© 2016. Published by The Company of Biologists Ltd | Journal of Cell Science (2016) 129, 2343-2353 doi:10.1242/jcs.185546

RESEARCH ARTICLE The EZH1–SUZ12 complex positively regulates the transcription of NF-κB target genes through interaction with UXT Shuai-Kun Su, Chun-Yuan Li, Pin-Ji Lei, Xiang Wang, Quan-Yi Zhao, Yang Cai, Zhen Wang, Lianyun Li* and Min Wu*

ABSTRACT zeste 2 polycomb repressive complex 2 subunit (EZH2) (Shen et al., Drosophila Unlike other members of the polycomb group protein family, EZH1 2008). EZH2 mimics the function of E(Z) in and is the has been shown to positively associate with active transcription on a main methyltransferase for H3K27 in mammalian cells (Shen et al., genome-wide scale. However, the underlying mechanism for this 2008). However, controversial reports exist about EZH1, which behavior still remains elusive. Here, we report that EZH1 physically means its activities have been a puzzle for a long time. Compared interacts with UXT, a small chaperon-like transcription co-activator. with EZH2, EZH1 has lower enzymatic activity, suggesting it might UXT specifically interacts with EZH1 and SUZ12, but not EED. have distinct functions (Margueron et al., 2008). Several groups Similar to upon knockdown of UXT, knockdown of EZH1 or SUZ12 reported that EZH1 methylates histone H3K27 and compensates for through RNA interference in the cell impairs the transcriptional the functions of EZH2 upon its absence (Bae et al., 2015; Hidalgo activation of nuclear factor (NF)-κB target genes induced by TNFα. et al., 2012; Shen et al., 2008). Recently, Mousavi et al. have EZH1 deficiency also increases TNFα-induced cell death. reported that EZH1 behaves as a positive regulator for transcription Interestingly, chromatin immunoprecipitation and the following next- and is required for proper recruitment of RNA polymerase II (Pol II) generation sequencing analysis show that H3K27 mono-, di- and to target genes (Mousavi et al., 2012). Another study has tri-methylation on NF-κB target genes are not affected in EZH1-or demonstrated that EZH1 forms two different complexes with UXT-deficient cells. EZH1 also does not affect the translocation of the distinguished functions (Xu et al., 2015). One is similar to the p65 subunit of NF-κB (also known as RELA) from the cytosol to the classical PRC2 complex containing EZH1, embryonic ectoderm nucleus. Instead, EZH1 and SUZ12 regulate the recruitment of p65 development (EED) and SUZ12, and the second complex is only and RNA Pol II to target genes. Taken together, our study shows that with SUZ12 (Xu et al., 2015). The former complex inhibits EZH1 and SUZ12 act as positive regulators for NF-κB signaling and transcription, while the latter one activates transcription (Xu et al., demonstrates that EZH1, SUZ12 and UXT work synergistically to 2015). However, the detailed mechanisms of how the two EZH1 regulate pathway activation in the nucleus. complexes regulate transcription remain elusive. The nuclear factor κ-light-chain-enhancer of activated B cells KEY WORDS: EZH1, UXT, Histone methylation, Transcription (NF-κB) signaling pathway is a well-studied pathway involved regulation, NF-κB signaling pathway in inflammation, immunity and anti-apoptosis (Iwai, 2012; Oeckinghaus et al., 2011). It is activated upon many extracellular INTRODUCTION signals, and tumor necrosis factor (TNFα) has been used as one of The epigenetic control of chromatin is crucial for transcription the typical molecules to activate the pathway (Hu et al., 2014; Wang regulation and the proper response to extracellular signals in the cell. et al., 2012; Zhang et al., 2014). Although the regulations of NF-κB The polycomb repressive complexes 1 and 2 (PRC1 and PRC2) are pathway in the cytosol have been extensively studied, the events the most studied protein complexes among all the Polycomb group after NF-κB enters the nucleus are still not clear and have emerged (PcG) proteins and have long been considered as transcription as being crucial regulatory steps for pathway activation. Multiple repressors (Campos et al., 2014; Conway et al., 2015; Margueron and histone methyltransferases are involved in pathway regulation. We Reinberg, 2011). PRC1 inhibits transcription by regulating mono- have previously found that lysine (K)-specific methyltransferase ubiquitination on histone H2A, whereas PRC2 is responsible for the 2A (KMT2A; also known as MLL1), a histone H3K4 methylation on histone H3K27 (Margueron and Reinberg, 2011). methyltransferase, selectively regulates the activation of NF-κB In Drosophila, three major subunits form PRC2 complex, target genes (Wang et al., 2012). EZH2 regulates the activation of including Enhancer of zeste [E(z)], extra sexcombs (ESC) and NF-κB pathway by distinct mechanisms in different cell lines, suppressor of zeste 12 (SUZ12), among which E(Z) is the enzyme which shows the complexity of epigenetic regulators (Lee et al., for H3K27 methylation (Lanzuolo and Orlando, 2012). In 2011). The NF-κB molecule itself is also regulated by protein mammals, two close homologs exist for E(Z), enhancer of zeste 1 methylation. SET-domain-containing (lysine methyltransferase) 7 polycomb repressive complex 2 subunit (EZH1) and enhancer of (SET7; also known as SET9 and SETD7) methylates NF-κBat Lys314 and Lys315 and promotes NF-κB degradation (Yang et al., 2009). However, SET7 also methylates Lys37 on NF-κB and Department of Biochemistry and Molecular Biology, College of Life Sciences, selectively activates target genes (Ea and Baltimore, 2009). Besides Hubei Provincial Key Laboratory of Developmentally Originated Disease, Wuhan University, Wuhan, Hubei 430072, China. protein methylation, multiple transcription co-regulators have been found to be involved in the activation of NF-κB target genes. For *Authors for correspondence ([email protected]; [email protected]) example, the ubiquitously expressed prefoldin-like chaperone UXT M.W., 0000-0003-1372-4764 (also known as STAP1 and ART-27), a small chaperone-like protein, has been reported to interact with the p65 subunit of NF-κB

