Original Article

Oncogene (2009) 28, 651–661; doi:10.1038/onc.2008.419; published online 17 November 2008

There is a Retraction (20 October 2011) associated with this article.

This article was republished in Volume 30: Issue 42 (20 October 2011).

Requirement for chromatin-remodeling complex in novel tumor suppressor HIC1-mediated transcriptional repression and growth control

B Zhang1, K J Chambers1, D Leprince2, D V Faller1 and S Wang1

  1. 1Cancer Research Center, Boston University School of Medicine, Boston, MA, USA
  2. 2CNRS UMR 8161, Institut de, Biologie de Lille, Institut Pasteur de Lille, Lille Cedex, France

Correspondence: Professor S Wang, Cancer Research Center, R-906, Boston University School of Medicine, 715 Albany Street, Boston, MA 02118, USA. E-mail: sw184@bu.edu

Received 10 July 2008; Revised 19 September 2008; Accepted 10 October 2008; Published online 17 November 2008.

Top

Abstract

HIC1 is a newly discovered tumor suppressor and transcriptional repressor that is frequently silenced in human tumors. HIC1 protein expression has been linked to better outcomes in breast cancers. The molecular mechanism underlying HIC1-mediated transcriptional and growth suppression, and the relevant targets of HIC1-mediated transcriptional modulation, is currently unclear. We have identified an HIC1 DNA-binding site in E2F-responsive gene promoters and demonstrate that HIC1 targets E2F-responsive genes for transcriptional regulation and growth suppression. We and others have recently discovered that Brg1, a central component of the SWI/SNF chromatin-remodeling family, is required for the transcriptional regulation of multiple cell cycle control-related genes, including E2F-responsive promoters. We studied HIC1 interactions with, and dependence upon, Brg1 activity, and found that HIC1 can recruit Brg1 to E2F-responsive promoters and that its transcriptional repression of these genes is dependent upon Brg1. These data indicate that HIC1 is a central molecule in a novel mechanism controlling cell growth and that the disruption of this HIC1-mediated pathway may lead to abnormal cell proliferation and, ultimately, cancer.

Keywords:

HIC1, E2F1, Brg1, SWI/SNF, transcription, cell cycle

Top

Introduction

Brg1 is a central ATP-dependent chromatin-remodeling molecule of the SWI/SNF complex family that was found to be associated with transcriptional regulation of multiple genes critical to cell cycle control and growth regulation (Dunaief et al., 1994; Dasgupta et al., 2004; Wang et al., 2004; Giacinti and Giordano, 2006; Inayoshi et al., 2006; Zhang et al., 2007). Earlier studies revealed that Brg1 contains conserved SWI/SNF family domains that are essential for cell growth and transcriptional activation. Other evidences have demonstrated that Brg1 is required for the activation of certain genes, and Brg1 was initially considered to be a transcription activator (Dunaief et al., 1994).

Subsequent investigations, however, led to the discovery of Brg1 involvement in transcriptional repression. Brg1 was found to associate with Rb and its family members, and to participate in the repression of E2F-responsive genes and growth suppression (Dunaief et al., 1994; Betz et al., 2002). We have recently established that Brg1 is also required for the transcriptional repression mediated by prohibitin (Wang et al., 2004). The potential importance of Brg1 in the development of cancer, including breast cancer, was further supported by the discovery of mutations of Brg1 in many cancer cells (Strobeck et al., 2001).

The molecular mechanisms of Brg1-mediated transcriptional repression have been recently investigated. We and others have discovered that the Brg1 and Brm ATPase components of SWI/SNF complexes are recruited by repressors of E2F to E2F-responsive promoters and are required for the repression of E2F-mediated transcription and growth control (Wang et al., 2002b, 2004; Banine et al., 2005).

To further investigate the precise molecular mechanism involved in the Brg1-mediated transcriptional regulation, we attempted to identify additional proteins that may associate with Brg1 and discovered that a novel tumor suppressor, HIC1, interacts with Brg1. HIC1 is a newly discovered tumor suppressor and transcriptional repressor (Bertrand et al., 2004) that is found to be silenced in certain cancers, due, in many cases, to the methylation of its promoter (Chen et al., 2005; Aggerholm et al., 2006; Britschgi et al., 2006). Recent data indicate that HIC1 collaborates with p53 and plays critical roles in the regulation of cell growth and death (Chen and Baylin, 2005). Disruption of HIC1 predisposes mice to a gender-dependent spectrum of malignant tumors (Chen et al., 2003). A clinical study demonstrated that the expression of HIC1 is associated with a good outcome in human breast cancers (Nicoll et al., 2001), suggesting the potential importance of HIC1 as a target for diagnosis, prognosis and treatment of breast cancer. The molecular mechanism of HIC1-mediated transcriptional repression is not yet defined. A screen using synthesized oligos identified a potential HIC1 consensus sequence (Pinte et al, 2004b).

The genes that are targeted by the HIC1 tumor suppressor have not been fully defined. A recent study indicated that HIC1 binds to the promoter of the class III histone deacetylase Sirt1 at a GGCA motif and regulates its transcription, which in turn modulates p53-dependent DNA-damage responses (Chen et al., 2005). We show here that Brg1 is indeed required for HIC1-mediated transcriptional regulation of Sirt1. Upon discovering that HIC1 interacts with Brg1, a known regulator of transcription and growth, we speculated that the HIC1 tumor suppressor may require the SWI/SNF family to target the cell cycle control machinery through E2F. We show here that HIC1 targets E2F-responsive genes for transcriptional regulation and growth suppression, and this repression is dependent upon Brg1. Depletion of HIC1 by small interfering RNA (siRNA) resulted in the release of growth arrest. In light of the recent studies suggesting that G1–S phase cell cycle transit does not appear to be regulated by Rb under normal circumstances (Takahashi et al., 2000), alternative mechanisms in the normal cell cycle control therefore remain to be discovered (Stevaux et al., 2002). Our novel finding of E2F regulation by HIC1 in collaboration with SWI/SNF family members may represent one new level of control at this critical transition point.

