparison) and in adenovirus-infected A5-1 cells (A5-1 +Dox GFP versus p16).

Close modal
Fig. 4.

RB target validation and regulation. A, Rat-16 (Lanes 1 and 2) and A5-1 (Lanes 3 and 4) cells were cultured in the presence (Lanes 1 and 3) or absence (Lanes 2 and 4) of Dox for 24 h. Cells were harvested, and equal total protein was resolved by SDS-PAGE. The indicated proteins were detected by immunoblotting. B, A5-1 cells were synchronized in S phase using aphidicolin and then cultured for an additional 24 h in the presence or absence of Dox (as described in Fig. 3 A). Cells were harvested, equal total protein was resolved by SDS-PAGE, and the indicated proteins were detected by immunoblotting. C, U2OS cells were either mock-infected or infected with a p16ink4a-encoding adenovirus. Left panels, cells were collected and analyzed for cell cycle distribution by flow cytometry. Right panel, cell lysates were prepared from cells, equal total protein was resolved by SDS-PAGE, and the indicated proteins were detected by immunoblotting. D, asynchronously growing A5-1 cells were cotransfected with cytomegalovirus-β-galactosidase and 3xE2F-Luc reporter plasmids (12). The cells were then cultured in the presence or absence of Dox and either mock-infected or infected with adenoviruses encoding GFP or E2F2 as indicated. Lysates were prepared from the cells and used in analysis of β-galactosidase and luciferase activity. The luciferase activity was normalized to β-galactosidase, with mock-infected +Dox set to 100. Data shown are from two independent experiments. E, A2-4 cells were either mock-infected or infected with an E2F2-encoding adenovirus. Left panels, cells were harvested 24 h postinfection and analyzed for cell cycle distribution by flow cytometry. Right panel, A2-4 cells were either mock-infected (Lanes 1 and 2) or infected with E2F2-encoding adenovirus (Lanes 3 and 4) in the presence (Lanes 1 and 3) or absence of Dox (Lanes 2 and 4). Cells were harvested 24 h postinfection, and the indicated proteins were detected by immunoblotting.

Fig. 4.

RB target validation and regulation. A, Rat-16 (Lanes 1 and 2) and A5-1 (Lanes 3 and 4) cells were cultured in the presence (Lanes 1 and 3) or absence (Lanes 2 and 4) of Dox for 24 h. Cells were harvested, and equal total protein was resolved by SDS-PAGE. The indicated proteins were detected by immunoblotting. B, A5-1 cells were synchronized in S phase using aphidicolin and then cultured for an additional 24 h in the presence or absence of Dox (as described in Fig. 3 A). Cells were harvested, equal total protein was resolved by SDS-PAGE, and the indicated proteins were detected by immunoblotting. C, U2OS cells were either mock-infected or infected with a p16ink4a-encoding adenovirus. Left panels, cells were collected and analyzed for cell cycle distribution by flow cytometry. Right panel, cell lysates were prepared from cells, equal total protein was resolved by SDS-PAGE, and the indicated proteins were detected by immunoblotting. D, asynchronously growing A5-1 cells were cotransfected with cytomegalovirus-β-galactosidase and 3xE2F-Luc reporter plasmids (12). The cells were then cultured in the presence or absence of Dox and either mock-infected or infected with adenoviruses encoding GFP or E2F2 as indicated. Lysates were prepared from the cells and used in analysis of β-galactosidase and luciferase activity. The luciferase activity was normalized to β-galactosidase, with mock-infected +Dox set to 100. Data shown are from two independent experiments. E, A2-4 cells were either mock-infected or infected with an E2F2-encoding adenovirus. Left panels, cells were harvested 24 h postinfection and analyzed for cell cycle distribution by flow cytometry. Right panel, A2-4 cells were either mock-infected (Lanes 1 and 2) or infected with E2F2-encoding adenovirus (Lanes 3 and 4) in the presence (Lanes 1 and 3) or absence of Dox (Lanes 2 and 4). Cells were harvested 24 h postinfection, and the indicated proteins were detected by immunoblotting.