Received 30 December 2015; Accepted 20 April 2016 (also known as RELA) and functions as a transcription co-activator, Journal of Cell Science

2343 RESEARCH ARTICLE Journal of Cell Science (2016) 129, 2343-2353 doi:10.1242/jcs.185546 as well as a cytoplasmic regulator for anti-viral pathway (Huang genes (Sun et al., 2007). Given that EZH1 interacts with UXT, we et al., 2011, 2012; Sun et al., 2007). However, how UXT contributes studied whether EZH1 also regulates the NF-κB pathway. UXT was to the activation of NF-κB target genes is still not clear. knocked down by small interfering RNA (siRNA) in the HCT116 In this study, we discovered that UXT physically interacts with cell, and the TNFα-induced expression of NFKBIA (also known as EZH1 and SUZ12, but not with EED or EZH2. EZH1 and UXT are IκBα), CXCL8 and TNFAIP3 (also known as A20), three typical required for p65 recruitment to NF-κB target genes and the NF-κB target genes, were significantly downregulated, as reported subsequent induction of expression. EZH1 and UXT are associated previously (Fig. 3A; Sun et al., 2007). Then, we knocked down with Pol II, and regulate its loading to chromatin. Our study EZH1 in HCT116 with two different siRNAs. Similar to upon UXT indicates that UXT, EZH1 and SUZ12 together help to link p65 with knockdown, the induction of the above three genes was greatly Pol II and regulate the activation of NF-κB target genes. impaired, indicating that EZH1 positively regulates the activation of NF-κB target genes (Fig. 3B). We also knocked down EZH1 in RESULTS RKO, HEK293 and U2OS cell lines, and observed the same results The physical interaction between EZH1 and UXT (Fig. S1A–C). To investigate whether EZH1 regulates the activation To investigate the role of EZH1 in regulating transcription, a yeast of NF-κB target genes during the anti-viral response, Sendai virus two-hybrid screen was performed with full-length EZH1 as the bait. (SeV) was used to treat HCT116, and we found that the expression Multiple clones were identified encoding the open reading frame of of NF-κB downstream genes was significantly impaired after EZH1 UXT, which has been shown to be a transcription co-factor and to knockdown (Fig. S1D). To further confirm the role of EZH1 in NF- interact with multiple transcription factors (Carter et al., 2014; Li κB signaling, we profiled the gene expression pattern after EZH1 et al., 2014; McGilvray et al., 2007; Sun et al., 2007). We speculated knockdown with next-generation sequencing. EZH1 deficiency that UXT might regulate transcription by bridging the master impaired the expression of almost all the TNFα-induced genes, transcription factors and epigenetic regulators and continued with similar to the effect of UXT (Fig. 3C), whereas EZH1 only regulated the following studies. the basal expression of a small portion of genes (Table S1). This is EZH1 UXT We cloned the full-length cDNA of and into different from some of the other epigenetic regulators, such as mammalian expression vectors, and performed immunoprecipitation KMT2A, which selectively regulates the activation of NF-κB target – with anti-FLAG or anti-HA antibodies. HA EZH1 successfully genes by modifying H3K4 methylation (Wang et al., 2012), – pulled down FLAG UXT and vice versa (Fig. 1A). To examine the suggesting that EZH1 might regulate NF-κB signaling through a interaction between endogenous proteins, we generated antibodies distinctive mechanism. against the two proteins and performed immunoprecipitation. The Interestingly, when analyzing the signaling pathway after UXT results showed that endogenous EZH1 and UXT are associated with knockdown, the differentially expressed genes were found to be each other (Fig. 1B). To further characterize whether EZH1 directly specifically enriched in the gene ontology terms ubiquitin-mediated interacts with UTX, both proteins were expressed in bacteria and proteolysis, cell cycle, p53 signaling etc, suggesting that UXT – purified with GST or His affinity resins. Then His UXT was bound to might play roles in cancer-related signaling pathways (Fig. S2). – – Ni resins, which successfully pulled down GST EZH1. His SPOP To further explore the roles of other PRC2 subunits in NF-κB was used as a negative control (Fig. 1C). These results indicate that signaling, we knocked down SUZ12 or EED and analyzed the EZH1 and UXT directly interact with each other. Given that UXT is a expression of NF-κB target genes. Interestingly, SUZ12 α small protein and contains only one prefoldin domain, we knockdown, but not EED knockdown, impaired the activation of just mapped the interacting domain in EZH1. A series of EZH1 NFKBIA and CXCL8 (Fig. 3D). The above results suggest that the truncations were generated (Fig. 1D) and co-immunoprecipitation EZH1–SUZ12 complex positively regulates the activation of NF- assays indicated that the fragment from 430 to 480 residues in EZH1, κB target genes, probably through interaction with UXT, whereas which represents the SANT domain, is crucial for the interaction EED is not involved. (Fig. 1E–G).