Top

Results

HIC1 interacts with SWI/SNF family members

Earlier studies demonstrated the involvement of SWI/SNF complex in the Rb-mediated E2F regulation (Strobeck et al., 2001). We recently discovered that Brg1 also participates in the regulation of transcription by E2F through an Rb-independent mechanism (Fusaro et al., 2002; Wang et al., 2002b, 2004). We extended our studies to investigate the molecular basis of Brg1-mediated growth suppression and transcriptional repression by identifying additional Brg1-interacting factors. HIC1 was of particular interest because of its newly discovered roles in transcriptional regulation and its potential links with cancer development. Furthermore, HIC1 expression is associated with good outcomes in breast cancer and we have recently reported that Brg1 is required in tamoxifen-induced growth suppression of breast cancer cells (Nicoll et al., 2001; Wang et al., 2004). We first tested whether HIC1 can physically interact with Brg1. HIC1-associated protein complexes were immunoprecipitated from human primary fibroblast cell extracts (Wang et al., 1998) with an anti-HIC1 antibody or control (anti-Myb) antibody, followed by immunoblot (IB) analysis using an anti-Brg1 antibody. As shown in Figure 1a, Brg1 was detected in the HIC1 immunoprecipitates, indicating the ability of these molecules to associate. A reciprocal immunoprecipitation (IP)/IB was also performed that confirmed the association (data not shown). To further verify this novel interaction, IP/IB was also performed in MCF7 cells (which do not express HIC1) transfected with a Flag-tagged HIC1, the expression of which was confirmed by IB analysis using an anti-Flag antibody (Figure 1b). The extracts of the Flag-HIC1-transfected MCF7 cells were then IP using anti-tubulin (control) or anti-Flag antibody followed by IB analysis using anti-Brg1 or control (anti-Myb) antibody. Brg1 was detected in the Flag-HIC1 immunoprecipitates, confirming the association of these molecules (Figure 1c). The reciprocal IP/IB was also performed, which confirmed the interactions (data not shown).

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

HIC1 interacts with Brg1. (a) Extracts of HSF8 cells were analysed by immunoprecipitation (IP), PAGE and immunoblotting (IB) using the indicated antibodies. CE represents cell extract before IP. (b) Extracts from MCF7 cells (control) or MCF7 cells transfected with Flag-tagged HIC1 (Flag-HIC1) were analysed by PAGE and IB, using the indicated antibodies. (c) Cell extracts of MCF7 cells transfected with Flag-HIC1 shown in panel b were immunoprecipitated by the indicated antibodies, followed by PAGE and IB using the antibodies shown. (d) The 35S-labeled proteins indicated were incubated with GST or the indicated fusion proteins, followed by pull-down analysis using standard protocols. RL represents the 35S-labeled proteins without binding. GST, glutathione S-transferase; PAGE, polyacrylamide gel electrophoresis.

Full figure and legend (66K)

To further confirm the association between HIC1 and Brg1 and determine whether it is direct, we performed an in vitro glutathione S-transferase (GST) pull-down assay, following protocols we have reported earlier (Wang et al., 1998, 1999a, 2002a). As shown in Figure 1d, a direct association between HIC1 and GST-Brg1 was detected. There was no association between HIC1 and the control GST or GST-DP1 (an E2F1-binding partner) protein, demonstrating the specificity of the association. The GST pull-down assay was repeated in the reciprocal fashion, which confirmed the association (Figure 1d).

We and others have recently established that Brg1 is a novel regulator of E2F activity (Dunaief et al., 1994; Fusaro et al., 2002; Wang et al., 2004). We therefore determined whether HIC1, as a transcriptional regulator and binding partner of Brg1, can also modulate E2F-mediated transcription. Reporter assays in HSF8 cells were performed to address this question. Transfection of HIC1 (expression confirmed by IB analysis, data not shown) repressed the positive control, a luciferase reporter the activity of which is controlled by HIC1-binding sites (Pinte et al., 2004a; Figure 2a); ectopic expression of HIC1 also effectively repressed an E2F-responsive luciferase reporter, to a similar extent as did the established repressors of E2F-mediated transcription (Rb and prohibitin; Figure 2a). HIC1 did not affect the control cFos gene activity, as shown in Figure 2a, confirming the specificity of its actions.

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

HIC1 represses E2F-mediated transcription of endogenous genes. (a) HSF8 cells were transfected with empty vector (control) or the indicated expression vector plasmids. The cells were collected 72h after transfection and the transcriptional activity of endogenous Sirt1, E2F1 and cFos (negative control) genes was analysed by quantitative RT–PCR, using standard protocols of the supplier. The experiments were repeated four times, yielding comparable results, and error bars shown represent the s.e.m. from all four assays. β-Actin expression was examined in each experiment as an internal control and used for normalization. The relative expression in the control non-transfected cells was assigned a value of 10. (b) The occupancy of HIC1 at the E2F1 and cFos promoters in unsynchronized actively growing HSF8 cells was analysed by ChIP assay with quantitative PCR, using an anti-HIC antibody (or anti-tubulin antibody as a specificity control). Relative quantitation of input DNA (not immunoprecipitated) is shown as ‘total.’ All the experiments were repeated four times and the averages are shown, with error bars representing s.e.m. ChIP, chromatin immunoprecipitation; RT–PCR, reverse transcriptase PCR.

