The transcriptional repressor NIPP1 is an essential player in EZH2-mediated gene silencing

Abstract

EZH2 is a Polycomb group (PcG) protein that promotes the late-stage development of cancer by silencing a specific set of genes, at least in part through trimethylation of associated histone H3 on Lys 27 (H3K27). Nuclear inhibitor of protein phosphatase-1 (NIPP1) is a ubiquitously expressed transcriptional repressor that has binding sites for the EZH2 interactor EED. Here, we examine the contribution of NIPP1 to EZH2-mediated gene silencing. Studies on NIPP1-deficient cells disclose a widespread and essential role of NIPP1 in the trimethylation of H3K27 by EZH2, not only in the onset of this trimethylation during embryonic development, but also in the maintenance of this repressive mark in proliferating cells. Consistent with this notion, EZH2 and NIPP1 silence a common set of genes, as revealed by gene-expression profiling, and NIPP1 is associated with established Polycomb target genes and with genomic regions that are enriched in Polycomb targets. Furthermore, most NIPP1 target genes are trimethylated on H3K27 and the knockdown of either NIPP1 or EZH2 is often associated with a loss of this modification. Our data reveal that NIPP1 is required for the global trimethylation of H3K27 and is implicated in gene silencing by EZH2.

Introduction

The Polycomb Group (PcG) proteins were originally discovered in Drosophila as repressors of the homeotic genes (Cao and Zhang, 2004a; Ringrose and Paro, 2004). In mammals, the PcG proteins are implicated in the silencing of genes that are important for embryonic development, cell proliferation and differentiation (Boyer et al., 2006; Bracken et al., 2006; Lee et al., 2006). In addition, mammalian PcG proteins are involved in X-chromosome inactivation and the imprinting of autosomal loci (reviewed in Cao and Zhang, 2004a). PcG proteins form two major multimeric complexes, known as the Polycomb Repressive Complexes (PRC) 1 and 2 (Ringrose and Paro, 2004). The PRC2 complex consists in mammals of a core of the proteins EZH2, EED, SUZ12 and, at least with some purification procedures, also RbAp46/48 and AEBP2. EZH2 functions as a methyltransferase for Lys27 of histone H3 (H3K27). Trimethylated H3K27 serves as a binding site for the chromodomain-containing Polycomb protein, a component of PRC1. This binding would stabilize the PcG complex and cause gene repression (reviewed by Schwartz and Pirotta, 2007). The PRC1 complex ensures the maintenance of gene repression by mechanisms that are not yet completely understood but may involve such diverse processes as the blocking of SWI/SNF-chromatin remodeling complexes (Shao et al., 1999), the inhibition of the transcription-initiation machinery (Dellino et al., 2004) and/or the ubiquitylation of histone H2A on K119 (Cao et al., 2005). An additional mechanism for PRC-mediated gene silencing involves the recruitment of DNA methyltransferases by EZH2 to specific target genes (Viré et al., 2006). A further layer in the complexity of gene regulation by EZH2 comes from the recent finding that this PRC2 component also functions as a transcriptional activator by a mechanism that does not require its catalytic domain (Shi et al., 2007).

PRC2 components have been implicated in the development of cancer (Sparmann and van Lohuizen, 2006). For example, SUZ12 is overexpressed in colon and breast cancers (Kirmizis et al., 2003). EZH2 is upregulated in many cancers, including Hodgkin lymphoma, prostate and breast cancer (van Kemenade et al., 2001; Bracken et al., 2003; Kleer et al., 2003). Moreover, EZH2 expression is associated with poor prognosis and is an indication for the metastatic character of the disease (Varambally et al., 2002). Intriguingly, in human tumors the PRC2 complex has a different substrate specificity and contains a distinct EED isoform as well as the sirtuin histone deacetylase (Kuzmichev et al., 2005).

Nuclear inhibitor of protein phosphatase-1 (NIPP1), encoded by the PPP1R8 gene, is a ubiquitously expressed protein in metazoans and plants but not in yeast (Ceulemans et al., 2002). The knockout of NIPP1 in mice is embryonic lethal before gastrulation and NIPP1−/− cell lines are not viable (Van Eynde et al., 2004). NIPP1 binds in vivo to protein Ser/Thr kinase MELK (Vulsteke et al., 2004) and protein Ser/Thr phosphatase-1 (Jagiello et al., 2000), but the physiological implication of these interactions is not known. In addition, NIPP1 interacts with RNA and the essential pre-mRNA splicing factors CDC5L and SAP155 (Jagiello et al., 1997; Jin et al., 1999; Boudrez et al., 2000, 2002). Consistent with the latter findings, NIPP1 was reported to be associated with nuclear storage sites of splicing factors as well as with spliceosomes, the RNA–protein complexes that catalyse pre-mRNA splicing (Beullens and Bollen, 2002). Furthermore, NIPP1 is required for a late step of spliceosome assembly in nuclear splicing extracts. Interestingly, NIPP1 also interacts with the PRC2 component EED and functions as a transcriptional repressor in transient transfection assays by a mechanism that does not require functional interaction sites for CDC5L, SAP155, MELK or PP1 (Jin et al., 2003).

In the present study we show that NIPP1 is essential for PRC2-mediated gene silencing. We demonstrate that NIPP1 is necessary for the initiation and maintenance of global H3K27 trimethylation in vivo and that NIPP1 and EZH2 silence a common set of genes. Moreover, we show that NIPP1 is associated with Polycomb target genes and is required for the trimethylation of the associated nucleosomes on H3K27.