Close modal

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

Supported by funding to E. S. K. and B. J. A. from the NIEHS-Comparative Mouse Genomic Center (Grant U01-E511308) and Center for Environmental Genetics (Grant P30-ES06096).

3

The abbreviations used are: CDK, cyclin-dependent kinase; GFP, green fluorescent protein; EST, expressed sequence tag; Dox, doxycycline; DHFR, dihydrofolate reductase; PCNA, proliferating cell nuclear antigen; APH, aphidicolin; WT, wild-type large pocket RB.

4

http://genet.chmcc.org.

Table 1
graphic
graphic
graphic
graphic
graphic
graphic

Annotated gene list of 251 targest from two repression clusters containing targets repressed by RB in all conditions. Fold repression and error were calculated from replicated independent experiments. Genes which have been previously described as being activated [P = Polager et al., 2002 (51); I = Isida et al., 2001 (44); M = Muller et al., 2001 (50)] or bound [R = Ren et al., 2002 (53); W = Weinmann et al., 2002 (52)] by E2F are indicated.

We are grateful to Dr. Karen Knudsen, Dr. Christopher Mayhew, Craig Burd, Christin Petre, and Sejal Ranmal for comments on the manuscript and all members of the Knudsens’ laboratories for thought-provoking discussions.

1
Mittnacht S. Control of pRB phosphorylation.
Curr. Opin. Genet. Dev.
,
8
:
21
-27,  
1998
.
2
Wang J. Y., Knudsen E. S., Welch P. J. The retinoblastoma tumor suppressor protein.
Adv. Cancer Res.
,
64
:
25
-85,  
1994
.
3
Harbour J. W., Luo R. X., Dei Santi A., Postigo A. A., Dean D. C. Cdk phosphorylation triggers sequential intramolecular interactions that progressively block Rb functions as cells move through G1.
Cell
,
98
:
859
-869,  
1999
.
4
Lundberg A. S., Weinberg R. A. Functional inactivation of the retinoblastoma protein requires sequential modification by at least two distinct cyclin-cdk complexes.
Mol. Cell. Biol.
,
18
:
753
-761,  
1998
.
5
Bremner R., Cohen B. L., Sopta M., Hamel R. A., Ingles C. J., Gallie B. L., Phillips R. A. Direct transcriptional repression by pRB and its reversal by specific cyclins.
Mol. Cell. Biol.
,
15
:
3256
-3265,  
1995
.
6
Knudsen E. S., Wang J. Y. Differential regulation of retinoblastoma protein function by specific Cdk phosphorylation sites.
J. Biol. Chem.
,
271
:
8313
-8320,  
1996
.
7
Hinds P. W., Mittnacht S., Dulic V., Arnold A., Reed S. I., Weinberg R. A. Regulation of retinoblastoma protein functions by ectopic expression of human cyclins.
Cell
,
70
:
993
-1006,  
1992
.
8
Zhang H. S., Postigo A. A., Dean D. C. Active transcriptional repression by the Rb-E2F complex mediates G1 arrest triggered by p16INK4a, TGFβ, and contact inhibition.
Cell
,
97
:
53
-61,  
1999
.
9
Chew Y. P., Ellis M., Wilkie S., Mittnacht S. pRB phosphorylation mutants reveal role of pRB in regulating S phase completion by a mechanism independent of E2F.
Oncogene
,
17
:
2177
-2186,  
1998
.
10
Knudsen E. S., Wang J. Y. Dual mechanisms for the inhibition of E2F binding to RB by cyclin-dependent kinase-mediated RB phosphorylation.
Mol. Cell. Biol.
,
17
:
5771
-5783,  
1997
.
11
