Epigenetic Regulation in Prostate Cancer Progression

Epigenetic Regulation in Prostate Cancer Progression Purpose of Review An important number of newly identified molecular alterations in prostate cancer affect gene encoding master regulators of chromatin biology epigenetic regulation. This review will provide an updated view of the key epigenetic mecha- nisms underlying prostate cancer progression, therapy resistance, and potential actionable mechanisms and biomarkers. Recent Findings Key players in chromatin biology and epigenetic master regulators has been recently described to be crucially altered in metastatic CRPC and tumors that progress to AR independency. As such, epigenetic dysregulation represents a driving mechanism in the reprograming of prostate cancer cells as they lose AR-imposed identity. Summary Chromatin integrity and accessibility for transcriptional regulation are key features altered in cancer progression, and particularly relevant in nuclear hormone receptor-driven tumors like prostate cancer. Understanding how chromatin remodeling dictates prostate development and how its deregulation contributes to prostate cancer onset and progression may improve risk stratification and treatment selection for prostate cancer patients. . . . . . Keywords Prostate cancer Epigenetics Transcriptional regulation Chromatin biology Androgen receptor Drug targets Introduction patients, who develop castration-resistant prostate tumors (CRPC) for which limited treatment options exist. Moreover, Prostate cancer has traditionally been seen as an aging-asso- under the CRPC definition, a pool of diverse disease presen- ciated, low mutational load tumor with a tendency for geno- tations with variable outcomes exists, including neuroendo- mic rearrangements and a particular dependency on the activ- crine tumors. ity of the androgen receptor (AR). As such, treatment strate- Massive parallel sequencing of hundreds of tumor speci- gies have been focused on targeting the AR axis, either mens from prostate cancer patients at different stages of cancer through inhibiting steroidogenic pathways and the production progression has provided an accurate picture of the landscape of testosterone, or by antagonizing the AR itself to prevent its of genetic alterations that accompany cancer evolution in the nuclear translocation and the activation of its transcriptional prostate. Yet, despite several molecular classification systems network. While these strategies have doubtlessly improved for prostate tumors have been proposed, clear association with survival for prostate cancer patients, they are not curative in risk stratification remains to be provided. On the other hand, many cases, and resistance eventually occurs in about 30% of whether these genetic classifiers predict treatment outcome and to what extent genetic alterations in prostate cancer can be exploited for personalized therapies is yet to be proven. This article is part of the Topical Collection on Molecular Biology of Interestingly, together with well-known drivers of cancer pro- Prostate Cancer gression, an important number of new alterations have been described, with an intriguing enrichment of those affecting * Alvaro Aytes aaytes@idibell.cat key players in chromatin biology and epigenetic master regu- lators (see a summary in Table 1). This is particularly relevant Programs of Molecular Mechanisms and Experimental Therapeutics in metastatic CRPC and tumors that have transitioned to AR- in Oncology (ONCOBell), Catalan Institute of Oncology, Bellvitge independent phenotypes after progressing on the newest Institute for Biomedical Research, Granvia de l’Hopitalet, 199 antiandrogen drugs. 08908, L’Hospitalet de Llobregat, 08907 Barcelona, Spain 2 Here, we introduce key concepts to understand how epige- Programs of Cancer Therapeutics Resistance (ProCURE), Catalan netic dysregulation is a plausible driving mechanism in the Institute of Oncology, Bellvitge Institute for Biomedical Research, L’Hospitalet de Llobregat, 08907 Barcelona, Spain 102 Curr Mol Bio Rep (2018) 4:101–115 Table 1 Summary of epigenetic Gene name Function in prostate cancer References master regulators implicated in prostate cancer Methyltransferases NSD2 H3K36 di-methyltransferase. Promotes prostate cancer [1�� , 2, 3] tumorigenesis and progression. It is overexpressed in metastatic stage and associated with biochemical recurrence EZH2 H3K27 di- and tri-methyltransferase. Member of the polycomb [4, 5] repressive complex 2, crucial driver of prostate oncogenesis SUV39H1 (KMT1A) H3K9 tri methyltransferase. Enhance prostate cancer cell [6, 7] migration and invasion SETDB1 (KMT1E) SUV39H2 (KMT1B) H3K9 tri methyltransferase increases androgen-dependent tran- [8] scriptional activity by interacting with the AR SMYD3 H3K4 di- and methyltransferase, promotes cell proliferation and [9, 10]. migration PRMT5 Drives prostate cancer cell growth through epigenetic [11, 12� , inactivation of several tumor suppressors through histone 13] arginine methylation at H4R3. Enhances AR-targeted gene expression Demethylases LSD1 H3K9 and H3K4 demethylase involved in prostate cancer [14�� , 15, recurrence, CRPC, and poor survival. Regulates AR 16� ] transcriptional activity in a context-dependent manner JARID1B (KDM5B) H3K4 mono, di-, and tri-demethylase. AR coactivator regulating [17, 18], its transcriptional activity. Upregulated in prostate cancer tissues JARID1C (KDM5C) H3K4 di- and tri-demethylase overexpressed in prostate cancer. [19]. Proposed as a predictive marker for therapy failure in patients after prostatectomy JARID1D (KMD5D) H3K4 di- and tri-demethylase. Suppress invasion and progres- [20, 21] sion of prostate cancer. Low levels were associated with poor prognosis and resistance to docetaxel PHF8 H3K9, H3K27, and H4K20 demethylase. Transcriptional [22–29] coactivator of AR. Promotes prostate cancer cell proliferation, migration, invasion, and neuroendocrine differentiation. Its expression highly correlated with poor prognosis and is induced by hypoxia JMJD2A (KDM4A) H3K9 and H3K36 tri demethylases. Modulates AR [30, 31] JMJD2C (KDM4C) transcriptional activity stimulating ligand-independent gene transcription via H3K9 demethylation JMJD1A (KDM3A) H3K9 mono- and di-demethylase. Regulates AR activity by re- [32, 33] cruitment to target genes only in the presence of androgens JMJD2B (KDM4B), H3K9 tri-demethylase, AR coactivator. Regulates AR [34] transcriptional activity via demethylation activity and via inhibition of ubiquitination and increased AR stability JMJD3 (KDM6B) H3K27 di- and tri-demethylase overexpressed in metastatic [35]. prostate cancer DNA methylation DNMTs Control of transcriptional program during prostate cancer and [36] CRPC progression GSTP1 Silencing of GSP1 upon promoter hypermethylation is a [37–39] potential prognostic biomarker and occurs early during prostate carcinogenesis Histone acetylation P300 Histone acetyltransferase. Besides canonical histone acetylation [40, 41] activity, it acetylates the AR and enhances its transcriptional activity (coactivator) and drives prostate cancer growth SIRT1 Histone deacetylase; regulates cellular growth through AR [42, 43]. deacetylation SIRT2 Histone deacetylase; its downregulation has been associated with [44] increased acetylated H3K18 and poorer outcome and decreased sensitivity to androgen deprivation therapy BET bromodomain epigenetic readers BRD4 Curr Mol Bio Rep (2018) 4:101–115 103 Table 1 (continued) Gene name Function in prostate cancer References Bromodomain and extra-terminal protein, interacts with AR and [45� , promote its activity and antiandrogen resistance 46–48] TRIM24 Epigenetic reader and transcription co-regulator, overexpressed [49]. in CRPC and associated to disease recurrence. Required for prostate cancer cell proliferation in CRPC CHD1 H3K4me2-3 epigenetic reader whose loss is related with prostate [50, 51] cancer aggressiveness and DNA repair defects, thus sensitizing tumor cells to PARP inhibitors Pioneer transcription factors FOXA1 FOXA1 activity on chromatin results in increased accessibility [52, 53] and increased chromatin-bound AR. High FOXA1 expression leads to a restricted AR cistrome regulation. FOXA1 also has the potential to reprogram GATA2 GATA2 GATA2 activity in human prostate cancer is strongly associated [53–55] to AR levels and is hence considered a prostate cancer oncogene Epigenetic regulators of lineage plasticity SOX2 Overexpressed TF in prostate cancer, regulating CRPC [56–61, proliferation, and evasion of apoptosis. Promotes tumor 62�� , metastasis by inducing EMT. Associated to NEPC emergence 63�� ] MYC Master regulator of prostate cancer transcriptional program. [64, 65] Associated with prostate cancer recurrence and poor prognosis MYCN Driver of NEPC by inducing an EZH2-mediated transcriptional [64, 66] program Oncogenic pathways Hsp90 Initiates ERK signaling and leads to the recruitment of EZH2 to [67]. the E-cadherin promoter and repression of E-cadherin expression, driving EMT and invasion in prostate cancer cells DAB2IP Tumor suppressor Ras-GAP. Negatively controls Ras-dependent [68–70]. mitogenic signals and modulates TNFα/NF-κB, WNT/β-catenin, PI3K/AKT, and androgen receptors path- ways RB1 This tumor suppressor gene is commonly loss in metastatic and [71, 72, antiandrogen resistant prostate cancer and NEPC. Directly 63�� ] repress the expression of Sox2 and Ezh2 ACK1 Tyrosine kinase correlated with poor prognosis and interacts with [72–74] AR to drive ADT resistance and CRPC growth. Regulates transcription of AR and AR-v7 via epigenetic regulation reprograming of prostate cancer cells as they lose AR- numerous other key genes have been implicated in DNA imposed identity. Beyond reviewing the current status of epi- methylation changes. In fact, the promoter of the Androgen genetic biomarkers and classifiers and their clinical impact, Receptor (AR) itself appears to be hypermethylated in up to we will discuss the scientific basis for therapeutic targeting 30% of CRPCs, resulting in the loss of AR expression [76]. master regulators of chromatin remodeling and integrity and Moreover, PTEN silencing is often a consequence of promoter the current state of epigenetic drugs for prostate cancer. CpG islands hypermethylation [77], while hypermethylation of the p16 tumor suppressor gene is associated with a prolif- erative advantage, thus contributing to carcinogenesis and dis- DNA Methylation and Histone Modifications ease progression [78]. Similarly, the hypomethylation and in Prostate Carcinogenesis consequent upregulation of genes like heparanase and uroki- nase plasminogen activator (uPA) was reported to contribute Perturbed DNA methylation patterns have long been reported to tumor cell invasion and metastasis [79]. More globally, during prostate cancer progression [75]. Among the most DNA methylation signatures have been identified and pro- well-described alterations is the GSTP1 promoter hyperme- posed as molecular biomarkers of prostate cancer progression thylation and subsequent silencing [37], which is thought to and treatment response [80]. occur early during prostate carcinogenesis [38] and has thus Histone modifications also play an important role in the been proposed as a potential prognostic biomarker [39]. Yet, progression of many tumor types including prostate cancer. 104 Curr Mol Bio Rep (2018) 4:101–115 Lysine methyltransferases (KMT) and demethylases (KDM) progression of prostate cancer cells; thus, it is highly down- are important epigenetic histone modifiers implicated in the regulated in metastatic prostate tumors and those low levels control of gene transcriptional regulation as well as in non- were associated with poor prognosis [20]. In addition, KDM5 histone protein posttranslational modifications and activity loss has been associated with resistance to docetaxel in pros- modulation [81]. More specifically, SUV39H1 (KMT1A) tate cancer [21]. The PHD-finger protein 8 (PHF8) is a histone and SETDB1 (KMT1E) have been shown to enhance prostate demethylase and a transcriptional coactivator of AR via cancer cell migration and invasion and to be upregulated in H4K20 demethylation [28]. Its expression, highly correlated human prostate cancer specimens, and hence suggested as with poor prognosis, is induced by hypoxia and promotes potential therapeutic targets [6], while SUV39H2 (KMT1B) prostate cancer cell proliferation, migration and invasion interacts with the AR to increase androgen-dependent tran- [28], and neuroendocrine differentiation [29]. scriptional activity [8]. Furthermore, levels of SETDB1 have been recently associated with prognosis and the development The Histone Methyltransferase NSD2 of bone metastases from prostate cancer [7]. Similarly, SET and MYND domain-containing protein 3 (SMYD3) has also NSD2 (nuclear receptor binding SET domain protein 2), also been identified as an upregulated H3 and H4 lysine methyl- known as WHSC1 (Wolf-Hirschhorn syndrome candidate 1) transferase promoting cell proliferation and migration, thus and MMSET (multiple myeloma SET domain), is a member emerging as a predictive marker of prostate cancer [10]. of the histone methyltransferase NSD family of proteins also Alternatively, protein arginine methyltransferase 5 (PRMT5) including NSD1 and NSD3. NSD2 catalyzes the was described as a prostate cancer oncogene driving cancer dimethylation of histone H3 at lysine 36 (H3K36me2), a per- cell growth through epigenetic inactivation of several tumor missive mark associated with open chromation conformation suppressors [11] through histone arginine methylation at andactivegenetranscription [85]. NSD2 was first linked to H4R3. PRMT5 has also recently been shown to enhance oncogenesis by the involvement in the t(4; 14) translocation AR-targeted gene expression by arginine methylation and in- identified in up to 20% of multiple myeloma patients [86]. In teraction with the transcription factor Sp1 [13]. the past years, NSD2 has been shown to be overexpressed in a Demethylases also play an important role in prostate cancer variety of solid tumors including prostate cancer, where it has development. Lysine-specific demethylase 1 been found overexpressed in metastatic PCa compared to pri- (LSD1/KDM1A) has been proposed as an oncogene whose mary tumors and is associated with biochemical recurrence overexpression has been positively correlated with the malig- [1�� ]. Further In vitro studies strengthened the role of NSD2 nancy of many cancer types, including prostate [14�� , 82], in prostate cancer tumorigenesis; it has been shown that NSD2 promoting carcinogenesis by multiple mechanisms. modulates Twist family bHLH transcription factor 1 Increased LSD1 expression is associated with prostate cancer (TWIST1) to promote epithelial to mesenchymal transition recurrence and poor survival and appears to have distinct and invasiveness in prostate cancer cell lines [2]. Moreover, functions in androgen-dependent [14�� , 83] and refractory Asangani and colleagues had reported that EZH2 mediates the prostate cancer [15]. Recently, it was discovered that LSD1 overexpression of NSD2 and that the oncogenic properties of is a co-regulator of vitamin D receptor activity in prostate EZH2 are NSD2 dependent [3]. Interestingly, transcriptional cancer and its expression is correlated with shorter targets of NSD2 in prostate cancer cells are highly enriched progression-free survival in primary and metastatic patients for components of the NF-kB-network, including IL-6, IL-8, [84]. In a recent study, it was found that LSD1-mediated epi- survivin/Birc5, and VEGFA. In fact, NSD2 has been linked to genetic reprogramming drives CRPC and was associated with constitutive activation of NF-kB signaling in CRPC, promot- the activation of CENPE, which was regulated by the co- ing cancer cell proliferation and survival via an autocrine pos- binding of LSD1 and AR to its promoter region, which was itive loop in which NSD2 expression is in turn stimulated by associated with loss of RB1 [16� ]. inflammatory cytokines, such as TNFα and IL-6, via NF-kB The overexpression of other histone demethylases (HDMs) [87]. has also been observed in prostate cancer. An exhaustive func- Very recently, work from Li and collaborators showed that tional screen [27] identified 32 enzymes belonging to the fam- NSD2 is