Full figure and legend (80K)

The molecular mechanism of HIC1-mediated transcriptional repression is not yet elucidated. A recent investigation identified a potential HIC1 consensus DNA-binding sequence (GGCA) (Pinte et al, 2004b). The functional importance of this consensus sequence is further suggested by recent reports that demonstrated that the SIRT1 gene is targeted by HIC1 at this consensus site (Chen et al., 2005). To test whether the repression of E2F-responsive transcription by HIC1 involves a direct association with the gene promoter, we first surveyed the human E2F1 promoter sequence. A sequence matching the consensus HIC1-binding site (GGCA) was identified on the E2F1 promoter (position 1644; GenBank accession no. U47675). We therefore tested whether HIC1 can indeed bind to the E2F1 promoter, using a chromatin immunoprecipitation (ChIP) assay (Wang et al., 2004). Unsynchronized, actively growing HSF8 cells were treated to cross-link DNA and protein, followed by IP with anti-HIC1 antibody or control antibody. The cross-linking was then reversed and the DNA recovered from the immunoprecipitates was analysed by PCR, using a pair of primers that span the E2F/HIC1 consensus site. As shown in Figure 2b, an amplified product was detected in the HIC1 immunoprecipitates of HSF8 cells for the E2F1 promoter, but not for the cFos promoter (a non-E2F-responsive promoter, used here as a control), indicating that HIC1 can indeed associate specifically with an E2F-responsive promoter. A ChIP assay on the same E2F promoter using an irrelevant control antibody (anti-tubulin) failed to generate any product, confirming the specificity of the ChIP assay (Figure 2b).

Brg1 is required for HIC1-mediated transcriptional repression

We next examined whether Brg1 might participate in HIC1-mediated transcriptional regulation. The first studies were carried out in SW13 cells, which lack Brg1 (Wang et al., 2002b). Expression of HIC1 in the Brg1-negative SW13 cells failed to repress the activity of either the endogenous Sirt1 (an HIC1-responsive gene) or E2F1 (an E2F-responsive gene) gene (Figure 3a). Co-expression of Brg1 together with HIC1, however, induced strong repression of both genes (Figure 3a), which correlated with the recruitment of HIC1 and Brg1 to the promoters of these genes (Figure 3b, ChIP, HIC1, Brg1). Expression of HIC1 and Brg1 after transfection in these studies was verified by IB (not shown). Transfection of Brg1 alone, without HIC1, produced only modest repression, suggesting that Brg1 is required for HIC1-mediated transcriptional repression (Figure 3).

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Brg1 is required for HIC1-mediated transcriptional repression. (a) Brg1-negative SW13 cells were transfected with an empty vector (control) or the indicated expression vector plasmids. The cells were collected 72h after transfection and the expression of the Sirt1, E2F1 and cFos (negative control) genes was analysed by quantitative RT–PCR. The relative expression of these genes was calculated and normalized to the expression of β-actin. Error bars represent the s.e.m. from four independent assays. (b) Promoter association of Brg1 and HIC1 was examined by quantitative ChIP and double-ChIP assays, using the indicated antibodies. Relative association, as reflected by the quantitative PCR values, was calculated and normalized to the input DNA results. Error bars represent s.e.m. from four independent assays. ChIP, chromatin immunoprecipitation; RT–PCR, reverse transcriptase PCR.

Full figure and legend (105K)

To test whether the HIC1 and Brg1 co-occupy these HIC1-responsive promoters, we performed a sequential double-ChIP assay that is a slight modification of the established ChIP protocol we have reported (Wang et al., 2002b, 2004), on the same endogenous gene promoters, using an anti-HIC1 antibody for the first round of IP and an anti-Brg1 antibody for the second round of IP, followed by amplification and quantitation. The reciprocal double-ChIP assay was also performed. These assays of promoter occupancy indicated co-occupancy by HIC1 and Brg1 on these HIC1-responsive promoters (data not shown).

To further verify the necessity of Brg1 in HIC1-mediated transcriptional repression, the activity of the endogenous Sirt1 and E2F1 gene promoters was evaluated after repression of Brg1, using siRNA knockdown and the dominant-negative Brg1 strategy we have established earlier (Wang et al., 2002b, 2004). As shown in Figure 4a, the activity of both the endogenous Sirt1 and E2F1 genes was effectively repressed when HIC1 was overexpressed. This repression, however, was reversed by the co-transfection of a dominant-negative Brg1 or a Brg1 siRNA, which depleted Brg1 in the cells (knockdown confirmed by IB shown in Figure 4b, and further verified by quantitative reverse transcriptase PCR (RT–PCR) analysis, Figure 4c), confirming the necessity of Brg1 in the HIC1-mediated transcriptional repression of endogenous HIC1-responsive promoters. HIC1 expression failed to affect a non-E2F-responsive promoter (cFos) (Figure 4a). The experiment was repeated using three different siRNAs specific for the Brg1 gene, all of which generated comparable results (data from one siRNA is shown in Figure 4).

Figure 4.
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Knockdown of Brg1 abrogates HIC1-mediated transcriptional repression. (a) HSF8 cells were transfected with the empty control vector, HIC1, a dominant-negative Brg1 vector or a vector-based Brg1 siRNA, followed by quantitation of transcripts to measure the activity of the endogenous HIC1-responsive promoters, and the non-responsive control promoter (cFos), as indicated. (b) The cells were analysed by immunoblot using the indicated antibodies, which confirmed the depletion of Brg1 by the transfected siRNA. (c) Quantitative RT–PCR was performed for the genes indicated, which further confirmed the depletion of Brg1 by the siRNA. As a control, the tubulin gene was not affected by the Brg1 siRNA, verifying the specificity. A second Brg1-specific siRNA, purchased from Santa Cruz, produced similar results (data not shown). RT–PCR, reverse transcriptase PCR; siRNA, small interfering RNA.