Results

NIPP1 is required for the global trimethylation of H3K27

To delineate the importance of NIPP1 for PcG-mediated gene silencing in vivo, we first examined the effect of a loss of NIPP1 on the trimethylation of H3K27. Since the knockout of NIPP1 in mice was embryonic lethal between 6.5 and 7.5 dpc (days post-coitus) and NIPP1−/− ES cells could not be obtained (Van Eynde et al., 2004), we performed immunostainings with anti-H3K27Me3 antibodies on outgrowths of NIPP1wt and NIPP1−/− blastocysts (Figure 1) as well as on paraffin-embedded NIPP1wt and NIPP1−/− embryos at 6.5 dpc (not illustrated). Immunofluorescence studies revealed that the trophoblast giant cells of outgrowths of NIPP1−/− blastocysts had a normal level of EZH2 (Figure 1a), but were severely deficient in H3K27Me3 (Figure 1b). The level of H3K27Me3 in NIPP1−/− trophoblast giant cells only amounted to 19.4±1.9% (n=21) of that in control cells (n=21), as detected by fluorescence intensity analysis, using the histogram function of the LSM510 microscope. In contrast, the trimethylation of histone H3 on Lys9, which is also associated with gene silencing but is not mediated by the PRC2 complex, was not affected in the NIPP1−/− outgrowths (Figure 1c). Consistent with these data, immunostainings of sections from paraffin-embedded 6.5 dpc embryos revealed a decreased trimethylation of H3K27 in the NIPP1−/− embryos, as compared to that of the wild-type E6.5 embryos, whereas the surrounding maternal tissue in the NIPP1−/− embryos remained unaffected (not shown). Since it has been demonstrated that EZH2 can be inactivated through phosphorylation of Ser21 by protein kinase B (Cha et al., 2005), we wondered whether the deficient trimethylation of H3K27 in NIPP1−/− cells could perhaps be explained by the hyperphosphorylation of EZH2 on Ser21. This was a particularly attractive hypothesis because NIPP1 is associated with protein phosphatase-1 and PP1 interactors often function as substrate-targeting subunits (Ceulemans and Bollen, 2004). If NIPP1 would target EZH2 for dephosphorylation by associated PP1, a loss of NIPP1 would be expected to result in the hyperphosphorylation of EZH2. However, using phospho-Ser21-specific EZH2 antibodies (Cha et al., 2005), we could not detect a different phosphorylation level of EZH2 in outgrowths of NIPP1wt and NIPP1−/− blastocysts (Figure 1d).

Figure 1
figure1

NIPP1 is required for H3K27 trimethylation during embryonic development (ad) Blastocysts from intercrosses of NIPP1−/+ mice were isolated and cultured in TX-WES medium. After 4 days, NIPP1wt and NIPP1−/− blastocyst outgrowths were stained for NIPP1 (a–d), EZH2 (a), H3K27Me3 (b), H3K9Me3 (c) and/or phospho-Ser21-EZH2 (d), as indicated, and for DNA by DAPI staining (ad).

Since it has been shown that the inactive X-chromosome (Xi) is enriched in H3K27 trimethylation and that EZH2 and EED are recruited to the Xi (Plath et al., 2003; Silva et al., 2003), we have also examined by immunostainings whether NIPP1 was present on the Xi. While we did see that EZH2 was enriched on the Xi in female trophoblast giant cells, this was clearly not the case for NIPP1 (Supplementary Figure S1), indicating that NIPP1 may not be implicated in X-inactivation.

The above data indicated that NIPP1 is needed for the initiation of H3K27 trimethylation during early embryonic development. To test whether NIPP1 is also required for the maintenance of H3K27 trimethylation in cultured cells, we performed an RNAi-mediated knockdown of NIPP1 in PC-3 (Figure 2) and U2OS cells (not shown), which are prostate cancer and osteosarcoma cells, respectively. The RNAi-mediated knockdown of NIPP1 in PC-3 cells was associated with growth inhibition, as evidenced by the lower number of cells (Figure 2a) and by MTT cell proliferation assays (not shown), and is in accordance with the established role of NIPP1 in cell proliferation (Van Eynde et al., 2004). A growth inhibition was also observed following the knockdown of EZH2 (Figure 2a), in agreement with observations by Varambally et al. (2002). The knockdowns of NIPP1 and EZH2 in these experiments were verified by both quantitative RT–PCR (qRT–PCR) (Figure 2b) and immunoblot analysis (Figure 2c). The loss of NIPP1 did not have major effects on the EZH2 transcript (Figure 2b) and protein levels (Figure 2c), and vice versa. However, the knockdown of either NIPP1 or EZH2 resulted in a decreased trimethylation of H3K27 (Figure 2d), which was quantified by scanning of immunoblots to 50±5% (n=4) and 27±5% (n=3) of the level in control cells, respectively. In contrast, the loss of NIPP1 or EZH2 had no clear effect on the di/trimethylation of H3K4 and H4K20, which represent PcG-independent marks of active and inactive chromatin, respectively (Figure 2d). Collectively, the above data strongly suggest that NIPP1 has a widespread and essential function in the trimethylation of H3K27 by PRC2, both during development and in cultured cells.

Figure 2
figure2

NIPP1 is required for H3K27 trimethylation in tissue culture cells (a) Representative photos of the male PC-3 cells taken 48 h after knockdown with control, NIPP1 or EZH2 siRNA. The remaining cells after the knockdown of NIPP1 or EZH2 were alive, as shown by the formation of purple formazan crystals in the MTT (tetrazolium salt) assay (not shown) (b, c) siRNA-mediated knockdown of NIPP1 and EZH2 in PC-3 cells was verified by quantitative RT–PCR (qRT–PCR) (n=3) (b) and by immunoblotting (c). SIPP1, a nuclear scaffold protein, served as a loading control. Control siRNA is a scrambled version of the NIPP1 siRNA. (d) Immunoblot analysis of formaldehyde cross-linked PC-3 cell lysates after knockdown with control, NIPP1 or EZH2 siRNA with the indicated antibodies. Histone H3 served as a loading control.

