Acetylation of CCAR2 Establishes a BET/BRD9 Acetyl Switch In

Published OnlineFirst January 14, 2019; DOI: 10.1158/0008-5472.CAN-18-2003

Cancer Genome and Epigenome Research

Acetylation of CCAR2 Establishes a BET/BRD9 Acetyl Switch in Response to Combined Deacetylase and Bromodomain Inhibition Praveen Rajendran1, Gavin Johnson1,LiLi1, Ying-Shiuan Chen1, Mohaiza Dashwood1, Nhung Nguyen1, Ahmet Ulusan1, Furkan Ertem1, Mutian Zhang1, Jia Li1, Deqiang Sun1, Yun Huang1, Shan Wang1, Hon-Chiu Leung2, David Lieberman3, Laura Beaver4, Emily Ho4, Mark Bedford5, Kyle Chang5, Eduardo Vilar5, and Roderick Dashwood1,5

Abstract

There continues to be interest in targeting epigenetic "read- members of the bromodomain and extraterminal domain ers, writers, and erasers" for the treatment of cancer and other (BET) family. Treatment with the BET inhibitor JQ1 synergized pathologies. However, a mechanistic understanding is fre- with sulforaphane in colon cancer cells and suppressed tumor quently lacking for the synergy observed when combining development effectively in a preclinical model of colorectal deacetylase and bromodomain inhibitors. Here we identify cancer. Studies with sulforaphaneþJQ1 in combination impli- cell cycle and apoptosis regulator 2 (CCAR2) as an early target cated a BET/BRD9 acetyl switch and a shift in the pool of acetyl for acetylation in colon cancer cells treated with sulforaphane. "reader" proteins in favor of BRD9-regulated target genes. N-terminal acetylation of CCAR2 diminished its interactions with histone deacetylase 3 and b-catenin, interfering with Wnt Signiﬁcance: These results highlight the competition that coactivator functions of CCAR2, including in cells harboring exists among the "readers" of acetylated histone and nonhis- genetically encoded CCAR2 acetylation. Protein domain tone proteins and provide a mechanistic basis for potential arrays and pull-down assays identiﬁed acetyl "reader" proteins new therapeutic avenues involving epigenetic combination that recognized CCAR2 acetylation sites, including BRD9 and treatments.

Introduction 3 (HDAC3), while also interacting with b-catenin to stabilize b-catenin/Tcf complexes in the nucleus (6, 7). In doing so, CCAR2 Cell cycle and apoptosis regulator 2 (CCAR2), also known as serves as a coactivator of Wnt signaling, a well-studied pathway in DBC1/KIAA1967, has gained attention as a "master regulator" of disease and development (8). metabolism, aging, and cancer (1–4). This designation derives Our attention was drawn to CCAR2 based on two converging from the interactions of CCAR2 with protein partners that observations. First, when CCAR2 is overexpressed in colon exert critical roles in physiology and pathophysiology, including tumors, the corresponding patients exhibit signiﬁcantly reduced Sirtuin 1 (SIRT1) and CHK2, linking CCAR2 to p53 function and survival (7). Second, as reported here, CCAR2 was identiﬁed as an DNA repair (1–7). Less is known about the N-terminal region of early target for acetylation by sulforaphane, an agent that causes CCAR2 that associates with, and inhibits, histone deacetylase inhibition and turnover of HDAC3 in colon cancer cells (9–12). Notably, when sulforaphane was combined with JQ1, an inhib-

1 itor of the bromodomain and extraterminal domain (BET) fam- Center for Epigenetics & Disease Prevention, Texas A&M College of Medicine, – Houston, Texas. 2Mass Spectrometry-Proteomics Core, Baylor College of Med- ily (13 15), CCAR2 no longer served as an effective coactivator of icine, Houston, Texas. 3Division of Gastroenterology and Hepatology, Oregon Wnt/b-catenin signaling in vitro and in vivo. Health & Science University, Portland, Oregon. 4College of Public Health and There is growing interest in targeting epigenetic "readers, Human Sciences, Oregon State University, Corvallis, Oregon. 5The University of writers, and erasers" deregulated in cancer and other patholo- Texas MD Anderson Cancer Center, Houston, Texas. gies (13–16). This investigation combined sulforaphaneþJQ1 Note: Supplementary data for this article are available at Cancer Research to affect CCAR2 acetylation, and in so doing provided new Online (http://cancerres.aacrjournals.org/). mechanistic insights into the competition that exists among the P. Rajendran and G. Johnson contributed equally and are the co-ﬁrst authors of "readers" of acetylated histone and nonhistone proteins that are this article. regulated during epigenetic combination therapies. Corresponding Authors: Roderick Dashwood, Texas A&M Health Sci Center and MD Anderson Cancer Center, 2121 West Holcombe Blvd., Houston, TX 77030. Phone: 713-677-7806; Fax: 713-677-7784; E-mail: [email protected]; Materials and Methods and Praveen Rajendran, Director, Antibody Biopharmaceutics Core, Center for Cells and treatments Epigenetics & Disease Prevention, Texas A&M Health Sci Center, Houston, TX HCT116, SW480 (human colon cancer cells), and 77030. Phone: 713-677-7803; E-mail: [email protected] CCD841 (nontransformed colonic epithelial cells) were from doi: 10.1158/0008-5472.CAN-18-2003 ATCC, and used within 10–15 passages from receipt. Each cell 2019 American Association for Cancer Research. line was conﬁrmed independently to be of human origin, with no