Full figure and legend (75K)

As shown in Figure 3, HIC1 and Brg1 both associate with the endogenous Sirt1 and E2F1 gene promoters in SW13 cells. We and others have established earlier that Brg1 and its SWI/SNF co-regulatory components are recruited to targeted promoters by certain transcription factors (Wang et al., 2002b, 2004). We therefore tested whether the requirement for Brg1 in HIC1-mediated transcriptional repression correlates with its recruitment by HIC1 at responsive promoters, and our results show that promoter occupancy of HIC1 and Brg1 was induced significantly when HIC1 was ectopically expressed by transfection and the transcriptional activity of these responsive genes was coincidently repressed, suggesting that HIC1 may recruit Brg1 for transcriptional repression (Figure 5a). Ectopic expression of HIC1 did not affect SP1 association with the cFos promoter (control), confirming the specificity (Figure 5b).

Figure 5.
Figure 5 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

HIC induces the recruitment of Brg1 to responsive genes. (a) HSF8 cells were transfected with empty vector (control), an HIC1 expression vector and/or a vector-based Brg1 siRNA. The cells were collected 72h after transfection and analysed by quantitative ChIP assay of the Sirt1, E2F1 or cFos (control) gene, using the indicated antibodies. (b) As a positive control, SP1 association with the cFos promoter in the same cells explained above was analysed using quantitative ChIP assay. ChIP, chromatin immunoprecipitation; siRNA, small interfering RNA.

Full figure and legend (72K)

To further elucidate the role of a Brg1-containing complex in HIC1-mediated transcriptional repression, we investigated the necessity of HIC1 in the recruitment of Brg1 to responsive promoters. As shown in Figures 6a and b, transfection of a vector-based siRNA depleted HIC1 from HSF8 cells. This HIC1 depletion induced an increase in transcripts from the endogenous E2F1 gene (Figure 6c). This release of transcriptional repression in a responsive gene when HIC1 was depleted correlated with decreases in Brg1 occupancy at the promoter of the E2F1 gene (Figure 6c). The experiments were repeated on another HIC1-responsive gene, the Sirt1, which produced similar results (data not shown). Depletion of HIC1 using a second independent siRNA against HIC1 (depletion of HIC1 again confirmed by both IB and quantitative RT–PCR) produced similar results (data not shown).

Figure 6.
Figure 6 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

HIC1 is required for the recruitment of Brg1 to responsive genes. (a) HSF8 cells were transfected with empty vector (control), vector-based HIC1 siRNA or control siRNA. The cells were collected 72h after transfection followed by immunoblot analysis using the indicated antibodies (tubulin as control), which confirmed the depletion of HIC1 by siRNA. (b) The cells were analysed by quantitative RT–PCR for the indicated genes (tubulin as control), which further verified the depletion of HIC1 by siRNA. (c) Quantitative ChIP assay using the indicated antibodies and quantitative RT–PCR were performed for the E2F1 gene. ChIP, chromatin immunoprecipitation; RT–PCR, reverse transcriptase PCR; siRNA, small interfering RNA.

Full figure and legend (85K)

HIC1 is required for growth arrest

A recent report linked expression of the HIC1 protein with a good clinical outcome in breast cancer, suggesting a role for HIC1 in growth control (Nicoll et al., 2001). The regulation of E2F transcriptional activity by HIC1 demonstrated above, along with a report showing the regulation of p53 by HIC1, further suggested a potential role for HIC1 in cell cycle control. We therefore tested the activity of HIC1 in growth regulation. Human primary fibroblast cells (HSF8) were stably transfected with either vector-based control siRNA or a specific HIC1 siRNA that depleted cellular HIC1 (confirmed by IB analysis and quantitative RT–PCR, data not shown). These transfected HSF8 cells were exposed to serum starvation, achieved by withdrawing serum for 72h. The cell cycle profiles of these cells were then analysed by propidium iodide staining and flow cytometry. Serum starvation induced a G0/G1 arrest in the control cells (Figure 7, compare panels 1, 2 and 3) as expected, but this growth arrest did not occur in those HSF8 cells in which HIC1 had been depleted, indicating the necessity of HIC1 in serum deprivation-induced growth arrest at the restriction point (Figure 7, panel 4).

Figure 7.
Figure 7 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

HIC1 is required for serum deprivation-induced growth arrest. HSF8 cells were stably transfected with an empty vector, or control siRNA, or HIC1 siRNA. The cells were serum starved for 72h and their cell cycle profiles were analysed by propidium iodide staining and flow cytometry. The experiments were independently repeated four times, yielding similar results. Profiles from one representative experiment are shown, with the percentages of cells in each phase of the cell cycle derived from the four independent experiments indicated. siRNA, small interfering RNA.

Full figure and legend (69K)

To further verify the necessity of HIC1 in the induction of growth suppression, a colony-formation assay was performed. Serum starvation dramatically repressed the number of colonies generated by the control cells (transfected with an empty vector or a control siRNA), demonstrating the expected growth-inhibitory effect (Table 1). However, as shown in Table 1, this suppression of growth by serum starvation was released when the level of HIC1 protein was suppressed by transfection of HIC1 siRNA (suppression of HIC1 levels in these cells was confirmed by IB analysis and by quantitative RT–PCR, data not shown).