UXT specifically interacts with EZH1 and SUZ12, but not EED UXT and EZH1 do not regulate the global H3K27 methylation One recent paper has demonstrated that two different EZH1 EZH2 is the major H3K27 methyltransferase in the mammalian cell complexes exist in the cell, one with EED and SUZ12, and and EZH1 has been proposed to compensate for its function under another with SUZ12 only (Xu et al., 2015). To study which complex certain circumstances (Margueron et al., 2008; Shen et al., 2008). We is associated with UXT, we performed immunoprecipitation with investigated the impact of UXT on global H3K27 methylation by UXT the anti-UXT antibody. The result indicated that EZH1 and SUZ12 western blotting. was knocked down by two different siRNAs are associated with UXT, but not EED, suggesting UXT is and the global H3K27me1, H3K27me2 and H3K27me3 levels were associated only with the EZH1–SUZ12 complex (Fig. 2A). To measured with corresponding antibodies. Multiple experiments were further confirm this result, exogenously expressed or endogenous performed and all of them indicated that the global levels of the three SUZ12 was pulled down with the corresponding antibody and UXT methylation statuses were not significantly affected (Fig. 4A). The EZH1 was found to be associated with SUZ12 (Fig. 2B,C). By contrast, similar results were observed with two different siRNAs immunoprecipitation of EED did not pull down UXT, nor did the (Fig. 4B), whereas depletion of EZH2 successfully decreased the reciprocal UXT immunoprecipitation (Fig. 2D). The assay was global H3K27me3 levels (Fig. S3A). This is consistent with the repeated more than three times in both directions to confirm that previous report that EZH2, but not EZH1, is the major enzyme endogenous EED and UXT are not associated with each other. involved in H3K27 methylation (Margueron et al., 2008). These results suggest that UXT specifically interacts with the κ EZH1–SUZ12 complex, but not with EED. EZH1 and UXT do not regulate H3K27 methylation on NF- B target genes The regulation of NF-κB pathway by EZH1 and SUZ12 Although EZH1 and UXT do not regulate the global H3K27 Previously, UXT has been reported to interact with p65 as a co- methylation level, it was still possible that they regulated the activator to transcriptionally regulate the expression of NF-κB target methylation on NF-κB target genes. We investigated the H3K27 Journal of Cell Science

2344 RESEARCH ARTICLE Journal of Cell Science (2016) 129, 2343-2353 doi:10.1242/jcs.185546

Fig. 1. The interaction between UXT and EZH1. (A) 293FRT cells were transfected with HA–EZH1 (H-EZH1) and FLAG–UXT (F-UXT). At 48 h after transfection, cells were immunoprecipitated (IP) and immunoblotted with the indicated antibodies. (B) HCT116 cells were immunoprecipitated and immunoblotted with the indicated antibodies. (C) In vitro pulldown assay with recombinant GST–EZH1 and His–UXT, with His-SPOP as a negative control. (D) Schematic illustration of EZH1 and its mutants used in this study. NLS, nuclear localization sequence. (E–G) FLAG–UXT was expressed in 293FRT cells with HA-tagged EZH1 or its deletion mutants. Cell lysates were immunoprecipitated with anti-FLAG and immunoblotted with anti-HA. Each experiment was repeated at least three times. methylation status on NFKBIA and CXCL8 by performing a H3K27me2 and H3K27me3 ChIP-Seq analysis. Consistent with the chromatin immunoprecipitation (ChIP) assay. Surprisingly, above result, although EZH1 or UXT deficiency altered H3K27 repeated experiments supported that the levels of H3K27me1, methylation on a few genes, the average levels on NF-κB target H3K27me2 and H3K27me3 on these genes were not significantly genes were not changed significantly (Fig. 4C). Actually, the reduced after EZH1 or UXT knockdown, although their expression average signals of all three forms of H3K27 methylation were quite decreased (Fig. S3B). Here, GAPDH and MYOD1 were used low on NF-κB target genes, in comparison with other genes as controls to ensure each experiment was performed correctly (Fig. 4C), which hints that H3K27 methylations might not be crucial

(Fig. S3B). To further confirm this result, we performed H3K27me1, in regulating the expression of NF-κB target genes here. Journal of Cell Science

2345 RESEARCH ARTICLE Journal of Cell Science (2016) 129, 2343-2353 doi:10.1242/jcs.185546

Fig. 2. UXT specifically interacts with EZH1 and SUZ12, but not with EED. (A) HCT116 cells were immunoprecipitated (Co-IP) and immunoblotted with the indicated antibodies. ns, non-specific bands. (B) 293FRT cells were transfected with HA–SUZ12 (H-SUZ12) and FLAG–UXT (F-UXT). Cell lysates were immunoprecipitated with anti-HA and immunoblotted with anti-FLAG antibody. (C,D) HCT116 cells were treated with 10 ng/ml TNFα for 2 h. Cell lysates were immunoprecipitated with anti-SUZ12 or anti-EED antibodies, and immunoblotted with anti-UXT. Each experiment was repeated at least three times.