918 Cancer Res; 79(5) March 1, 2019

BET/BRD9 Acetyl Switch on CCAR2

mammalian interspecies contamination, and with the correct D1, matrix metalloproteinase 7 (MMP7), PARP, caspase-3, Pin1, genetic profile based on allele-specific markers (IDEXX Bioana- Lamin, and b-actin primary antibodies were from sources lytics Radil; refs. 17, 18). Cells were cultured in McCoy's 5A media reported (9–11, 17, 24). IHC followed the general procedures (Invitrogen) or EMEM (Invitrogen), supplemented with 10% FBS described elsewhere (17, 24). and 1% penicillin/streptomycin, at 37C in a humidified chamber with 5% CO2. All cells were tested routinely for Mycoplasma by Proximity ligation assays DAPI staining, and by using a PCR-based methodology (19). Protein–protein interactions were examined in situ, in cell- JQ1 was purchased from MedChem Express, whereas the other based assays and tissue sections, using the Duolink PLA Fluores- test agents were from the sources noted elsewhere (11). Nominal cence Protocol (Sigma-Aldrich), according to the manufacturer's concentrations were as follows: 15 mmol/L sulforaphane, recommendations. 6-methylsulfinylhexyl isothiocyanate (6-SFN), 9-methylsulfinyl- nonyl isothiocyanate (9-SFN), and allyl isothiocyanate (AITC); Pulldown assays 1 mmol/L trichostatin A (TSA); 10 mmol/L sodium butyrate Immunoprecipitation (IP) methodologies were as reported for (NaB); and 1 mmol/L valproic acid (VPA). For combination index endogenous proteins (10, 11), or Myc-, GST-, and HA-tagged (CI) experiments, sulforaphane, JQ1, and suberoylanilide hydro- proteins (5, 25–27). xamic acid (SAHA) were tested in the range 2–17 mmol/L, 1–60 mmol/L, and 0.1–2 mmol/L, respectively, with DMSO as Mass spectrometry vehicle. In most experiments, cells were treated with test agents 24 Acetylation sites on CCAR2 were identified following the hours after seeding (10, 11), except in HDAC3 siRNA knockdown general approach reported previously (5). In brief, 24 hours after assays, which were conducted according to a published method- seeding, HCT116 cells were treated with sulforaphane or DMSO, ology (11, 12). HDAC3 siRNA (Trilencer-27) and control siRNA and 6 hours later the cell lysates were subjected to IP using CCAR2 were procured from Origene, and cells were transfected with antibody. Following SDS-PAGE separation, the CCAR2 band was RNAiMAX Reagent (Invitrogen) for 24–48 hours, using the man- excised from the gel and digested overnight with trypsin prior to ufacturer's protocol. Two of the target siRNAs, designated as extraction and analysis on an Eksigent cHiPLC with nanoLC siRNA(1) and siRNA(3), produced the most efficient knockdown linked via a nanoflex to an ABSCIEX TripleTOF 5600 of HDAC3, and the data are shown in the corresponding figures. Mass Spectrometer (Mass Spectrometry-Proteomics Core, Baylor Unless indicated otherwise, whole-cell lysates or nuclear and College of Medicine, Houston, TX). Peaks Studio version 7.0 cytoplasmic fractions (10) were harvested 6 hours after treatment (Bioinformatics Solutions Inc.) was used to match spectra to with test agents, followed by RNA or protein expression analyses. peptides using the NCBI nonredundant database, including con- Additional experiments involved CCAR2 deletion from colon sideration of lysine acetylation. Modified peptides were verified cancer cells via CRISPR/Cas9 genome editing (20, 21). The PX459 by manual inspection of MS/MS data. Vector Control (Addgene) included a nontargeting gRNA sequence integrated into the vector. For reintroduction of CCAR2 RNA analyses into CCAR2-null cells, transient transfection was conducted using RNA-sequencing (RNA-seq) and bioinformatics analyses were expression constructs for wild-type (WT) protein or acetylation as reported (28) for adenomatous colon polyps from patients mutants. In the latter case, a Q5 Site-Directed Mutagenesis Kit with familial adenomatous polyposis (FAP; GSE88945 and (New England Biolabs) was used to convert Lys to Arg, starting GSE106500) and the polyposis in rat colon (Pirc) preclinical with CCAR2 plasmid pcDNA Myc DBC1 (Addgene plasmid no. model (29). Library preparation via a NEBNext Ultra Directional 35096; ref. 22), with confirmation by direct sequencing. RNA Library Prep Kit was followed by Illumina sequencing on a NextSeq 500/550 instrument (Illumina). Real-time reverse tran- Genetically encoded acetylation of CCAR2 scription quantitative PCR (RT-qPCR) was conducted according A system for genetically encoded Lys acetylation on his- to a reported methodology (28). tones (23) was adopted for CCAR2. In brief, CCAR2 and 3xHA were PCR amplified and subcloned into pGEM-9Zf() Docking in silico (Promega catalog no. P2391) to generate HA-CCAR2. HA-CCAR2 After multiple sequence alignment (30), docking of BRD2, was restriction cloned into pE337, replacing H3.3-HA. Q5 site- BRD3, BRD4, and BRD9 was performed using AutoDock directed mutagenesis was used to convert Lys to TAG stop codons Vina (31), on CCAR2 structures predicted via SWISS-MOD- at defined sites in CCAR2. Plasmids pE312 and pE337-HA- EL (32). Ligand–protein interactions were analyzed using PDBe- CCAR2 WT and Lys mutants were stably expressed in CCAR2- PISA (33, 34) and LPC/CSU (35). Initial work-up confirmed that null HCT116 cells using Super PiggyBac Transposase vector (SBI the docking of JQ1 with BRD2, BRD3, and BRD4 corresponded catalog no. PB210PA-1) and selected with puromycin and neo- favorably with the reported orientations (13). mycin. Cells were treated for 24 hours with 10 mmol/L N-acetyl-L- lysine to express acetylated CCAR2 (Sigma, catalog no. A4021). Chromatin immunoprecipitation The ChIP-IT Express Enzymatic Kit (Active Motif) was used, as Immunoblotting and IHC reported previously (12). Following drug treatment, HCT116 cells Immunoblotting (IB) used published procedures for whole- were cross-linked with formaldehyde and homogenized to isolate cell lysates, nuclear/cytoplasmic fractions, and tissue lysates of the nuclear fraction. DNA fragmentation was performed using a colon tumors or normal colon biopsies (9–11, 17, 24). Antibody Biorupter for 15 cycles of 20 seconds each. Ten microliters of to CCAR2 was from Bethyl Laboratories, whereas acetyl-lysine fragmented chromatin was kept as input, while the remainder was (Ac-Lys), histone H3, histone H4, histone H4K12-acetylated subjected to IP with anti-CCAR2 (Cell Signaling Technology), (H4K12ac), 14-3-3, RAD54, HDAC3, b-catenin, c-Myc, cyclin BRD9 (Active Motif), or BRD3 (Active Motif) antibodies. After

www.aacrjournals.org Cancer Res; 79(5) March 1, 2019 919

Rajendran et al.

reversing the cross-linking, and proteinase treatment, DNA was calipers. In rat experiments, Pirc males (29) at 5 months of age purified using the QIAquick PCR Purification Kit (Qiagen). PCR were assigned to study groups (3–4/group), and 2 months later was run on a Roche Light Cycler 480 II with preincubation for 5 occluding colon polyps were resected (36). Rats were then treated minutes at 95 C, then 55 cycles at 95 C for 10 seconds, 60 C for for 5 weeks with test agents, as follows: sulforaphane, 400 parts 10 seconds, and 72 C for 10 seconds. Each experiment was per million in AIN93 diet; JQ1, 12.5 mg/kg body weight via twice repeated at least twice. Primer sequences were as follows: weekly intraperitoneal injection; sulforaphane þ JQ1, at the doses of the individual compounds, or vehicle. The study was termi- E1 (Forward primer) TTGTCGCAGGTATGCTGAGTC nated 2 months after polypectomy, and gastrointestinal lesions E1 (Reverse primer) TGTGATTACCCAGGCACACT were enumerated prior to IB and RNA-seq, as reported previous- E2 (Forward primer) TCCTGAGTCACGGAGTTGTCT ly (28). To our knowledge, this is the first report to examine E2 (Reverse primer) TGCGATCTTCAGAGGGCCTA secondary prevention in a murine model of FAP, following E3 (Forward primer) GAGATTACAGGGAGTGGCAGTG surgical intervention. E3 (Reverse primer) TGGAAACTCAGATACTCCTGGG E4 (Forward primer) CTCCCGAGGGCGATAAAAGG Statistical analysis E4 (Reverse primer) GGATGTTTGCTGGAACGCTG Results are representative of findings from at least three inde- Promoter (Forward primer) TGCATGACCGCATTTCCAATA pendent experiments, expressed as mean SE, unless indicated Promoter (Reverse primer) CGGACAAACCGGACGTTTAATTC otherwise. Student t test was used for paired comparisons, whereas multiple groups were subjected to ANOVA and Bonferroni Preclinical experiments test (GraphPad Prism v5.04). Statistical significance was shown All studies were approved by the Institutional Animal Care and in the corresponding figures, as follows: , P < 0.05; , P < 0.01; Use Committee. For xenograft experiments, 5 106 cells (SW480 , P < 001; and , P < 0.0001. CCAR2 CRISPR/Cas9 knockout or vector controls) were injected into either flank of male athymic nude mice (Envigo). After 10 days, animals were randomized as follows (n ¼ 5 mice/group): Results sulforaphane, 100 mg/kg body weight via daily oral gavage; JQ1, Novel acetylation sites are produced on CCAR2 by sulforaphane 50 mg/kg body weight, twice weekly intraperitoneal injection; Whole-cell lysates were prepared from HCT116 colon sulforaphaneþJQ1, at doses of the individual compounds, or cancer cells incubated with sulforaphane for 6 hours, and vehicle. Tumor volumes were measured twice per week using anti-acetyl-lysine (Ac-Lys) antibody was used to