NIPP1 and EZH2 contribute to the silencing of a common set of genes

Since both NIPP1 and EZH2 turned out to be essential for global H3K27 trimethylation, a key step in the PcG-mediated silencing of genes, we performed microarray analyses to find out whether the loss of NIPP1 or EZH2 results in the upregulation of a common set of genes. In four independent experiments, PC-3 cells were transfected with either control siRNA or siRNA duplexes for the knockdown of either EZH2 or NIPP1. To minimize indirect effects, RNA was already isolated 48 h after transfection. At this time, the knockdown of NIPP1 or EZH2 was verified by quantitative RT–PCR and amounted to 73±7 and 82±3% (n=4), respectively (not illustrated). The isolated RNA pools were labeled with a fluorescent dye and were hybridized onto Whole Human Genome Oligo microarrays from Agilent (Santa Clara, CA, USA). These chips contain ca. 44 000 60-mer oligonucleotides, 26 902 of which mapped unambiguously to ca. 17 000 annotated genes. Using paired SAM (significance analysis of microarray) (Tusher et al., 2001), we identified 1622 and 2072 genes with an expression level that differed significantly (P<0.05) between NIPP1 and EZH2 knockdowns, respectively, and the siRNA control (Supplementary Tables S1 and S2). At the more stringent P-value cutoff of 0.01, the loss of NIPP1 or EZH2 resulted in the significant upregulation of 281 and 319 genes, and the downregulation of 409 and 175 genes, respectively (Figure 3c). The successful knockdown of NIPP1 or EZH2 was confirmed by their microarray readings, and NIPP1 and EZH2 did not repress each other (Figure 3a).

Figure 3
figure3

NIPP1 and EZH2 silence a common set of genes. (a) Microarray data of the expression level of NIPP1 and EZH2 in PC-3 cells 48 h after transfection with control siRNA oligos or siRNAs against NIPP1 and EZH2. (b) Heat map diagram depicting the 50 most upregulated (red) and downregulated (green) genes after the knockdown of NIPP1, ranked according to fold changes. The panel also shows the corresponding changes after the knockdown of EZH2. The columns (1–4) represent four independent experiments. The numbers on the color scheme represent the fold changes in gene expression (log2 ratio) in the knockdown samples, as compared to the control siRNA. (c) Venn diagram illustrating the overlap in genes (P<0.01) increased (left panel) or decreased (right panel) after the depletion of EZH2 or NIPP1. (d) Residual plot of the linear regression of the log2-transformed knockdown to control ratios for EZH2 versus those for NIPP1 including all the data points with significantly (P<0.05) altered expression after NIPP1 knockdown. The data points are ranked in function of the fold change. To illustrate the residual shift, separate regression lines for the residuals were added for downregulated and upregulated data points. (e) Scatter plots of the log2-transformed knockdown to control ratios for EZH2 versus those for NIPP1, including the data points with significantly (P<0.05) upregulated (left panel) or downregulated (right panel) expression after NIPP1 knockdown. The Pearson's correlation coefficient (r) with confidence interval is shown in each plot.

Strikingly, the majority of the 50 genes with the highest increase in expression after the knockdown of NIPP1, was also upregulated following the knockdown of EZH2 (Figure 3b). By contrast, the 50 genes that were the most negatively affected by a loss of NIPP1 were only occasionally downregulated after an EZH2 knockdown. In total, 121 (43%) of the genes that were significantly (P<0.01) upregulated after a knockdown of NIPP1 also displayed a significantly (P<0.01) increased expression after a loss of EZH2 (Figure 3c). However, only 25 (6%) of the genes that were significantly repressed by the knockdown of NIPP1 were also significantly downregulated following the loss of EZH2. Figure 3c also shows that there is a large group of genes (384 genes, P<0.01) that are downregulated following the knockdown of NIPP1 but are not significantly affected by the knockdown of EZH2.

As illustrated by a residual plot (Figure 3d), the linear regression after log2 transformation of the EZH2 knockdown to control ratios in function of NIPP1 knockdown to control ratios for all data points with significantly altered expression after an NIPP1 knockdown, supported separate regression studies of the upregulated and downregulated genes. In these separate analyses, the genes that were significantly upregulated by a loss of NIPP1 tended to be proportionately affected by a knockdown of EZH2, in contrast to the downregulated genes (r=0.59±0.01 and r=0.01±0.01, respectively, with 95% confidence interval boundaries defined by a Fisher's z-test; Figure 3e). In summary, our microarray data indicate that NIPP1 and EZH2 silence a common set of genes but activate a distinct set of genes.

We have subsequently corroborated the expression array data by qRT–PCR (Figure 4). The samples that were used for these experiments were obtained independently of the ones obtained for the microarray analyses. The selected genes showed an increased expression following the knockdown of either NIPP1 or EZH2 (Figure 4a), only an increased expression after the knockdown of NIPP1 (Figure 4b), a decreased expression following the knockdown of either NIPP1 or EZH2 (Figure 4c) or only a decreased expression after the knockdown of NIPP1 (Figure 4d).

Figure 4
figure4

Validation of the microarray results by quantitative RT–PCR (qRT–PCR) analysis of RNAi experiments, independent of the ones used for the microarray analysis. mRNA was prepared from PC-3 cells transfected with control (Ctr), NIPP1 or EZH2 siRNAs. ACTIN was used as a control for normalization. (a) Genes significantly upregulated in NIPP1 and EZH2 knockdown, (b) genes significantly upregulated only in NIPP1 knockdown, (c) genes significantly downregulated in NIPP1 and EZH2 knockdown, (d) genes significantly downregulated only in NIPP1 knockdown.

NIPP1 is associated with PcG target genes

To examine whether NIPP1, like the PRC2 core components, is associated with PcG target genes, we performed chromatin immunoprecipitations (ChIP). To validate the used NIPP1 antibodies we first examined the association of NIPP1 with MYT1, a well-established PRC2 target gene (Kirmizis et al., 2004; Bracken et al., 2006; Viré et al., 2006). In these experiments, EZH2 served as a positive control and glyceraldehyde-3-phosphate dehydrogenase as a negative control. NIPP1 was clearly enriched on the MYT1 gene, as detected by both agarose gel electrophoresis (Figures 5a and b) and qPCR (Figure 5c). Importantly, the enrichment of NIPP1 on MYT1 (Figure 5c) and the immunodetection of NIPP1 (Figure 5d) could be competed for by the addition of the immunogenic peptide NIPP1-(335–351), attesting to the specificity of the used NIPP1 antibody.