activated in PTEN null tumors by the AKT pathway ily of JmjC domain-containing histone demethylases as criti- and that its expression is required for metastatic progression. cal for prostate cancer proliferation and survival. KDM5 fam- Mechanistically, AKT-mediated phosphorylation of NSD2 Cdt2 ily members are H3K4 demethylases; JARID1B (KDM5B) is prevents its degradation by CRL4 E3 ligase leading to upregulated in prostate cancer tissues and acts as an AR coac- NSD2 stabilization and overexpression. By directly inducing tivator [17], while JARID1C (KDM5C), overexpressed in RICTOR expression, NSD2 mediates a positive feedback prostate cancer, emerged as a predictive marker for therapy loop sustaining AKT signaling [1�� ]. failure in patients after prostatectomy [19]. JARID1D Finally, NSD2 has been shown to physically interact with (KMD5D) was found to suppress the invasion and the AR DNA-binding domain and to be recruited to the Curr Mol Bio Rep (2018) 4:101–115 105 enhancer region of the PSA gene and enhance AR transcrip- ubiquitination and increased AR stability [34]. Finally, JMJD3 tional activity [88], suggesting that NSD2 might be implicated (KDM6B) is progressively overexpressed in metastatic pros- in resistant to ADTor androgen signaling inhibition. Of note is tate cancer [35]. the recent identification of NSD2 as a candidate gene promot- ing androgen independence through an unbiased insertional Histone Acetylation and AR mutagenesis screen [89]. In fact, unpublished data and data from our laboratory currently under peer-review for publica- Acetylated chromatin is generally associated to active tran- tion strongly suggest that NSD2 is an actionable mechanism scription and the enzymes regulating this process are histone in CRPC. acetyltransferases (HAT) and deacetylases (HDAC). Accordingly, acetylated histone H3 in the vicinity of AR- bound chromatin has been shown to reduce androgen depen- Epigenetic Control of Androgen Receptor dence in castration resistance models [92, 93]. That is the case Activity for canonical HAT like p300 and CREB-binding protein, which, besides canonical histone acetylation activities, have Histone modifying enzymes, and LSD1 in particular, are been shown to acetylate the AR and enhance its transcriptional among the best-known modulators of AR transcriptional ac- activity [40]. Importantly, two groups have recently indepen- tivity. LSD1 is an important enzyme involved in AR regula- dently developed small molecule inhibitors targeting tion and prostate cancer that interacts with AR and can stim- p300/CBP. Lasko and colleagues reported a selective catalytic ulate [14�� ] or suppress [15] the transcriptional expression p300/CBP inhibitor able to downregulate the AR transcrip- depending on promoter/enhancer context. This interaction tional program both in castration-sensitive and castration- promotes ligand-dependent transcription of AR target genes, resistant prostate tumors and to inhibit tumor growth in resulting in enhanced tumor cell growth. Its coactivator activ- CRPC xenograft models [94], while Jin and colleagues found ity seems to be associated with H3K9me1,2 demethylation that targeting the p300/CBP bromodomain had remarkably leading to transcriptional de-repression of AR target genes similar effetcs [41]. More broadly, a recent study highlights [14 ]. Intriguingly, LSD1 also plays a role as co-repressor, the important role of histone acetylation in prostate cancer via H3K4me1,2 demethylation [90] and the recruitment of co- beyond active promoters via activation of AR associated en- repressor complexes. This highlights the dual role of many hancers and the increase in chromatin accessibility [95� ]. chromatin remodelers and may explain why translating them Conversely, a variety of HDACs are also capable of to new therapeutics has so far been limited. A possible way deacetylating the AR and inhibit its activity, for example via forward may be to define the context specificities for this regulation of heat-shock protein 90 (Hsp90), a chaperone con- duality. For example, it has been shown that in high androgen trolling AR nuclear localization and activation through its levels, AR recruits LSD1 to mediate AR gene silencing [15]; acetylation/deacetylation, or sirtuin 1 (SIRT1), which regu- however, this negative feedback loop is apparently disrupted lates cellular growth through AR deacetylation [42, 43]. In in CRPC, where low androgen levels promote AR overex- fact, acetylation of H3K18, putatively via downregulation of pression. Additionally, post-transcriptional modifications can SIRT2 deacetylase, has been associated to poorer outcome regulate LSD1 activity and may become better targets; LSD1 and decreased sensitivity to androgen deprivation therapy phosphorylation [91] results in a switch of substrate from (ADT). Finally, at the mechanistic level, the Wu lab has re- H3K4me1,2 to H3K9me1,2, and the promotion of its coacti- cently demonstrated that HDAC inhibitors can suppress vator activity. Jumonji C domain-containing trimethyl lysine HMGA-driven EMT, reduce tumor growth and metastasis demethylases JMJD2A (KDM4A) and JMJD2C (KDM4C) and, importantly, resensitize prostate cancer cells to [96]. also play a significant role in modulating AR transcriptional activity [30, 31], stimulating ligand-independent gene tran- scription via H3K9 demethylation. On the contrary, The Role of EZH2/Polycomb Repressive JMJD1A (KDM3A) recruitment to target genes only occurs Complex in Prostate Cancer in the presence of androgens, regulating AR activity and iden- tifying KDM3A-dependent genes involved in androgen re- The enhancer of zeste homolog 2 (EZH2) is a critical member sponse, hypoxia, glycolysis, and lipid metabolism [33], again of the Polycomb Repressive Complex 2 (PRC2) that regulates evidencing the complex balance between chromatin modify- histone methylation mainly via lysine 27 at histone H3 ing enzymes in controlling different but interconnected cellu- (H3K27), a modification associated to transcriptional silenc- lar processes. Of note is the case of JMJD2B (KDM4B), ing [97] that is found upregulated in many tumor types. In which is an AR coactivator, emerging as a suitable therapeutic prostate cancer, its elevated expression associates with poorer target for the treatment of prostate cancer. JMJD2B controls outcomes and has therefore been proposed as an oncogene [4, AR transcriptional activity via demethylation and inhibition of 98]. Amajor functionofEZH2istorepresslineage- �� 106 Curr Mol Bio Rep (2018) 4:101–115 specifying factors, thereby promoting stemness features [99], finding by Zuber and colleagues with implications in risk epithelial-mesenchymal transition (EMT), and ultimately met- assessment shows that tissue-specific SNPs in super- astatic progression [100]. A wealth of recent evidence has enhancer sequence bound by BRD4 are significantly associ- confirmed these previous observation in the prostate cancer ated with increased prostate cancer risk and show better en- field. Back-to-back recent articles in Science by the Sawyers richment for risk loci than AR [110]. and Goodrich groups demonstrated that lineage plasticity and BRD4 physically interacts with high-affinity with the N- neuroendocrine differentiation in androgen independence is terminal domain of AR leading to AR translocation into the partly driven by Ezh2 and Sox2 in prostate cancer mouse nucleus and AR recruitment to target loci, promoting AR ac- models carrying loss of function alleles for p53 and Rb tumor tivity and expression of AR target genes in CRPC [45� ]. A suppressors [62�� , 63�� ]. This came to confirm two previous recent study showed that the small molecule BET inhibitor reports by Dardenne and colleagues [64] and by Xu and col- ABBV-075 could disrupt the recruitment of BRD4 at enhanc- leagues [101] showing that N-myc induces EZH2-driven neu- er of AR target genes and repress their expression, whithout roendocrine prostate cancer [64] and it cooperates with E2F1 affecting AR protein levels [111]. Moreover, BET proteins in castration resistance [101]. have a role in resistance to antiandrogens and BET inhibitors Yet, EZH2 has also PRC2-independent roles as coactivator can effectively resensitize resistant tumors to enzalutamide of transcription factors, including an AKT-dependent methyl- [112]. One of these mechanisms of resistance to antiandrogens ation of the AR, via PI3K/AKT phosphorylation of EZH2 at is the upregulation of the glucocorticoid receptor (GR), and serine 21 [102], and modulation of AR recruitment to its target the co-option of the AR regulon, thus favoring CRPC progres- sites [103�� ]. Not surprisingly, EZH2 inhibitors are the focus sion by overcoming AR dependency [46, 47, 113]. of intensive development and have been widely tested in vivo Beyond AR signaling, BRD4 has been shown to bind to the [5] and in clinical trials (see Table 2 for details). Beyond a truncated ERG (ERGΔ39) encoded by the TMPRSS2-ERG promising drug target, EZH2 and TOP2A have been proposed fusion, co-regulating the expression of ERG target genes in as prognostic as well as predictive biomarkers of treatment CRPC, thereby stimulating cell growth and invasion [114]. response against EZH2 inhibitors [104]. Additionally, SPOP, an E3 ligase substrate binding protein frequently mutated in prostate cancer, was also reported to target BET proteins for ubiquitination-mediated degradation. Bromodomain-Containing Proteins Interestingly, SPOP mutants fail to ubiquitinate BET proteins, in Prostate Cancer leading to their stabilization and to resistance to BET inhibi- tors [48, 115]. This mechanism of resistance causes activation Bromodomain-containing proteins are chromatin readers that of AKT-mTORC1 signaling and consequently resistance to BET inhibitors can be overcome by combination with AKT recognized mono-acetylated histones and trigger chromatin remodeling to initiate transcription. Mutations and deregula- inhibitors [116]. tion of BRD-containing proteins is a common feature of a It is well known that one of the major aging-associated variety of cancers. More than 50% of primary and metastatic drivers of prostate carcinogenesis is oxidative stress and its prostate tumors and more than 70% of neuroendocrine pros- impact on DNA [117]. Interestingly, Hussong and colleagues tate cancer present genomic alterations in any of the 42 known have recently established a link between BRD4 and oxidative BRD-containing proteins [105]. Further, BRD-containing stress response genes in prostate cancer, such as the KEAP1/ proteins have a diversity of catalytic and scaffolding functions NRF2 axis and HMOX1, and reactive oxygen species (ROS) and may act as transcription factors, transcriptional co-factors production [118]. recruiting other proteins in the transcriptional complex, meth- Other than BET, several BRD-containing proteins have yltransferases, HATs, Helicases, and ATP-dependent chroma- been associated to prostate cancer progression and are at dif- tin remodelers, therefore playing a central role in gene expres- ferent validation stages for therapeutic targets in mCRPC. sion regulation [106]. TRIM24, tripartite motif-containing protein 24, is an epige- The subgroup of BET proteins (bromodomain and extra- netic reader and transcription co-regulator overexpressed in terminal), and in particular BRD4, have been the best charac- CRPC and associated to disease recurrence. Recurrent SPOP terized in prostate cancer, and several inhibitors of BET mutants stabilize TRIM24 [119], enhancing AR signaling and bromodomains have been developed and are currently in clin- promoting tumor growth via binding with the proteins TIP60 ical trial (see Table 2). The conserved BET family includes and BRD7 [120], which has led to the proposition of TRIM24 BRD4, BRD2, BRD3, expressed ubiquitously, and BRDT, as an essential gene for prostate cancer cell proliferation in specifically expressed in the testis. BRD4 recognizes acetylat- CRPC [49]. ed lysines at enhancers/superenhancer [107�� , 108�� ] and re- Finally, the role of chromodomain proteins, and in partic- cruits the elongation factor P-TEFb and stimulates RNA po- ular chromodomain helicase DNA-binding protein 1 (CHD1), lymerase II-dependent transcription [109]. A provocative new has in the recent years been elucidated in the context of Curr Mol Bio Rep (2018) 4:101–115 107 Table 2 Clinical trials for epigenetic drugs including prostate cancer patients Trial ID Drug Phase Conditions Patients Status BET bromodomain inhibitors NCT02259114 OTX015/MK-8628 I NUT midline carcinoma, triple negative 47 Completed breast cancer, non-small cell lung cancer (rearranged ALK or mut KRAS), CPRC, pancreatic ductal adenocarcinoma NCT02698176 I NUT midline carcinoma, triple negative breast 13 Terminated cancer, non-small cell lung cancer, CRPC NCT01987362 I Solid Tumors 120 Active NCT02711956 ZEN003694 I Metastatic CRPC (+enzalutamide) 58 Recruiting NCT02705469 I Metastatic CRPC 44 Active NCT03266159 GSK525762 II Solid tumors 150 Not recruiting NCT02419417 BMS-986158 I/II Advanced solid tumors 150 Recruiting NCT02391480 ABBV-075 I Advanced cancer, breast cancer, non-small, 150 Recruiting ell lung cancer, acute myeloid leukemia, multiple myeloma, prostate cancer, small-cell lung cancer, non-Hodgkins lymphoma NCT02630251 GSK2820151 I Advanced or recurrent solid tumors 60 Recruiting NCT02369029 BAY 1238097 I Neoplasms 8 Terminated NCT02431260 INCB054329 I/II Advanced cancer 69 Active, not recruiting NCT02711137 INCB057643 I/II Advanced cancer 230 Recruiting NCT02607228 GS-5829 I/II Metastatic CRPC (+enzalutamide) 132 Recruiting NCT02711137 INCB057643 I/II Advanced solid tumors and hematologic 420 Recruiting malignancy (+abiraterone) EZH2 and PRC1/2 inhibitors NCT03213665 Tazemetostat II Advanced solid tumors, non-Hodgkin 49 Recruiting lymphoma, or histiocytic (EZH2, SMARCB1, or SMARCA4 mutations) NCT01897571 I/II Advanced solid tumors 420 Recruiting NCT02875548 II Advanced solid tumors 300 Recruiting NCT03217253 I Metastatic malignant solid neoplasm 48 Not recruiting PRMT5 inhibitor NCT02900651 MAK683 I/II Diffuse large B cell lymphoma, 113 Recruiting advanced solid tumors LSD1/KDM1A inhibitors NCT02712905 INCB059872 I/II Advanced cancer 180 Recruiting DNMT inhibitors NCT01118741 Disulfiram Prostate cancer 19 Completed NCT00503984 Azacitidine I/II Metastatic CRPC (+docetaxel, prednisone) 22 Terminated NCT00384839 II CRPC 53 Completed NCT02998567 Guadecitabine I Non-small cell lung cancer, CRPC 35 Not yet recruiting (+pembrolizumab) HDAC inhibitors NCT01075308 Pracinostat (SB939) II Metastatic CRPC 32 Completed NCT00670553 I Prostate cancer, head and neck 7 Completed cancer, esophageal cancer NCT00878436 Panobinostat (LBH589) I/II CRPC (+bicalutamide) 52 Completed NCT00667862 II Metastatic CRPC 35 Completed NCT00663832 I CRPC (+docetaxel and prednisone) 44 Completed NCT00493766 I CRPC (+docetaxel and prednisone) 16 Terminated NCT00419536 I CRPC (+docetaxel and prednisone) 108 Terminated NCT00330161 Vorinostat (SAHA, MK0683) II Metastatic CRPC 29 Completed NCT01174199 I Metastatic CRPC 13 Terminated NCT00589472 II Primary prostate cancer (+bicalutamide.) 