A B C HCT116 IP: Ac-Lys Input controls IP: Ac-Lys (6 h) Input controls IB: IP: Ac-Lys (6 h) CCAR2 Cytoplasmic Nuclear

Nuclear CCAR2 Cytoplasmic Histone H3 CCAR2 HDAC3 14-3-3 Cyclin D1 RAD51 DMSO SFN NaB TSA DMSO SFN NaB TSA (6 h) −SFN+SFN −SFN+SFN No Ab Input controls SFN SFN SFN SFN HCT116 Cells SFN AITC AITC - HCT116 Cells - 9-SFN 9- 6 6 DMSO DMSO

HCT116 cells D E HCT116 F G CCAR2 Cells seeded Protein mass spectrometry K97ac K916ac 24 h K54ac NLS LZ EF Hand Coiled coil β-Catenin IP: CCAR2 N C Ac-Lys Bands Histone H4 1 202 264 748704 794 923 −SFN +SFN (6 h) excised kDa HDAC3, IgG CCAR2 120 β-Catenin IP:CCAR2 Interactions HDAC3 S1 RNA Binding IgGH 50 Input SDS-PAGE controls CCAR2 IgG L 20 β-Actin Excise CCAR2 + − + − + − + − SFN Negative HDAC3 Mass spectrometry siRNA siRNA IP: CCAR2 Inputs

Figure 1. CCAR2 is an early target for acetylation in sulforaphane (SFN)-treated colon cancer cells. A, HCT116 cells were treated with test agents, and 6 hours later, cell lysates were subjected to IP with Ac-Lys antibody. B, Protocol from A applied to nuclear and cytoplasmic extracts. C, Protocol from A repeated with sulforaphane analogs 6-SFN, 9-SFN, and AITC. D, siRNA-mediated knockdown of HDAC3 and IP/IB of cell lysates, as indicated. For HDAC3 knockdown in SW480 cells, refer to Supplementary Fig. S2. E and F, After IP and SDS-PAGE, CCAR2 was excised from the gel, digested with trypsin, and analyzed by protein mass spectrometry. G, Positions of sulforaphane-induced acetylation sites. NLS, nuclear localization signal; LZ, leucine zipper. The IP and IB data shown in each panel are from a single experiment in each case and are representative of the ﬁndings from three or more independent experiments.

920 Cancer Res; 79(5) March 1, 2019 Cancer Research

BET/BRD9 Acetyl Switch on CCAR2

immunoprecipitate endogenous acetylated proteins. IB con- acetylation status of b-catenin (Fig. 1D). Thus, CCAR2 acet- firmed the increased histone acetylation after treatment with ylation can occur in colon cancer cells in the absence of pan-HDAC inhibitors NaB and TSA, but not by sulforaphane similar changes to other nonhistone proteins, including at 6 hours (Fig. 1A, red arrow). Under the same conditions, b-catenin and HDAC3. robust acetylation of CCAR2 was detected in sulforaphane- CCAR2 was pulled down from colon cancer cells (Fig. 1E), treated cells (Fig. 1A, red box), comparable with that of NaB and after protein separation the CCAR2 band was excised from and TSA. None of the test agents caused 14–3–3 acetylation. the gel (Fig. 1F) and subjected to tandem mass spectrometry. Similar observations were made when SW480 cells were Following sulforaphane treatment, three novel acetylation sites treated with sulforaphane, TSA, SAHA, NaB, or VPA (Supple- were identified on CCAR2 at K54, K97, and K916 (Fig. 1G; mentary Fig. S1); all of the test agents except sulforaphane Supplementary Fig. S3). N-terminal acetylation sites were with- (arrow) caused increased histone acetylation at 6 hours, and in a region that interacts with HDAC3 and b-catenin, whereas all of the compounds, including sulforaphane (box), pro- C-terminal acetylation was adjacent to a coiled coil domain duced a marked increase in CCAR2 acetylation without (Fig. 1G). affecting the acetylation status of its paralog, CCAR1. Acet- ylation of CCAR2 at 6 hours occurred in both the cytoplasmic CCAR2 acetylation interferes with Wnt coactivator functions and nuclear compartments (Fig. 1B), whereas sulforaphane To examine the functional consequences of sulforaphane- had no effect on the acetylation status of cyclin D1. Structural induced CCAR2 acetylation, we first deleted CCAR2 from colon analogsofsulforaphanethatalsowerereportedtoinhibit cancer cells using CRISPR/Cas9 (Fig. 2A and B). Clones that HDAC activity and to turnover HDAC3 protein (11), namely lacked CCAR2 protein had reduced growth rates compared with 6-SFN and 9-SFN, similarly increased the acetylation status of the vector controls in vitro (Fig.2A),andwheninjectedinto CCAR2 without affecting a negative control, RAD51 (Fig. 1C). nude mice (Fig. 2C). CCAR2 was then reintroduced into Increased CCAR2 acetylation was not observed for AITC, CCAR2-null cells via transient transfection of the correspond- which lacks HDAC inhibitory activity in colon cancer ing expression constructs, either as WT CCAR2 or as acetylation cells (11). No HDAC3 acetylation was detected under con- mutants. IP with Ac-Lys antibody revealed low basal acetylation ditions in which HDAC3 protein levels were reduced by for WT CCAR2 in the vehicle controls, which was increased sulforaphane/6-SFN/9-SFN at 6 hours (Fig. 1C, dashed box). after sulforaphane treatment, but was less marked for the SiRNA-mediated knockdown of HDAC3 recapitulated the acetylation mutants K97R and K916R (Fig. 2D). Starting with induction of CCAR2 acetylation in colon cancer cells CCAR2 null cells, reintroduction of acetylation mutants K54R, (Fig. 1D; Supplementary Fig. S2), without changing the K97R, and K916R, or the double-mutant K54R/K97R, had no

Figure 2. CCAR2 acetylation lowers oncogene expression in colon cancer cells. A, Deletion of CCAR2, with each line signifying a different clone, and each datapoint representing mean SD (n ¼ 3). B, Confirmation by IB of CCAR2 loss after CRISPR/Cas9 genome editing. V, vector; KO, knockout. C, Xenograft studies in mice. Each datapoint represents mean SD (n ¼ 5). D, CCAR2 null colon cancer cells transiently transfected with vector control, or expression constructs for WT or acetylation mutants of CCAR2. After 24 hours, cells were treated with sulforaphane or vehicle, and 6 hours later, cell lysates were subjected to IP/IB, as indicated. E, After treatment of colon cancer cells as indicated in D, total RNA was isolated and RT-qPCR was performed for MYC normalized to GAPDH. Data bars, mean SD (n ¼ 3). , P < 0.01; , P < 0.001, significant difference from the corresponding vector/KO control. The data shown in each panel are from a single experiment in each case and are representative of the findings from two or more independent experiments.

www.aacrjournals.org Cancer Res; 79(5) March 1, 2019 921

Rajendran et al.