Figure 5
figure5

Validation of NIPP1 antibody for ChIP experiments. ChIP assays were performed in PC-3 cells using antibodies against EZH2 (a) and NIPP1 (b, c). Primers specific for MYT1 and GAPDH were used to amplify the immunoprecipitated DNA by PCR. The DNA was visualized by ethidium bromide staining of the agarose gels (a, b) or by quantitative PCR (qPCR) (c). Three dilutions of the input chromatin were used as positive controls to demonstrate linear amplification (a, b). (c) The results presented in (c) are means±s.e. (n=3–5). A control ChIP was performed with a NIPP1 antibody that was preincubated with 3 mM of the immunogenic peptide. (d) HEK293T lysates were resolved by electrophoresis, transferred to PVDF membranes and probed with NIPP1 antibody, or NIPP1 antibody preincubated with 10 μM of the immunogenic NIPP1 peptide (NIPP1-(335–351)).

As an extension of our analysis of the association of NIPP1 with PcG target genes, we selected a set of 21 genes that showed a significantly altered expression in PC-3 cells after the knockdown of NIPP1 and/or EZH2 (Figure 3), and that were recently identified as likely PcG target genes because of trimethylation on H3K27 and association with Suz12 (Boyer et al., 2006; Bracken et al., 2006; Lee et al., 2006). In addition, we included in our analysis four established EZH2 target genes (MYT1, WNT1, KCNA1 and MSMB) and CDC6 as negative control (Kirmizis et al., 2004; Bracken et al., 2006; Viré et al., 2006, Beke et al., 2007). All these genes were classified in eight groups (A–H), based on their expression response following the knockdown of NIPP1 (Figure 6a) or EZH2 (Figure 6b). A ChIP analysis in PC-3 cells was performed with NIPP1, EZH2 and H3K27Me3 antibodies, and using a single primer set to amplify up to 250 bp mostly mapped to 1000–2000 nucleotides upstream of the transcriptional start sites. Rabbit anti-mouse immunoglobulines were used as a control. Using a cutoff of twofold enrichment, as compared to CDC6 (last lane), NIPP1 was associated with a majority of the selected genes (Figure 6c). Importantly, EZH2 was nearly always associated with the same loci (Figure 6d). That not all selected gene fragments showed an enrichment for NIPP1 and EZH2 was not unexpected since only a single primer set was used for our analysis and since it is well known that the targeting of PRC2 complexes can be limited to specific gene fragments (Bracken et al., 2006; Beke et al., 2007). Importantly, 18 of the 25 selected gene fragments were more than 10-fold enriched for trimethylation of the associated nucleosomes on H3K27, but the extent of trimethylation varied considerably between the different genes (Figure 6e). Intriguingly, NIPP1 and EZH2 were also associated with some genes that were downregulated following the knockdown of NIPP1 and/or EZH2 (groups C, D and F), and at least some of these genes were trimethylated on H3K27. This is consistent with recent findings that some PcG targets are activated by Polycomb signaling (Pasini et al., 2007; Shi et al., 2007).

Figure 6
figure6

NIPP1 is associated with PcG target genes. A total of 25 likely PcG target genes were divided in 8 groups (ah), based on their expression pattern in PC-3 cells following the knockdown of NIPP1 (a) or EZH2 (b), as derived from the experiments described in Figure 3. The expression level of KCNA1 and MSMB (class H) was under detection limit of the microarray reading after knockdown of NIPP1 or EZH2 in PC-3 cells. The numbers on the y axis represent the fold changes in gene expression (log2 ratio), as compared to the control siRNA. ChIP analyses using antibodies against NIPP1 (c), EZH2 (d), H3K27Me3 (e) and rabbit anti-mouse immunoglobulines (IgG, control) were performed in PC-3 cells. The precipitated DNA was amplified by real-time quantitative PCR (qPCR) using primers specific to the promoters (between 1 and 2 kb upstream of the transcriptional start site) of the indicated genes. The primer sets for MYT1 and MSMB were located at about 400 and 6000 bp downstream of the transcriptional start site, respectively. CDC6 was used as negative control. The data are expressed as fold enrichment (means±s.e.; n=3–6), as compared to the negative control with IgGs.

To confirm the association of NIPP1 with PcG target genes in a different cell line, and with a distinct and unbiased approach, we performed a ChIP on chip analysis in HeLa cells. This technique involves a ChIP followed by a microarray analysis of the precipitated DNA (Kirmizis and Farnham, 2004). Following a ChIP with NIPP1 antibodies or control IgGs, the precipitated DNA was linker-mediated (LM)-PCR-amplified (Kirmizis et al., 2004) and hybridized onto CGH (comparative genomic hybridization)-1Mb-3K-2 arrays. These arrays consist of 3434 BACs of 150 kb on average, covering the full human genome with a 1 Mb resolution. All the BACs were represented twice on the array and two separate hybridizations were done with color flip. NIPP1-associated DNA was found to be enriched at least 1.3-fold (Student's t-test; P<0.05) on 44 BACs and 32 of these comprised one or more genes (Table 1). Eight BACs did not contain any gene but were flanked within 50 kb by at least one gene. Since PcG-mediated gene silencing often involves large chromatin fragments (100–200 kb), the flanking genes also represent candidate NIPP1 target genes. Finally, only 4 BACs did not contain any gene and were not flanked by genes within 50 kb. In total, 121 genes were identified in the NIPP1-associated BACs or within 50 kb of their flanking sequences. Of these, 35 (29%) were identified as likely PcG target genes based on recently published genome-wide PcG targetscreens (Boyer et al., 2006; Bracken et al., 2006; Lee et al., 2006; Squazzo et al., 2006). This number is much higher than expected from the random distribution of PcG target genes, which, according to recent estimates (Bracken et al., 2006; Lee et al., 2006), represents at the very most 10% of all genes. Collectively, our data show that NIPP1 binds in vivo to genomic regions that are enriched in PcG target genes.