Knudsen E. S., Buckmaster C., Chen T. T., Feramisco J. R., Wang J. Y. Inhibition of DNA synthesis by RB: effects on G1/S transition and S-phase progression.
Genes Dev.
,
12
:
2278
-2292,  
1998
.
12
Knudsen K. E., Booth D., Naderi S., Sever-Chroneos Z., Fribourg A. F., Hunton I. C., Feramisco J. R., Wang J. Y., Knudsen E. S. RB-dependent S-phase response to DNA damage.
Mol. Cell. Biol.
,
20
:
7751
-7763,  
2000
.
13
Lukas J., Sorensen C. S., Lukas C., Santoni-Rugiu E., Bartek J. p16INK4a, but not constitutively active pRb, can impose a sustained G1 arrest: molecular mechanisms and implications for oncogenesis.
Oncogene
,
18
:
3930
-3935,  
1999
.
14
Lukas J., Herzinger T., Hansen K., Moroni M. C., Resnitzky D., Helin K., Reed S. I., Bartek J. Cyclin E-induced S phase without activation of the pRb/E2F pathway.
Genes Dev.
,
11
:
1479
-1492,  
1997
.
15
Kaelin W. G., Jr. Recent insights into the functions of the retinoblastoma susceptibility gene product.
Cancer Invest.
,
15
:
243
-254,  
1997
.
16
Bartek J., Bartkova J., Lukas J. The retinoblastoma protein pathway in cell cycle control and cancer.
Exp. Cell Res.
,
237
:
1
-6,  
1997
.
17
Sherr C. J. Cancer cell cycles.
Science (Wash. DC)
,
274
:
1672
-1677,  
1996
.
18
Bartkova J., Lukas J., Bartek J. Aberrations of the G1- and G1/S-regulating genes in human cancer.
Prog. Cell Cycle Res.
,
3
:
211
-220,  
1997
.
19
Reed S. I. Control of the G1/S transition.
Cancer Surv.
,
29
:
7
-23,  
1997
.
20
Sellers W. R., Kaelin W. G., Jr. Role of the retinoblastoma protein in the pathogenesis of human cancer.
J. Clin. Oncol.
,
15
:
3301
-3312,  
1997
.
21
Weinberg R. A. The retinoblastoma protein and cell cycle control.
Cell
,
81
:
323
-330,  
1995
.
22
Nevins J. R. E2F: a link between the Rb tumor suppressor protein and viral oncoproteins.
Science (Wash. DC)
,
258
:
424
-429,  
1992
.
23
Morris E. J., Dyson N. J. Retinoblastoma protein partners.
Adv. Cancer Res.
,
82
:
1
-54,  
2001
.
24
Harbour J. W., Dean D. C. Chromatin remodeling and Rb activity.
Curr. Opin. Cell Biol.
,
12
:
685
-689,  
2000
.
25
Harbour J. W., Dean D. C. The Rb/E2F pathway: expanding roles and emerging paradigms.
Genes Dev.
,
14
:
2393
-2409,  
2000
.
26
Dyson N. The regulation of E2F by pRB-family proteins.
Genes Dev.
,
12
:
2245
-2262,  
1998
.
27
Weintraub S. J., Prater C. A., Dean D. C. Retinoblastoma protein switches the E2F site from positive to negative element.
Nature (Lond.)
,
358
:
259
-261,  
1992
.
28
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
,
92
:
11544
-11548,  
1995
.
29
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 (Lond.)
,
375
:
812
-815,  
1995
.
30
Zhang H. S., Dean D. C. Rb-mediated chromatin structure regulation and transcriptional repression.
Oncogene
,
20
:
3134
-3138,  
2001
.
31
Zhang H. S., Gavin M., Dahiya A., Postigo A. A., Ma D., Luo R. X., Harbour J. W., Dean D. C. Exit from G1 and S phase of the cell cycle is regulated by repressor complexes containing HDAC-Rb-hSWI/SNF and Rb-hSWI/SNF.
Cell
,
101
:
79
-89,  
2000
.
32
Strobeck M. W., Knudsen K. E., Fribourg A. F., DeCristofaro M. F., Weissman B. E., Imbalzano A. N., Knudsen E. S. BRG-1 is required for RB-mediated cell cycle arrest.
Proc. Natl. Acad. Sci. USA