AB C % Nuclear −SFN +SFN interactions MMP7 % Nuclear MFAP2 −SFN +SFN FAM60A interactions 50 GAL Wnt Signaling BMP4 0.10 EFNB1 9 * PRKCG 25 0.0 PPP1R1A MET 6 * − MYC 0.10 ENPP1 NES = –2.78 MAPRE2 3 0 − 0.20 FDR = 0.0017 TNC CCAR2/HDAC3, 24 h (PLA) −SFN +SFN

PLA2G10 CCD841 Cells CCND1 0 TSPAN5 Enrichment score Enrichment FXYD3 IP: β-Catenin IP: HDAC3 MMP9 D −SFN +SFN CD24 KRTAP3-1 DAB2 15 IB: CCAR2 IB: CCAR2 FSCN1 SERPINA5 IFIH1 10 HSD3B7 IB: β-Catenin IB: Pin1 SEMA3C LGR5 5 ** IgG AOAH IB: HDAC3 CRMP1 Cells HCT116 IB: HDAC3 EFNB1 0 MSX2 −SFN +SFN −SFN +SFN CEACAM1 CLDN1 RAC2 30 SLC7A2 -TAG SGK1 E -TAG JAG1 20 * CKMT2 K54 K54

PLAU -CCAR2 CXCL1 10 HA- HA-CCAR2 − KITLG HA HA- SFN +SFN INHBB AIM1 SW480 Cells 0 – + – + – + – + Ac-Lys (24 h) FGF9 IP: HA-Tag ABCA1 −SFN +SFN Row Z-score DKK1 IB: HA-Tag MYB ENC1 Proximity ligation assays (PLA) on endogenous IP: HA-Tag 21 0–1 –2 CXXC4 IB: β-Catenin LRG1 CCAR2/β-catenin at 24 h in colon cancer cells IB: β-Catenin IB: Lamin Nuclear Cytoplasmic

Figure 3. CCAR2 protein interactions are disrupted by sulforaphane (SFN) treatment, and genetically encoded acetylation of CCAR2 Lys 54 blocks b-catenin interactions. A, RNA-seq data from HCT116 colon cancer cells, 6 hours after treatment with sulforaphane or vehicle. Each column is a biological replicate (n ¼ 3). Wnt signaling was among the top five cancer-related pathways altered by sulforaphane (49), and gene set enrichment analysis prioritized 118 Wnt-related genes among 22,727 genes in the dataset. B, PLA identified endogenous interactions of CCAR2 and b-catenin proteins. Cells were imaged 24 hours after treatment with sulforaphane or vehicle. C, The approach in B was used to examine endogenous CCAR2/HDAC3 interactions. Data bars designate mean SD (n ¼ 3); , P < 0.05; , P < 0.01, compared with vehicle. D, Twenty-four hours after treating HCT116 cells with sulforaphane or vehicle, nuclear extracts were subjected to IP/IB. E, A system for genetically encoding lysine modifications on histones (23) was used to engineer K54 acetylation on CCAR2. Nuclear and cytoplasmic extracts were subjected to IP/IB with the antibodies shown, 24 hours after addition of Ac-Lys, to trigger the designed acetylation on CCAR2. The data shown in each panel are from a single experiment in each case and are representative of the findings from two or more independent experiments.

effect on MYC expression, whereas reintroduction of WT HDAC3 in the nuclear compartment (10), was used as a control in CCAR2 increased MYC levels significantly (Fig. 2E). Similar some experiments. We conclude that CCAR2/HDAC3/b-catenin results were obtained for MMP7 (Supplementary Fig. S4A), interactions are disrupted in sulforaphane-treated colon cancer and while this was reversed by sulforaphane treatment follow- cells, interfering with the Wnt coactivator role of CCAR2 (Sup- ing transient transfection of WT CCAR2, acetylation mutants plementary Fig. S5). such as K54R were resistant to sulforaphane (Supplementary Next, a system for genetically encoding lysine modifications on Fig. S4B). histones (23) was used, for the first time, to engineer acetylation In addition to MYC and MMP7, RNA-seq revealed a suite of sites on a nonhistone protein, CCAR2. Starting with CCAR2-null Wnt/b-catenin target genes downregulated in sulforaphane- colon cancer cells, stable clones were generated containing treated colon cancer cells (Fig. 3A). Next, we examined direct HA-tagged CCAR2 or HA-tagged CCAR2-K54-TAG (abbreviated interactions between endogenous CCAR2 and b-catenin proteins hereafter as HA-CCAR2 and HA-K54-TAG), the "TAG" premature via PLA (37, 38). In CCD841 nontransformed colonic epithelial stop codon preventing protein expression in the absence of acetyl- cells, CCAR2/b-catenin interactions were detected at low levels, lysine (Ac-Lys) reagent (23). Twenty-four hours after the addition whereas numerous interactions were observed in HCT116 and of Ac-Lys, nuclear and cytoplasmic extracts were subjected to IP SW480 cells (Fig. 3B, red dots). After sulforaphane treatment, with an antibody to the HA-tag on CCAR2, followed by IB with the fewer CCAR2/b-catenin interactions were detected, especially in same antibody (Fig. 3E). In cells stably transfected with HA-K54- the nucleus. Using the same approach, we also detected dimin- TAG, no band was detected until the addition of Ac-Lys reagent, ished CCAR2/HDAC3 nuclear interactions (Fig. 3C). consistent with the formation of genetically encoded CCAR2-K54 To corroborate these findings, we pulled down endogenous acetylated protein (Fig. 3E, dotted boxes). In the presence of b-catenin or HDAC3 from nuclear extracts of colon cancer Ac-Lys reagent, IP with HA antibody followed by IB for b-catenin cells (Fig. 3D), and confirmed that interactions with CCAR2 revealed a strong band in the nuclear compartment of cells stably were reduced markedly after sulforaphane treatment (red boxes). transfected with HA-CCAR2 but not HA-K54-TAG (Fig. 3E, solid Peptidyl-prolyl cis/trans isomerase 1 (Pin1), which interacts with boxes). Thus, acetylation of Lys 54 on CCAR2 was sufficient to

922 Cancer Res; 79(5) March 1, 2019 Cancer Research

BET/BRD9 Acetyl Switch on CCAR2

Figure 4. Acetylation sites on CCAR2 are recognized by acetyl "reader" proteins. A and B, Protein arrays comprising all BROMO/PHD/PWWP/YEATS domains were screened using peptide mimetics of acetylated CCAR2 (A), followed by GST-pulldown assays (B) for validation. C, Viability of colon cancer cells. SFN, sulforaphane. D, Mice were injected in either flank with parental SW480 cells or SW480 CCAR2 KO cells, and 10 days later, animals were treated with test agents (see Materials and Methods). Datapoints, mean SD (n ¼ 5); , P < 0.05; , P < 0.01 compared with vehicle time-point. E, Working model of acetyl readers and CCAR2interactions.Thedatashownineachfigure panel are from a single experiment in each case and are representative of the findings from two or more independent experiments.