EZH1 does not affect the translocation of RELA to nucleus whether EZH1 is bound to NF-κB target genes on chromatin. The above data demonstrate that the regulation of NF-κB signaling Because we do not have a ChIP grade anti-EZH1 antibody, we by UXT and EZH1 is not dependent on H3K37 methylation. A generated a stable HCT116 cell line expressing HA-tagged EZH1 previous study has reported that UXT deficiency impairs the (Fig. S4A) and performed a ChIP assay with anti-HA antibody translocation of p65 from the cytosol to the nucleus (Sun et al., (Fig. 5G). The results demonstrated that EZH1 binds to the 2007), we started to investigate whether EZH1 regulates NF-κB promoters of NFKBIA and CXCL8 on chromatin even without pathway through a similar mechanism. We firstly confirmed the role TNFα treatment, and when EZH1 was knocked down, the signal of UXT by western blotting after fractionation. After UXT significantly decreased (Fig. 5G). knockdown, the p65 level in the nucleus was modestly reduced Next, a p65 ChIP assay was performed in HCT116 cells. When compared with the control (Fig. 5A). Then the effect of EZH1 UXT was knocked down, the recruitment of p65 to NFKBIA and knockdown was investigated. Interestingly, EZH1 knockdown did CXCL8 promoters was greatly impaired, consistent with the previous not affect the p65 level in the nucleus (Fig. 5B). report (Fig. 5H; Sun et al., 2007). When EZH1 was knocked down, Then we examined the subcellular localization of EZH1. UXT the same result was observed (Fig. 5H). The result was then confirmed was localized both in the cytosol and the nucleus (Fig. 5C), as in 293FRT cells (Fig. S4B), as well as after virus treatment reported previously (Sun et al., 2007). However, the results of (Fig. S4C). These results demonstrate that EZH1 regulates the western blotting after fractionation and fluorescent staining both induced transcription of NF-κB target genes not through modifying indicated that EZH1 was mostly localized in the nucleus (Fig. 5C, histones, but by modulating the recruitment of NF-κB to target genes. D). Hence, it is highly possible that EZH1 interacts with UXT and We further investigated whether p65 also regulates EZH1 binding regulates the expression of NF-κB target genes only in the nucleus. to target genes. The RELA gene was knocked down in the HA– EZH1 stable cell line and a ChIP assay was performed with anti-HA EZH1 regulates the recruitment of p65 to target genes antibody. The data indicated that TNFα treatment increases EZH1 We next assessed whether EZH1 regulates the recruitment of p65 to binding to target genes and p65 deficiency impairs the binding chromatin. First, we studied whether EZH1 interacts with p65. As (Fig. 5I). All the above data suggest that p65 and EZH1 probably reported previously, UXT interacts with p65 (Fig. 5E; Sun et al., work synergistically to regulate the activation of their target genes. 2007). We did not detect an interaction between EZH1 and p65 at the endogenous level; however, upon co-expressing EZH1 and p65 EZH1 and UXT regulates Pol II recruitment to NF-κB target in the cell, we did observe an interaction between the exogenous genes proteins (Fig. 5F). This suggests that the interaction between EZH1 Pol II is the sole enzyme responsible for mRNA transcription and p65 is quite weak and perhaps not direct. Then we studied and in a previous study it has been reported EZH1 is associated Journal of Cell Science

2346 RESEARCH ARTICLE Journal of Cell Science (2016) 129, 2343-2353 doi:10.1242/jcs.185546

Fig. 3. The expression of NF-κB target genes is regulated by EZH1 and SUZ12. (A) HCT116 cells were transfected with siRNA for UXT (siUXT). At 48 h after transfection, cells were stimulated by TNFα for the indicated times. The mRNA levels of NFKBIA, CXCL8 and TNFAIP3 were measured by quantitative RT-PCR. (B) EZH1 was knocked down by two independent siRNAs (siEZH1.1 or siEZH1.2), and an experiment as in A was performed. (C) HCT116 cells were transfected with the indicated siRNA. At 48 h after transfection, cells were treated with TNFα for 2 h. Gene expression profile was assayed by RNA- sequencing. The heat map shows the mRNA level of the TNFα-induced differentially expressed genes (DEG) in the indicated cells. (D) HCT116 cells were transfected with the indicated siRNAs. At 48 h after transfection, cells were treated with TNFα for 2 h. Endogenous mRNA expression of NFKBIA and CXCL8 were measured by quantitative real-time RT-PCR. *P<0.05; **P<0.01 (t-test). Results in A, B and D are mean±s.d. for three experiments. siNC, control siRNA. Journal of Cell Science

2347 RESEARCH ARTICLE Journal of Cell Science (2016) 129, 2343-2353 doi:10.1242/jcs.185546

Fig. 4. EZH1 and UXT do not regulate H3K27 methylation on NF-κB target genes. (A,B) UXT (siUXT.1 or siUXT.2) and EZH1 (siEZH1.1 or siEZH1.2) were knocked down in HCT116. After 72 h, cells were harvested and lysates were analyzed by western blotting. (C) UXT and EZH1 were knocked down in HCT116 and a ChIP-Seq assay was performed. Profiles of histone H3K27me1, me2 and me3 across the gene bodies of TNFα-induced genes are shown. TSS, transcriptional start site; TES, transcriptional termination site. siNC, control siRNA.