As demonstrated above, HIC1 associates with the E2F1 promoter. As HIC1 is required for the growth suppression induced by serum deprivation (Table 1 and Figure 7), we hypothesized that the recruitment of HIC1 to responsive promoters may be regulated in a cell cycle-specific manner. To test this possibility, we prepared a G0 cell population and a late G1 population of HSF8 cells by serum starvation and release, respectively (Wang et al., 1998). E2F1 and HIC1 gene expression was assayed by quantitation of transcript levels. HIC1 association with the E2F1 promoter was also analysed by ChIP assay. As shown in Figure 8a, a strong signal for HIC1 occupancy was detected in the G0 stage, a time when E2F1 promoter activity is repressed, but not in late G1 phase cells, when the E2F1 promoter is transcriptionally active (confirmed by transcript levels in Figure 8b), indicating the cell cycle-specific recruitment of HIC1. Conversely, recruitment of E2F1 to the promoter was detected in the late G1, but not the G0, cell population, as expected (Figure 8a). Quantitative RT–PCR revealed that the level of expression of HIC1 is also cell cycle regulated and correlated temporally with its pattern of association with E2F-responsive promoters (Figure 8b). As HIC1 requires Brg1 for the repression of E2F1 gene expression, and the two molecules co-occupy HIC1-responsive promoters simultaneously, we tested for co-occupancy during different stages of the cell cycle by sequential (double) ChIP assay. As described above, the first IP was performed using an anti-HIC1 antibody, followed by a second round of IP on the HIC1 immunoprecipitates, using an anti-Brg1 antibody, followed by amplification and quantitation. As shown in Figure 8a, a signal was detected in G0 cells by the double-ChIP assay, but not in the late G1 population, indicating HIC1/Brg1 co-occupancy on the E2F1 promoter at the G0 stage of the cell cycle. We further confirmed this result by the reciprocal IPs, using an anti-Brg1 antibody IP first, followed by an anti-HIC1 IP and amplification (Figure 8a). Interestingly, recruitment of Brg1 to the promoter was also detected in the late G1 cell population, although no HIC1/Brg1 double occupancy was observed at that stage of the cell cycle, suggesting that Brg1, in the absence of HIC1, also participates in the transcriptional activation of the E2F1 gene. Double-ChIP assay for E2F1 and HIC1 was also performed and revealed that these two molecules never co-occupied the E2F1 promoter. A control experiment, demonstrating the expected co-occupancy by E2F1 and DP1 in late G1 phase, validated the double-ChIP assay (Figure 8).

Figure 8.
Figure 8 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Cell cycle dependency of HIC1 versus E2F promoter occupancy. (a) HSF8 cells were serum starved for 72h (G0), and one portion was then treated with serum for 12h (late G1). The E2F1 and HIC1 genes in these HSF8 cells were analysed for transcription factor occupancy by quantitative ChIP or double-ChIP assays, using the indicated antibodies. (b) The E2F1 and HIC1 genes in these HSF8 cells were analysed for activity by quantitation of transcripts. The experiments were repeated four times, which yielded comparable results, and the error bars shown represent the s.e.m. from all four assays. ChIP, chromatin immunoprecipitation.

Full figure and legend (70K)

Top

Discussion

The newly discovered tumor suppressor HIC1 serves as a transcriptional repressor (Pinte et al., 2004b; Chen et al., 2005) and growth repressor, but the molecular mechanisms involved in HIC1-mediated transcriptional repression or growth control have not yet been fully defined. Recent reports indicate that HIC1 regulates p53 through the repression of Sirt1 transcription (Chen and Baylin, 2005; Chen et al., 2005), suggesting the possibility that HIC1 might suppress cell proliferation through control of p53 activity. It has been well established that E2F collaborates with p53 in growth control, and a recent investigation revealed that Sirt1 modulates the E2F pathway by targeting the Rb tumor suppressor family (Wong and Weber, 2007). Ever since the seminal discovery of the functional interaction between the Rb tumor suppressor and the E2F transcriptional factors, the role of E2F has become a central focus of study regarding cell cycle regulation and cancer development (Nevins, 2001). Despite the tremendous advances made in this field, however, the precise mechanisms involved in the regulation of E2F transcriptional activity are not yet fully elucidated. Earlier studies indicated the important roles of Rb and its family members in E2F regulation. Yet, more recent technological advances allowing promoter-specific examination of regulatory complexes, such as ChIP assays, have produced findings that cast doubt on the traditional model in which Rb plays the central role in E2F regulation. For example, Rb is not found to associate with E2F-responsive promoters during normal cell cycle progression, and has therefore been theorized to serve as an E2F repressor only in specialized stress situations (Takahashi et al., 2000). These new developments inevitably raise the question as to what molecule(s) serves as the regulator of E2F-responsive genes during normal cell cycle progression, if not Rb. The findings elucidated in this study indicate that the novel tumor suppressor HIC1 could serve as the alternative repressor of E2F-responsive genes.

As discussed above, HIC1 is a transcriptional regulator that might cooperate with p53 for growth suppression (Chen et al., 2004, 2005; Britschgi et al., 2006). The data presented in this report reveal that HIC1 can also regulate E2F-responsive genes directly and repress cell growth. Significantly, depletion of HIC1 resulted in the abrogation of normal G1/S arrest, demonstrating the strict necessity and fundamental role of HIC1 in growth suppression in response to growth factor deprivation.

The molecular mechanism underlying HIC1-mediated transcriptional regulation is not yet clear. Recent biochemical studies identified HIC1-binding consensus sites (GGCA) (Pinte et al, 2004b; Britschgi et al., 2006). These potential HIC1-binding sites were later discovered in the Sirt1 promoter, which was recently established to be HIC1 responsive (Chen et al., 2005; Britschgi et al., 2006). Our results now establish that the E2F1 gene, which encodes an HIC1 consensus binding site in its promoter, is also targeted by HIC1. This discovery suggests a novel critical regulatory mechanism in E2F regulation. We have initiated studies (beyond the scope of this report) to further define this repressive mechanism.