block its interactions with b-catenin in the nuclear compartment. gesting a scenario in which acetylated CCAR2 competes with Colon cancer cells also were generated containing stably trans- acetylated histones for the binding of acetyl readers and their fected HA-K97-TAG; no CCAR2-K97ac band was detected after inhibitors, such as JQ1. Ac-Lys treatment (Supplementary Fig. S6). The K97ac site may be JQ1 exhibited strong synergy with sulforaphane in colon destabilizing to CCAR2 under the conditions used, in the absence cancer cells, the CI of 0.25 (Fig. 4C) being comparable with of K54ac. 0.13 and 0.33, respectively, for JQ1 plus SAHA or 6-SFN (Supplementary Fig. S8A). Cotreatment with JQ1 plus sulfo- CCAR2 acetyl "readers" include BET family members and BRD9 raphaneor6-SFNincreasedthelevelsofcleavedPARPand On the basis of the CCAR2 acetylation sites observed after cleaved caspase-3, indicating enhanced apoptosis in colon sulforaphane treatment (Fig. 1G), we turned our attention to cancer cells (Supplementary Fig. S8B, dotted boxes), consistent the acetyl "reader" proteins. Biotin-tagged peptide mimetics of with prior studies using sulforaphane and 6-SFN alone (10– CCAR2 were screened (25–27) via protein arrays comprising all 12). Next, we took SW480 cells that are reported to be resistant known acetyl readers (Fig. 4A), followed by GST-pulldown to JQ1 (39), but also exhibited reduced CCAR2/b-catenin assays for validation (Fig. 4B). CCAR2-K97ac peptide was interactions after sulforaphane treatment (Fig. 3B), and exam- recognized by bromodomains of ASH1L and BAZ1A, whereas ined their growth in nude mice. As expected for a JQ1-resistant CCAR2-K916ac peptide interacted with bromodomains of cell line (39), JQ1 alone had no effect, but JQ1 enhanced the ASH1L, BRDT, BRD2, BRD3, and BRD9. The arrays also impli- tumor suppressive actions of sulforaphane in vivo (Fig. 4D, left; cated BRD7 interacting with CCAR2-K916ac peptide, but this JQ1þsulforaphane; , P < 0.01) despite an apparent lack of was not corroborated in pulldown experiments (Fig. 4B, right). synergy under the conditions employed. When mice were Protein arrays did not recognize a peptide mimetic for CCAR2- injected with SW480 CCAR2-null cells, as before (Fig. 2C), no K54ac (Fig. 4A, left). inhibition was observed for sulforaphane, JQ1, or sulforapha- The BET members BRD2, BRD3, and BRDT, along with neþJQ1 (Fig. 4D, right). Thus, sulforaphane required the pres- BRD4, interact with high speciﬁcity on the arrays with JQ1 ence of CCAR2, and circumvented resistance mechanisms in (M.T. Bedford, manuscript in preparation). However, CCAR2 JQ1-resistant colon cancer cells (39, 40). Our working model peptide mimetics did not interact with BRD4 on the arrays, (Fig. 4E) proposes a shift to increased BRD9/CCAR2-containing despite favorable docking scores in silico (Supplementary Fig. chromatin complexes as a basis for the synergistic interactions S7). Docking scores supported the preferred interactions of BET of sulforaphaneþJQ1 in colon cancer cells, supported by bio- members and BRD9 with CCAR2K916ac versus H4K16ac, sug- informatics data, see below.

www.aacrjournals.org Cancer Res; 79(5) March 1, 2019 923

Rajendran et al.

CCAR2 High AB60 C

50 Vehicle 40 SFN 30 JQ1 20 Before After One week CCAR2 Low SFN+JQ1

Transcripts per million, RNA-seq FAP FAP Pirc Pirc 567 89 months Normal Polyp Normal Polyp Polypectomy

RNA-seq, Pirc colon polyps E RNA-seq, Pirc colon polyps Pla2g10 D Plaur F CCAR2 Sox17 JQ1 Crmp1 Plcb1 Cyclin D1 Birc5 4– Vehicle Ppard Inhbb MMP7 Ccnd1 Ccl2 β Serpina3n -Actin 0 – Sgk1 Lyz2 Jag1 Vehicle SFN JQ1 SFN+JQ1 SFN+JQ1 Mmp7 Abcb1a G −4– Tff3 Rac2 200 Mmp9 Cldn1 Slc7a2 175 PC2: 25% Variance PC2: 25% −8 – Isl1 Gsta4 Lcn2 150 SFN Cxcl13 Dpep1 −12– Sox2 125 ** JQ1 JQ1 SFN SFN Lix1 Gast Vehicle Vehicle after polypectomy 100 PC1: 29% Variance Row Z-score Gal Percent tumor growth growth Percent tumor SFN+JQ1 SFN+JQ1 Vehicle SFN JQ1 SFN+JQ1 21 0–1 –2

Figure 5. SulforaphaneþJQ1 protect in a murine model of FAP. A, RNA-seq data (28) mined for CCAR2 levels in adenomatous colon polyps from patients with FAP and the Pirc rat (29). Each datapoint designates an individual polyp or a normal-looking colonic mucosa sample. B, Polypectomy in the Pirc model (36). C, At 5 months of age, Pirc males (3–4/group) were assigned to different arms of the study, and 2 months thereafter, occluding colon polyps were resected. Rats were then treated for 5 weeks with sulforaphane (SFN), JQ1, sulforaphaneþJQ1, or vehicle (see Materials and Methods). The study was terminated 2 months after polypectomy, and duplicate lesions in each group provided data for RNA-seq (D and E) and IB (F). The IB data are from a single experiment and are representative of the findings from two independent experiments. Principle component analysis (D) of 1,436 differentially expressed genes identified Wnt targets downregulated by JQ1þsulforaphane (E). G, Compared with vehicle, JQ1þsulforaphane inhibited the growth of colon polyps significantly (, P < 0.01).

Cooperative inhibition by sulforaphaneþJQ1 in a genetic with respect to Wnt genes downregulated by sulforaphane model of colorectal cancer (Fig. 3A), and these observations were extended to JQ1 and We next reexamined RNA-seq data from a recent study (28) sulforaphaneþJQ1 groups (Fig. 5E). IB of tissue lysates from andobservedthestratification of CCAR2 in colon adenomas Pirc colon tumors also showed reduced expression of target of patients with FAP (Fig. 5A). Subjects with high CCAR2 proteins such as CCAR2, cyclin D1, and MMP7, especially levels in colon adenomas also had high CCAR2 expression in for sulforaphaneþJQ1 in combination (Fig. 5F, red box). normal-looking tissues, whereas patients with lower CCAR2 Consistent with these molecular changes, sulforaphaneþJQ1 levels in colon adenomas had reduced CCAR2 expression in suppressed colon tumor growth significantly, exceeding the normal-looking tissues. Normal-looking colon in patients inhibition observed for sulforaphane or JQ1 alone (Fig. 5G; with FAP, and in preclinical models of FAP, is rarely "normal" , P < 0.01). due to the presence of microadenomas and other preneoplastic On the basis of the working model (Fig. 4E), we took the entire lesions. RNA-seq dataset (Fig. 6A), and prioritized 104 combination- Interestingly, the "CCAR2-high" molecular phenotype also specific "cooperativity/synergy" candidates among 324 genes was detected (Fig. 5A) in the Pirc model of FAP (29). We in the sulforaphaneþJQ1 group (Fig. 6B, green circle). In addition resected occluding polyps in the rat (Fig. 5B), as reported to Wnt, top cancer-specific pathways included hypoxia, p53, previously (36), and animals were then treated with sulforaph- inflammation, reactive oxygen species, KRAS, RB, and apoptosis ane, JQ1, sulforaphaneþJQ1, or vehicle (Fig. 5C). When the (Fig. 6C), and the most upregulated and downregulated genes study was terminated, 2 months after polypectomy, paired were identified (Fig. 6D). Notably, when all 104 sulforapha- colon polyps in each group were subjected to RNA-seq neþJQ1 "cooperativity/synergy" genes were interrogated together and IB. RNA-seq segregated the groups based on principle with available chromatin immunoprecipitation (ChIP)-seq data componentanalysesof1,436genes in the dataset (Fig. 5D). for BRD9 (GSM2092891), BRD9 was localized at the correspond- Notably, RNA-seq recapitulated findings from cell-based assays ing transcription start sites (Fig. 6E, red line). No corresponding