Table 1 Mapping of NIPP1 target regions by a ChIP on chip, using CGH-1Mb-3K-2 arrays

NIPP1 is required for the H3K27 trimethylation of many of its target genes

Next, we examined whether NIPP1 target genes are trimethylated on H3K27 by an NIPP1 and EZH2-dependent mechanism. For these experiments we selected five genes that were trimethylated on H3K27, each from a distinct group, as defined in Figure 6. The RNAi-mediated knockdown of either NIPP1 or EZH2 resulted in a loss of trimethylation of three out of these five genes (Figure 7), consistent with the notion that both NIPP1 and EZH2 are required for the initiation and/or maintenance of H3K27 trimethylation. That not all selected genes showed a decreased trimethylation on H3K27 is in accordance with published data showing that the loss of this histone modification within the time frame of EZH2 knockdown experiments (48 h) can be limited to nucleosomes associated with specific gene fragments (Cao and Zhang, 2004b; Beke et al., 2007). Intriguingly, a decreased trimethylation on H3K27 can be associated with an increased expression (RPS6KC1, AGPAT4) as well as with a decreased expression (COL2A1), consistent with recently published data (Pasini et al., 2007; Ringrose, 2007; Shi et al., 2007).

Figure 7
figure7

Effect of the loss of NIPP1 or EZH2 on H3K27 trimethylation of NIPP1 target genes. ChIP experiments with H3K27Me3 antibodies were performed in PC-3 cells for the indicated five polycomb target genes, each from a different group in Figure 6, before and after the knockdown of either NIPP1 or EZH2. The ChIPs were performed as indicated in the legend of Figure 6. The results are expressed as a percentage±s.e. (n=3) of the data obtained with the control siRNA.

Discussion

NIPP1 is required for PRC2-mediated gene silencing

We have obtained various independent lines of evidence that are consistent with an essential role for NIPP1 in the initiation and maintenance of PcG-mediated gene silencing. Firstly, the knockouts of NIPP1 (Figure 1; Van Eynde et al., 2004), EZH2 (O’Carroll et al., 2001) or EED (Faust et al., 1995) in mice are all associated with a similar phenotype, namely a deficient early embryonic trimethylation of H3K27 and an embryonic lethality at or around gastrulation (6.5–8.5 dpc). Similarly, knockdown experiments in cultured cells disclosed an essential function for all of these proteins in the trimethylation of H3K27 and in proliferation (see Figure 2 for NIPP1). Secondly, NIPP1 and EZH2 silence a common set of genes (Figure 3) and NIPP1 is associated with at least a large subset of established PcG target genes (Figures 5 and 6, Table 1). Thirdly, in a subset of the Polycomb target genes, the knockdown of either NIPP1 or EZH2 is correlated with a decreased trimethylation of H3K27 (Figure 7).

What is the role of NIPP1 in PcG-mediated gene silencing?

NIPP1 does not appear to be required for the synthesis or stability of EZH2, as indicated by the normal level of EZH2 in NIPP1-deficient cells (Figure 2). Although we cannot rule out the possibility that NIPP1 affects the substrate specificity or specific activity of EZH2 as an H3K27 methyltransferase, it is known that an in vitro reconstituted complex of EZH2, EED and SUZ12 is active (Cao and Zhang, 2004b). We have also excluded the possibility that NIPP1 promotes the activation of EZH2 via dephosphorylation of phospho-Ser21 by NIPP1-associated PP1 (Figure 1d). Since NIPP1 is part of a macromolecular complex that contains EZH2 and EED (Roy et al., 2007), has distinct interaction sites for EED (Jin et al., 2003) and EZH2 (Roy et al., 2007) and is also a nucleic acid-binding protein (Jagiello et al., 1997; Jin et al., 1999), this leads to the enticing hypothesis that NIPP1 plays a role in the targeting of the PRC2 complex to at least a large subset of its target genes.

Our previous studies identified NIPP1 as a pre-mRNA splicing factor that interacts in vivo with the splicing factors Cdc5L and SAP155, and is needed for a late step of spliceosome assembly (Boudrez et al., 2000, 2002; Beullens and Bollen, 2002). We currently do not know whether the splicing function of NIPP1 is somehow linked to its transcriptional repressor function. While our study is the first implicating a pre-mRNA splicing factor in H3K27 trimethylation, there are other reports linking pre-mRNA processing factors to histone methylation. For example, the NIPP1 ligand and U2 snRNP component SAP155 was recently reported to be required for PcG-mediated gene silencing and to interact with PRC1 components (Isono et al., 2005). Since NIPP1 interacts with PRC2 components (Jin et al., 2003; Roy et al., 2007) as well as with phosphorylated forms of SAP155 (Boudrez et al., 2002), it is tempting to speculate that NIPP1 contributes to interactions between PRC1 and PRC2.

Does NIPP1 have a role in oncogenesis?

NIPP1 is required for the trimethylation of H3K27 by the oncogene EZH2 (this work) and interacts with protein kinase MELK (Vulsteke et al.,. 2004), which has also been identified as an oncogene. Indeed, EZH2 and MELK are required for cell proliferation (Varambally et al., 2002; Bracken et al., 2003; Gray et al., 2005; Lin et al., 2007) and their expression level is increased, often as a result of gene amplification, in a wide range of cancers, which correlates with the aggressiveness of cancer and a poor prognosis (Varambally et al., 2002; Kleer et al., 2003; Stanbrough et al., 2006; Lin et al., 2007). MELK has also been shown to control stem cell proliferation but it is not known whether this contributes to its putative role in oncogenesis. The neoplastic properties of EZH2 require its catalytic activity, indicating that it promotes tumorigenesis by repressing specific tumor suppressor genes (Beke et al., 2007). Our finding that NIPP1 is also required for cell proliferation and for global trimethylation of H3K27 indicates that it may itself also be an oncogene. It will therefore be important to examine the concentration and regulation of NIPP1 in cancer and to explore whether NIPP1 acts tumorigenic.

In conclusion, we have identified NIPP1 as a novel player in PRC2-mediated gene silencing, in that it is required for the initiation and maintenance of global trimethylation of H3K27 by EZH2. Additional work is needed to elucidate the mechanistic details and to explore the role of other NIPP1 ligands in this process.