19 Completed 108 Curr Mol Bio Rep (2018) 4:101–115 Table 2 (continued) Trial ID Drug Phase Conditions Patients Status NCT00565227 I Non-small-cell lung carcinoma, prostate 12 Terminated cancer, bladder cancer, urothelial carcinoma NCT00511576 Mocetinostat (MGCD0103) I Breast cancer, lung cancer, prostate cancer, 54 Terminated gastric cancer (+docetaxel) NCT00020579 Entinostat (MS-275) I Advanced solid tumors, lymphoma 75 Completed NCT00413075 Belinostat (PXD101) I Advanced solid tumors, lymphoma 121 Completed NCT00413322 I Advanced solid tumors (+5-fluorouracil) 35 Completed prostate cancer progression. This H3K4me2-3 epigenetic to be quite different. GATA2 depletion did not seem to reader has been reported mutated in 43% of Gleason 7 have a reprogramming effect on AR binding sites and in or higher prostate cancer tumors, associated with ETS fact correlated with a downregulation in AR expression. gene fusion negative status [121] and its loss related Accordingly, GATA2 activity in human prostate cancer with prostate cancer aggressiveness [50] and DNA re- is strongly associated to AR levels and is hence consid- pair defects, hence sensitizing tumor cells to PARP in- ered a prostate cancer oncogene. Provocatively, it was hibitors [51]. More recently, Zhao and colleagues at the found that FOXA1 also has the potential to reprogram DePinho laboratory have demonstrated in PTEN null GATA2 and act as a pioneering effect for both AR and GATA2, suggesting that FOXA1 regulates a transcrip- prostate tumors that CHD1 depletion dramatically sup- pressed cell proliferation, survival, and tumorigenic po- tional network that controls AR-mediated gene expres- tential by activating the pro-tumorigenic TNF-NF-κB sion in prostate cancer [53]. gene network [122]. Lineage Plasticity in Prostate Cancer Stem Pioneer Factors in Prostate Cancer Cells Progression Aside from their ability to induce pluripotency, the Yamanaka Different from other DNA bound proteins and tran- factors (OCT4, SOX2, KLF4, and c-MYC) [130], and other scription factors, pioneer factors can access their targets reprograming factors like NANOG or LIN28, have been in nucleosomes and in highly compacted chromatin re- widely implicated in tumorigenesis in various cancers includ- gions, facilitating chromatin accessibility and the re- ing the prostate. cruitment of additional TFs and co-TF and the tran- SOX2 is required for survival, pluripotency, scriptional machinery [123]. Among paradigmatic clonogenicity, and self-renewal of ESCs. A relationship pioneering factors are some of the members of the between SOX2 overexpression in tumorigenesis has GATA and FoxA gene families, known mainly for their been established in different types of cancer, including key role as chromatin-factors during early development prostate [56] and its expression linked to tumor grade [124–127]. [58]. SOX2 is an epigenetic reprogramming factor and The best-known pioneering factor for its role in pros- oncogene shown to regulate androgen-independent tate cancer is FOXA1. Through the interaction and re- CRPC proliferation and evasion of apoptosis [57, 58] cruitment of AR to chromatin site, FoxA1 defines and andtopromote tumormetastasisbyinducing EMT controls the AR cistrome resulting in context-dependent [59]. Further evidence suggests that SOX2 promotes positive or negative regulation [52, 55, 128, 129]. In self-renewal of the CSCs population by acting down- particular, because FOXA1 activity on chromatin re- stream of EGFR [131]. Importantly, in the recent years, sults in increased accessibility [52] and increased SOX2 activity has been tightly associated to neuroen- chromatin-bound AR, high FOXA1 expression leads docrine transdifferentiation from prostate adenocarcino- to a restricted AR cistrome regulation [53]. ma cells and the subsequent androgen independence of GATA genes, and GATA2 in particular, have proved neuroendocrine prostate cancer phenotypes (NEPC). to be crucial for prostate development via modulating While the exact mechanisms remain unclear, substantial AR function [54, 55]. However, despite the role is com- progress was made over the last couple of years. In parable to that of FoxA1, the mechanisms have shown particular, Russo and colleagues showed that SOX2 Curr Mol Bio Rep (2018) 4:101–115 109 was expressed in NEPC murine models [60] whereas contributor of aggressiveness via the activation of EMT tran- others found its expression restricted to NEPC areas scriptional programs. Nolan and colleagues proposed a model of advanced human prostate cancer [61]. Recent studies in which the secreted extracellular protein Hsp90 initiates by Bishop and collaborators at the Zoubeidi laboratory ERK signaling and leads to the recruitment of EZH2 to the have shown that SOX2 is transcriptionally regulated by E-cadherin promoter and repression of E-cadherin expression, neural transcription factor BRN2 [132��], which in turn driving epithelial to mesenchymal transition (EMT) and inva- is negatively suppressed by the AR, hence revealing an sion in prostate cancer cells [67]. Additionally, DAB2IP (dis- AR-dependent suppression of cell differentiation to- abled homolog 2 interacting protein) is a tumor suppressor ward a neuroendocrine AR-independent phenotype. Ras-GAP that negatively controls Ras-dependent mitogenic Additional support to the central role of SOX2 in the signals and modulates TNFα/NF-κB, WNT/β-catenin, emergence of NEPC and AR-independence after PI3K/AKT, and androgen receptors pathways [68–70]. Enzalutamide treatment came from studies at the Ku EZH2-induced DAB2IP silencing activates Ras and NF- and Mu and collaborators at the Sawyers and kappaB and triggers metastasis [141, 142]. Data from our Goodrich laboratories [62��, 63��]. laboratory showed that concomitant activation of the PI3K c-MYC (MYC) is a well-known oncogene proposed and MAPk pathways in mice results in highly aggressive as a marker of disease progression in prostate cancer and fully metastatic tumors that are inherently castration resis- [133] and associated with prostate cancer recurrence tant [143, 144]. Interestingly, by targeting the PI3K/MAPk and poor prognosis [134]. MYC activation cooperates pathways with small molecules in vivo, we demonstrated that with loss of PTEN to drive prostate cancer progression the drug response network was highly enriched in epigenetic [135] and metastasis [136]. MYC proteins also drive modulators, including SUV39H1, WHSC1, TOP2A, or epigenetic activation of gene expression in prostate UHRF1 [145], suggesting that epigenetic control of gene ex- cancer; the PRC2 member EZH2 is directly upregulated pression plays a central role in the aggressive phenotype im- by MYC [137] and MYCN, which was shown to be a posed by the activation of Ras signaling. Accordingly, we driver of NEPC [66] by inducing an EZH2-mediated have found that a core signature of chromatin modifiers and transcriptional program [64]. Additionally, MYC ex- DNMTs drive the cancer cell intrinsic mechanisms of metas- pression was found to be regulated by the histone tasis and CRPC (unpublished). demethylase JMJD1A, controlling proliferation and sur- The retinoblastoma tumor suppressor gene RB1 is more vival of prostate cancer cells [138]. MYC also regulates commonly loss in metastatic and antiandrogen resistant pros- the expression of histone demethylases PHF8 and tate cancer (74% of cases) and NEPC (90% of cases) [71] than KDMA3 in NEPC and CRPC [29]. Interestingly, while it is in primary tumors (34% of cases) [72]. It has been recently described an activity of Rb1 in the epigenetic regulation of AR signaling in the normal prostate represses MYC expression, its expression is stimulated by AR during expression, since RB1 directly repress the expression of Sox2 tumorigenesis, [139, 140]. It was also recently reported and Ezh2. Consequently, Rb1 loss in prostate cancer lead to that MYC overexpression deregulates the AR transcrip- EZH2 and Sox2 increase and gene expression widespread tional program by altering AR chromatin occupancy changes that leads toward a stem cell-like state that would and H3K4me1 and H3K27me3 marks distribution, an- facilitate the onset of metastasis, neuroendocrine tagonizing clinically relevant AR target genes such as transdifferentiation, and the acquisition of ADT resistance. PSA [65]. The authors show that Ezh2 inhibition restores enzalutamide sensitivity in NEPC variants and recurrent prostate cancer cells by opposing lineage transformation [63�� ]. Oncogenic Pathways Involved in Epigenetic Furthermore, mutations in TP53 and RB1 tumor suppressor Regulations genes can promote a cellular plasticity state mediated by in- creased expression of SOX2 that, when it is compromised Together with the AR, the oncogenic pathways most frequent- with antiandrogen therapy promotes resistance through line- ly altered in prostate cancer onset and progression are the RB, age switching [62�� ]. It has also recently been shown that the PI3K/AKT, and Ras/Raf pathways due to mutations in several Hedgehog (HH) signaling pathway and SOX2 co-operate in members [72]. While the Ras/Raf pathway is activated in 43% androgen-independent prostate cancer to promote carcinogen- of primary and 90% of metastatic prostate cancer, the trigger- esis [146]. ing mechanisms remain to be fully understood. The Whitte The PTEN/PI3K/AKT pathway is altered in 42% of prima- laboratory demonstrated a synergistic interaction between Ras ry and 100% of metastatic cases; loss of PTEN and activation pathway activation and AR signaling that leads to elevated of the PI3K/AKT signaling pathway are hallmarks of prostate EZH2 expression and expand prostate cancer progenitor cells cancer, and cooperate with the activation of the RAS/MAPK pathway to promote EMT and metastatic CRPC development. in vivo. It has been long suggested that this pathway is a major 110 Curr Mol Bio Rep (2018) 4:101–115 Epigenetically, it has also been shown that PTEN depletion Drug Development on Epigenetic Regulators contributes to a switch from a global H3K27 acetylatilation to as Therapeutic Targets H3K27 trimethylation, resulting in increased expression of EZH2 and decrease of the target genes DAB2PI together with Mounting evidence from basic and preclinical investigations KIP1 CIP1 negative regulator of cell growth p27 and p21 [147]. suggest that targeting key components of the epigenetic ma- As mentioned above, increased AKT activity phosphorylates chinery will have clinical benefit for cancer patients including Cdt2 NSD2 at S172, preventing its degradation by CRL4 E3 prostate cancer ones. Yet, clinical development for those ther- ligase, hence leading to its stabilization, which in turn apies is still very limited. On the one hand, this may be partly upregulates RICTOR (mTORC2). This results in further en- due to the inherent difficulty in targeting nuclear effector hancement of AKT signaling in a AKT/NSD2/mTORC2 pos- mechanisms. On the other hand, the fact that most epigenetic itive feedback loop that sustains AKT signaling [1 ]. master regulators exert their functions over an extensive tran- Constitutive activation of TGF-β signaling is a well- scriptional network in a context-dependent manner makes it recognized mechanism for induction of EMT and prostate particularly challenging to achieve cancer cell specificity, thus cancer metastasis development. TGF-β1-induced EMT in resulting in significant toxicity. Despite these limitations, a prostate cancer is mediated by the histone methyltransfer- number of drugs are currently in clinical trials at different ase RbBP5. RbBP5 is a conserved component of the phases, being BET bromodomain inhibitors, HMT/HDMT COMPASS/-like complex, which catalyzes the inhibitors, DNMT inhibitors, and HDAC inhibitors the focus trimethylation of histone H3 lysine 4 that is considered of most intense drug development efforts. Table 2 summarizes an epigenetic mark of actively transcribed genes. RbBP5 the most relevant ongoing or recently completed clinical trials activity is in turn modulated by the binding of SMAD2/3, a involving epigenetic drugs. downstream signaling factor to the TGF-beta pathway, to the Snail promoter [148]. Snail activates the EMT process by inhibiting transcription of E-cadherin via the recruit- ment to its promoter of the polycomb repressive complex Conclusion 2 (PRC2) and the histone methylstranferase G9a, leading to repressive H3K27 and H3K9 methylation [149, 150]. In view of the accumulated evidenced supporting the key An interesting new perspective was provided recently role of the dysregulated epigenome to prostate cancer onset linking ERG signaling with TGF-β. Data suggest that and progression, three mechanisms emerge as the most sig- ERG regulates the transcription of the transcription factor nificant contributors. First, a number of alterations in epi- SOX4 and together they cooperate in TGF-β1-induced genetic master regulators result in enhanced transcriptional EMT of prostate cancer cells [151]. This is not surprising activity and pro-oncogenic role of the Androgen Receptor taking into account that the oncogenic role of SOX4 has signaling. This is largely mediated by either remodeling of been proposed in several other tumor types. In particular, the chromatin to facilitate AR binding and assembly of the SOX4 regulates EZH2 expression and chromating remod- transcriptional complex and posttranslational modifications eling, and is a key component of the PI3K/AKT pathway in in the AR itself or essential co-factors resulting in gain of prostate cancer. In fact, SOX4 inhibition reduces AKT and function features. Secondly, the aberrant activation of tran- β-catenin pathways activation and decreases prostate can- scriptional programs tightly associated to developmental cer invasiveness through positive feedback loop between pathways and stem features, either via alterations in SOX4 and PI3K-AKT-mTOR [152]. pioneering factors or pluripotency master regulators, con- Finally, a tyrosine kinase, namely ACK1, has been found tributes to the acquisition of treatment-resistant phenotypes to link oncogenic signaling with epigenetic regulation. that are highly aggressive. Finally, a significant number of ACK1 was found upregulated in primary PCa and CRPC alterations in epigenetic master regulators also result in the [72, 73], correlated with poor prognosis and reported to activation of oncogenic signaling pathways that contribute interact with AR to drive ADT resistance and CRPC growth to the aggressiveness and androgen independence in ad- [74]. A recent study demostrates that ACK1 regulates tran- vanced prostate tumors. In summary, the epigenome is scription of AR and AR-v7 via epigenetic regulation. In emerging as an attractive and plausible target for anticancer particular, ACK1 would phosphorylate histone H4 up- therapy in general and prostate cancer in particular. While stream of the AR transcription start site, recruiting the drug development is still limited, and faces inherent chal- WRD5/MLL2 complex, therefore mediating H3K4 lenges associated with the unique nature of these targets, it trymethylation and transcriptional activation. Inhibition of seems evident that efficacy of such treatments will be max- ACK1 with a small molecule inhibitor confirms that this imized in combination with standard of care treatments for epigenetic activity is required to maintain AR transcription which most lethal prostate cancer ultimately develop resis- and CRPC tumor growth [153]. tant mechanism. �� Curr Mol Bio Rep (2018) 4:101–115 111 Acknowledgements This research is supported by funding from the 9. Vieira FQ, Costa-Pinheiro P, Almeida-Rios D, Graca I, Monteiro- Spanish Ministry of Economy, Industry and Competitiveness (MINECO), Reis S, Simoes-Sousa S, et al. SMYD3 contributes to a more which is part of the State Research Agency, through the projects PI16/01070 aggressive phenotype of prostate cancer and targets cyclin D2 and CP15/00090. We also acknowledge the support from the European through H4K20me3. Oncotarget. 2015;6(15):13644–57. Association of Urology Research Foundation (EAURF/407003/XH), and 10. Huang L, Xu AM. SET and MYND domain containing protein 3 the Fundacion BBVA-Young Investigator Award. We thank CERCA in cancer. Am J Transl Res. 2017;9(1):1–14. Program / Generalitat de Catalunya for the institutional support. 11. Stopa N, Krebs JE, Shechter D. The PRMT5 arginine methyltrans- ferase: many roles in development, cancer and beyond. Cell Mol Life Sci. 2015;72(11):2041–59. Compliance with Ethical Standards 12.� Mounir Z, Korn JM, Westerling T, Lin F, Kirby CA, Schirle M, et al. ERG signaling in prostate cancer is driven through PRMT5- Conflict of Interest Katia Ruggero, Sonia Farran-Matas, Adrian dependent methylation of the Androgen Receptor. Elife. Martinez-Tebar, and Alvaro Aytes declare no conflicts of interest. 2016;16(5):13964. This paper links for the first time epigenet- ically driven posttranslational modifications in the AR and Human and Animal Rights and Informed Consent This article does not ERG transcriptional expression. contain any studies with human or animal subjects performed by any of 13. Deng X, Shao G, Zhang HT, Li C, Zhang D, Cheng L, et al. the authors. Protein arginine methyltransferase 5 functions as an epigenetic activator of the androgen receptor to promote prostate cancer cell Open Access This article is distributed under the terms of the Creative growth. Oncogene. 2017;36(9):1223–31. Commons Attribution 4.0 International License (http:// 14.