924 Cancer Res; 79(5) March 1, 2019 Cancer Research

BET/BRD9 Acetyl Switch on CCAR2

RNA-seq, Pirc colon polyps A B D SFN+JQ1-Specific genes Gene Fold-change SFN JQ1 Misp3 3.95 Krt39 3.19 16 Anxa13 2.76 Upregulated 117 360 Apobec2 2.63 Sorcs1 2.52 86 Krt20 −2.02 52 82 Myorg −2.25 Ubash3b −2.40 Hsd3b1 −2.49 104 Igfbp1 −2.54 Lix1 −2.54 Mettl7b −2.55 SFN+JQ1 Tmem14a −2.64 Csf3 −2.71 Downregulated Chga −2.85 C Klk8 −3.08 Pathway enrichment (SFN+JQ1) Tnnl2 −3.38 Shh −3.69 HYPOXIA Reg3b −3.90 p53 PATHWAY Tubal3 −4.14 INFLAMMATORY RESPONSE − REACTIVE OXYGEN SPECIES Prl2a1 5.56 − APOPTOSIS Ercc2 6.25 KRAS PATHWAY RB PATHWAY E 0.05 SFN+JQ1-Specific (n = 104) JQ1 JQ1 SFN SFN PKCA PATHWAY 0.04 WNT PATHWAY Vehicle Vehicle Row Z-score KINASE ACTIVITY 0.03 SFN+JQ1 SFN+JQ1 0.02 21 0–1 –2 01234567 Random gene-set (n = 104) 0.01 – – −Log(FDR) – –

BRD9 ChIP Signal BRD9 ChIP −2 Kb Start Gene body End 2 Kb

Figure 6. RNA-seq prioritizes HDAC þ BET "cooperativity" genes in Pirc colon tumors. A, Heatmaps for groups analyzed in duplicate (1,436 differentially expressed genes, no cutoff applied). B, Number of DEGs compared with vehicle controls. C, Pathway enrichment analysis for sulforaphane (SFN)þJQ1 differentially expressed genes. D, Highly upregulated and downregulated sulforaphaneþJQ1–speciﬁc genes. E, Human genome reference GRCh38 was interrogated using BRD9 ChIP-seq data downloaded from GSM2092891. The proﬁle was plotted using Mmint, with the red line representing the average BRD9 signal for sulforaphaneþJQ1 "synergy/cooperativity" genes. The corresponding BRD9 signal also was examined for a set of 104 randomly selected genes, not among the "synergy/ cooperativity" candidates (green line). Start, transcription start sites; End, transcription stop sites (n ¼ 104).

BRD9 signal was detected at transcription start sites of 104 "CCAR2-high" molecular phenotype was observed, which is randomly selected genes (Fig. 6E, green line), implicating BRD9 noteworthy given that CCAR2 overexpression is associated with enrichment on the promoters of sulforaphaneþJQ1 cooperativ- poor prognosis in patients with colorectal cancer (7). "CCAR2- ity/synergy genes as being mechanistically relevant. high" adenomatous polyps from the screening colonoscopy trial had elevated expression of b-catenin and its downstream targets, such as, MMP7, c-Myc, and cyclin D1, and increased Discussion b-catenin/CCAR2 interactions were detected by PLA (Supple- Acetylation of CCAR2 by hMOF at K112/K215 sites is known mentary Fig. S9). to displace SIRT1 (5), and we speculated that novel N-terminal The C-terminal K916 CCAR2 acetylation site appears to be acetylation sites identified here might similarly interfere with distant from N-terminal K54/K97 acetylation sites that overlap b-catenin interactions. Consistent with this idea, we observed with HDAC3/b-catenin interacting domains (Fig. 1G). However, reduced nuclear CCAR2/b-catenin interactions coinciding with as a protein with structural flexibility (41), circumstances might downregulation of multiple Wnt targets. Expression of a genet- dictate that the ends become aligned, for example after binding ically encoded K54ac site on CCAR2 was sufficient to block its lysine methyltransferase ASH1L, which interacted with acetylated interactions with b-catenin, indicating a key role for this post- peptides from N- and C-terminal regions of CCAR2 (Fig. 4B), or translational modification in regulating b-catenin associations. BAZ1A. BAZ1A is a noncatalytic ISWI subunit that associates Furthermore, the K54 acetylation mutant interfered with the relatively weakly with acetylated histones, but is critical for DNA ability of sulforaphane to reduce downstream targets of b-cate- damage recovery (42), which is a key function of CCAR2 (43, 44). nin, such as MMP7 (Supplementary Fig. S4B). Xenograft studies An intriguing question is whether BAZ1A and ASH1L interact in mice indicated that CCAR2 was required for tumor growth preferentially with acetylated nonhistone proteins such as inhibition by sulforaphaneþJQ1 in vivo, and we extended these CCAR2, affecting gene expression changes as members of specific observations to the Pirc model, showing suppression of ade- chromatin remodeling complexes in response to sulforapha- nomatous colon polyps by sulforaphaneþJQ1 in the rat. In Pirc neþJQ1 treatment. In this context, competition between BET colon polyps and in a subset of adenomas from patients with members and BRD9 for the K916ac site on CCAR2 would shift FAP (Fig. 5A), as well as in adenomatous polyps from a in favor of CCAR2/BRD9 complexes after sulforaphaneþJQ1 screening colonoscopy trial (Supplementary Fig. S9), a treatment (Fig. 4E).

www.aacrjournals.org Cancer Res; 79(5) March 1, 2019 925

Rajendran et al.