The requirement for Brg1 in HIC1-mediated transcriptional regulation and growth suppression demonstrates the critical role of chromatin-remodeling molecules in this machinery. ChIP assay results (shown in Figure 3) indicate that HIC1 may recruit Brg1 only after binding to the responsive promoters; this result suggests a novel regulatory mechanism that achieves transcriptional regulation by temporally regulated association between HIC1 and its co-repressors on targeted promoters. Brg1 is the central core catalytic molecule in SWI/SNF complexes, which contain up to 10 other different proteins, varying according to their activity and specific promoter occupancy. The precise mechanisms involved in the SWI/SNF-mediated transcriptional regulation have been under intensive investigation since the discovery of their involvement in E2F regulation. Brg1 and its related molecules were initially discovered for their contribution to transcriptional activation. We and others, however, have established that this group of chromatin-remodeling molecules is also involved in transcriptional repression (Fusaro et al., 2002; Wang et al., 2004). Interestingly, we have recently reported that Brg1 and its related molecules can perform as either suppressors or activators of transcription on the same promoter, depending upon the composition of the complexes they form, suggesting their possible role as a mode-changing mechanism to switch promoters between activated and repressed states. For example, Brg1 was found to be recruited and required for the activation of ER-mediated transcription in response to estrogen. Paradoxically, Brg1 was also recruited and required for the repression of the ER-mediated transcription in response to estrogen antagonists (Zhang et al., 2007). We found here that Brg1 occupies the E2F1 promoter during both G0 and G1 stages of cell cycle, times when the promoter is either repressed or active, respectively (shown in Figure 8). This finding further indicates the potential involvement of Brg1 in diverse and opposite aspects of the transcriptional regulation of E2F1. The molecular mechanisms underlying these dichotomous actions of Brg1 are currently under investigation.

The importance of the E2F pathway in cancer development has been well established. However, the mechanisms linking cancer development and E2F-controlled cell cycle progression have not been fully elucidated. The discoveries presented in this report link the novel tumor suppressor HIC1 with the regulation of the E2F node. HIC1 was initially identified based on its inactivation in multiple cancers (Melki et al., 1999; Nicoll et al., 2001; Waha et al., 2004; Aggerholm et al., 2006). Recent studies demonstrated a clear correlation between HIC1 expression and good outcomes of human breast cancers. Our data indicate that HIC1 is a critical molecule required for appropriate cellular growth suppression in response to physiological antiproliferative signals such as growth factor availability, and may mediate this cell cycle control through its regulation of the E2F node.

Top

Materials and methods

Transfections

HSF-8, MCF7 and SW13 cells were transfected by calcium phosphate precipitation following standard protocols. A total of 2μg of the plasmids were used for a 10cm dish, unless noted otherwise. Plasmids pCDNA3-Raf-1 (Wang et al., 1999b), E2Luc, pSVRb (Wang et al., 1999b) and pCDNA3FlagHIC1 (wild-type HIC1) have been described earlier (Guerardel et al., 1999; Deltour et al., 2002; Bertrand et al., 2004; Lefebvre et al., 2004; Pinte et al., 2004b; Chopin and Leprince, 2006; Stankovic-Valentin et al., 2006). Vector-based pRNAT-U6.1/Neo-HIC1-siRNA and control vector were purchased from GenScript Corporation (Piscataway, NJ, USA) and transfected into cells according to the supplier's protocols. Brg1 siRNA has been described earlier (Fusaro et al., 2002; Wang et al., 2004). Sequences used for HIC1 siRNA were (1) ACGTGATCATCGTGGTGCAGA and (2) CAACCUGUACGUGUGCAUU.

Immunoprecipitation and immunoblotting

Immunoprecipitation and IB procedures have been described earlier (Fusaro et al., 2002; Wang et al., 2004). Flag antibody (CAT no. F-3165) was purchased from Sigma (St Louis, MO, USA). Prohibitin antibody (CAT no. 11-14-10) was purchased from Lab Vision Products, Thermo Fisher Scientific (Fremont, CA, USA). The following antibodies were purchased from Santa Cruz Biotechnology ( Santa Cruz, CA, USA), Brg1 (CAT no. sc-17796), HIC1 (CAT no. sc-28703) and tubulin (CAT no. sc-8035).

GST pull-down assay

GST pull-down assay was described earlier (Wang et al., 1998). GST-HIC1 was produced by cloning the synthesized full-length human HIC1 into the pGEX-4T vector (GE Healthcare Life Sciences Inc.; CAT no. 27-4580-01) at BamHI and XhoI sites (confirmed by sequence analysis). Full-length human Brg1, pE2F1 and DP1 vectors were described earlier (Wang et al., 1999a; Zhang et al., 2007).

Quantitative PCR and RT–PCR

Experiments were performed by following protocols we have recently reported (Dai et al., 2007, 2008). Briefly, quantitative PCR was performed using Applied Biosystems Power SYBR Green PCR master mix (for ChIP assay) and Power SYBR Green RT−PCR reagents kit (for RT–PCR). DNA quantities were measured by 7500 Fast Real-time PCR system (Applied Biosystems, Foster City, CA, USA) by the 2ΔΔCt cycle threshold method. Results were normalized to account for variation in inputs and irrelevant controls to determine the relative quantities. All the experiments were repeated four times. Quantitative PCR conditions were as follows: 95°C, 10min; 59°C, 1min; 60°C, 1min, for 40 cycles. The following RT–PCR primers were purchased from Santa Cruz: HIC-1(h)-PR (CAT no. sc-37712-PR); Brg-1(h)-PR (CAT no. sc-29827-PR); SIRT1(h)-PR (CAT no. sc-40986-PR); and E2F-1(h)-PR (CAT no. sc-29297).

ChIP assays

The detailed protocol for the ChIP assay of the endogenous E2F1 gene promoter was described earlier (Wang et al., 2002b, 2004; Dai et al., 2007, 2008). Primers for the cFos and β-actin genes (control) were described before (Dai et al., 2008). Double-ChIP assay was performed by same procedure, except that a second round of IP was performed on the immunoprecipitated products of the first IP, before quantitative PCR amplification.

PCR primers for ChIP assay on E2F1 promoter: 5′-GCAGCCAATTGTGGCGGC-3′; 5′-GACGCTCACGGCCCG-3′.