�� Metzger E, Wissmann M, Yin N, Muller JM, Schneider R, Peters creativecommons.org/licenses/by/4.0/), which permits unrestricted use, AH, et al. LSD1 demethylates repressive histone marks to pro- distribution, and reproduction in any medium, provided you give appro- mote androgen-receptor-dependent transcription. Nature. priate credit to the original author(s) and the source, provide a link to the 2005;437(7057):436–9. This paper links demethylation of a Creative Commons license, and indicate if changes were made. repressive histone mark with androgen-receptor-dependent gene activation. 15. Cai C, He HH, Chen S, Coleman I, Wang H, Fang Z, et al. Androgen receptor gene expression in prostate cancer is directly References suppressed by the androgen receptor through recruitment of lysine-specific demethylase 1. Cancer Cell. 2011;20(4):457–71. 16.� Liang Y, Ahmed M, Guo H, Soares F, Hua JT, Gao S, et al. LSD1- Papers of particular interest, published recently, have been mediated epigenetic reprogramming drives CENPE expression highlighted as: and prostate cancer progression. Cancer Res. 2017;77(20):5479– � Of importance 90. This study shows that LSD1-mediated epigenetic reprogramming drives CRPC and was associated with the �� Of major importance activation of CENPE and other cell-cycle genes. 17. Xiang Y, Zhu Z, Han G, Ye X, Xu B, Peng Z, et al. JARID1B is a 1.�� Li N, Xue W, Yuan H, Dong B, Ding Y, Liu Y, et al. AKT- histone H3 lysine 4 demethylase up-regulated in prostate cancer. mediated stabilization of histone methyltransferase WHSC1 pro- Proc Natl Acad Sci U S A. 2007;104(49):19226–31. motes prostate cancer metastasis. J Clin Invest. 2017;127(4): 18. Lu W, Liu S, Li B, Xie Y, Adhiambo C, Yang Q, et al. SKP2 1284–302. This work demonstrates that NSD2 is activated in inactivation suppresses prostate tumorigenesis by mediating PTEN null tumors by the AKT pathway and that its expres- JARID1B ubiquitination. Oncotarget. 2015;6(2):771–88. sion is required for metastatic progression. 19. Stein J, Majores M, Rohde M, Lim S, Schneider S, Krappe E, et al. 2. Ezponda T, Popovic R, Shah MY, Martinez-Garcia E, Zheng Y, DM5C is overexpressed in prostate cancer and is a prognostic Min DJ, et al. The histone methyltransferase MMSET/WHSC1 marker for prostate-specific antigen-relapse following radical activates TWIST1 to promote an epithelial-mesenchymal transi- prostatectomy. Am J Pathol. 2014;184(9):2430–7. tion and invasive properties of prostate cancer. Oncogene. 20. Li N, Dhar SS, Chen TY, Kan PY, Wei Y, Kim JH, et al. JARID1D 2013;32(23):2882–90. is a suppressor and prognostic marker of prostate cancer invasion 3. Asangani IA, Ateeq B, Cao Q, Dodson L, Pandhi M, Kunju LP, and metastasis. Cancer Res. 2016;76(4):831–43. et al. Characterization of the EZH2-MMSET histone methyltrans- 21. Komura K, Jeong SH, Hinohara K, Qu F, Wang X, Hiraki M, et al. ferase regulatory axis in cancer. Mol Cell. 2013;49(1):80–93. Resistance to docetaxel in prostate cancer is associated with an- 4. Yang YA, Yu J. EZH2, an epigenetic driver of prostate cancer. drogen receptor activation and loss of KDM5D expression. Proc Protein Cell. 2013;4(5):331–41. Natl Acad Sci U S A. 2016;113(22):6259–64. 5. Wu C, Jin X, Yang J, Yang Y, He Y, Ding L, et al. Inhibition of 22. Yu L, Wang Y, Huang S, Wang J, Deng Z, Zhang Q, et al. EZH2 by chemo- and radiotherapy agents and small molecule Structural insights into a novel histone demethylase PHF8. Cell inhibitors induces cell death in castration-resistant prostate cancer. Res. 2010;20(2):166–73. Oncotarget. 2016;7(3):3440–52. 23. Qi HH, Sarkissian M, Hu G-Q, Wang Z, Bhattacharjee A, Gordon 6. Yu T, Wang C, Yang J, Guo Y, Wu Y, Li X. Metformin inhibits DB, et al. Histone H4K20/H3K9 demethylase PHF8 regulates SUV39H1-mediated migration of prostate cancer cells. zebrafish brain and craniofacial development. Nature. 2010;466: Oncogene. 2017;6(5):):28. 503–7. 7. Li Q, Li Y, Wang Y, Cui Z, Gong L, Qu Z, et al. Quantitative 24. Qiu J, Shi G, Jia Y, Li J, Wu M, Li J, et al. The X-linked mental proteomic study of human prostate cancer cells with different met- retardation gene PHF8 is a histone demethylase involved in neu- astatic potentials. Int J Oncol. 2016;48(4):1437–46. ronal differentiation. Cell Res. 2010;20:908–18. 8. Askew EB, Bai S, Parris AB, Minges JT, Wilson EM. Androgen 25. Liu W, Tanasa B, Tyurina OV, Zhou TY, Gassmann R, Liu WT, receptor regulation by histone methyltransferase suppressor of et al. PHF8 mediates histone H4 lysine 20 demethylation events variegation 3-9 homolog 2 and melanoma antigen-A11. Mol involved in cell cycle progression. Nature. 2010;466(7305):508– Cell Endocrinol. 2017;443:42–51. 12. 112 Curr Mol Bio Rep (2018) 4:101–115 26. Horton JR, Upadhyay AK, Qi HH, Zhang X, Shi Y, Cheng X. 44. Damodaran S, Damaschke N, Gawdzik J, Yang B, Shi C, Allen GO, et al. Dysregulation of Sirtuin 2 (SIRT2) and histone H3K18 Enzymatic and structural insights for substrate specificity of a family of jumonji histone lysine demethylases. Nat Struct Amp acetylation pathways associates with adverse prostate cancer out- Mol Biol. 2009;17:38. comes. BMC Cancer. 2017;17(1):874. 27. Bjorkman M, Ostling P, Harma V, Virtanen J, Mpindi JP, Rantala 45.� Asangani IA, Dommeti VL, Wang X, Malik R, Cieslik M, Yang J, et al. Systematic knockdown of epigenetic enzymes identifies a R, et al. Therapeutic targeting of BET bromodomain proteins in novel histone demethylase PHF8 overexpressed in prostate cancer castration-resistant prostate cancer. Nature. 2014;510(7504):278– with an impact on cell proliferation, migration and invasion. 82. This study reports that BRD4 physically interacts with Oncogene. 2012;31(29):3444–56. high-affinity with the N-terminal domain of AR leading to AR translocation into the nucleus and AR recruitment to tar- 28. Tong D, Liu Q, Liu G, Yuan W, Wang L, Guo Y, et al. The HIF/ get loci, promoting AR activity and expression of AR target PHF8/AR axis promotes prostate cancer progression. Oncogene. genes in CRPC. 2016;5(12):e283. 29. Maina PK, Shao P, Liu Q, Fazli L, Tyler S, Nasir M, et al. c-MYC 46. Shah N, Wang P, Wongvipat J, Karthaus WR, Abida W, Armenia drives histone demethylase PHF8 during neuroendocrine differen- J, et al. Regulation of the glucocorticoid receptor via a BET- tiation and in castration-resistant prostate cancer. Oncotarget. dependent enhancer drives antiandrogen resistance in prostate 2016;7(46):75585–602. cancer. elife. 2017;6 47. Li J, Alyamani M, Zhang A, Chang KH, Berk M, Li Z, et al. 30. Shin S, Janknecht R. Activation of androgen receptor by histone Aberrant corticosteroid metabolism in tumor cells enables GR demethylases JMJD2A and JMJD2D. Biochem Biophys Res takeover in enzalutamide resistant prostate cancer. elife. 2017;6 Commun. 2007;359(3):742–6. 31. Wissmann M, Yin N, Muller JM, Greschik H, Fodor BD, 48. Dai X, Gan W, Li X, Wang S, Zhang W, Huang L, et al. Prostate Jenuwein T, et al. Cooperative demethylation by JMJD2C and cancer-associated SPOP mutations confer resistance to BET inhib- LSD1 promotes androgen receptor-dependent gene expression. itors through stabilization of BRD4. Nat Med. 2017;23(9):1063– Nat Cell Biol. 2007;9(3):347–53. 71. 32. Yamane K, Toumazou C, Tsukada Y, Erdjument-Bromage H, 49. Groner AC, Cato L, de Tribolet-Hardy J, Bernasocchi T, Tempst P, Wong J, et al. JHDM2A, a JmjC-containing H3K9 Janouskova H, Melchers D, et al. TRIM24 is an oncogenic tran- demethylase, facilitates transcription activation by androgen re- scriptional activator in prostate cancer. Cancer Cell. 2016;29(6): ceptor. Cell. 2006;125(3):483–95. 846–58. 50. Rodrigues LU, Rider L, Nieto C, Romero L, Karimpour-Fard A, 33. Wilson S, Fan L, Sahgal N, Qi J, Filipp FV. The histone Loda M, et al. Coordinate loss of MAP3K7 and CHD1 promotes demethylase KDM3A regulates the transcriptional program of aggressive prostate cancer. Cancer Res. 2015;75(6):1021–34. the androgen receptor in prostate cancer cells. Oncotarget. 2017;8(18):30328–43. 51. Kari V, Mansour WY, Raul SK, Baumgart SJ, Mund A, Grade M, 34. Coffey K, Rogerson L, Ryan-Munden C, Alkharaif D, Stockley J, et al. Loss of CHD1 causes DNA repair defects and enhances Heer R, et al. The lysine demethylase, KDM4B, is a key molecule prostate cancer therapeutic responsiveness. EMBO Rep. in androgen receptor signalling and turnover. Nucleic Acids Res. 2016;17(11):1609–23. 2013;41(8):4433–46. 52. Jin HJ, Zhao JC, Wu L, Kim J, Yu J. Cooperativity and equilibri- um with FOXA1 define the androgen receptor transcriptional pro- 35. Xiang Y, Zhu Z, Han G, Lin H, Xu L, Chen CD. JMJD3 is a gram. Nat Commun. 2014;5(3972) histone H3K27 demethylase. Cell Res. 2007;17(10):850–7. 36. Chin SP, Dickinson JL, Holloway AF. Epigenetic regulation of 53. Zhao JC, Fong KW, Jin HJ, Yang YA, Kim J, Yu J. FOXA1 acts prostate cancer. Clin Epigenetics. 2011;2(2):151–69. upstream of GATA2 and AR in hormonal regulation of gene ex- pression. Oncogene. 2016;35(33):4335–44. 37. Lin PC, Giannopoulou EG, Park K, Mosquera JM, Sboner A, Tewari AK, et al. Epigenomic alterations in localized and ad- 54. Lupien M, Eeckhoute J, Meyer CA, Wang Q, Zhang Y, Li W, et al. vanced prostate cancer. Neoplasia. 2013;15(4):373–83. Fox A1 translates epigenetic signatures into enhancer-driven line- age-specific transcription. Cell. 2008;132(6):958–70. 38. Kang GH, Lee S, Lee HJ, Hwang KS. Aberrant CpG island hyper- 55. Wang Q, Li W, Liu XS, Carroll JS, Jänne OA, Keeton EK, et al. A methylation of multiple genes in prostate cancer and prostatic hierarchical network of transcription factors governs androgen intraepithelial neoplasia. J Pathol. 2004;202(2):233–40. receptor-dependent prostate cancer growth. Mol Cell. 39. Mahon KL, Qu W, Devaney J, Paul C, Castillo L, Wykes RJ, et al. 2007;27(3):380–92. Methylated glutathione S-transferase 1 (mGSTP1) is a potential plasma free DNA epigenetic marker of prognosis and response to 56. Wong OG, Huo Z, Siu MK, Zhang H, Jiang L, Wong ES, et al. chemotherapy in castrate-resistant prostate cancer. Br J Cancer. Hypermethylation of SOX2 promoter in endometrial carcinogen- 2014;111(9):1802–9. esis. Obstet Gynecol Int. 2010;10(682504):9. 40. Fu M, Wang C, Reutens AT, Wang J, Angeletti RH, Siconolfi- 57. Lin F, Lin P, Zhao D, Chen Y, Xiao L, Qin W, et al. Sox 2 targets Baez L, et al. p300 and p300/cAMP-response element-binding cyclinE, p 27 and survivin to regulate androgen-independent hu- protein-associated factor acetylate the androgen receptor at sites man prostate cancer cell proliferation and apoptosis. Cell Prolif. governing hormone-dependent transactivation. J Biol Chem. 2012;45(3):207–16. 2000;275(27):20853–60. 58. Kregel S, Kiriluk KJ, Rosen AM, Cai Y, Reyes EE, Otto KB, et al. 41. Jin L, Garcia J, Chan E, de la Cruz C, Segal E, Merchant M, et al. Sox2 is an androgen receptor-repressed gene that promotes Therapeutic targeting of the CBP/p300 bromodomain blocks the castration-resistant prostate cancer. PLoS One. 2013;8(1):11. growth of castration-resistant prostate cancer. Cancer Res. 59. Li X, Xu Y, Chen Y, Chen S, Jia X, Sun T, et al. SOX2 promotes 2017;77(20):5564–75. tumor metastasis by stimulating epithelial-to-mesenchymal transi- 42. Fu M, Liu M, Sauve AA, Jiao X, Zhang X, Wu X, et al. Hormonal tion via regulation of WNT/beta-catenin signal network. Cancer control of androgen receptor function through SIRT1. Mol Cell Lett. 2013;336(2):379–89. Biol. 2006;26(21):8122–35. 60. Russo MV, Esposito S, Tupone MG, Manzoli L, Airoldi I, Pompa 43. Lavery DN, Bevan CL. Androgen receptor signalling in prostate P, et al. SOX2 boosts major tumor progression genes in prostate cancer: the functional consequences of acetylation. J Biomed cancer and is a functional biomarker of lymph node metastasis. Biotechnol. 2011;862125(10):28. Oncotarget. 2016;7(11):12372–85. Curr Mol Bio Rep (2018) 4:101–115 113 61. Esposito S, Russo MV, Airoldi I, Tupone MG, Sorrentino C, 76. Suzuki H, Ueda T, Ichikawa T, Ito H. Androgen receptor involve- ment in the progression of prostate cancer. Endocr Relat Cancer. Barbarito G, et al. SNAI2/Slug gene is silenced in prostate cancer and regulates neuroendocrine differentiation, metastasis- 2003;10(2):209–16. suppressor and pluripotency gene expression. Oncotarget. 77. Suzuki H, Freije D, Nusskern DR, Okami K, Cairns P, Sidransky 2015;6(19):17121–34. D, et al. Interfocal heterogeneity of PTEN/MMAC1 gene alter- ations in multiple metastatic prostate cancer tissues. Cancer Res. 62.�� Mu P, Zhang Z, Benelli M, Karthaus WR, Hoover E, Chen CC, 1998;58(2):204–9. et al. SOX2 promotes lineage plasticity and antiandrogen resis- 78. Jarrard DF, Bova GS, Ewing CM, Pin SS, Nguyen SH, Baylin SB, tance in TP53- and RB1-deficient prostate cancer. Science. et al. Deletional, mutational, and methylation analyses of CDKN2 2017;355(6320):84–8. This work demonstrates that lineage (p16/MTS1) in primary and metastatic prostate cancer. Genes plasticity and neuroendocrine differentiation in androgen in- Chromosomes Cancer. 1997;19(2):90–6. dependence is partly driven by Sox2 in prostate cancer mouse 79. Pakneshan P, Xing RH, Rabbani SA. Methylation status of uPA models carrying loss of function alleles for p53 and Rb tumor promoter as a molecular mechanism regulating prostate cancer suppressors. invasion and growth in vitro and in vivo. FASEB J. 2003;17(9): 63.�� Ku SY, Rosario S, Wang Y, Mu P, Seshadri M, Goodrich ZW, et al. 1081–8. Rb1 and Trp53 cooperate to suppress prostate cancer lineage plas- 80. Ashour N, Angulo JC, Andres G, Alelu R, Gonzalez-Corpas A, ticity, metastasis, and antiandrogen resistance. Science. Toledo MV, et al. A DNA hypermethylation profile reveals new 2017;355(6320):78–83. This work shows that Sox2 and Ezh2 potential biomarkers for prostate cancer diagnosis and prognosis. mediates suppression of prostate cancer lineage plasticity in Prostate. 2014;74(12):1171–82. Rb1 and p53 null tumours. Ezh2 inhibition restores 81. Coffey K, Rogerson L, Ryan-Munden C, Alkharaif D, Stockley J, enzalutamide sensitivity in NEPC variants and recurrent Heer R, et al. The lysine demethylase, KDM4B, is a key molecule prostate cancer cells by opposing lineage transformation. in androgen receptor signalling and turnover. Nucleic Acids Res. 64. Dardenne E, Beltran H, Benelli M, Gayvert K, Berger A, Puca L, 2013;41(8):4433–46. et al. N-Myc induces an EZH2-mediated transcriptional program 82. Hayami S, Kelly JD, Cho HS, Yoshimatsu M, Unoki M, Tsunoda driving neuroendocrine prostate Cancer. Cancer Cell. 2016;30(4): T, et al. Overexpression of LSD1 contributes to human carcino- 563–77. genesis through chromatin regulation in various cancers. Int J 65. Barfeld SJ, Urbanucci A, Itkonen HM, Fazli L, Hicks JL, Thiede Cancer. 2011;128(3):574–86. B, et al. C-Myc Antagonises the transcriptional activity of the 83. Kahl P, Gullotti L, Heukamp LC, Wolf S, Friedrichs N, Vorreuther androgen receptor in prostate cancer affecting key gene networks. R, et al. Androgen receptor coactivators lysine-specific histone EBioMedicine. 2017;18:83–93. demethylase 1 and four and a half LIM domain protein 2 predict 66. Lee JK, Phillips JW, Smith BA, Park JW, Stoyanova T, McCaffrey risk of prostate cancer recurrence. Cancer Res. 2006;66(23): EF, et al. N-Myc drives neuroendocrine prostate cancer initiated 11341–7. from human prostate epithelial cells. Cancer Cell. 2016;29(4): 84. Battaglia S, Karasik E, Gillard B, Williams J, Winchester T, Moser 536–47. MT, et al. LSD1 dual function in mediating epigenetic corruption 67. Nolan KD, Franco OE, Hance MW, Hayward SW, Isaacs JS. of the vitamin D signaling in prostate cancer. Clin Epigenetics. Tumor-secreted Hsp90 subverts polycomb function to drive pros- 2017;9(82):017–0382. tate tumor growth and invasion. J Biol Chem. 2015;290(13): 85. Li Y, Trojer P, Xu CF, Cheung P, Kuo A, Drury WJ, 3rd, et al. The 8271–82. target of the NSD family of histone lysine methyltransferases de- 68. Bellazzo A, Di Minin G, Collavin L. Block one, unleash a hun- pends on the nature of the substrate. J Biol Chem 2009; 284(49): dred. Mechanisms of DAB2IP inactivation in cancer. Cell Death 34283–34295. Differ. 2017;24(1):15–25. 86. Keats JJ, Maxwell CA, Taylor BJ, Hendzel MJ, Chesi M, 69. Wu K, Liu J, Tseng SF, Gore C, Ning Z, Sharifi N, et al. The role Bergsagel PL, et al. Overexpression of transcripts originating of DAB2IP in androgen receptor activation during prostate cancer from the MMSET locus characterizes all t(4;14)(p16;q32)-posi- progression. Oncogene. 