Materials and methods

Antibodies

Polyclonal anti-H3K27Me3 (07–449) and anti-H3K9Me3 (07–442) antibodies were obtained from Upstate (Dundee, UK), and anti-Histone H3 (ab1791), anti-H3K4Me2 (ab7766) and anti-H4K20Me3 (ab9053) antibodies from Abcam (Cambridge, UK). For the immunostainings, goat anti-NIPP1 (ab5300) was delivered by Abcam, mouse anti-EZH2 (AC22) was purchased from Cell Signaling technology (Danvers, MA, USA) and rabbit anti-phospho-Ser21-EZH2 from Bethyl Laboratories (Montgomery, TX, USA). Rabbit anti-mouse immunoglobulins (control IgG) were purchased from DakoCytomation (Gostrup, Denmark). Synthetic fragments of human EZH2 (14-IndexTermCWRKRVKSEYMRLRQLK-30), human SIPP1 (627-IndexTermDDVYEAFMKEMEGLL-641) and human NIPP1 (341-IndexTermPGKKPTPSLLI-351), coupled to keyhole limpet hemocyanin, were used to generate polyclonal antibodies in rabbits. The antibodies were affinity-purified on the bovine serum albumin-coupled peptides linked to CNBr-activated Sepharose 4B.

siRNA

Sense and antisense RNA oligonucleotides were chemically synthesized and annealed to form siRNA duplexes targeted against human NIPP1 (IndexTermGGAACCUCACAAGCCUCAGCAAAUU; Stealth RNAi, Invitrogen, Paisley, UK) and human EZH2 (IndexTermGACUCUGAAUGCAGUUGCUdTdT; Dharmacon, Chicago, IL, USA). A scrambled version of the NIPP1 siRNA duplex (IndexTermGGAACUCGAACCUCCACGAACAAUU; Stealth RNAi, Invitrogen) was used as negative control. PC-3 cells were transfected using Lipofectamine 2000 (Invitrogen), samples for immunobot analysis and qRT–PCR were taken at the indicated time points.

Quantitative RT–PCR

Total RNA was isolated using the Genelute Mammalian Total RNA Miniprep kit (Sigma, St Louis, MO, USA). RNA (1–5 μg) was reverse-transcribed with oligo dT primer (Sigma) and the M-MulV reverse transcriptase (Fermentas, GmBH, St Leon-Rot, Germany). About 1.5% of this cDNA was PCR-amplified in duplicate or triplicate using a Platinum SYBR Green qPCR SuperMix-UDG (Invitrogen) in a Rotorgene detection system (Corbett Research, Cambridge, UK). The quantitative PCR reactions were performed under conditions that were standardized for each primer pair. To compare the relative amount of target in different samples, all values were normalized to the housekeeping gene ACTIN. The sequences of the primers are available on request.

Gene expression analysis

Total RNA was extracted from PC-3 cells transfected with a control siRNA or with siRNAs for NIPP1 or EZH2, as described above. For each treatment, RNA was prepared from four independent experiments. Targets for microarray hybridization were generated from the RNA according to the supplier's instructions (Agilent). The Whole Human Genome Oligo Array (Agilent) was used for gene expression profiling. The microarray data were analysed as described in detail in the online section within the Supplementary Materials and Methods section. All microarray data are available at ArrayExpress under the accession number e-tabm-128 (password=spiachle).

Chromatin immunoprecipitations

ChIP reactions were performed according to the protocol of Upstate, with some modifications as described in the section with Supplementary information. For normal chromatin IPs, the immunoprecipitated DNA was quantified by real-time qPCR. The sequences of the PCR primers are available upon request. For the ChIP on chip analysis, the immunoprecipitated DNA was amplified by LM-PCR, as described previously (Kirmizis and Farnham, 2004), and hybridized to the human CGH-1Mb-3K-2 array manufactured by the VIB Microarray Facility in Belgium (http://www.microarrays.be/). More detailed information on the analysis of the data of ChIP–chip is described in the online section within Supplementary Materials and methods.

Cell culture and immunostainings

Male PC-3 cells were cultured as monolayers in 50% Dulbecco's modified Eagle's medium (DMEM) and 50% Ham's F12 with 10% fetal calf serum (FCS). HEK293T and HeLa (female) cells were cultured in DMEM with 10% FCS. Blastocyst outgrowth experiments and immunostainings were performed as described by Van Eynde et al. (2004). Microscopy was performed with a confocal microscope (Zeiss LSM 510).

References

  1. Beke L, Nuytten M, Van Eynde A, Beullens M, Bollen M . (2007). The gene encoding the prostatic tumor suppressor PSP94 is a target for repression by the Polycomb group protein EZH2. Oncogene 26: 4590–4595.

    CAS  Article  Google Scholar 

  2. Beullens M, Bollen M . (2002). The protein phosphatase-1 regulator NIPP1 is also a splicing factor involved in a late step of spliceosome assembly. J Biol Chem 277: 19855–19860.

    CAS  Article  Google Scholar 

  3. Boudrez A, Beullens M, Groenen P, Van Eynde A, Vulsteke V, Jagiello I et al. (2000). NIPP1-mediated interaction of protein phosphatase-1 with CDC5L, a regulator of pre-mRNA splicing and mitotic entry. J Biol Chem 275: 25411–25417.

    CAS  Article  Google Scholar 

  4. Boudrez A, Beullens M, Waelkens E, Stalmans W, Bollen M . (2002). Phosphorylation-dependent interaction between the splicing factors SAP155 and NIPP1. J Biol Chem 277: 31834–31841.

    CAS  Article  Google Scholar 

  5. Boyer LA, Plath K, Zeitlinger J, Brambrink T, Medeiros LA, Lee TI et al. (2006). Polycomb complexes repress developmental regulators in murine embryonic stem cells. Nature 441: 349–353.

    CAS  Article  Google Scholar 

  6. Bracken AP, Dietrich N, Pasini D, Hansen KH, Helin K . (2006). Genome-wide mapping of Polycomb target genes unravels their roles in cell fate transitions. Genes Dev 20: 1123–1136.

    CAS  Article  Google Scholar 

  7. Bracken AP, Pasini D, Capra M, Prosperini E, Colli E, Helin K . (2003). EZH2 is downstream of the pRB-E2F pathway, essential for proliferation and amplified in cancer. EMBO J 22: 5323–5335.