,
97
:
7748
-7753,  
2000
.
33
Meloni A. R., Smith E. J., Nevins J. R. A mechanism for Rb/p130-mediated transcription repression involving recruitment of the CtBP corepressor.
Proc. Natl. Acad. Sci. USA
,
96
:
9574
-9579,  
1999
.
34
Lai A., Kennedy B. K., Barbie D. A., Bertos N. R., Yang X. J., Theberge M. C., Tsai S. C., Seto E., Zhang Y., Kuzmichev A., Lane W. S., Reinberg D., Harlow E., Branton P. E. RBP1 recruits the mSIN3-histone deacetylase complex to the pocket of retinoblastoma tumor suppressor family proteins found in limited discrete regions of the nucleus at growth arrest.
Mol. Cell. Biol.
,
21
:
2918
-2132,  
2001
.
35
Dahiya A., Wong S., Gonzalo S., Gavin M., Dean D. C. Linking the rb and polycomb pathways.
Mol. Cell
,
8
:
557
-569,  
2001
.
36
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 (Lond.)
,
391
:
597
-601,  
1998
.
37
Luo R. X., Postigo A. A., Dean D. C. Rb interacts with histone deacetylase to repress transcription.
Cell
,
92
:
463
-473,  
1998
.
38
He S., Cook B. L., Deverman B. E., Weihe U., Zhang F., Prachand V., Zheng J., Weinstraub S. J. E2F is required to prevent inappropriate S-phase entry of mammalian cells.
Mol. Cell. Biol.
,
20
:
363
-371,  
2000
.
39
Qin X. Q., Livingston D. M., Ewen M., Sellers W. R., Arany Z., Kaelin W. G., Jr. The transcription factor E2F-1 is a downstream target of RB action.
Mol. Cell. Biol.
,
15
:
742
-755,  
1995
.
40
Ohtani K., DeGregori J., Nevins J. R. Regulation of the cyclin E gene by transcription factor E2F1.
Proc. Natl. Acad. Sci. USA
,
92
:
12146
-12150,  
1995
.
41
DeGregori J., Kowalik T., Nevins J. R. Cellular targets for activation by the E2F1 transcription factor include DNA synthesis- and G1/S-regulatory genes.
Mol. Cell. Biol.
,
15
:
4215
-4224,  
1995
.
42
Slansky J. E., Farnham P. J. Introduction to the E2F family: protein structure and gene regulation.
Curr. Top. Microbiol. Immunol.
,
208
:
1
-30,  
1996
.
43
Nevins J. R., DeGregori J., Jakoi L., Leone G. Functional analysis of E2F transcription factor.
Methods Enzymol.
,
283
:
205
-219,  
1997
.
44
Ishida S., Huang E., Zuzan H., Spang R., Leone G., West M., Nevins J. R. Role for E2F in control of both DNA replication and mitotic functions as revealed from DNA microarray analysis.
Mol. Cell. Biol.
,
21
:
4684
-4699,  
2001
.
45
Helin K. Regulation of cell proliferation by the E2F transcription factors.
Curr. Opin. Genet. Dev.
,
8
:
28
-35,  
1998
.
46
Dalton S. Cell cycle regulation of the human cdc2 gene.
EMBO J.
,
11
:
1797
-1804,  
1992
.
47
Geng Y., Eaton E. N., Picon M., Roberts J. M., Lundberg A. S., Gifford A., Sardet C., Weinberg R. A. Regulation of cyclin E transcription by E2Fs and retinoblastoma protein.
Oncogene
,
12
:
1173
-1180,  
1996
.
48
Knudsen K. E., Weber E., Arden K. C., Cavenee W. K., Feramisco J. R., Knudsen E. S. The retinoblastoma tumor suppressor inhibits cellular proliferation through two distinct mechanisms: inhibition of cell cycle progression and induction of cell death.
Oncogene
,
18
:
5239
-5245,  
1999
.
49
Qin X. Q., Chittenden T., Livingston D. M., Kaelin W. G., Jr. Identification of a growth suppression domain within the retinoblastoma gene product.