with Pol II (Mousavi et al., 2012). To investigate whether after TNFα treatment (Fig. 7B). NF-κB signaling inhibits SUZ12 and UXT are associated with Pol II, we performed apoptosis through the activation of downstream anti-apoptosis immunoprecipitation with a widely used commercial monoclonal genes. We surveyed our RNA-seq data and confirmed by RT-PCR antibody 8WG16 in HCT116, and found that EZH1, UXT and that TNFα activates the expression of BIRC2 (encoding cIAP1) SUZ12 were all associated with Pol II, but that EED was not and BIRC3 (encoding cIAP2), two well-known anti-apoptotic (Fig. 6A,B). The recruitment of Pol II to NFKBIA and CXCL8 was genes, in the studied HCT116 cell line (Fig. S4E). Further study also greatly reduced in the absence of EZH1, UXT or SUZ12, but showed that EZH1 and UXT both regulate the expression of not EED (Fig. 6C,D). The ChIP data was confirmed in 293FRT BIRC2 and BIRC3 induced by TNFα (Fig. 7C). These results cells (Fig. S4D). These results are consistent with our previous suggest that EZH1 and UXT increase the sensitivity of the cells to data, suggesting that EZH1, SUZ12 and UXT function apoptosis, probably through regulating the expression of BIRC2 synergistically in regulating NF-κB signaling, but that EED and BIRC3. does not. DISCUSSION EZH1 deficiency increases the sensitivity to TNFα-induced The epigenetic regulation of signaling pathways has emerged to be apoptosis one of the crucial steps in the cell for responding to the extra- and TNFα treatment not only activates NF-κB signaling, but also the intra-cellular signals. Polycomb group proteins are key factors in apoptosis pathway. The activation of NF-κB pathway strongly determining cell status and transcriptional programs. However, the inhibits apoptosis, so the inhibition of NF-κB activation often functions and mechanisms of EZH1 have remained controversial. In enhances TNFα-induced apoptosis. To further confirm the role of the current study, we demonstrate that EZH1 regulates the EZH1 in TNFα signaling, we studied its effect on the apoptosis transcription of NF-κB target genes by interacting with UXT, a induced by TNFα. EZH1 or UXT knockdown alone slightly small protein functioning as a transcription co-activator. EED is increased the percentage of sub-G1 cells, which represents dead required for PRC2 activity (Montgomery et al., 2005, 2007), but cells (Fig. 7A). With TNFα treatment, EZH1 or UXT knockdown because UXT here is just associated with EZH1–SUZ12, and not significantly increased the percentage of apoptotic cells (Fig. 7A). EED, it is reasonable that H3K27 methylation is not involved in the To further confirm this result, we used annexin V and propidium transcription activation regulated by UXT and EZH1. Instead, both iodide to double label the cell, and flow cytometry analysis EZH1 and UXT are required for the expression of NF-κB target confirmed that EZH1 and UXT deficiency increased the cell death genes, and proper p65 recruitment and Pol II loading to their target Journal of Cell Science

2348 RESEARCH ARTICLE Journal of Cell Science (2016) 129, 2343-2353 doi:10.1242/jcs.185546

Fig. 5. EZH1 and UXT regulate the recruitment of p65 to target genes. (A,B) HCT116 cells were transfected with the UXT (siUXT) or EZH1 (siEZH1) siRNAs and treated with 10 ng/ml TNFα for the time indicated. Cytoplasmic and nuclear fractions were prepared and immunoblotted with the indicated antibodies. (C) Cytoplasmic and nuclear fractions of HCT116 cells were prepared and assayed by western blotting. (D) FLAG–EZH1 was expressed in the cell, and immunofluorescent microscopy was performed. (E) Cells were treated with 10 ng/ml TNFα for 2 h, and immunoprecipitation (IP) was performed with anti-UXT antibody. (F) 293FRT cells were transfected with indicated plasmids, and, after 48 h, cells were treated with TNFα for 2 h. Immunoprecipitation and western blotting were performed as indicated. (G) A stable cell line expressing HA-tagged EZH1 (H-EZH1) was transfected with EZH1 siRNAs. After 72 h, cells were harvested for real-time RT-PCR (right) or ChIP assays (left). (H) HCT116 cells were transfected with the indicated siRNAs and, 72 h later, cells were induced with 10 ng/ml TNFα for 2 h. The amount of p65 on NFKBIA and CXCL8 promoters was assayed by ChIP. (I) RELA was knocked down by siRNA in the stable cell line expressing HA–EZH1 and ChIP analysis was performed with anti-HA antibody. The inset shows a western blot of the p65 level after siRNA knockdown. *P<0.05;

**P<0.01 (t-test). Results in G–I are mean±s.d. for three experiments. siNC, control siRNA. Journal of Cell Science

2349 RESEARCH ARTICLE Journal of Cell Science (2016) 129, 2343-2353 doi:10.1242/jcs.185546

Fig. 6. Recruitment of RNA Pol II to NF-κB target genes regulated by EZH1 and UXT. (A) HCT116 cells were treated with or without TNFα, and lysates were immunoprecipitated with 8WG16 and immunoblotted with the indicated antibodies. (B) Lysates as in A were immunoprecipitated with 8WG16 and immunoblotted with SUZ12 and EED antibodies. (C,D) Control siRNA (siNC) or siRNA against UXT (siUXT) or EZH1 (siEZH1) were transfected into HCT116 cells as indicated. Cells were treated with 10 ng/ml TNFα 2 h later. RNA Pol II bound on the promoters of NFKBIA and CXCL8 was examined by a ChIP assay with 8WG16. *P<0.05; **P<0.01 (t-test). Results in C and D are mean±s.d. for three experiments. genes. Here, UXT, EZH1 and SUZ12 seem to act like a bridge to MATERIALS AND METHODS link NF-κB and Pol II together. Cell lines and antibodies The functional difference between EZH2 and EZH1 on The 293FRT cell line was purchased from Invitrogen, and HEK293 transcription regulation is also puzzling. Here, we show that and HCT116 cell lines were purchased from Cell Bank of Chinese Academy ’ ’ EZH1 positively regulates transcription in an EED- and H3K27- of Sciences. Cells were grown in Dulbecco s modified Eagle s medium methylation-independent manner. Our data are consistent with the (DMEM) (4.5 g/l D-glucose) with L-glutamine, 25 mM HEPES and 10% fetal bovine serum (FBS, HyClone) and a penicillin–streptomycin recent report by Xu et al. (2015). In fact, EZH2 sometimes also supplement. The stable cell line containing HA-tagged EZH1 was positively regulates transcription independently of its enzyme constructed with a lentivirus system (pHAGE, psPAX2 and pMG2.G) activity in a context-dependent manner (Lee et al., 2011; Xu in HCT116 cells. The antibodies against FLAG (1:1000 for immunoblotting, et al., 2012). Our study is helpful to understand the detailed 1:500 for immunoprecipitation; M2, Sigma), HA (1:1000 for mechanisms of how EZH1 and EZH2 regulate gene expression by immunoblotting, 1:500 for immunoprecipitation; clone CB051, Origene), different mechanisms. GST (1:1000 for immunoblotting; Abmart, M20007), His (1:1000 for UXT has been reported to regulate multiple signaling pathways, immunoblotting; Abmart, M20001), SUZ12 (1:1000 for immunoblotting, making it an important regulator for inducible transcription (Carter 1:500 for immunoprecipitation; clone D39F6, CST 3737), EED (1:1000 et al., 2014; Li et al., 2014; McGilvray et al., 2007; Sun et al., 2007). for immunoblotting, 1:500 for immunoprecipitation; Proteintech, 16818-1- AP), H3K27me1 (1:1000 for immunoblotting, 1:250 for ChIP; Millipore 07- Meanwhile, EZH1 is widely associated with positive transcription 448), H3K27me2 (1:1000 for immunoblotting, 1:250 for ChIP; ACTIVE genome. Hence, it is highly possible that UXT and EZH1 use the same MOTIF 39919), H3K27me3 (1:1000 for immunoblotting, 1:250 for ChIP; mechanisms to regulate other signaling pathways. It will be interesting ABclonal A2363, Millipore 07-449), H3 (1:1000 for immunoblotting; to study whether EZH1 regulates other UXT-dependent pathways. ABclonal A2348), NFKBIA (1:1000 for immunoblotting; Epitomics, Journal of Cell Science