    CAS  Article  Google Scholar 

  8. Cao R, Zhang Y . (2004a). The functions of E(Z)/EZH2-mediated methylation of lysine 27 in histone H3. Curr Opin Genet Dev 14: 155–164.

    CAS  Article  Google Scholar 

  9. Cao R, Zhang Y . (2004b). SUZ12 is required for both the histone methyltransferase activity and the silencing function of the EED-EZH2 complex. Mol Cell 15: 57–67.

    CAS  Article  Google Scholar 

  10. Cao R, Tsukada Y, Zhang Y . (2005). Role of Bmi-1 and Ring1A in H2A ubiquitylation and Hox gene silencing. Mol Cell 20: 845–854.

    CAS  Article  Google Scholar 

  11. Ceulemans H, Bollen M . (2004). Functional diversity of protein phosphatase-1, a cellular economizer and reset button. Physiol Rev 84: 1–39.

    CAS  Article  Google Scholar 

  12. Ceulemans H, Stalmans W, Bollen M . (2002). Regulator-driven functional diversification of protein phosphatase-1 in eukaryotic evolution. Bioessays 24: 371–381.

    CAS  Article  Google Scholar 

  13. Cha TL, Zhou BP, Xia W, Wu Y, Yang CC, Chen CT et al. (2005). Akt-mediated phosphorylation of EZH2 suppresses methylation of lysine 27 in histone H3. Science 310: 306–310.

    CAS  Article  Google Scholar 

  14. Dellino GI, Schwartz YB, Farkas G, McCabe D, Elgin SC, Pirrotta V . (2004). Polycomb silencing blocks transcription initiation. Mol Cell 13: 887–893.

    CAS  Article  Google Scholar 

  15. Faust C, Schumacher A, Holdener B, Magnuson T . (1995). The eed mutation disrupts anterior mesoderm production in mice. Development 121: 273–285.

    CAS  Google Scholar 

  16. Gray D, Jubb AM, Hogue D, Dowd P, Kljavin N, Yi S et al. (2005). Maternal embryonic leucine zipper kinase/murine protein serine-threonine kinase 38 is a promising therapeutic target for multiple cancers. Cancer Res 65: 9751–9761.

    CAS  Article  Google Scholar 

  17. Isono K, Mizutani-Koseki Y, Komori T, Schmidt-Zachmann MS, Koseki H . (2005). Mammalian polycomb-mediated repression of Hox genes requires the essential spliceosomal protein Sf3b1. Genes Dev 19: 536–541.

    CAS  Article  Google Scholar 

  18. Jagiello I, Beullens M, Vulsteke V, Wera S, Sohlberg B, Stalmans W et al. (1997). NIPP-1, a nuclear inhibitory subunit of protein phosphatase-1, has RNA-binding properties. J Biol Chem 272: 22067–22071.

    CAS  Article  Google Scholar 

  19. Jagiello I, Van Eynde A, Vulsteke V, Beullens M, Boudrez A, Keppens S et al. (2000). Nuclear and subnuclear targeting sequences of the protein phosphatase-1 regulator NIPP1. J Cell Sci 113 (Part 21): 3761–3768.

    CAS  Google Scholar 

  20. Jin Q, Beullens M, Jagiello I, Van Eynde A, Vulsteke V, Stalmans W et al. (1999). Mapping of the RNA-binding and endoribonuclease domains of NIPP1, a nuclear targeting subunit of protein phosphatase 1. Biochem J 342 (Part 1): 13–19.

    CAS  Article  Google Scholar 

  21. Jin Q, Van Eynde A, Beullens M, Roy N, Thiel G, Stalmans W et al. (2003). The protein phosphatase-1 (PP1) regulator, nuclear inhibitor of PP1 (NIPP1), interacts with the polycomb group protein, embryonic ectoderm development (EED), and functions as a transcriptional repressor. J Biol Chem 278: 30677–30685.

    CAS  Article  Google Scholar 

  22. Kirmizis A, Bartley SM, Farnham PJ . (2003). Identification of the polycomb group protein SU(Z)12 as a potential molecular target for human cancer therapy. Mol Cancer Ther 2: 113–121.

    CAS  Google Scholar 

  23. Kirmizis A, Bartley SM, Kuzmichev A, Margueron R, Reinberg D, Green R et al. (2004). Silencing of human polycomb target genes is associated with methylation of histone H3 Lys 27. Genes Dev 18: 1592–1605.

    CAS  Article  Google Scholar 

  24. Kirmizis A, Farnham PJ . (2004). Genomic approaches that aid in the identification of transcription factor target genes. Exp Biol Med (Maywood) 229: 705–721.

    CAS  Article  Google Scholar 

  25. Kleer CG, Cao Q, Varambally S, Shen R, Ota I, Tomlins SA et al. (2003). EZH2 is a marker of aggressive breast cancer and promotes neoplastic transformation of breast epithelial cells. Proc Natl Acad Sci USA 100: 11606–11611.

    CAS  Article  Google Scholar 

  26. Kuzmichev A, Margueron R, Vaquero A, Preissner TS, Scher M, Kirmizis A et al. (2005). Composition and histone substrates of polycomb repressive group complexes change during cellular differentiation. Proc Natl Acad Sci USA 102: 1859–1864.

    CAS  Article  Google Scholar 

  27. Lee TI, Jenner RG, Boyer LA, Guenther MG, Levine SS, Kumar RM et al. (2006). Control of developmental regulators by Polycomb in human embryonic stem cells. Cell 125: 301–313.

    CAS  Article  Google Scholar 

  28. Lin ML, Park JH, Nishidate T, Nakamura Y, Katagiri T . (2007). Involvement of maternal embryonic leucine zipper kinase (MELK) in mammary carcinogenesis through interaction with Bcl-G, a pro-apoptotic member of the Bcl-2 family. Breast Cancer Res 9: R17.