The latter working model derives from three interrelated Authors' Contributions observations: (i) like BRD2 and BRD3, BRD9 interacts favor- Conception and design: P. Rajendran, G. Johnson, R. Dashwood ably with CCAR2-K916ac (Fig. 4B); (ii) unlike BRD2 and BRD3, Development of methodology: P. Rajendran, G. Johnson, F. Ertem, BRD9 is not subject to inhibition by JQ1; and (iii) BRD9 is a H.-C. Leung, M. Bedford Acquisition of data (provided animals, acquired and managed patients, required subunit of SWI–SNF complexes (45). We speculate provided facilities, etc.): P. Rajendran, G. Johnson, L. Li, Y.-S. Chen, that sulforaphane-induced acetylation sites on CCAR2 might M. Dashwood, N. Nguyen, A. Ulusan, H.-C. Leung, D. Lieberman, exert distinct functions, with K54ac for b-catenin displacement, L. Beaver, M. Bedford, E. Vilar K97ac for ASH1L/BAZ1A-mediated chromatin interactions, and Analysis and interpretation of data (e.g., statistical analysis, biostatistics, K916ac as an acetyl switch between BET versus BRD9 functions. computational analysis): P. Rajendran, G. Johnson, L. Li, N. Nguyen, This does not preclude JQ1 also inhibiting BET acetyl readers M. Zhang, J. Li, D. Sun, S. Wang, H.-C. Leung, K. Chang, R. Dashwood Writing, review, and/or revision of the manuscript: P. Rajendran, on histones (Fig. 4E) to affect changes in gene expression (13– G. Johnson, D. Sun, S. Wang, D. Lieberman, L. Beaver, E. Ho, 16). The possibility that sulforaphane and JQ1 might interact E. Vilar, R. Dashwood synergistically at the level of MYC transcription was investigat- Administrative, technical, or material support (i.e., reporting or organizing ed via ChIP assays, with the following observations: (i) CCAR2 data, constructing databases): L. Li, D. Sun, Y. Huang, K. Chang interactions were conﬁrmed on promoter and super-enhancer Study supervision: P. Rajendran, R. Dashwood regions, (ii) these interactions were almost completely inhibited by the combination of sulforaphaneþJQ1 (Supplementary Acknowledgments Fig. S10A), and (iii) BRD3 interactions on super-enhancer We thank R. Jaimes, L. Chew, and A. Khan for technical assistance. Dr. O. Hiraike (University of Tokyo, Japan) provided a Myc-DBC expression regions also were reduced, to a lesser degree, by JQ1 alone construct, whereas plasmids pE312 (pPB 4xU25C EF1 AcKRS-TAGT2A- (Supplementary Fig. S10B). Dendra2 IRES Puro) and pE337 (pPB 4xU25C EF1 H33 3xHA IRES Neo) were Finally, as an HDAC3-interacting protein, CCAR2 might be from Dr. J. Chin (MRC Laboratory of Molecular Biology, Cambridge, UK). targeted using HDAC3-selective inhibitors (46, 47), although Protein arrays were run by C. Sagum in the Protein Array and Analysis Core, these agents have yet to enter clinical trials. One approach to supported by Cancer Prevention & Research Institute of Texas grant no. enhancing efﬁcacy might involve modifying sulforaphane as a RP130432. L.M. Lui provided technical help with mass spectrometry (Protein Mass Spectrometry Core, Baylor College of Medicine), and N. Otto performed lead compound (48, 49), and combining with improved, second- IHC in the MD Anderson Pathology & Imaging Core. Initial RNA-seq was generation bromodomain inhibitors (50, 51). This strategy could conducted at the Center for Genome Research and Biocomputing at Oregon provide further insights into the "cooperativity/synergy" candi- State University (Corvallis, OR). This work was supported by grants CA090890 date genes prioritized here, and the associated regulatory path- and CA122959 from the NCI, the John S. Dunn Foundation, and a Chancellor's ways to be targeted in future clinical trials. We conclude that Research Initiative. Funding also was provided by grants R25TCA057730, JQ1þsulforaphane interferes with the Wnt coactivator role of CA208461, and CA016672 and a gift from the Feinberg Family to E. Vilar. CCAR2, and shifts the pool of acetyl readers in favor of BRD9- The costs of publication of this article were defrayed in part by the payment regulated genes, providing a mechanistic basis for new therapeutic of page charges. This article must therefore be hereby marked advertisement in þ avenues combining HDAC3 BET inhibition. accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Disclosure of Potential Conflicts of Interest Received June 29, 2018; revised October 19, 2018; accepted January 9, 2019; No potential conflicts of interest were disclosed. published first January 14, 2019.

References 1. Kim J-E, Chen J, Lou Z. DBC1 is a negative regulator of SIRT1. Nature 2008; 10. RajendranP,DelageB,DashwoodWM,YuT-W,WuthB,WilliamsDE, 451:583–6. et al. Histone deacetylase turnover and recovery in sulforaphane-treated 2. Escande C, Chini CCS, Nin V, Dykhouse KM, Novak CM, Levine J, et al. colon cancer cells: competing actions of 14-3-3 and Pin1 in HDAC3/ Deleted in breast cancer-1 regulates SIRT1 activity and contributes to high- SMRT corepressor complex dissociation/reassembly. Mol Cancer 2011; fat diet-induced liver steatosis in mice. J Clin Invest 2010;120:545–58. 10:68. 3. Li J, Bonkowski MS, Moniot S, Zhang D, Hubbard BP, Ling AJY, et al. A 11. Rajendran P, Kidane AI, Yu T-W, Dashwood W-M, Bisson WH, Lohr€ CV, conserved NADþ binding pocket that regulates protein-protein interac- et al. HDAC turnover, CtIP acetylation and dysregulated DNA damage tions during aging. Science 2017;355:1312–7. signaling in colon cancer cells treated with sulforaphane and related dietary 4. Giguere SSB, Guise AJ, Jean Beltran PM, Joshi PM, Greco TM, Quach OL, isothiocyanates. Epigenetics 2013;8:612–23. et al. The proteomic profile of deleted in breast cancer 1 (DBC1) interactions 12. Rajendran P, Dashwood W-M, Li L, Kang Y, Kim E, Johnson G, et al. Nrf2 points to a multifaceted regulation of gene expression. Mol Cell Proteomics status affects tumor growth, HDAC3 gene promoter associations, and the 2016;15:791–809. response to sulforaphane in the colon. Clin Epigenetics 2015;7:102. 5. Zheng H, Yang L, Peng L, Izumi V, Koomen J, Seto E, et al. hMOF acetylation 13. Filippakopoulos P, Qi J, Picaud S, Shen Y, Smith WB, Fedorov O, et al. of DBC1/CCAR2 prevents binding and inhibition of SirT1. Mol Cell Biol Selective inhibition of BET bromodomains. Nature 2010;468:1067–73. 2013;33:4960–70. 14. Mazur PK, Herner A, Mello SS, Wirth M, Hausmann S, Sanchez-Rivera FJ, 6. Chini CCS, Escande C, Nin V, Chini EN. HDAC3 is negatively regulated by et al. Combined inhibition of BET family proteins and histone deacetylases the nuclear protein DBC1. J Biol Chem 2010;285:40830–7. as a potential epigenetics-based therapy for pancreatic ductal adenocarci- 7. Yu EJ, Kim SH, Kim HJ, Heo K, Ou CY, Stallcup MR, et al. Positive regulation noma. Nat Med 2015;21:1163–71. of b-catenin-PROX1 signaling axis by DBC1 in colon cancer progression. 15. Fujisawa T, Filippakopoulos P. Functions of bromodomain-containing Oncogene 2016;35:3410–8. proteins and their roles in homeostasis and cancer. Nat Rev Mol Cell Biol 8. Nusse R, Clevers H. Wnt/b-Catenin signaling, disease, and emerging 2017;18:246–62. therapeutic modalities. Cell 2017;169:985–99. 16. Kleppe M, Koche R, Zou L, van Galen P, Hill CE, Dong L, et al. Dual 9. Myzak MC, Karplus PA, Chung F-L, Dashwood RH. A novel mechanism of targeting of oncogenic activation and inflammatory signaling increases chemoprotection by sulforaphane: inhibition of histone deacetylase. therapeutic efficacy in myeloproliferative neoplasms. Cancer Cell 2018;33: Cancer Res 2004;64:5767–74. 29–43.