2014;33(15):1954–63. tive multiple myeloma patients. Blood. 2005;105(10):4060–9. 70. Tsai YS, Lai CL, Lai CH, Chang KH, Wu K, Tseng SF, et al. The 87. Yang P, Guo L, Duan ZJ, Tepper CG, Xue L, Chen X, et al. role of homeostatic regulation between tumor suppressor DAB2IP Histone methyltransferase NSD2/MMSET mediates constitutive and oncogenic Skp2 in prostate cancer growth. Oncotarget. NF-kappaB signaling for cancer cell proliferation, survival, and 2014;5(15):6425–36. tumor growth via a feed-forward loop. Mol Cell Biol. 71. Tan HL, Sood A, Rahimi HA, Wang W, Gupta N, Hicks J, et al. 2012;32(15):3121–31. Rb loss is characteristic of prostatic small cell neuroendocrine 88. Kang HB, Choi Y, Lee JM, Choi KC, Kim HC, Yoo JY, et al. The carcinoma. Clin Cancer Res. 2014;20(4):890–903. histone methyltransferase, NSD2, enhances androgen receptor- mediated transcription. FEBS Lett. 2009;583(12):1880–6. 72. Taylor BS, Schultz N, Hieronymus H, Gopalan A, Xiao Y, Carver 89. Nalla AK, Williams TF, Collins CP, Rae DT, Trobridge GD. BS, et al. Integrative genomic profiling of human prostate cancer. Lentiviral vector-mediated insertional mutagenesis screen iden- Cancer Cell. 2010;18(1):11–22. tifies genes that influence androgen independent prostate cancer 73. van der Horst EH, Degenhardt YY, Strelow A, Slavin A, Chinn L, progression and predict clinical outcome. Mol Carcinog. Orf J, et al. Metastatic properties and genomic amplification of the 2016;55(11):1761–71. tyrosine kinase gene ACK1. Proc Natl Acad Sci U S A. 90. Shi Y, Lan F, Matson C, Mulligan P, Whetstine JR, Cole PA, et al. 2005;102(44):15901–6. Histone demethylation mediated by the nuclear amine oxidase 74. Mahajan NP, Liu Y, Majumder S, Warren MR, Parker CE, Mohler homolog LSD1. Cell. 2004;119(7):941–53. JL, et al. Activated Cdc42-associated kinase Ack1 promotes pros- 91. Metzger E, Yin N, Wissmann M, Kunowska N, Fischer K, tate cancer progression via androgen receptor tyrosine phosphor- Friedrichs N, et al. Phosphorylation of histone H3 at threonine ylation. Proc Natl Acad Sci U S A. 2007;104(20):8438–43. 11 establishes a novel chromatin mark for transcriptional regula- 75. Nelson WG, De Marzo AM, Yegnasubramanian S. Epigenetic tion. Nat Cell Biol. 2008;10(1):53–60. alterations in human prostate cancers. Endocrinology. 92. Jia L, Berman BP, Jariwala U, Yan X, Cogan JP, Walters A, et al. 2009;150(9):3991–4002. Genomic androgen receptor-occupied regions with different 114 Curr Mol Bio Rep (2018) 4:101–115 functions, defined by histone acetylation, coregulators and tran- enhancers and superenhancers through BET bromodomain inhibition is a promising therapeutic option in cancer and scriptional capacity. PLoS One. 2008;3(11):10. 93. Jia L, Shen HC, Wantroba M, Khalid O, Liang G, Wang Q, et al. highlight the key role of this epigenetic master regulator in cancer progression Locus-wide chromatin remodeling and enhanced androgen receptor-mediated transcription in recurrent prostate tumor cells. 109. Jang MK, Mochizuki K, Zhou M, Jeong HS, Brady JN, Ozato K. Mol Cell Biol. 2006;26(19):7331–41. The bromodomain protein Brd4 is a positive regulatory compo- 94. Lasko LM, Jakob CG, Edalji RP, Qiu W, Montgomery D, nent of P-TEFb and stimulates RNA polymerase II-dependent Digiammarino EL, et al. Discovery of a selective catalytic p300/ transcription. Mol Cell. 2005;19(4):523–34. CBP inhibitor that targets lineage-specific tumours. Nature. 110. Zuber V, Bettella F, Witoelar A, Andreassen OA, Mills IG, 2017;550(7674):128–32. Urbanucci A. Bromodomain protein 4 discriminates tissue- 95.� Valdes-Mora F, Gould CM, Colino-Sanguino Y, Qu W, Song JZ, specific super-enhancers containing disease-specific susceptibility Taylor KM, et al. Acetylated histone variant H2A.Z is involved in loci in prostate and breast cancer. BMC Genomics. 2017;18(1): the activation of neo-enhancers in prostate cancer. Nat Commun. 2017;8(1):1346. This study highlights the important role of 111. Faivre EJ, Wilcox D, Lin X, Hessler P, Torrent M, He W, et al. histone acetylation in prostate cancer beyond active pro- Exploitation of castration-resistant prostate cancer transcription moters via activation of AR associated enhancers and the in- factor dependencies by the novel BET inhibitor ABBV-075. Mol crease in chromatin accessibility Cancer Res. 2017;15(1):35–44. 96. Ruscetti M, Dadashian EL, Guo W, Quach B, Mulholland DJ, 112. Asangani IA, Wilder-Romans K, Dommeti VL, Krishnamurthy Park JW, et al. HDAC inhibition impedes epithelial- PM, Apel IJ, Escara-Wilke J, et al. BET Bromodomain inhibitors mesenchymal plasticity and suppresses metastatic, castration- enhance efficacy and disrupt resistance to AR antagonists in the resistant prostate cancer. Oncogene. 2016;35(29):3781–95. treatment of prostate cancer. Mol Cancer Res. 2016;14(4):324–31. 97. Di Croce L, Helin K. Transcriptional regulation by polycomb 113. Arora VK, Schenkein E, Murali R, Subudhi SK, Wongvipat J, group proteins. Nat Struct Mol Biol. 2013;20(10):1147–55. Balbas MD, et al. Glucocorticoid receptor confers resistance to 98. Melling N, Thomsen E, Tsourlakis MC, Kluth M, Hube-Magg C, antiandrogens by bypassing androgen receptor blockade. Cell. Minner S, et al. Overexpression of enhancer of zeste homolog 2 2013;155(6):1309–22. (EZH2) characterizes an aggressive subset of prostate cancers and 114. Blee AM, Liu S, Wang L, Huang H. BET bromodomain-mediated predicts patient prognosis independently from pre- and postoper- interaction between ERG and BRD4 promotes prostate cancer cell atively assessed clinicopathological parameters. Carcinogenesis. invasion. Oncotarget. 2016;7(25):38319–32. 2015;36(11):1333–40. 115. Janouskova H, El Tekle G, Bellini E, Udeshi ND, Rinaldi A, 99. Lee TI, Jenner RG, Boyer LA, Guenther MG, Levine SS, Kumar Ulbricht A, et al. Opposing effects of cancer-type-specific SPOP RM, et al. Control of developmental regulators by Polycomb in mutants on BET protein degradation and sensitivity to BET inhib- human embryonic stem cells. Cell. 2006;125(2):301–13. itors. Nat Med. 2017;23(9):1046–54. 100. Liu L, Xu Z, Zhong L, Wang H, Jiang S, Long Q, et al. Enhancer 116. Zhang P, Wang D, Zhao Y, Ren S, Gao K, Ye Z, et al. Intrinsic of zeste homolog 2 (EZH2) promotes tumour cell migration and BET inhibitor resistance in SPOP-mutated prostate cancer is me- invasion via epigenetic repression of E-cadherin in renal cell car- diated by BET protein stabilization and AKT-mTORC1 activation. cinoma. BJU Int. 2016;117(2):351–62. Nat Med. 2017;23(9):1055–62. 101. Xu H, Xu K, He HH, Zang C, Chen CH, Chen Y, et al. Integrative 117. Gupta-Elera G, Garrett AR, Robison RA, O'Neill KL. The role of analysis reveals the transcriptional collaboration between EZH2 oxidative stress in prostate cancer. Eur J Cancer Prev. 2012;21(2): and E2F1 in the regulation of cancer-related gene expression. Mol 155–62. Cancer Res. 2016;14(2):163–72. 102. Cha TL, Zhou BP, Xia W, Wu Y, Yang CC, Chen CT, et al. Akt- 118. Hussong M, Borno ST, Kerick M, Wunderlich A, Franz A, mediated phosphorylation of EZH2 suppresses methylation of ly- Sultmann H, et al. The bromodomain protein BRD4 regulates sine 27 in histone H3. Science. 2005;310(5746):306–10. the KEAP1/NRF2-dependent oxidative stress response. Cell 103.�� Xu K, Wu ZJ, Groner AC, He HH, Cai C, Lis RT, et al. EZH2 Death Dis. 2014;5:e1195. oncogenic activity in castration-resistant prostate cancer cells is 119. Theurillat JP, Udeshi ND, Errington WJ, Svinkina T, Baca SC, Polycomb-independent. Science. 2012;338(6113):1465–9. This Pop M, et al. Prostate cancer. Ubiquitylome analysis identifies paper provides compelling evidence that the oncogenic role dysregulation of effector substrates in SPOP-mutant prostate can- of Ezh2 in CRPC is partly independent of its repressive func- cer. Science. 2014;346(6205):85–9. tion and acts via coactivator functions with key transcription 120. Kikuchi M, Okumura F, Tsukiyama T, Watanabe M, Miyajima N, factors including AR Tanaka J, et al. TRIM24 mediates ligand-dependent activation of 104. Labbe DP, Sweeney CJ, Brown M, Galbo P, Rosario S, Wadosky androgen receptor and is repressed by a bromodomain-containing KM, et al. TOP2A and EZH2 provide early detection of an ag- protein, BRD7, in prostate cancer cells. Biochim Biophys Acta. gressive prostate cancer subgroup. Clin Cancer Res. 2017;23(22): 2009;1793(12):1828–36. 7072–83. 121. Grasso CS, Wu YM, Robinson DR, Cao X, Dhanasekaran SM, 105. Urbanucci A, Mills IG. Bromodomain-containing proteins in Khan AP, et al. The mutational landscape of lethal castration- prostate cancer. Mol Cell Endocrinol. 2017; resistant prostate cancer. Nature. 2012;487(7406):239–43. 106. Fujisawa T, Filippakopoulos P. Functions of bromodomain- 122. Zhao D, Lu X, Wang G, Lan Z, Liao W, Li J, et al. Synthetic containing proteins and their roles in homeostasis and cancer. essentiality of chromatin remodelling factor CHD1 in PTEN- Nat Rev Mol Cell Biol. 2017;18(4):246–62. deficient cancer. Nature. 2017;542(7642):484–8. 107.�� Loven J, Hoke HA, Lin CY, Lau A, Orlando DA, Vakoc CR, et al. 123. Zaret KS, Carroll JS. Pioneer transcription factors: establishing Selective inhibition of tumor oncogenes by disruption of super- competence for gene expression. Genes Dev. 2011;25(21):2227– enhancers. Cell 2013; 153(2):320–334. 108.�� Whyte WA, Orlando DA, Hnisz D, Abraham BJ, Lin CY, Kagey 124. Lee CS, Friedman JR, Fulmer JT, Kaestner KH. The initiation of MH, et al. Master transcription factors and mediator establish liver development is dependent on Foxa transcription factors. super-enhancers at key cell identity genes. Cell. 2013;153(2): Nature. 2005;435(7044):944–7. 307–19. This two papers provide evidence that disruption of Curr Mol Bio Rep (2018) 4:101–115 115 125. Watt AJ, Zhao R, Li J, Duncan SA. Development of the mamma- 140. Antony L, van der Schoor F, Dalrymple SL, Isaacs JT. Androgen receptor (AR) suppresses normal human prostate epithelial cell lian liver and ventral pancreas is dependent on GATA4. BMC Dev Biol. 2007;7(1):37. proliferation via AR/beta-catenin/TCF-4 complex inhibition of c- 126. Zhao R, Watt AJ, Li J, Luebke-Wheeler J, Morrisey EE, Duncan MYC transcription. Prostate. 2014;74(11):1118–31. SA. GATA6 is essential for embryonic development of the liver 141. Chen H, Tu SW, Hsieh JT. Down-regulation of human DAB2IP but dispensable for early heart formation. Mol Cell Biol. gene expression mediated by polycomb Ezh2 complex and his- 2005;25(7):2622–31. tone deacetylase in prostate cancer. J Biol Chem. 2005;280(23): 127. Holtzinger A, Evans T. Gata4 regulates the formation of multiple 22437–44. organs. Development. 2005;132(17):4005–14. 142. Min J, Zaslavsky A, Fedele G, McLaughlin SK, Reczek EE, De 128. Wang Q, Li W, Zhang Y, Yuan X, Xu K, Yu J, et al. Androgen Raedt T, et al. An oncogene-tumor suppressor cascade drives met- receptor regulates a distinct transcription program in androgen- astatic prostate cancer by coordinately activating Ras and nuclear independent prostate cancer. Cell. 2009;138(2):245–56. factor-kappaB. Nat Med. 2010;16(3):286–94. 129. Wang D, Garcia-Bassets I, Benner C, Li W, Su X, Zhou Y, et al. 143. Aytes A, Mitrofanova A, Lefebvre C, Alvarez MJ, Castillo-Martin Reprogramming transcription by distinct classes of enhancers M, Zheng T, et al. Cross-species regulatory network analysis iden- functionally defined by eRNA. Nature. 2011;474(7351):390–4. tifies a synergistic interaction between FOXM1 and CENPF that 130. Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka T, Tomoda drives prostate cancer malignancy. Cancer Cell. 2014;25(5):638– K, et al. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell. 2007;131(5):861–72. 144. Aytes A, Mitrofanova A, Kinkade CW, Lefebvre C, Lei M, Phelan 131. Rybak AP, Tang D. SOX2 plays a critical role in EGFR-mediated V, et al. ETV4 promotes metastasis in response to activation of self-renewal of human prostate cancer stem-like cells. Cell Signal. PI3-kinase and Ras signaling in a mouse model of advanced pros- 2013;25(12):2734–42. tate cancer. Proc Natl Acad Sci U S A. 2013;110(37):E3506–15. 132.�� Bishop JL, Thaper D, Vahid S, Davies A, Ketola K, Kuruma H, 145. Mitrofanova A, Aytes A, Zou M, Shen MM, Abate-Shen C, et al. The master neural transcription factor BRN2 is an androgen Califano A. Predicting drug response in human prostate cancer receptor-suppressed driver of neuroendocrine differentiation in from preclinical analysis of in vivo mouse models. Cell Rep. prostate cancer. Cancer Discov. 2017;7(1):54–71. This works 2015;12(12):2060–71. provides a mechanistic model for neuroendocrine differentia- 146. Kar S, Sengupta D, Deb M, Pradhan N, Patra SK. SOX2 function tion after AR signaling blockade using patiend-derived models and Hedgehog signaling pathway are co-conspirators in promot- 133. Zeng W, Sun H, Meng F, Liu Z, Xiong J, Zhou S, et al. Nuclear C- ing androgen independent prostate cancer. Biochim Biophys Acta. MYC expression level is associated with disease progression and 2017;1863(1):253–65. potentially predictive of two year overall survival in prostate can- 147. Ding L, Chen S, Liu P, Pan Y, Zhong J, Regan KM, et al. CBP loss cer. Int J Clin Exp Pathol. 2015;8(2):1878–88. cooperates with PTEN haploinsufficiency to drive prostate cancer: 134. Rye M, Bertilsson H, Drabløs F, Angelsen A, Bathen T, Tessem implications for epigenetic therapy. Cancer Res. 2014;74(7): M-B. Gene signatures ESC, MYC and ERG-fusion are early 2050–61. markers of a potentially dangerous subtype of prostate can- 148. Li D, Sun H, Sun WJ, Bao HB, Si SH, Fan JL, et al. Role of cer2014. 50 p. RbBP5 and H3K4me3 in the vicinity of Snail transcription start 135. Kim J, Roh M, Doubinskaia I, Algarroba GN, Eltoum IE, site during epithelial-mesenchymal transition in prostate cancer Abdulkadir SA. A mouse model of heterogeneous, c-MYC- cell. Oncotarget. 2016;7(40):65553–67. initiated prostate cancer with loss of Pten and p53. Oncogene. 149. Dong C, Wu Y, Yao J, Wang Y, Yu Y, Rychahou PG, et al. G9a 2012;31(3):322–32. interacts with Snail and is critical for Snail-mediated E-cadherin 136. Hubbard GK, Mutton LN, Khalili M, McMullin RP, Hicks JL, repression in human breast cancer. J Clin Invest. 2012;122(4): Bianchi-Frias D, et al. Combined MYC activation and Pten loss 1469–86. are sufficient to create genomic instability and lethal metastatic 150. Liu S, Ye D, Guo W, Yu W, He Y, Hu J, et al. G9a is essential for prostate cancer. Cancer Res. 2016;76(2):283–92. EMT-mediated metastasis and maintenance of cancer stem cell- 137. Koh CM, Iwata T, Zheng Q, Bethel C, Yegnasubramanian S, De like characters in head and neck squamous cell carcinoma. Marzo AM. Myc enforces overexpression of EZH2 in early pros- Oncotarget. 2015;6(9):6887–901. tatic neoplasia via transcriptional and post-transcriptional mecha- 151. Liu Y, Zeng S, Jiang X, Lai D, Su Z. SOX4 induces tumor inva- nisms. Oncotarget. 2011;2(9):669–83. sion by targeting EMT-related pathway in prostate cancer. Tumour 138. Fan L, Peng G, Sahgal N, Fazli L, Gleave M, Zhang Y, et al. Biol. 2017;39(5):1010428317694539. Regulation of c-Myc expression by the histone demethylase 152. Bilir B, Osunkoya AO, WGt W, Sannigrahi S, Lefebvre V, JMJD1A is essential for prostate cancer cell growth and survival. Metzger D, et al. SOX4 is essential for prostate tumorigenesis Oncogene. 2016;35(19):2441–52. initiated by PTEN ablation. Cancer Res. 2016;76(5):1112–21. 139. Vander Griend DJ, Litvinov IV, Isaacs JT. Conversion of androgen receptor signaling from a growth suppressor in normal prostate 153. Mahajan K, Malla P, Lawrence HR, Chen Z, Kumar-Sinha C, epithelial cells to an oncogene in prostate cancer cells involves a Malik R, et al. ACK1/TNK2 regulates histone H4 Tyr88- gain of function in c-Myc regulation. Int J Biol Sci. 2014;10(6): phosphorylation and AR gene expression in castration-resistant 627–42. prostate cancer. Cancer Cell. 2017;31(6):790–803.e8. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Current Molecular Biology Reports Springer Journals

Epigenetic Regulation in Prostate Cancer Progression

Free
15 pages
Loading next page...
 
/lp/springer_journal/epigenetic-regulation-in-prostate-cancer-progression-UuWvOnyYJV
Publisher
Springer International Publishing
Copyright
Copyright © 2018 by The Author(s)
Subject
Biomedicine; Molecular Medicine; Human Genetics
eISSN
2198-6428
D.O.I.