2350 RESEARCH ARTICLE Journal of Cell Science (2016) 129, 2343-2353 doi:10.1242/jcs.185546

Fig. 7. TNFα-induced apoptosis is regulated by EZH1. (A) HCT116 cells were transfected with UXT (siUXT) or EZH1 (siEZH1) siRNAs, and then treated with or without TNFα (0.05 ng/ml) for 12 h. Cell death was measured with propidium iodide (PI) staining followed by flow cytometry. The statistical calculation of the proportion of sub-G1 cells is shown in the right panel. (B) Cells were prepared as in A. Cells were double stained with annexin-V–FITC and propidium iodide, and then assayed by flow cytometry. The statistical calculation of the proportion of annexin-V-positive cells is shown in the right panel. (C) UXT or EZH1 was knocked down in HCT116 cells. The expression of BIRC2 and BIRC3 after TNFα treatment was measured by quantitative RT-PCR. *P<0.05; **P<0.01 (t-test). Results are mean ±s.d. for three experiments. siNC, control siRNA.

1130-1), 8WG16 (1:2000 for immunoblotting, 1:500 for ChIP; Cell fractionation Covance) and p65 (1:1000 for immunoblotting, 1:250 for ChIP; Abcam, Cells were harvested and spun down in cold PBS. 10 volumes of buffer A ab7970) were purchased from indicated companies. The antibodies (10 mM Tris-HCl pH 7.4, 5 mM MgCl2, 10 mM NaCl, 1 mM DTT, against EZH1 and UXT were raised against full-length EZH1 or UXT proteinase inhibitors) was added to the cells, which were then incubated on expressed in bacteria using the pET30-C plasmid. The siRNA information is ice for 10 min. Next, 0.5 volumes of buffer B (10 mM Tris-HCl pH 7.4, in Table S2. 5 mM MgCl2, 10 mM NaCl, 1 mM DTT, proteinase inhibitors, 10% NP-40) Journal of Cell Science