926 Cancer Res; 79(5) March 1, 2019 Cancer Research

BET/BRD9 Acetyl Switch on CCAR2

17. Wang R, Dashwood W-M, Nian H, Lohr€ CV, Fischer KA, Tsuchiya N, et al. 36. Ertem F, Dashwood WM, Rajendran P, Raju G, Rashid A, Dashwood R. NADPH oxidase overexpression in human colon cancers and rat colon Development of a murine colonoscopic polypectomy model (with tumors induced by 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine videos). Gastrointest Endosc 2016;83:1272–6. (PhIP). Int J Cancer 2011;128:2581–90. 37. Soderberg€ O, Gullberg M, Jarvius M, Ridderstrale K, Leuchowius K-J, Jarvius 18. Parasramka MA, Dashwood WM, Wang R, Saeed HH, Williams DE, Ho E, J, et al. Direct observation of individual endogenous protein complexes et al. A role for low-abundance miRNAs in colon cancer: the miR-206/ in situ by proximity ligation. Nat Methods 2006;3:995–1000. Kruppel-like€ factor 4 (KLF4) axis. Clin Epigenetics 2012;4:16. 38. Blokzijl A, Nong R, Darmanis S, Hertz E, Landegren U, Kamali-Moghad- 19. Young L, Sung J, Stacey G, Masters JR. Detection of mycoplasma in cell dam M. Protein biomarker validation via proximity ligation assays. cultures. Nat Protoc 2010;5:929–34. Biochim Biophys Acta 2014;1844:933–9. 20. Cong L, Ran FA, Cox D, Lin S, Barretto R, Habib N, et al. Multiplex genome 39. TogelL,NightingaleR,ChuehAC,JayachandranA,TranH,PhesseT,€ engineering using CRISPR/Cas systems. Science 2013;339:819–23. et al. Dual targeting of bromodomain and extraterminal domain pro- 21. Mali P, Yang L, Esvelt KM, Aach J, Guell M, DiCarlo JE, et al. RNA-guided teins, and WNT or MAPK signaling, inhibits c-MYC expression and human genome engineering via Cas9. Science 2013;339:823–6. proliferation of colorectal cancer cells. Mol Cancer Ther 2016;15: 22. Hiraike H, Wada-Hiraike O, Nakagawa S, Koyama S, Miyamoto Y, Sone K, 1217–26. et al. Identification of DBC1 as a transcriptional repressor for BRCA1. 40. Engelke CG, Chinnaiyan AM. aBETting therapeutic resistance by Wnt Br J Cancer 2010;102:1061–7. signaling. Cell Res 2015;25:1187–8. 23. Els€asser SJ, Ernst RJ, Walker OS, Chin JW. Genetic code expansion in stable 41. Brunquell J, Yuan J, Erwin A, Westerheide SD, Xue B. DBC1/CCAR2 and cell lines enables encoded chromatin modification. Nat Methods 2016;13: CCAR1 are largely disordered proteins that have evolved from one com- 158–64. mon ancestor. Biomed Res Int 2014;2014:418458. 24. Wang R, Lohr€ CV, Fischer K, Dashwood WM, Greenwood JA, Ho E, et al. 42. Oppikofer M, Sagolla M, Haley B, Zhang H-M, Kummerfeld SK, Sudhamsu Epigenetic inactivation of endothelin-2 and endothelin-3 in colon cancer. J, et al. Non-canonical reader modules of BAZ1A promote recovery from Int J Cancer 2013;132:1004–12. DNA damage. Nat Commun 2017;8:862. 25. Gates LA, Shi J, Rohira AD, Feng Q, Zhu B, Bedford MT, et al. Acetylation on 43. Magni M, Ruscica V, Restelli M, Fontanella E, Buscemi G, Zannini L. histone H3 lysine 9 mediates a switch from transcription initiation to CCAR2/DBC1 is required for Chk2-dependent KAP1 phosphorylation and elongation. J Biol Chem 2017;292:14456–72. repair of DNA damage. Oncotarget 2015;6:17817–31. 26. Kim J, Daniel J, Espejo A, Lake A, Krishna M, Xia L, et al. Tudor, MBT and 44. Lopez-Saavedra A, Gomez-Cabello D, Domínguez-Sanchez MS, Mejías- chromo domains gauge the degree of lysine methylation. EMBO Rep 2006; Navarro F, Fernandez-Avila MJ, Dinant C, et al. A genome-wide screening 7:397–403. uncovers the role of CCAR2 as an antagonist of DNA end resection. 27. Espejo A, Bedford MT. Protein-domain microarrays. Methods Mol Biol Nat Commun 2016;7:12364. 2004;264:173–81. 45. Hohmann AF, Martin LJ, Minder JL, Roe J-S, Shi J, Steurer S, et al. Sensitivity 28. Ertem FU, Zhang W, Chang K, Mohaiza Dashwood W, Rajendran P, Sun D, and engineered resistance of myeloid leukemia cells to BRD9 inhibition. et al. Oncogenic targets Mmp7, S100a9, Nppb and Aldh1a3 from tran- Nat Chem Biol 2016;12:672–9. scriptome profiling of FAP and Pirc adenomas are downregulated in 46. McLeod AB, Stice JP, Wardell SE, Alley HM, Chang C-Y, McDonnell DP. response to tumor suppression by Clotam. Int J Cancer 2017;140:460–8. Validation of histone deacetylase 3 as a therapeutic target in castration- 29. Amos-Landgraf JM, Kwong LN, Kendziorski CM, Reichelderfer M, Tor- resistant prostate cancer. Prostate 2018;78:266–77. realba J, Weichert J, et al. A target-selected Apc-mutant rat kindred enhances 47. Harada T, Ohguchi H, Grondin Y, Kikuchi S, Sagawa M, Tai YT, et al. the modeling of familial human colon cancer. Proc Natl Acad Sci U S A HDAC3 regulates DNMT1 expression in multiple myeloma: therapeutic 2007;104:4036–41. implications. Leukemia 2017;31:2670–7. € 30. Gille C, Fahling M, Weyand B, Wieland T, Gille A. Alignment-annotator web 48. Okonkwo A, Mitra J, Johnson GS, Li L, Dashwood WM, Hegde ML, et al. server: rendering and annotating sequence alignments. Nucleic Acids Res Heterocyclic analogs of sulforaphane trigger DNA damage and impede 2014;42:W3–6. DNA repair in colon cancer cells: interplay of HATs and HDACs. 31. Trott O, Olson AJ. AutoDock Vina: improving the speed and accuracy of Mol Nutr Food Res 2018;62:e1800228. docking with a new scoring function, efficient optimization, and multi- 49. Johnson GS, Li J, Beaver LM, Dashwood WM, Sun D, Rajendran P, et al. A threading. J Comput Chem 2010;31:455–61. functional pseudogene, NMRAL2P, is regulated by Nrf2 and serves as a 32. Biasini M, Bienert S, Waterhouse A, Arnold K, Studer G, Schmidt T, et al. coactivator of NQO1 in sulforaphane-treated colon cancer cells. SWISS-MODEL: modelling protein tertiary and quaternary structure using Mol Nutr Food Res 2017;61. doi: 10.1002/mnfr.201600769. evolutionary information. Nucleic Acids Res 2014;42:W252–8. 50. Cheung K, Lu G, Sharma R, Vincek A, Zhang R, Plotnikov AN, et al. BET 33. Paxman JJ, Heras B. Bioinformatics tools and resources for analyzing N-terminal bromodomain inhibition selectively blocks Th17 cell differ- protein structures. Methods Mol Biol 2017;1549:209–20. entiation and ameliorates colitis in mice. Proc Natl Acad Sci U S A 2017; 34. Krissinel E, Henrick K. Inference of macromolecular assemblies from 114:2952–7. crystalline state. J Mol Biol. 2007;372:774–97. 51. Xu L, Chen Y, Mayakonda A, Koh L, Chong YK, Buckley DL, et al. 35. Sobolev V, Sorokine A, Prilusky J, Abola EE, Edelman M. Automated Targetable BET proteins- and E2F1-dependent transcriptional program analysis of interatomic contacts in proteins. Bioinformatics 1999;15: maintains the malignancy of glioblastoma. Proc Natl Acad Sci U S A 327–32. 2018;115:E5086–95.

www.aacrjournals.org Cancer Res; 79(5) March 1, 2019 927

Acetylation of CCAR2 Establishes a BET/BRD9 Acetyl Switch in Response to Combined Deacetylase and Bromodomain Inhibition

Praveen Rajendran, Gavin Johnson, Li Li, et al.

Cancer Res 2019;79:918-927. Published OnlineFirst January 14, 2019.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-18-2003

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2019/01/12/0008-5472.CAN-18-2003.DC1

Cited articles This article cites 50 articles, 13 of which you can access for free at: http://cancerres.aacrjournals.org/content/79/5/918.full#ref-list-1

Citing articles This article has been cited by 2 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/79/5/918.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/79/5/918. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.