    Article  Google Scholar 

  29. O’Carroll D, Erhardt S, Pagani M, Barton SC, Surani MA, Jenuwein T . (2001). The polycomb-group gene Ezh2 is required for early mouse development. Mol Cell Biol 21: 4330–4336.

    Article  Google Scholar 

  30. Pasini D, Bracken AP, Hansen JB, Capillo M, Helin K . (2007). The polycomb group protein Suz12 is required for embryonic stem cell differentiation. Mol Cell Biol 27: 3769–3779.

    CAS  Article  Google Scholar 

  31. Plath K, Fang J, Mlynarczyk-Evans SK, Cao R, Worringer KA, Wang H et al. (2003). Role of histone H3 lysine 27 methylation in X inactivation. Science 300: 131–135.

    CAS  Article  Google Scholar 

  32. Ringrose L . (2007). Polycomb comes of age: genome-wide profiling of target sites. Curr Opin Cell Biol 19: 290–297.

    CAS  Article  Google Scholar 

  33. Ringrose L, Paro R . (2004). Epigenetic regulation of cellular memory by the Polycomb and Trithorax group proteins. Annu Rev Genet 38: 413–443.

    CAS  Article  Google Scholar 

  34. Roy N, Van Eynde A, Beke L, Nuytten M, Bollen M . (2007). The transcriptional repression by NIPP1 is mediated by Polycomb group proteins. Biochim Biophys Acta, doi:10.1016/j.bbaexp.2007.07.004.

    CAS  Article  Google Scholar 

  35. Schwartz YB, Pirrotta V . (2007). Polycomb silencing mechanisms and the management of genomic programmes. Nat Rev Genet 8: 9–22.

    CAS  Article  Google Scholar 

  36. Shao Z, Raible F, Mollaaghababa R, Guyon JR, Wu CT, Bender W et al. (1999). Stabilization of chromatin structure by PRC1, a Polycomb complex. Cell 98: 37–46.

    CAS  Article  Google Scholar 

  37. Shi B, Liang J, Yang X, Wang Y, Zhao Y, Wu H et al. (2007). Integration of estrogen and wnt signaling circuits by the polycomb group protein ezh2 in breast cancer cells. Mol Cell Biol 27: 5105–5119.

    CAS  Article  Google Scholar 

  38. Silva J, Mak W, Zvetkova I, Appanah R, Nesterova TB, Webster Z et al. (2003). Establishment of histone h3 methylation on the inactive X chromosome requires transient recruitment of Eed-Enx1 polycomb group complexes. Dev Cell 4: 481–495.

    CAS  Article  Google Scholar 

  39. Sparmann A, van Lohuizen M . (2006). Polycomb silencers control cell fate, development and cancer. Nat Rev Cancer 6: 846–856.

    CAS  Article  Google Scholar 

  40. Squazzo SL, O’Geen H, Komashko VM, Krig SR, Jin VX, Jang SW et al. (2006). Suz12 binds to silenced regions of the genome in a cell-type-specific manner. Genome Res 16: 890–900.

    CAS  Article  Google Scholar 

  41. Stanbrough M, Bubley GJ, Ross K, Golub TR, Rubin MA, Penning TM et al. (2006). Increased expression of genes converting adrenal androgens to testosterone in androgen-independent prostate cancer. Cancer Res 66: 2815–2825.

    CAS  Article  Google Scholar 

  42. Tusher VG, Tibshirani R, Chu G . (2001). Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci USA 98: 5116–5121.

    CAS  Article  Google Scholar 

  43. Van Eynde A, Nuytten M, Dewerchin M, Schoonjans L, Keppens S, Beullens M et al. (2004). The nuclear scaffold protein NIPP1 is essential for early embryonic development and cell proliferation. Mol Cell Biol 24: 5863–5874.

    CAS  Article  Google Scholar 

  44. van Kemenade FJ, Raaphorst FM, Blokzijl T, Fieret E, Hamer KM, Satijn DP et al. (2001). Coexpression of BMI-1 and EZH2 polycomb-group proteins is associated with cycling cells and degree of malignancy in B-cell non-Hodgkin lymphoma. Blood 97: 3896–3901.

    CAS  Article  Google Scholar 

  45. Varambally S, Dhanasekaran SM, Zhou M, Barrette TR, Kumar-Sinha C, Sanda MG et al. (2002). The polycomb group protein EZH2 is involved in progression of prostate cancer. Nature 419: 624–629.

    CAS  Article  Google Scholar 

  46. Viré E, Brenner C, Deplus R, Blanchon L, Fraga M, Didelot C et al. (2006). The Polycomb group protein EZH2 directly controls DNA methylation. Nature 439: 871–874.

    Article  Google Scholar 

  47. Vulsteke V, Beullens M, Boudrez A, Keppens S, Van Eynde A, Rider MH et al. (2004). Inhibition of spliceosome assembly by the cell cycle-regulated protein kinase MELK and involvement of splicing factor NIPP1. J Biol Chem 279: 8642–8647.

    CAS  Article  Google Scholar 

  48. Zeidler M, Varambally S, Cao Q, Chinnaiyan AM, Ferguson DO, Merajver SD et al. (2005). The Polycomb group protein EZH2 impairs DNA repair in breast epithelial cells. Neoplasia 7: 1011–1019.

    CAS  Article  Google Scholar 

Download references

Acknowledgements

Fabienne Withof provided expert technical assistance. This work was financially supported by the Fund for Scientific Research-Flanders (Grant G.0290.05 and ZKB6003-01-W01), a Flemish Concerted Research Action and the Prime Minister's office (IAP/V-05). The microscopy was performed in the Cell Imaging Core Facility of KULeuven.

Author information

Affiliations

Authors

Corresponding author

Correspondence to A Van Eynde.

Additional information

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc).

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Nuytten, M., Beke, L., Van Eynde, A. et al. The transcriptional repressor NIPP1 is an essential player in EZH2-mediated gene silencing. Oncogene 27, 1449–1460 (2008). https://doi.org/10.1038/sj.onc.1210774

Download citation

Keywords

  • EZH2
  • gene silencing
  • NIPP1
  • polycomb group proteins
  • PRC2
  • transcriptional repression

Further reading