10.1007/s40610-018-0095-9
Publisher site
See Article on Publisher Site

Abstract

Purpose of Review An important number of newly identified molecular alterations in prostate cancer affect gene encoding master regulators of chromatin biology epigenetic regulation. This review will provide an updated view of the key epigenetic mecha- nisms underlying prostate cancer progression, therapy resistance, and potential actionable mechanisms and biomarkers. Recent Findings Key players in chromatin biology and epigenetic master regulators has been recently described to be crucially altered in metastatic CRPC and tumors that progress to AR independency. As such, epigenetic dysregulation represents a driving mechanism in the reprograming of prostate cancer cells as they lose AR-imposed identity. Summary Chromatin integrity and accessibility for transcriptional regulation are key features altered in cancer progression, and particularly relevant in nuclear hormone receptor-driven tumors like prostate cancer. Understanding how chromatin remodeling dictates prostate development and how its deregulation contributes to prostate cancer onset and progression may improve risk stratification and treatment selection for prostate cancer patients. . . . . . Keywords Prostate cancer Epigenetics Transcriptional regulation Chromatin biology Androgen receptor Drug targets Introduction patients, who develop castration-resistant prostate tumors (CRPC) for which limited treatment options exist. Moreover, Prostate cancer has traditionally been seen as an aging-asso- under the CRPC definition, a pool of diverse disease presen- ciated, low mutational load tumor with a tendency for geno- tations with variable outcomes exists, including neuroendo- mic rearrangements and a particular dependency on the activ- crine tumors. ity of the androgen receptor (AR). As such, treatment strate- Massive parallel sequencing of hundreds of tumor speci- gies have been focused on targeting the AR axis, either mens from prostate cancer patients at different stages of cancer through inhibiting steroidogenic pathways and the production progression has provided an accurate picture of the landscape of testosterone, or by antagonizing the AR itself to prevent its of genetic alterations that accompany cancer evolution in the nuclear translocation and the activation of its transcriptional prostate. Yet, despite several molecular classification systems network. While these strategies have doubtlessly improved for prostate tumors have been proposed, clear association with survival for prostate cancer patients, they are not curative in risk stratification remains to be provided. On the other hand, many cases, and resistance eventually occurs in about 30% of whether these genetic classifiers predict treatment outcome and to what extent genetic alterations in prostate cancer can be exploited for personalized therapies is yet to be proven. This article is part of the Topical Collection on Molecular Biology of Interestingly, together with well-known drivers of cancer pro- Prostate Cancer gression, an important number of new alterations have been described, with an intriguing enrichment of those affecting * Alvaro Aytes aaytes@idibell.cat key players in chromatin biology and epigenetic master regu- lators (see a summary in Table 1). This is particularly relevant Programs of Molecular Mechanisms and Experimental Therapeutics in metastatic CRPC and tumors that have transitioned to AR- in Oncology (ONCOBell), Catalan Institute of Oncology, Bellvitge independent phenotypes after progressing on the newest Institute for Biomedical Research, Granvia de l’Hopitalet, 199 antiandrogen drugs. 08908, L’Hospitalet de Llobregat, 08907 Barcelona, Spain 2 Here, we introduce key concepts to understand how epige- Programs of Cancer Therapeutics Resistance (ProCURE), Catalan netic dysregulation is a plausible driving mechanism in the Institute of Oncology, Bellvitge Institute for Biomedical Research, L’Hospitalet de Llobregat, 08907 Barcelona, Spain 102 Curr Mol Bio Rep (2018) 4:101–115 Table 1 Summary of epigenetic Gene name Function in prostate cancer References master regulators implicated in prostate cancer Methyltransferases NSD2 H3K36 di-methyltransferase. Promotes prostate cancer [1�� , 2, 3] tumorigenesis and progression. It is overexpressed in metastatic stage and associated with biochemical recurrence EZH2 H3K27 di- and tri-methyltransferase. Member of the polycomb [4, 5] repressive complex 2, crucial driver of prostate oncogenesis SUV39H1 (KMT1A) H3K9 tri methyltransferase. Enhance prostate cancer cell [6, 7] migration and invasion SETDB1 (KMT1E) SUV39H2 (KMT1B) H3K9 tri methyltransferase increases androgen-dependent tran- [8] scriptional activity by interacting with the AR SMYD3 H3K4 di- and methyltransferase, promotes cell proliferation and [9, 10]. migration PRMT5 Drives prostate cancer cell growth through epigenetic [11, 12� , inactivation of several tumor suppressors through histone 13] arginine methylation at H4R3. Enhances AR-targeted gene expression Demethylases LSD1 H3K9 and H3K4 demethylase involved in prostate cancer [14�� , 15, recurrence, CRPC, and poor survival. Regulates AR 16� ] transcriptional activity in a context-dependent manner JARID1B (KDM5B) H3K4 mono, di-, and tri-demethylase. AR coactivator regulating [17, 18], its transcriptional activity. Upregulated in prostate cancer tissues JARID1C (KDM5C) H3K4 di- and tri-demethylase overexpressed in prostate cancer. [19]. Proposed as a predictive marker for therapy failure in patients after prostatectomy JARID1D (KMD5D) H3K4 di- and tri-demethylase. Suppress invasion and progres- [20, 21] sion of prostate cancer. Low levels were associated with poor prognosis and resistance to docetaxel PHF8 H3K9, H3K27, and H4K20 demethylase. Transcriptional [22–29] coactivator of AR. Promotes prostate cancer cell proliferation, migration, invasion, and neuroendocrine differentiation. Its expression highly correlated with poor prognosis and is induced by hypoxia JMJD2A (KDM4A) H3K9 and H3K36 tri demethylases. Modulates AR [30, 31] JMJD2C (KDM4C) transcriptional activity stimulating ligand-independent gene transcription via H3K9 demethylation JMJD1A (KDM3A) H3K9 mono- and di-demethylase. Regulates AR activity by re- [32, 33] cruitment to target genes only in the presence of androgens JMJD2B (KDM4B), H3K9 tri-demethylase, AR coactivator. Regulates AR [34] transcriptional activity via demethylation activity and via inhibition of ubiquitination and increased AR stability JMJD3 (KDM6B) H3K27 di- and tri-demethylase overexpressed in metastatic [35]. prostate cancer DNA methylation DNMTs Control of transcriptional program during prostate cancer and [36] CRPC progression GSTP1 Silencing of GSP1 upon promoter hypermethylation is a [37–39] potential prognostic biomarker and occurs early during prostate carcinogenesis Histone acetylation P300 Histone acetyltransferase. Besides canonical histone acetylation [40, 41] activity, it acetylates the AR and enhances its transcriptional activity (coactivator) and drives prostate cancer growth SIRT1 Histone deacetylase; regulates cellular growth through AR [42, 43]. deacetylation SIRT2 Histone deacetylase; its downregulation has been associated with [44] increased acetylated H3K18 and poorer outcome and decreased sensitivity to androgen deprivation therapy BET bromodomain epigenetic readers BRD4 Curr Mol Bio Rep (2018) 4:101–115 103 Table 1 (continued) Gene name Function in prostate cancer References Bromodomain and extra-terminal protein, interacts with AR and [45� , promote its activity and antiandrogen resistance 46–48] TRIM24 Epigenetic reader and transcription co-regulator, overexpressed [49]. in CRPC and associated to disease recurrence. Required for prostate cancer cell proliferation in CRPC CHD1 H3K4me2-3 epigenetic reader whose loss is related with prostate [50, 51] cancer aggressiveness and DNA repair defects, thus sensitizing tumor cells to PARP inhibitors Pioneer transcription factors FOXA1 FOXA1 activity on chromatin results in increased accessibility [52, 53] and increased chromatin-bound AR. High FOXA1 expression leads to a restricted AR cistrome regulation. FOXA1 also has the potential to reprogram GATA2 GATA2 GATA2 activity in human prostate cancer is strongly associated [53–55] to AR levels and is hence considered a prostate cancer oncogene Epigenetic regulators of lineage plasticity SOX2 Overexpressed TF in prostate cancer, regulating CRPC [56–61, proliferation, and evasion of apoptosis. Promotes tumor 62�� , metastasis by inducing EMT. Associated to NEPC emergence 63�� ] MYC Master regulator of prostate cancer transcriptional program. [64, 65] Associated with prostate cancer recurrence and poor prognosis MYCN Driver of NEPC by inducing an EZH2-mediated transcriptional [64, 66] program Oncogenic pathways Hsp90 Initiates ERK signaling and leads to the recruitment of EZH2 to [67]. the E-cadherin promoter and repression of E-cadherin expression, driving EMT and invasion in prostate cancer cells DAB2IP Tumor suppressor Ras-GAP. Negatively controls Ras-dependent [68–70]. mitogenic signals and modulates TNFα/NF-κB, WNT/β-catenin, PI3K/AKT, and androgen receptors path- ways RB1 This tumor suppressor gene is commonly loss in metastatic and [71, 72, antiandrogen resistant prostate cancer and NEPC. Directly 63�� ] repress the expression of Sox2 and Ezh2 ACK1 Tyrosine kinase correlated with poor prognosis and interacts with [72–74] AR to drive ADT resistance and CRPC growth. Regulates transcription of AR and AR-v7 via epigenetic regulation reprograming of prostate cancer cells as they lose AR- numerous other key genes have been implicated in DNA imposed identity. Beyond reviewing the current status of epi- methylation changes. In fact, the promoter of the Androgen genetic biomarkers and classifiers and their clinical impact, Receptor (AR) itself appears to be hypermethylated in up to we will discuss the scientific basis for therapeutic targeting 30% of CRPCs, resulting in the loss of AR expression [76]. master regulators of chromatin remodeling and integrity and Moreover, PTEN silencing is often a consequence of promoter the current state of epigenetic drugs for prostate cancer. CpG islands hypermethylation [77], while hypermethylation of the p16 tumor suppressor gene is associated with a prolif- erative advantage, thus contributing to carcinogenesis and dis- DNA Methylation and Histone Modifications ease progression [78]. Similarly, the hypomethylation and in Prostate Carcinogenesis consequent upregulation of genes like heparanase and uroki- nase plasminogen activator (uPA) was reported to contribute Perturbed DNA methylation patterns have long been reported to tumor cell invasion and metastasis [79]. More globally, during prostate cancer progression [75]. Among the most DNA methylation signatures have been identified and pro- well-described alterations is the GSTP1 promoter hyperme- posed as molecular biomarkers of prostate cancer progression thylation and subsequent silencing [37], which is thought to and treatment response [80]. occur early during prostate carcinogenesis [38] and has thus Histone modifications also play an important role in the been proposed as a potential prognostic biomarker [39]. Yet, progression of many tumor types including prostate cancer. 104 Curr Mol Bio Rep (2018) 4:101–115 Lysine methyltransferases (KMT) and demethylases (KDM) progression of prostate cancer cells; thus, it is highly down- are important epigenetic histone modifiers implicated in the regulated in metastatic prostate tumors and those low levels control of gene transcriptional regulation as well as in non- were associated with poor prognosis [20]. In addition, KDM5 histone protein posttranslational modifications and activity loss has been associated with resistance to docetaxel in pros- modulation [81]. More specifically, SUV39H1 (KMT1A) tate cancer [21]. The PHD-finger protein 8 (PHF8) is a histone and SETDB1 (KMT1E) have been shown to enhance prostate demethylase and a transcriptional coactivator of AR via cancer cell migration and invasion and to be upregulated in H4K20 demethylation [28]. Its expression, highly correlated human prostate cancer specimens, and hence suggested as with poor prognosis, is induced by hypoxia and promotes potential therapeutic targets [6], while SUV39H2 (KMT1B) prostate cancer cell proliferation, migration and invasion interacts with the AR to increase androgen-dependent tran- [28], and neuroendocrine differentiation [29]. scriptional activity [8]. Furthermore, levels of SETDB1 have been recently associated with prognosis and the development The Histone Methyltransferase NSD2 of bone metastases from prostate cancer [7]. Similarly, SET and MYND domain-containing protein 3 (SMYD3) has also NSD2 (nuclear receptor binding SET domain protein 2), also been identified as an upregulated H3 and H4 lysine methyl- known as WHSC1 (Wolf-Hirschhorn syndrome candidate 1) transferase promoting cell proliferation and migration, thus and MMSET (multiple myeloma SET domain), is a member emerging as a predictive marker of prostate cancer [10]. of the histone methyltransferase NSD family of proteins also Alternatively, protein arginine methyltransferase 5 (PRMT5) including NSD1 and NSD3. NSD2 catalyzes the was described as a prostate cancer oncogene driving cancer dimethylation of histone H3 at lysine 36 (H3K36me2), a per- cell growth through epigenetic inactivation of several tumor missive mark associated with open chromation conformation suppressors [11] through histone arginine methylation at andactivegenetranscription [85]. NSD2 was first linked to H4R3. PRMT5 has also recently been shown to enhance oncogenesis by the involvement in the t(4; 14) translocation AR-targeted gene expression by arginine methylation and in- identified in up to 20% of multiple myeloma patients [86]. In teraction with the transcription factor Sp1 [13]. the past years, NSD2 has been shown to be overexpressed in a Demethylases also play an important role in prostate cancer variety of solid tumors including prostate cancer, where it has development. Lysine-specific demethylase 1 been found overexpressed in metastatic PCa compared to pri- (LSD1/KDM1A) has been proposed as an oncogene whose mary tumors and is associated with biochemical recurrence overexpression has been positively correlated with the malig- [1�� ]. Further In vitro studies strengthened the role of NSD2 nancy of many cancer types, including prostate [14�� , 82], in prostate cancer tumorigenesis; it has been shown that NSD2 promoting carcinogenesis by multiple mechanisms. modulates Twist family bHLH transcription factor 1 Increased LSD1 expression is associated with prostate cancer (TWIST1) to promote epithelial to mesenchymal transition recurrence and poor survival and appears to have distinct and invasiveness in prostate cancer cell lines [2]. Moreover, functions in androgen-dependent [14�� , 83] and refractory Asangani and colleagues had reported that EZH2 mediates the prostate cancer [15]. Recently, it was discovered that LSD1 overexpression of NSD2 and that the oncogenic properties of is a co-regulator of vitamin D receptor activity in prostate EZH2 are NSD2 dependent [3]. Interestingly, transcriptional cancer and its expression is correlated with shorter targets of NSD2 in prostate cancer cells are highly enriched progression-free survival in primary and metastatic patients for components of the NF-kB-network, including IL-6, IL-8, [84]. In a recent study, it was found that LSD1-mediated epi- survivin/Birc5, and VEGFA. In fact, NSD2 has been linked to genetic reprogramming drives CRPC and was associated with constitutive activation of NF-kB signaling in CRPC, promot- the activation of CENPE, which was regulated by the co- ing cancer cell proliferation and survival via an autocrine pos- binding of LSD1 and AR to its promoter region, which was itive loop in which NSD2 expression is in turn stimulated by associated with loss of RB1 [16� ]. inflammatory cytokines, such as TNFα and IL-6, via NF-kB The overexpression of other histone demethylases (HDMs) [87]. has also been observed in prostate cancer. An exhaustive func- Very recently, work from Li and collaborators showed that tional screen [27] identified 32 enzymes belonging to the fam- NSD2 is activated in PTEN null tumors by the AKT pathway ily of JmjC domain-containing histone demethylases as criti- and that its expression is required for metastatic progression. cal for prostate cancer proliferation and survival. KDM5 fam- Mechanistically, AKT-mediated phosphorylation of NSD2 Cdt2 ily members are H3K4 demethylases; JARID1B (KDM5B) is prevents its degradation by CRL4 E3 ligase leading to upregulated in prostate cancer tissues and acts as an AR coac- NSD2 stabilization and overexpression. By directly inducing tivator [17], while JARID1C (KDM5C), overexpressed in RICTOR expression, NSD2 mediates a positive feedback prostate cancer, emerged as a predictive marker for therapy loop sustaining AKT signaling [1�� ]. failure in patients after prostatectomy [19]. JARID1D Finally, NSD2 has been shown to physically interact with (KMD5D) was found to suppress the invasion and the AR DNA-binding domain and to be recruited to the Curr Mol Bio Rep (2018) 4:101–115 105 enhancer region of the PSA gene and enhance AR transcrip- ubiquitination and increased AR stability [34]. Finally, JMJD3 tional activity [88], suggesting that NSD2 might be implicated (KDM6B) is progressively overexpressed in metastatic pros- in resistant to ADTor androgen signaling inhibition. Of note is tate cancer [35]. the recent identification of NSD2 as a candidate gene promot- ing androgen independence through an unbiased insertional Histone Acetylation and AR mutagenesis screen [89]. In fact, unpublished data and data from our laboratory currently under peer-review for publica- Acetylated chromatin is generally associated to active tran- tion strongly suggest that NSD2 is an actionable mechanism scription and the enzymes regulating this process are histone in CRPC. acetyltransferases (HAT) and deacetylases (HDAC). Accordingly, acetylated histone H3 in the vicinity of AR- bound chromatin has been shown to reduce androgen depen- Epigenetic Control of Androgen Receptor dence in castration resistance models [92, 93]. That is the case Activity for canonical HAT like p300 and CREB-binding protein, which, besides canonical histone acetylation activities, have Histone modifying enzymes, and LSD1 in particular, are been shown to acetylate the AR and enhance its transcriptional among the best-known modulators of AR transcriptional ac- activity [40]. Importantly, two groups have recently indepen- tivity. LSD1 is an important enzyme involved in AR regula- dently developed small molecule inhibitors targeting tion and prostate cancer that interacts with AR and can stim- p300/CBP. Lasko and colleagues reported a selective catalytic ulate [14�� ] or suppress [15] the transcriptional expression p300/CBP inhibitor able to downregulate the AR transcrip- depending on promoter/enhancer context. This interaction tional program both in castration-sensitive and castration- promotes ligand-dependent