2351 RESEARCH ARTICLE Journal of Cell Science (2016) 129, 2343-2353 doi:10.1242/jcs.185546 was added to the cells, which were incubated on ice for 1 min. The cell KAPA Hyper Prep Kit. We used 10 ng of ChIP DNA as initiation, prepared suspension was vortexed for 5 s and centrifuged at 800 g for 5 min at 4°C. according to the manufacturer’s instructions. The ChIPed DNA from three The supernatant was collected as cytoplasm fraction. The above steps were biological replicates were mixed together and subjected to high-throughput repeated once more and the supernatant was discarded. The sediment was sequencing. ChIP DNA and matched input DNA were prepared for end suspended in 10 volumes of PBS as the nuclear fraction. SDS loading buffer repair and ‘A’-tailing, adaptor ligation and library amplification. ChIP-Seq was added to the cell fractions for western blotting. sequencing was performed by using an Illumina HiSeq2500 platform for 100-bp paired-end sequencing. Quality control of ChIP-seq data was Immunofluorescent staining performed using Fatsqc, and low-quality bases and library adaptors were Cells were cultured on coverslips and fixed with freezing methanol after trimmed. After quality control, data were mapped to hg19 genome reference washing twice in PBS. The coverslips were then washed three times by PBS by bowtie Tophat2 allowing a maximum of two mismatches. For histone and blocked in PBS with 1% BSA for 10 min. The coverslips were modification analysis, SICER software was used to call peaks and identify hybridized with primary and secondary antibodies for 1 hour each. Then the the differential modification region. coverslips were mounted with prolong anti-fade kit (Invitrogen) and observed with fluorescent microscopy. Acknowledgements We thank Dr Hong-bing Shu of Wuhan University and Dr Yan-Yi Wang of the Institute Immunoprecipitation of Viology for sharing reagents and plasmids. The cells were harvested and lysed in NP40 lysis buffer (50 mM Tris-HCl pH 7.4, 150 mM NaCl, 0.5% NP40) or high-salt lysis buffer (20 mM Competing interests The authors declare no competing or financial interests. HEPES pH 7.4, 10% glycerol, 0.35 M NaCl, 1 mM MgCl2, 0.5% Triton X-100, 1 mM DTT) with proteinase inhibitors. The supernatant was then incubated with protein G beads (GE Healthcare) and the desired antibody at Author contributions 4°C for 4 h. The beads were spun down and washed three times with lysis S.-K.S. performed most of the experiments. C.-Y.L. initiated the yeast two-hybrid screening and identified the EZH1–UXT interaction. P.-J.L. analyzed the data from buffer. The final drop of wash buffer was vacuumed out and SDS loading RNA-sequencing and ChIP-sequencing. X.W. and Q.-Y.Z. helped perform ChIP buffer was added to the beads, followed by western blotting. assays. Y.C. and Z.W. contributed to plasmid construction. M.W. and L.L. directed the project, conducted the experiments and wrote the manuscript. ChIP assay ChIP assays were performed as previously described (Wu et al., 2008). Funding 7 Briefly, ∼10 cells were fixed with 1% formaldehyde and quenched by This work was supported by grants from the National Basic Research Program of glycine. The cells were washed three times with PBS and then harvested in China [973 Program, grant number 2012CB518700]; and the National Natural ChIP lysis buffer (50 mM Tris-HCl, pH 8.0, 1% SDS and 5 mM EDTA). Science Foundation of China [grant numbers 31470771 to M.W., and 31221061, DNA was sonicated to 400–600 bp before extensive centrifugation. Four 31200653, 31370866, 31540013 to L.L.]. volumes of ChIP dilution buffer (20 mM Tris-HCl pH 8.0, 150 mM NaCl, 2 mM EDTA and 1% Triton X-100) was added to the supernatant. Data availability The resulting lysate was then incubated with protein G beads and antibodies The raw data from the next-generation sequencing have been uploaded to GEO at 4°C overnight. The beads were washed five times and DNA was eluted in database under the accession number GSE75217 (http://www.ncbi.nlm.nih.gov/ geo/query/acc.cgi?acc=GSE75217). ChIP elution buffer (0.1 M NaHCO3, 1% SDS and 30 μg/ml proteinase K). The elution was incubated at 65°C overnight and DNA was extracted with a Supplementary information DNA purification kit (Tiangen). The purified DNA was assayed by Supplementary information available online at quantitative PCR with Biorad MyIQ. Assays were repeated at least three http://jcs.biologists.org/lookup/suppl/doi:10.1242/jcs.185546/-/DC1 times. Data shown are mean±s.d. of representative experiments. The primer information is in Table S2. At least three biological replicates were analyzed References in each experiment. A t-test was used for statistical analysis. Bae, W. K., Kang, K., Yu, J. H., Yoo, K. H., Factor, V. M., Kaji, K., Matter, M., Thorgeirsson, S. and Hennighausen, L. (2015). The methyltransferases Reverse transcription and quantitative PCR enhancer of zeste homolog (EZH) 1 and EZH2 control hepatocyte homeostasis Cells were scraped down and collected by centrifugation. Total RNA was and regeneration. FASEB J. 29, 1653-1662. extracted with an RNA extraction kit (Yuanpinghao) according to the Campos, E. I., Stafford, J. M. and Reinberg, D. (2014). Epigenetic inheritance: Trends Cell Biol. manufacturer’s manual. Approximately 1 µg of total RNA was used for histone bookmarks across generations. 24, 664-674. Carter, D. R., Buckle, A. D., Tanaka, K., Perdomo, J. and Chong, B. H. (2014). reverse transcription with a first-strand cDNA synthesis kit (Toyobo). The β Art27 interacts with GATA4, FOG2 and NKX2.5 and is a novel co-repressor of amount of mRNA was assayed by quantitative PCR. -actin was used to cardiac genes. PLoS ONE 9, e95253. normalize the amount of each sample. Assays were repeated at least three Conway, E., Healy, E. and Bracken, A. P. (2015). PRC2 mediated H3K27 times. Data shown are mean±s.d. of one representative experiment. The methylations in cellular identity and cancer. Curr. Opin. Cell Biol. 37, 42-48. primer information is in Table S2. At least three biological replicates were Ea, C.-K. and Baltimore, D. (2009). Regulation of NF-kappaB activity through lysine analyzed in each experiment. A t-test was used for statistical analysis. monomethylation of p65. Proc. Natl. Acad. Sci. USA 106, 18972-18977. Hidalgo, I., Herrera-Merchan, A., Ligos, J. M., Carramolino, L., Nuñez, J., RNA-sequencing and data analysis Martinez, F., Dominguez, O., Torres, M. and Gonzalez, S. (2012). Ezh1 is The extracted mRNA from three biological replicates was subjected to high- required for hematopoietic stem cell maintenance and prevents senescence-like Cell Stem Cell throughput sequencing. The mRNA-seq library was constructed by using the cell cycle arrest. 11, 649-662. μ Hu, M.-M., Yang, Q., Zhang, J., Liu, S.-M., Zhang, Y., Lin, H., Huang, Z.-F., Wang, Illumina TruSeq library construction kit. 5 g total RNA was used for Y.-Y., Zhang, X.-D., Zhong, B. et al. (2014). TRIM38 inhibits TNFalpha- and IL- initiation, prepared according to the manufacturer’s instructions. mRNA-seq 1beta-triggered NF-kappaB activation by mediating lysosome-dependent libraries were sequenced using HiSeq2000 for 100-bp paired-end sequencing. degradation of TAB2/3. Proc. Natl. Acad. Sci. USA 111, 1509-1514. Quality control of mRNA-seq data was performed using Fatsqc, and low- Huang, Y., Chen, L., Zhou, Y., Liu, H., Yang, J., Liu, Z. and Wang, C. (2011). UXT- quality bases were trimmed. After quality control, data were mapped to hg19 V1 protects cells against TNF-induced apoptosis through modulating complex II genome reference by Tophat2 allowing a maximum of two mismatches. formation. Mol. Biol. Cell 22, 1389-1397. Cufflinks was used to assess the differentially expressed genes. Gene ontology Huang, Y., Liu, H., Ge, R., Zhou, Y., Lou, X. and Wang, C. (2012). UXT-V1 facilitates the formation of MAVS antiviral signalosome on mitochondria. analysis was performed using DAVID (http://david.abcc.ncifcrf.gov). J. Immunol. 188, 358-366. Iwai, K. (2012). Diverse ubiquitin signaling in NF-kappaB activation. Trends Cell ChIP-sequencing and data analysis Biol. 22, 355-364. ChIP was performed by using antibody against H3K27me1, H3K27me2 Lanzuolo, C. and Orlando, V. (2012). Memories from the polycomb group proteins. and H3K27me3. After ChIP, a sequence library was prepared by using a Annu. Rev. Genet. 46, 561-589. Journal of Cell Science