PCR primers for ChIP assay on Sirt1 promoter: 5′-TTGGTTGCCTAAAGTCACGCAGGT-3′; 5′-TGCCTGGAGCACAGCGTTTCTATC-3′

Colony-formation assay

The protocol for the colony-formation assay was described earlier (Fusaro et al., 2002; Wang et al., 2004). Briefly, HSF8 cells were cultured in 0% serum (serum starvation) or 10% fetal bovine serum, in RPMI. The cells were transfected with the vectors indicated in the text and transfectants were selected by culture in G418 (100ng/ml) for 14 days. The cells were then fixed and stained, and the colonies with 20 or more cells were scored.

Top

References

  1. Aggerholm A, Holm MS, Guldberg P, Olesen LH, Hokland P. (2006). Promoter hypermethylation of p15, hic1, cdh1, and er is frequent in myelodysplastic syndrome and predicts poor prognosis in early-stage patients. Eur J Haematol 76: 23–32. | Article | PubMed | ISI | ChemPort |
  2. Banine F, Bartlett C, Gunawardena R, Muchardt C, Yaniv M, Knudsen ES et al. (2005). Swi/snf chromatin-remodeling factors induce changes in DNA methylation to promote transcriptional activation. Cancer Res 65: 3542–3547. | Article | PubMed | ISI | ChemPort |
  3. Bertrand S, Pinte S, Stankovic-Valentin N, Deltour-Balerdi S, Guerardel C, Begue A et al. (2004). Identification and developmental expression of the zebrafish orthologue of the tumor suppressor gene hic1. Biochim Biophys Acta 1678: 57–66. | PubMed | ChemPort |
  4. Betz BL, Strobeck MW, Reisman DN, Knudsen ES, Weissman BE. (2002). Re-expression of hsnf5/ini1/baf47 in pediatric tumor cells leads to g1 arrest associated with induction of p16ink4a and activation of rb. Oncogene 21: 5193–5203. | Article | PubMed | ISI | ChemPort |
  5. Britschgi C, Rizzi M, Grob TJ, Tschan MP, Hugli B, Reddy VA et al. (2006). Identification of the p53 family-responsive element in the promoter region of the tumor suppressor gene hypermethylated in cancer 1. Oncogene 25: 2030–2039. | Article | PubMed | ChemPort |
  6. Chen W, Cooper TK, Zahnow CA, Overholtzer M, Zhao Z, Ladanyi M et al. (2004). Epigenetic and genetic loss of hic1 function accentuates the role of p53 in tumorigenesis. Cancer Cell 6: 387–398. | Article | PubMed | ISI | ChemPort |
  7. Chen WY, Baylin SB. (2005). Inactivation of tumor suppressor genes: choice between genetic and epigenetic routes. Cell Cycle 4: 10–12. | PubMed | ChemPort |
  8. Chen WY, Wang DH, Yen RC, Luo J, Gu W, Baylin SB. (2005). Tumor suppressor hic1 directly regulates sirt1 to modulate p53-dependent DNA-damage responses. Cell 123: 437–448. | Article | PubMed | ISI | ChemPort |
  9. Chen WY, Zeng X, Carter MG, Morrell CN, Chiu Yen RW, Esteller M et al. (2003). Heterozygous disruption of hic1 predisposes mice to a gender-dependent spectrum of malignant tumors. Nat Genet 33: 197–202. | Article | PubMed | ISI | ChemPort |
  10. Chopin V, Leprince D. (2006). [Chromosome arm 17p13.3: Could hic1 be the one?]. Med Sci (Paris) 22: 54–61. | PubMed |
  11. Dai Y, Ngo D, Forman LW, Qin DC, Jacob J, Faller DV. (2007). Sirtuin 1 is required for antagonist-induced transcriptional repression of androgen-responsive genes by the androgen receptor. Mol Endocrinol 21: 1807–1821. | Article | PubMed | ChemPort |
  12. Dai Y, Ngo D, Jacob J, Forman LW, Faller DV. (2008). Prohibitin and the swi/snf atpase subunit brg1 are required for effective androgen antagonist-mediated transcriptional repression of androgen receptor-regulated genes. Carcinogenesis 29: 1725–1733. | Article | PubMed | ChemPort |
  13. Dasgupta P, Sun J, Wang S, Fusaro G, Betts V, Padmanabhan J et al. (2004). Disruption of the rb–raf-1 interaction inhibits tumor growth and angiogenesis. Mol Cell Biol 24: 9527–9541. | Article | PubMed | ISI | ChemPort |
  14. Deltour S, Pinte S, Guerardel C, Wasylyk B, Leprince D. (2002). The human candidate tumor suppressor gene hic1 recruits ctbp through a degenerate gldlskk motif. Mol Cell Biol 22: 4890–4901. | Article | PubMed | ISI | ChemPort |
  15. Dunaief JL, Strober BE, Guha S, Khavari PA, Alin K, Luban J et al. (1994). The retinoblastoma protein and brg1 form a complex and cooperate to induce cell cycle arrest. Cell 79: 119–130. | Article | PubMed | ISI | ChemPort |
  16. Fusaro G, Wang S, Chellappan S. (2002). Differential regulation of rb family proteins and prohibitin during camptothecin-induced apoptosis. Oncogene 21: 4539–4548. | Article | PubMed | ISI | ChemPort |
  17. Giacinti C, Giordano A. (2006). Rb and cell cycle progression. Oncogene 25: 5220–5227. | Article | PubMed | ISI | ChemPort |
  18. Guerardel C, Deltour S, Leprince D. (1999). Evolutionary divergence in the broad complex, tramtrack and bric a brac/poxviruses and zinc finger domain from the candidate tumor suppressor gene hypermethylated in cancer. FEBS Lett 451: 253–256. | Article | PubMed | ChemPort |
  19. Inayoshi Y, Miyake K, Machida Y, Kaneoka H, Terajima M, Dohda T et al. (2006). Mammalian chromatin remodeling complex swi/snf is essential for enhanced expression of the albumin gene during liver development. J Biochem (Tokyo) 139: 177–188. | PubMed | ChemPort |