transcription of AR target genes, resistant prostate tumors and to inhibit tumor growth in resulting in enhanced tumor cell growth. Its coactivator activ- CRPC xenograft models [94], while Jin and colleagues found ity seems to be associated with H3K9me1,2 demethylation that targeting the p300/CBP bromodomain had remarkably leading to transcriptional de-repression of AR target genes similar effetcs [41]. More broadly, a recent study highlights [14 ]. Intriguingly, LSD1 also plays a role as co-repressor, the important role of histone acetylation in prostate cancer via H3K4me1,2 demethylation [90] and the recruitment of co- beyond active promoters via activation of AR associated en- repressor complexes. This highlights the dual role of many hancers and the increase in chromatin accessibility [95� ]. chromatin remodelers and may explain why translating them Conversely, a variety of HDACs are also capable of to new therapeutics has so far been limited. A possible way deacetylating the AR and inhibit its activity, for example via forward may be to define the context specificities for this regulation of heat-shock protein 90 (Hsp90), a chaperone con- duality. For example, it has been shown that in high androgen trolling AR nuclear localization and activation through its levels, AR recruits LSD1 to mediate AR gene silencing [15]; acetylation/deacetylation, or sirtuin 1 (SIRT1), which regu- however, this negative feedback loop is apparently disrupted lates cellular growth through AR deacetylation [42, 43]. In in CRPC, where low androgen levels promote AR overex- fact, acetylation of H3K18, putatively via downregulation of pression. Additionally, post-transcriptional modifications can SIRT2 deacetylase, has been associated to poorer outcome regulate LSD1 activity and may become better targets; LSD1 and decreased sensitivity to androgen deprivation therapy phosphorylation [91] results in a switch of substrate from (ADT). Finally, at the mechanistic level, the Wu lab has re- H3K4me1,2 to H3K9me1,2, and the promotion of its coacti- cently demonstrated that HDAC inhibitors can suppress vator activity. Jumonji C domain-containing trimethyl lysine HMGA-driven EMT, reduce tumor growth and metastasis demethylases JMJD2A (KDM4A) and JMJD2C (KDM4C) and, importantly, resensitize prostate cancer cells to [96]. also play a significant role in modulating AR transcriptional activity [30, 31], stimulating ligand-independent gene tran- scription via H3K9 demethylation. On the contrary, The Role of EZH2/Polycomb Repressive JMJD1A (KDM3A) recruitment to target genes only occurs Complex in Prostate Cancer in the presence of androgens, regulating AR activity and iden- tifying KDM3A-dependent genes involved in androgen re- The enhancer of zeste homolog 2 (EZH2) is a critical member sponse, hypoxia, glycolysis, and lipid metabolism [33], again of the Polycomb Repressive Complex 2 (PRC2) that regulates evidencing the complex balance between chromatin modify- histone methylation mainly via lysine 27 at histone H3 ing enzymes in controlling different but interconnected cellu- (H3K27), a modification associated to transcriptional silenc- lar processes. Of note is the case of JMJD2B (KDM4B), ing [97] that is found upregulated in many tumor types. In which is an AR coactivator, emerging as a suitable therapeutic prostate cancer, its elevated expression associates with poorer target for the treatment of prostate cancer. JMJD2B controls outcomes and has therefore been proposed as an oncogene [4, AR transcriptional activity via demethylation and inhibition of 98]. Amajor functionofEZH2istorepresslineage- �� 106 Curr Mol Bio Rep (2018) 4:101–115 specifying factors, thereby promoting stemness features [99], finding by Zuber and colleagues with implications in risk epithelial-mesenchymal transition (EMT), and ultimately met- assessment shows that tissue-specific SNPs in super- astatic progression [100]. A wealth of recent evidence has enhancer sequence bound by BRD4 are significantly associ- confirmed these previous observation in the prostate cancer ated with increased prostate cancer risk and show better en- field. Back-to-back recent articles in Science by the Sawyers richment for risk loci than AR [110]. and Goodrich groups demonstrated that lineage plasticity and BRD4 physically interacts with high-affinity with the N- neuroendocrine differentiation in androgen independence is terminal domain of AR leading to AR translocation into the partly driven by Ezh2 and Sox2 in prostate cancer mouse nucleus and AR recruitment to target loci, promoting AR ac- models carrying loss of function alleles for p53 and Rb tumor tivity and expression of AR target genes in CRPC [45� ]. A suppressors [62�� , 63�� ]. This came to confirm two previous recent study showed that the small molecule BET inhibitor reports by Dardenne and colleagues [64] and by Xu and col- ABBV-075 could disrupt the recruitment of BRD4 at enhanc- leagues [101] showing that N-myc induces EZH2-driven neu- er of AR target genes and repress their expression, whithout roendocrine prostate cancer [64] and it cooperates with E2F1 affecting AR protein levels [111]. Moreover, BET proteins in castration resistance [101]. have a role in resistance to antiandrogens and BET inhibitors Yet, EZH2 has also PRC2-independent roles as coactivator can effectively resensitize resistant tumors to enzalutamide of transcription factors, including an AKT-dependent methyl- [112]. One of these mechanisms of resistance to antiandrogens ation of the AR, via PI3K/AKT phosphorylation of EZH2 at is the upregulation of the glucocorticoid receptor (GR), and serine 21 [102], and modulation of AR recruitment to its target the co-option of the AR regulon, thus favoring CRPC progres- sites [103�� ]. Not surprisingly, EZH2 inhibitors are the focus sion by overcoming AR dependency [46, 47, 113]. of intensive development and have been widely tested in vivo Beyond AR signaling, BRD4 has been shown to bind to the [5] and in clinical trials (see Table 2 for details). Beyond a truncated ERG (ERGΔ39) encoded by the TMPRSS2-ERG promising drug target, EZH2 and TOP2A have been proposed fusion, co-regulating the expression of ERG target genes in as prognostic as well as predictive biomarkers of treatment CRPC, thereby stimulating cell growth and invasion [114]. response against EZH2 inhibitors [104]. Additionally, SPOP, an E3 ligase substrate binding protein frequently mutated in prostate cancer, was also reported to target BET proteins for ubiquitination-mediated degradation. Bromodomain-Containing Proteins Interestingly, SPOP mutants fail to ubiquitinate BET proteins, in Prostate Cancer leading to their stabilization and to resistance to BET inhibi- tors [48, 115]. This mechanism of resistance causes activation Bromodomain-containing proteins are chromatin readers that of AKT-mTORC1 signaling and consequently resistance to BET inhibitors can be overcome by combination with AKT recognized mono-acetylated histones and trigger chromatin remodeling to initiate transcription. Mutations and deregula- inhibitors [116]. tion of BRD-containing proteins is a common feature of a It is well known that one of the major aging-associated variety of cancers. More than 50% of primary and metastatic drivers of prostate carcinogenesis is oxidative stress and its prostate tumors and more than 70% of neuroendocrine pros- impact on DNA [117]. Interestingly, Hussong and colleagues tate cancer present genomic alterations in any of the 42 known have recently established a link between BRD4 and oxidative BRD-containing proteins [105]. Further, BRD-containing stress response genes in prostate cancer, such as the KEAP1/ proteins have a diversity of catalytic and scaffolding functions NRF2 axis and HMOX1, and reactive oxygen species (ROS) and may act as transcription factors, transcriptional co-factors production [118]. recruiting other proteins in the transcriptional complex, meth- Other than BET, several BRD-containing proteins have yltransferases, HATs, Helicases, and ATP-dependent chroma- been associated to prostate cancer progression and are at dif- tin remodelers, therefore playing a central role in gene expres- ferent validation stages for therapeutic targets in mCRPC. sion regulation [106]. TRIM24, tripartite motif-containing protein 24, is an epige- The subgroup of BET proteins (bromodomain and extra- netic reader and transcription co-regulator overexpressed in terminal), and in particular BRD4, have been the best charac- CRPC and associated to disease recurrence. Recurrent SPOP terized in prostate cancer, and several inhibitors of BET mutants stabilize TRIM24 [119], enhancing AR signaling and bromodomains have been developed and are currently in clin- promoting tumor growth via binding with the proteins TIP60 ical trial (see Table 2). The conserved BET family includes and BRD7 [120], which has led to the proposition of TRIM24 BRD4, BRD2, BRD3, expressed ubiquitously, and BRDT, as an essential gene for prostate cancer cell proliferation in specifically expressed in the testis. BRD4 recognizes acetylat- CRPC [49]. ed lysines at enhancers/superenhancer [107�� , 108�� ] and re- Finally, the role of chromodomain proteins, and in partic- cruits the elongation factor P-TEFb and stimulates RNA po- ular chromodomain helicase DNA-binding protein 1 (CHD1), lymerase II-dependent transcription [109]. A provocative new has in the recent years been elucidated in the context of Curr Mol Bio Rep (2018) 4:101–115 107 Table 2 Clinical trials for epigenetic drugs including prostate cancer patients Trial ID Drug Phase Conditions Patients Status BET bromodomain inhibitors NCT02259114 OTX015/MK-8628 I NUT midline carcinoma, triple negative 47 Completed breast cancer, non-small cell lung cancer (rearranged ALK or mut KRAS), CPRC, pancreatic ductal adenocarcinoma NCT02698176 I NUT midline carcinoma, triple negative breast 13 Terminated cancer, non-small cell lung cancer, CRPC NCT01987362 I Solid Tumors 120 Active NCT02711956 ZEN003694 I Metastatic CRPC (+enzalutamide) 58 Recruiting NCT02705469 I Metastatic CRPC 44 Active NCT03266159 GSK525762 II Solid tumors 150 Not recruiting NCT02419417 BMS-986158 I/II Advanced solid tumors 150 Recruiting NCT02391480 ABBV-075 I Advanced cancer, breast cancer, non-small, 150 Recruiting ell lung cancer, acute myeloid leukemia, multiple myeloma, prostate cancer, small-cell lung cancer, non-Hodgkins lymphoma NCT02630251 GSK2820151 I Advanced or recurrent solid tumors 60 Recruiting NCT02369029 BAY 1238097 I Neoplasms 8 Terminated NCT02431260 INCB054329 I/II Advanced cancer 69 Active, not recruiting NCT02711137 INCB057643 I/II Advanced cancer 230 Recruiting NCT02607228 GS-5829 I/II Metastatic CRPC (+enzalutamide) 132 Recruiting NCT02711137 INCB057643 I/II Advanced solid tumors and hematologic 420 Recruiting malignancy (+abiraterone) EZH2 and PRC1/2 inhibitors NCT03213665 Tazemetostat II Advanced solid tumors, non-Hodgkin 49 Recruiting lymphoma, or histiocytic (EZH2, SMARCB1, or SMARCA4 mutations) NCT01897571 I/II Advanced solid tumors 420 Recruiting NCT02875548 II Advanced solid tumors 300 Recruiting NCT03217253 I Metastatic malignant solid neoplasm 48 Not recruiting PRMT5 inhibitor NCT02900651 MAK683 I/II Diffuse large B cell lymphoma, 113 Recruiting advanced solid tumors LSD1/KDM1A inhibitors NCT02712905 INCB059872 I/II Advanced cancer 180 Recruiting DNMT inhibitors NCT01118741 Disulfiram Prostate cancer 19 Completed NCT00503984 Azacitidine I/II Metastatic CRPC (+docetaxel, prednisone) 22 Terminated NCT00384839 II CRPC 53 Completed NCT02998567 Guadecitabine I Non-small cell lung cancer, CRPC 35 Not yet recruiting (+pembrolizumab) HDAC inhibitors NCT01075308 Pracinostat (SB939) II Metastatic CRPC 32 Completed NCT00670553 I Prostate cancer, head and neck 7 Completed cancer, esophageal cancer NCT00878436 Panobinostat (LBH589) I/II CRPC (+bicalutamide) 52 Completed NCT00667862 II Metastatic CRPC 35 Completed NCT00663832 I CRPC (+docetaxel and prednisone) 44 Completed NCT00493766 I CRPC (+docetaxel and prednisone) 16 Terminated NCT00419536 I CRPC (+docetaxel and prednisone) 108 Terminated NCT00330161 Vorinostat (SAHA, MK0683) II Metastatic CRPC 29 Completed NCT01174199 I Metastatic CRPC 13 Terminated NCT00589472 II Primary prostate cancer (+bicalutamide.) 19 Completed 108 Curr Mol Bio Rep (2018) 4:101–115 Table 2 (continued) Trial ID Drug Phase Conditions Patients Status NCT00565227 I Non-small-cell lung carcinoma, prostate 12 Terminated cancer, bladder cancer, urothelial carcinoma NCT00511576 Mocetinostat (MGCD0103) I Breast cancer, lung cancer, prostate cancer, 54 Terminated gastric cancer (+docetaxel) NCT00020579 Entinostat (MS-275) I Advanced solid tumors, lymphoma 75 Completed NCT00413075 Belinostat (PXD101) I Advanced solid tumors, lymphoma 121 Completed NCT00413322 I Advanced solid tumors (+5-fluorouracil) 35 Completed prostate cancer progression. This H3K4me2-3 epigenetic to be quite different. GATA2 depletion did not seem to reader has been reported mutated in 43% of Gleason 7 have a reprogramming effect on AR binding sites and in or higher prostate cancer tumors, associated with ETS fact correlated with a downregulation in AR expression. gene fusion negative status [121] and its loss related Accordingly, GATA2 activity in human prostate cancer with prostate cancer aggressiveness [50] and DNA re- is strongly associated to AR levels and is hence consid- pair defects, hence sensitizing tumor cells to PARP in- ered a prostate cancer oncogene. Provocatively, it was hibitors [51]. More recently, Zhao and colleagues at the found that FOXA1 also has the potential to reprogram DePinho laboratory have demonstrated in PTEN null GATA2 and act as a pioneering effect for both AR and GATA2, suggesting that FOXA1 regulates a transcrip- prostate tumors that CHD1 depletion dramatically sup- pressed cell proliferation, survival, and tumorigenic po- tional network that controls AR-mediated gene expres- tential by activating the pro-tumorigenic TNF-NF-κB sion in prostate cancer [53]. gene network [122]. Lineage Plasticity in Prostate Cancer Stem Pioneer Factors in Prostate Cancer Cells Progression Aside from their ability to induce pluripotency, the Yamanaka Different from other DNA bound proteins and tran- factors (OCT4, SOX2, KLF4, and c-MYC) [130], and other scription factors, pioneer factors can access their targets reprograming factors like NANOG or LIN28, have been in nucleosomes and in highly compacted chromatin re- widely implicated in tumorigenesis in various cancers includ- gions, facilitating chromatin accessibility and the re- ing the prostate. cruitment of additional TFs and co-TF and the tran- SOX2 is required for survival, pluripotency, scriptional machinery [123]. Among paradigmatic clonogenicity, and self-renewal of ESCs. A relationship pioneering factors are some of the members of the between SOX2 overexpression in tumorigenesis has GATA and FoxA gene families, known mainly for their been established in different types of cancer, including key role as chromatin-factors during early development prostate [56] and its expression linked to tumor grade [124–127]. [58]. SOX2 is an epigenetic reprogramming factor and The best-known pioneering factor for its role in pros- oncogene shown to regulate androgen-independent tate cancer is FOXA1. Through the interaction and re- CRPC proliferation and evasion of apoptosis [57, 58] cruitment of AR to chromatin site, FoxA1 defines and andtopromote tumormetastasisbyinducing EMT controls the AR cistrome resulting in context-dependent [59]. Further evidence suggests that SOX2 promotes positive or negative regulation [52, 55, 128, 129]. In self-renewal of the CSCs population by acting down- particular, because FOXA1 activity on chromatin re- stream of EGFR [131]. Importantly, in the recent years, sults in increased accessibility [52] and increased SOX2 activity has been tightly associated to neuroen- chromatin-bound AR, high FOXA1 expression leads docrine transdifferentiation from prostate adenocarcino- to a restricted AR cistrome regulation [53]. ma cells and the subsequent androgen independence of GATA genes, and GATA2 in particular, have proved neuroendocrine prostate cancer phenotypes (NEPC). to be crucial for prostate development via modulating While the exact mechanisms remain unclear, substantial AR function [54, 55]. However, despite the role is com- progress was made over the last couple of years. In parable to that of FoxA1, the mechanisms have shown particular, Russo and colleagues showed that SOX2 Curr Mol Bio Rep (2018) 4:101–115 109 was expressed in NEPC murine models [60] whereas contributor of aggressiveness via the activation of EMT tran- others found its expression restricted to NEPC areas scriptional programs. Nolan and colleagues proposed a model of advanced human prostate cancer [61]. Recent studies in which the secreted extracellular protein Hsp90 initiates by Bishop and collaborators at the Zoubeidi laboratory ERK signaling and leads to the recruitment of EZH2 to the have shown that SOX2 is transcriptionally regulated by E-cadherin promoter and repression of E-cadherin expression, neural transcription factor BRN2 [132��], which in turn driving epithelial to mesenchymal transition (EMT) and inva- is negatively suppressed by the AR, hence revealing an sion in prostate cancer cells [67]. Additionally, DAB2IP (dis- AR-dependent suppression of cell differentiation to- abled homolog 2 interacting protein) is a tumor suppressor ward a neuroendocrine AR-independent phenotype. Ras-GAP that negatively controls Ras-dependent mitogenic Additional support to the central role of SOX2 in the signals and modulates TNFα/NF-κB, WNT/β-catenin, emergence of NEPC and AR-independence after PI3K/AKT, and androgen receptors pathways [68–70]. Enzalutamide treatment came from studies at the Ku EZH2-induced DAB2IP silencing activates Ras and NF- and Mu and collaborators at the Sawyers and kappaB and triggers metastasis [141, 142]. Data from our Goodrich laboratories [62��, 63��]. laboratory showed that concomitant activation of the PI3K c-MYC (MYC) is a well-known oncogene proposed and MAPk pathways in mice results in highly aggressive as a marker of disease progression in prostate cancer and fully metastatic tumors that are inherently castration resis- [133] and associated with prostate cancer recurrence tant [143, 144]. Interestingly, by targeting the PI3K/MAPk and poor prognosis [134]. MYC activation cooperates pathways with small molecules in vivo, we demonstrated that with loss of PTEN to drive prostate cancer progression the drug response network was highly enriched in epigenetic [135] and metastasis [136]. MYC proteins also drive modulators, including SUV39H1, WHSC1, TOP2A, or epigenetic activation of gene expression in prostate UHRF1 [145], suggesting that epigenetic control of gene ex- cancer; the PRC2 member EZH2 is directly upregulated pression plays a central role in the aggressive phenotype im- by MYC [137] and MYCN, which was shown to be a posed by the activation of Ras signaling. Accordingly, we driver of NEPC [66] by inducing an EZH2-mediated have found that a core signature of chromatin modifiers and transcriptional program [64]. Additionally, MYC