2352 RESEARCH ARTICLE Journal of Cell Science (2016) 129, 2343-2353 doi:10.1242/jcs.185546

Lee, S. T., Li, Z., Wu, Z., Aau, M., Guan, P., Karuturi, R. K. M., Liou, Y. C. and Yu, complements EZH2 in maintaining stem cell identity and executing pluripotency. Q. (2011). Context-specific regulation of NF-kappaB target gene expression by Mol. Cell 32, 491-502. EZH2 in breast cancers. Mol. Cell 43, 798-810. Sun, S., Tang, Y., Lou, X., Zhu, L., Yang, K., Zhang, B., Shi, H. and Wang, C. Li, W., Wang, L., Jiang, C., Li, H., Zhang, K., Xu, Y., Hao, Q., Li, M., Xue, X., Qin, X. (2007). UXT is a novel and essential cofactor in the NF-kappaB transcriptional et al. (2014). UXT is a novel regulatory factor of regulatory T cells associated with enhanceosome. J. Cell Biol. 178, 231-244. Foxp3. Eur. J. Immunol. 44, 533-544. Wang, X., Zhu, K., Li, S., Liao, Y., Du, R., Zhang, X., Shu, H.-B., Guo, A.-Y., Li, L. Margueron, R. and Reinberg, D. (2011). The Polycomb complex PRC2 and its and Wu, M. (2012). MLL1, a H3K4 methyltransferase, regulates the TNFalpha- Nature mark in life. 469, 343-349. stimulated activation of genes downstream of NF-kappaB. J. Cell Sci. 125, Margueron, R., Li, G., Sarma, K., Blais, A., Zavadil, J., Woodcock, C. L., 4058-4066. Dynlacht, B. D. and Reinberg, D. (2008). Ezh1 and Ezh2 maintain repressive Wu, M., Wang, P. F., Lee, J. S., Martin-Brown, S., Florens, L., Washburn, M. and chromatin through different mechanisms. Mol. Cell 32, 503-518. Shilatifard, A. (2008). Molecular regulation of H3K4 trimethylation by Wdr82, a McGilvray, R., Walker, M. and Bartholomew, C. (2007). UXT interacts with the component of human Set1/COMPASS. Mol. Cell. Biol. 28, 7337-7344. transcriptional repressor protein EVI1 and suppresses cell transformation. FEBS Xu, K., Wu, Z. J., Groner, A. C., He, H. H., Cai, C., Lis, R. T., Wu, X., Stack, E. C., J. 274, 3960-3971. Loda, M., Liu, T. et al. (2012). EZH2 oncogenic activity in castration-resistant Montgomery, N. D., Yee, D., Chen, A., Kalantry, S., Chamberlain, S. J., Otte, prostate cancer cells is Polycomb-independent. Science 338, 1465-1469. A. P. and Magnuson, T. (2005). The murine polycomb group protein Eed is Xu, J., Shao, Z., Li, D., Xie, H., Kim, W., Huang, J., Taylor, J. E., Pinello, L., Glass, required for global histone H3 lysine-27 methylation. Curr. Biol. 15, 942-947. Montgomery, N. D., Yee, D., Montgomery, S. A. and Magnuson, T. (2007). K., Jaffe, J. D. et al. (2015). Developmental control of polycomb subunit Mol. Molecular and functional mapping of EED motifs required for PRC2-dependent composition by GATA factors mediates a switch to non-canonical functions. histone methylation. J. Mol. Biol. 374, 1145-1157. Cell 57, 304-316. Mousavi, K., Zare, H., Wang, A. H. and Sartorelli, V. (2012). Polycomb protein Yang, X.-D., Huang, B., Li, M., Lamb, A., Kelleher, N. L. and Chen, L.-F. (2009). Ezh1 promotes RNA polymerase II elongation. Mol. Cell 45, 255-262. Negative regulation of NF-kappaB action by Set9-mediated lysine methylation of Oeckinghaus, A., Hayden, M. S. and Ghosh, S. (2011). Crosstalk in NF-kappaB the RelA subunit. EMBO J. 28, 1055-1066. signaling pathways. Nat. Immunol. 12, 695-708. Zhang, Y., Lei, C.-Q., Hu, Y.-H., Xia, T., Li, M., Zhong, B. and Shu, H.-B. (2014). Shen, X., Liu, Y., Hsu, Y.-J., Fujiwara, Y., Kim, J., Mao, X., Yuan, G.-C. and Kruppel-like factor 6 is a co-activator of NF-kappaB that mediates p65-dependent Orkin, S. H. (2008). EZH1 mediates methylation on histone H3 lysine 27 and transcription of selected downstream genes. J. Biol. Chem. 289, 12876-12885. Journal of Cell Science

2353