  20. Lefebvre T, Pinte S, Guerardel C, Deltour S, Martin-Soudant N, Slomianny MC et al. (2004). The tumor suppressor hic1 (hypermethylated in cancer 1) is o-glcnac glycosylated. Eur J Biochem 271: 3843–3854. | Article | PubMed | ChemPort |
  21. Melki JR, Vincent PC, Clark SJ. (1999). Cancer-specific region of hypermethylation identified within the hic1 putative tumour suppressor gene in acute myeloid leukaemia. Leukemia 13: 877–883. | Article | PubMed | ChemPort |
  22. Nevins JR. (2001). The rb/e2f pathway and cancer. Hum Mol Genet 10: 699–703. | Article | PubMed | ISI | ChemPort |
  23. Nicoll G, Crichton DN, McDowell HE, Kernohan N, Hupp TR, Thompson AM. (2001). Expression of the hypermethylated in cancer gene (hic-1) is associated with good outcome in human breast cancer. Br J Cancer 85: 1878–1882. | Article | PubMed | ISI | ChemPort |
  24. Pinte S, Guerardel C, Deltour-Balerdi S, Godwin AK, Leprince D. (2004a). Identification of a second g-c-rich promoter conserved in the human, murine and rat tumor suppressor genes hic1. Oncogene 23: 4023–4031. | Article | ISI | ChemPort |
  25. Pinte S, Stankovic-Valentin N, Deltour S, Rood BR, Guerardel C, Leprince D. (2004b). The tumor suppressor gene hic1 (hypermethylated in cancer 1) is a sequence-specific transcriptional repressor: definition of its consensus binding sequence and analysis of its DNA binding and repressive properties. J Biol Chem 279: 38313–38324. | Article | PubMed | ISI | ChemPort |
  26. Stankovic-Valentin N, Verger A, Deltour-Balerdi S, Quinlan KG, Crossley M, Leprince D. (2006). A l225a substitution in the human tumour suppressor hic1 abolishes its interaction with the corepressor ctbp. FEBS J 273: 2879–2890. | Article | PubMed | ChemPort |
  27. Stevaux O, Dimova D, Frolov MV, Taylor-Harding B, Morris E, Dyson N. (2002). Distinct mechanisms of e2f regulation by drosophila rbf1 and rbf2. EMBO J 21: 4927–4937. | Article | PubMed | ISI | ChemPort |
  28. Strobeck MW, DeCristofaro MF, Banine F, Weissman BE, Sherman LS, Knudsen ES. (2001). The brg-1 subunit of the swi/snf complex regulates cd44 expression. J Biol Chem 276: 9273–9278. | Article | PubMed | ISI | ChemPort |
  29. Takahashi Y, Rayman JB, Dynlacht BD. (2000). Analysis of promoter binding by the e2f and prb families in vivo: distinct e2f proteins mediate activation and repression. Genes Dev 14: 804–816. | PubMed | ISI | ChemPort |
  30. Waha A, Koch A, Hartmann W, Mack H, Schramm J, Sorensen N et al. (2004). Analysis of hic-1 methylation and transcription in human ependymomas. Int J Cancer 110: 542–549. | Article | PubMed | ChemPort |
  31. Wang S, Fusaro G, Padmanabhan J, Chellappan SP. (2002a). Prohibitin co-localizes with rb in the nucleus and recruits n-cor and hdac1 for transcriptional repression. Oncogene 21: 8388–8396. | Article | PubMed | ISI | ChemPort |
  32. Wang S, Ghosh RN, Chellappan SP. (1998). Raf-1 physically interacts with rb and regulates its function: a link between mitogenic signaling and cell cycle regulation. Mol Cell Biol 18: 7487–7498. | PubMed | ISI | ChemPort |
  33. Wang S, Nath N, Adlam M, Chellappan S. (1999a). Prohibitin, a potential tumor suppressor, interacts with rb and regulates e2f function. Oncogene 18: 3501–3510. | Article | PubMed | ISI | ChemPort |
  34. Wang S, Nath N, Minden A, Chellappan S. (1999b). Regulation of rb and e2f by signal transduction cascades: divergent effects of jnk1 and p38 kinases. EMBO J 18: 1559–1570. | Article | PubMed | ISI | ChemPort |
  35. Wong S, Weber JD. (2007). Deacetylation of the retinoblastoma tumour suppressor protein by sirt1. Biochem J 407: 451–460. | Article | PubMed | ChemPort |
  36. Wang S, Zhang B, Faller DV. (2002b). Prohibitin requires brg-1 and brm for the repression of e2f and cell growth. EMBO J 21: 3019–3028. | Article | PubMed | ISI | ChemPort |
  37. Wang S, Zhang B, Faller DV. (2004). Brg1/brm and prohibitin are required for growth suppression by estrogen antagonists. EMBO J 23: 2293–2303. | Article | PubMed | ISI | ChemPort |
  38. Zhang B, Chambers KJ, Faller DV, Wang S. (2007). Reprogramming of the swi/snf complex for co-activation or co-repression in prohibitin-mediated estrogen receptor regulation. Oncogene 26: 7153–7157. | Article | PubMed | ChemPort |
Top

Acknowledgements

We thank Dr Srikumar P Chellappan for his continuous support. This study was partially supported by grants from Susan G Komen Breast Cancer Foundation Research Award (BCTR0403163) (SW) and the National Cancer Institute ((CA102940) (SW) and (CA101992) (DVF)) and by the Karin Grunebaum Cancer Research Foundation (DVF). SW is the recipient of DOD/CDMRP 2008 Breast Cancer Concept Award, Carter Family Foundation for Melanoma Research grant award, a BUSM Department of Medicine Pilot Project Grant Award and an Aid for Cancer Research grant award.