Genes Dev.
,
6
:
953
-964,  
1992
.
50
Muller H., Bracken A. P., Vernell R., Moroni M. C., Christians F., Grassilli E., Prosperini E., Vigo E., Oliner J. D., Helin K. E2Fs regulate the expression of genes involved in differentiation, development, proliferation, and apoptosis.
Genes Dev.
,
15
:
267
-285,  
2001
.
51
Polager S., Kalma Y., Berkovich E., Ginsberg D. E2Fs up-regulate expression of genes involved in DNA replication, DNA repair and mitosis.
Oncogene
,
21
:
437
-446,  
2002
.
52
Weinmann A. S., Yan P. S., Oberley M. J., Huang T. H., Farnham P. J. Isolating human transcription factor targets by coupling chromatin immunoprecipitation and CpG island microarray analysis.
Genes Dev.
,
16
:
235
-244,  
2002
.
53
Ren B., Cam H., Takahashi Y., Volkert T., Terragni J., Young R. A., Dynlacht B. D. E2F integrates cell cycle progression with DNA repair, replication, and G2/M checkpoints.
Genes Dev.
,
16
:
245
-256,  
2002
.
54
Sever-Chroneos Z., Angus S. P., Fribourg A. F., Wan H., Todorov I., Knudsen K. E., Knudsen E. S. Retinoblastoma tumor suppressor protein signals through inhibition of cyclin-dependent kinase 2 activity to disrupt PCNA function in S phase.
Mol. Cell. Biol.
,
21
:
4032
-4045,  
2001
.
55
Lukas J., Parry D., Aagaard L., Mann D. J., Bartkova J., Strauss M., Peters G., Bartek J. Retinoblastoma-protein-dependent cell-cycle inhibition by the tumour suppressor p16.
Nature (Lond.)
,
375
:
503
-506,  
1995
.
56
Koh J., Enders G. H., Dynlacht B. D., Harlow E. Tumour-derived p16 alleles encoding proteins defective in cell-cycle inhibition.
Nature (Lond.)
,
375
:
506
-510,  
1995
.
57
Plumb M., Stein J., Stein G. Influence of DNA synthesis inhibition on the coordinate expression of core human histone genes during S phase.
Nucleic Acids Res.
,
11
:
7927
-7945,  
1983
.
58
Kaelin W. G., Jr. Alterations in G1/S cell-cycle control contributing to carcinogenesis.
Ann. N. Y. Acad. Sci.
,
833
:
29
-33,  
1997
.
59
DeGregori J., Leone G., Ohtani K., Miron A., Nevins J. R. E2F-1 accumulation bypasses a G1 arrest resulting from the inhibition of G1 cyclin-dependent kinase activity.
Genes Dev.
,
9
:
2873
-2887,  
1995
.
60
Bartek J., Lukas J. Pathways governing G1/S transition and their response to DNA damage.
FEBS Lett.
,
490
:
117
-122,  
2001
.
61
Flatt P. M., Tang L. J., Scatena C. D., Szak S. T., Pietenpol J. A. p53 regulation of G2 checkpoint is retinoblastoma protein dependent.
Mol. Cell. Biol.
,
20
:
4210
-4223,  
2000
.
62
Harrington E. A., Bruce J. L., Harlow E., Dyson N. pRB plays an essential role in cell cycle arrest induced by DNA damage.
Proc. Natl. Acad. Sci. USA
,
95
:
11945
-11950,  
1998
.
63
Cho R. J., Huang M., Campbell M. J., Dong H., Steinmetz L., Sapinoso L., Hampton G., Elledge S. J., Davis R. W., Lockhart D. J. Transcriptional regulation and function during the human cell cycle.
Nat. Genet.
,
27
:
48
-54,  
2001
.
64
Notterman D. A., Alon U., Sierk A. J., Levine A. J. Transcriptional gene expression profiles of colorectal adenoma, adenocarcinoma, and normal tissue examined by oligonucleotide arrays.
Cancer Res.
,
61
:
3124
-3130,  
2001
.