ex- DNMTs drive the cancer cell intrinsic mechanisms of metas- pression was found to be regulated by the histone tasis and CRPC (unpublished). demethylase JMJD1A, controlling proliferation and sur- The retinoblastoma tumor suppressor gene RB1 is more vival of prostate cancer cells [138]. MYC also regulates commonly loss in metastatic and antiandrogen resistant pros- the expression of histone demethylases PHF8 and tate cancer (74% of cases) and NEPC (90% of cases) [71] than KDMA3 in NEPC and CRPC [29]. Interestingly, while it is in primary tumors (34% of cases) [72]. It has been recently described an activity of Rb1 in the epigenetic regulation of AR signaling in the normal prostate represses MYC expression, its expression is stimulated by AR during expression, since RB1 directly repress the expression of Sox2 tumorigenesis, [139, 140]. It was also recently reported and Ezh2. Consequently, Rb1 loss in prostate cancer lead to that MYC overexpression deregulates the AR transcrip- EZH2 and Sox2 increase and gene expression widespread tional program by altering AR chromatin occupancy changes that leads toward a stem cell-like state that would and H3K4me1 and H3K27me3 marks distribution, an- facilitate the onset of metastasis, neuroendocrine tagonizing clinically relevant AR target genes such as transdifferentiation, and the acquisition of ADT resistance. PSA [65]. The authors show that Ezh2 inhibition restores enzalutamide sensitivity in NEPC variants and recurrent prostate cancer cells by opposing lineage transformation [63�� ]. Oncogenic Pathways Involved in Epigenetic Furthermore, mutations in TP53 and RB1 tumor suppressor Regulations genes can promote a cellular plasticity state mediated by in- creased expression of SOX2 that, when it is compromised Together with the AR, the oncogenic pathways most frequent- with antiandrogen therapy promotes resistance through line- ly altered in prostate cancer onset and progression are the RB, age switching [62�� ]. It has also recently been shown that the PI3K/AKT, and Ras/Raf pathways due to mutations in several Hedgehog (HH) signaling pathway and SOX2 co-operate in members [72]. While the Ras/Raf pathway is activated in 43% androgen-independent prostate cancer to promote carcinogen- of primary and 90% of metastatic prostate cancer, the trigger- esis [146]. ing mechanisms remain to be fully understood. The Whitte The PTEN/PI3K/AKT pathway is altered in 42% of prima- laboratory demonstrated a synergistic interaction between Ras ry and 100% of metastatic cases; loss of PTEN and activation pathway activation and AR signaling that leads to elevated of the PI3K/AKT signaling pathway are hallmarks of prostate EZH2 expression and expand prostate cancer progenitor cells cancer, and cooperate with the activation of the RAS/MAPK pathway to promote EMT and metastatic CRPC development. in vivo. It has been long suggested that this pathway is a major 110 Curr Mol Bio Rep (2018) 4:101–115 Epigenetically, it has also been shown that PTEN depletion Drug Development on Epigenetic Regulators contributes to a switch from a global H3K27 acetylatilation to as Therapeutic Targets H3K27 trimethylation, resulting in increased expression of EZH2 and decrease of the target genes DAB2PI together with Mounting evidence from basic and preclinical investigations KIP1 CIP1 negative regulator of cell growth p27 and p21 [147]. suggest that targeting key components of the epigenetic ma- As mentioned above, increased AKT activity phosphorylates chinery will have clinical benefit for cancer patients including Cdt2 NSD2 at S172, preventing its degradation by CRL4 E3 prostate cancer ones. Yet, clinical development for those ther- ligase, hence leading to its stabilization, which in turn apies is still very limited. On the one hand, this may be partly upregulates RICTOR (mTORC2). This results in further en- due to the inherent difficulty in targeting nuclear effector hancement of AKT signaling in a AKT/NSD2/mTORC2 pos- mechanisms. On the other hand, the fact that most epigenetic itive feedback loop that sustains AKT signaling [1 ]. master regulators exert their functions over an extensive tran- Constitutive activation of TGF-β signaling is a well- scriptional network in a context-dependent manner makes it recognized mechanism for induction of EMT and prostate particularly challenging to achieve cancer cell specificity, thus cancer metastasis development. TGF-β1-induced EMT in resulting in significant toxicity. Despite these limitations, a prostate cancer is mediated by the histone methyltransfer- number of drugs are currently in clinical trials at different ase RbBP5. RbBP5 is a conserved component of the phases, being BET bromodomain inhibitors, HMT/HDMT COMPASS/-like complex, which catalyzes the inhibitors, DNMT inhibitors, and HDAC inhibitors the focus trimethylation of histone H3 lysine 4 that is considered of most intense drug development efforts. Table 2 summarizes an epigenetic mark of actively transcribed genes. RbBP5 the most relevant ongoing or recently completed clinical trials activity is in turn modulated by the binding of SMAD2/3, a involving epigenetic drugs. downstream signaling factor to the TGF-beta pathway, to the Snail promoter [148]. Snail activates the EMT process by inhibiting transcription of E-cadherin via the recruit- ment to its promoter of the polycomb repressive complex Conclusion 2 (PRC2) and the histone methylstranferase G9a, leading to repressive H3K27 and H3K9 methylation [149, 150]. In view of the accumulated evidenced supporting the key An interesting new perspective was provided recently role of the dysregulated epigenome to prostate cancer onset linking ERG signaling with TGF-β. Data suggest that and progression, three mechanisms emerge as the most sig- ERG regulates the transcription of the transcription factor nificant contributors. First, a number of alterations in epi- SOX4 and together they cooperate in TGF-β1-induced genetic master regulators result in enhanced transcriptional EMT of prostate cancer cells [151]. This is not surprising activity and pro-oncogenic role of the Androgen Receptor taking into account that the oncogenic role of SOX4 has signaling. This is largely mediated by either remodeling of been proposed in several other tumor types. In particular, the chromatin to facilitate AR binding and assembly of the SOX4 regulates EZH2 expression and chromating remod- transcriptional complex and posttranslational modifications eling, and is a key component of the PI3K/AKT pathway in in the AR itself or essential co-factors resulting in gain of prostate cancer. In fact, SOX4 inhibition reduces AKT and function features. Secondly, the aberrant activation of tran- β-catenin pathways activation and decreases prostate can- scriptional programs tightly associated to developmental cer invasiveness through positive feedback loop between pathways and stem features, either via alterations in SOX4 and PI3K-AKT-mTOR [152]. pioneering factors or pluripotency master regulators, con- Finally, a tyrosine kinase, namely ACK1, has been found tributes to the acquisition of treatment-resistant phenotypes to link oncogenic signaling with epigenetic regulation. that are highly aggressive. Finally, a significant number of ACK1 was found upregulated in primary PCa and CRPC alterations in epigenetic master regulators also result in the [72, 73], correlated with poor prognosis and reported to activation of oncogenic signaling pathways that contribute interact with AR to drive ADT resistance and CRPC growth to the aggressiveness and androgen independence in ad- [74]. A recent study demostrates that ACK1 regulates tran- vanced prostate tumors. In summary, the epigenome is scription of AR and AR-v7 via epigenetic regulation. In emerging as an attractive and plausible target for anticancer particular, ACK1 would phosphorylate histone H4 up- therapy in general and prostate cancer in particular. While stream of the AR transcription start site, recruiting the drug development is still limited, and faces inherent chal- WRD5/MLL2 complex, therefore mediating H3K4 lenges associated with the unique nature of these targets, it trymethylation and transcriptional activation. Inhibition of seems evident that efficacy of such treatments will be max- ACK1 with a small molecule inhibitor confirms that this imized in combination with standard of care treatments for epigenetic activity is required to maintain AR transcription which most lethal prostate cancer ultimately develop resis- and CRPC tumor growth [153]. tant mechanism. �� Curr Mol Bio Rep (2018) 4:101–115 111 Acknowledgements This research is supported by funding from the 9. Vieira FQ, Costa-Pinheiro P, Almeida-Rios D, Graca I, Monteiro- Spanish Ministry of Economy, Industry and Competitiveness (MINECO), Reis S, Simoes-Sousa S, et al. SMYD3 contributes to a more which is part of the State Research Agency, through the projects PI16/01070 aggressive phenotype of prostate cancer and targets cyclin D2 and CP15/00090. We also acknowledge the support from the European through H4K20me3. Oncotarget. 2015;6(15):13644–57. Association of Urology Research Foundation (EAURF/407003/XH), and 10. Huang L, Xu AM. SET and MYND domain containing protein 3 the Fundacion BBVA-Young Investigator Award. We thank CERCA in cancer. Am J Transl Res. 2017;9(1):1–14. Program / Generalitat de Catalunya for the institutional support. 11. Stopa N, Krebs JE, Shechter D. The PRMT5 arginine methyltrans- ferase: many roles in development, cancer and beyond. Cell Mol Life Sci. 2015;72(11):2041–59. Compliance with Ethical Standards 12.� Mounir Z, Korn JM, Westerling T, Lin F, Kirby CA, Schirle M, et al. ERG signaling in prostate cancer is driven through PRMT5- Conflict of Interest Katia Ruggero, Sonia Farran-Matas, Adrian dependent methylation of the Androgen Receptor. Elife. Martinez-Tebar, and Alvaro Aytes declare no conflicts of interest. 2016;16(5):13964. This paper links for the first time epigenet- ically driven posttranslational modifications in the AR and Human and Animal Rights and Informed Consent This article does not ERG transcriptional expression. contain any studies with human or animal subjects performed by any of 13. Deng X, Shao G, Zhang HT, Li C, Zhang D, Cheng L, et al. the authors. Protein arginine methyltransferase 5 functions as an epigenetic activator of the androgen receptor to promote prostate cancer cell Open Access This article is distributed under the terms of the Creative growth. Oncogene. 2017;36(9):1223–31. Commons Attribution 4.0 International License (http:// 14.�� Metzger E, Wissmann M, Yin N, Muller JM, Schneider R, Peters creativecommons.org/licenses/by/4.0/), which permits unrestricted use, AH, et al. LSD1 demethylates repressive histone marks to pro- distribution, and reproduction in any medium, provided you give appro- mote androgen-receptor-dependent transcription. Nature. priate credit to the original author(s) and the source, provide a link to the 2005;437(7057):436–9. This paper links demethylation of a Creative Commons license, and indicate if changes were made. repressive histone mark with androgen-receptor-dependent gene activation. 15. Cai C, He HH, Chen S, Coleman I, Wang H, Fang Z, et al. Androgen receptor gene expression in prostate cancer is directly References suppressed by the androgen receptor through recruitment of lysine-specific demethylase 1. Cancer Cell. 2011;20(4):457–71. 16.� Liang Y, Ahmed M, Guo H, Soares F, Hua JT, Gao S, et al. LSD1- Papers of particular interest, published recently, have been mediated epigenetic reprogramming drives CENPE expression highlighted as: and prostate cancer progression. Cancer Res. 2017;77(20):5479– � Of importance 90. This study shows that LSD1-mediated epigenetic reprogramming drives CRPC and was associated with the �� Of major importance activation of CENPE and other cell-cycle genes. 17. Xiang Y, Zhu Z, Han G, Ye X, Xu B, Peng Z, et al. JARID1B is a 1.�� Li N, Xue W, Yuan H, Dong B, Ding Y, Liu Y, et al. AKT- histone H3 lysine 4 demethylase up-regulated in prostate cancer. mediated stabilization of histone methyltransferase WHSC1 pro- Proc Natl Acad Sci U S A. 2007;104(49):19226–31. motes prostate cancer metastasis. J Clin Invest. 2017;127(4): 18. Lu W, Liu S, Li B, Xie Y, Adhiambo C, Yang Q, et al. SKP2 1284–302. This work demonstrates that NSD2 is activated in inactivation suppresses prostate tumorigenesis by mediating PTEN null tumors by the AKT pathway and that its expres- JARID1B ubiquitination. Oncotarget. 2015;6(2):771–88. sion is required for metastatic progression. 19. Stein J, Majores M, Rohde M, Lim S, Schneider S, Krappe E, et al. 2. Ezponda T, Popovic R, Shah MY, Martinez-Garcia E, Zheng Y, DM5C is overexpressed in prostate cancer and is a prognostic Min DJ, et al. The histone methyltransferase MMSET/WHSC1 marker for prostate-specific antigen-relapse following radical activates TWIST1 to promote an epithelial-mesenchymal transi- prostatectomy. Am J Pathol. 2014;184(9):2430–7. tion and invasive properties of prostate cancer. Oncogene. 20. Li N, Dhar SS, Chen TY, Kan PY, Wei Y, Kim JH, et al. JARID1D 2013;32(23):2882–90. is a suppressor and prognostic marker of prostate cancer invasion 3. Asangani IA, Ateeq B, Cao Q, Dodson L, Pandhi M, Kunju LP, and metastasis. Cancer Res. 2016;76(4):831–43. et al. Characterization of the EZH2-MMSET histone methyltrans- 21. Komura K, Jeong SH, Hinohara K, Qu F, Wang X, Hiraki M, et al. ferase regulatory axis in cancer. Mol Cell. 2013;49(1):80–93. Resistance to docetaxel in prostate cancer is associated with an- 4. Yang YA, Yu J. EZH2, an epigenetic driver of prostate cancer. drogen receptor activation and loss of KDM5D expression. Proc Protein Cell. 2013;4(5):331–41. Natl Acad Sci U S A. 2016;113(22):6259–64. 5. Wu C, Jin X, Yang J, Yang Y, He Y, Ding L, et al. Inhibition of 22. Yu L, Wang Y, Huang S, Wang J, Deng Z, Zhang Q, et al. EZH2 by chemo- and radiotherapy agents and small molecule Structural insights into a novel histone demethylase PHF8. Cell inhibitors induces cell death in castration-resistant prostate cancer. Res. 2010;20(2):166–73. Oncotarget. 2016;7(3):3440–52. 23. Qi HH, Sarkissian M, Hu G-Q, Wang Z, Bhattacharjee A, Gordon 6. Yu T, Wang C, Yang J, Guo Y, Wu Y, Li X. Metformin inhibits DB, et al. Histone H4K20/H3K9 demethylase PHF8 regulates SUV39H1-mediated migration of prostate cancer cells. zebrafish brain and craniofacial development. Nature. 2010;466: Oncogene. 2017;6(5):):28. 503–7. 7. Li Q, Li Y, Wang Y, Cui Z, Gong L, Qu Z, et al. Quantitative 24. Qiu J, Shi G, Jia Y, Li J, Wu M, Li J, et al. The X-linked mental proteomic study of human prostate cancer cells with different met- retardation gene PHF8 is a histone demethylase involved in neu- astatic potentials. Int J Oncol. 2016;48(4):1437–46. ronal differentiation. Cell Res. 2010;20:908–18. 8. Askew EB, Bai S, Parris AB, Minges JT, Wilson EM. Androgen 25. Liu W, Tanasa B, Tyurina OV, Zhou TY, Gassmann R, Liu WT, receptor regulation by histone methyltransferase suppressor of et al. PHF8 mediates histone H4 lysine 20 demethylation events variegation 3-9 homolog 2 and melanoma antigen-A11. Mol involved in cell cycle progression. Nature. 2010;466(7305):508– Cell Endocrinol. 2017;443:42–51. 12. 112 Curr Mol Bio Rep (2018) 4:101–115 26. Horton JR, Upadhyay AK, Qi HH, Zhang X, Shi Y, Cheng X. 44. Damodaran S, Damaschke N, Gawdzik J, Yang B, Shi C, Allen GO, et al. Dysregulation of Sirtuin 2 (SIRT2) and histone H3K18 Enzymatic and structural insights for substrate specificity of a family of jumonji histone lysine demethylases. Nat Struct Amp acetylation pathways associates with adverse prostate cancer out- Mol Biol. 2009;17:38. comes. BMC Cancer. 2017;17(1):874. 27. Bjorkman M, Ostling P, Harma V, Virtanen J, Mpindi JP, Rantala 45.� Asangani IA, Dommeti VL, Wang X, Malik R, Cieslik M, Yang J, et al. Systematic knockdown of epigenetic enzymes identifies a R, et al. Therapeutic targeting of BET bromodomain proteins in novel histone demethylase PHF8 overexpressed in prostate cancer castration-resistant prostate cancer. Nature. 2014;510(7504):278– with an impact on cell proliferation, migration and invasion. 82. This study reports that BRD4 physically interacts with Oncogene. 2012;31(29):3444–56. high-affinity with the N-terminal domain of AR leading to AR translocation into the nucleus and AR recruitment to tar- 28. Tong D, Liu Q, Liu G, Yuan W, Wang L, Guo Y, et al. The HIF/ get loci, promoting AR activity and expression of AR target PHF8/AR axis promotes prostate cancer progression. Oncogene. genes in CRPC. 2016;5(12):e283. 29. Maina PK, Shao P, Liu Q, Fazli L, Tyler S, Nasir M, et al. c-MYC 46. Shah N, Wang P, Wongvipat J, Karthaus WR, Abida W, Armenia drives histone demethylase PHF8 during neuroendocrine differen- J, et al. Regulation of the glucocorticoid receptor via a BET- tiation and in castration-resistant prostate cancer. Oncotarget. dependent enhancer drives antiandrogen resistance in prostate 2016;7(46):75585–602. cancer. elife. 2017;6 47. Li J, Alyamani M, Zhang A, Chang KH, Berk M, Li Z, et al. 30. Shin S, Janknecht R. Activation of androgen receptor by histone Aberrant corticosteroid metabolism in tumor cells enables GR demethylases JMJD2A and JMJD2D. Biochem Biophys Res takeover in enzalutamide resistant prostate cancer. elife. 2017;6 Commun. 2007;359(3):742–6. 31. Wissmann M, Yin N, Muller JM, Greschik H, Fodor BD, 48. Dai X, Gan W, Li X, Wang S, Zhang W, Huang L, et al. Prostate Jenuwein T, et al. Cooperative demethylation by JMJD2C and cancer-associated SPOP mutations confer resistance to BET inhib- LSD1 promotes androgen receptor-dependent gene expression. itors through stabilization of BRD4. Nat Med. 2017;23(9):1063– Nat Cell Biol. 2007;9(3):347–53. 71. 32. Yamane K, Toumazou C, Tsukada Y, Erdjument-Bromage H, 49. Groner AC, Cato L, de Tribolet-Hardy J, Bernasocchi T, Tempst P, Wong J, et al. JHDM2A, a JmjC-containing H3K9 Janouskova H, Melchers D, et al. TRIM24 is an oncogenic tran- demethylase, facilitates transcription activation by androgen re- scriptional activator in prostate cancer. Cancer Cell. 2016;29(6): ceptor. Cell. 2006;125(3):483–95. 846–58. 50. Rodrigues LU, Rider L, Nieto C, Romero L, Karimpour-Fard A, 33. Wilson S, Fan L, Sahgal N, Qi J, Filipp FV. The histone Loda M, et al. Coordinate loss of MAP3K7 and CHD1 promotes demethylase KDM3A regulates the transcriptional program of aggressive prostate cancer. Cancer Res. 2015;75(6):1021–34. the androgen receptor in prostate cancer cells. Oncotarget. 2017;8(18):30328–43. 51. Kari V, Mansour WY, Raul SK, Baumgart SJ, Mund A, Grade M, 34. Coffey K, Rogerson L, Ryan-Munden C, Alkharaif D, Stockley J, et al. Loss of CHD1 causes DNA repair defects and enhances Heer R, et al. The lysine demethylase, KDM4B, is a key molecule prostate cancer therapeutic responsiveness. EMBO Rep. in androgen receptor signalling and turnover. Nucleic Acids Res. 2016;17(11):1609–23. 2013;41(8):4433–46. 52. Jin HJ, Zhao JC, Wu L, Kim J, Yu J. Cooperativity and equilibri- um with FOXA1 define the androgen receptor transcriptional pro- 35. Xiang Y, Zhu Z, Han G, Lin H, Xu L, Chen CD. JMJD3 is a gram. Nat Commun. 2014;5(3972) histone H3K27 demethylase. Cell Res. 2007;17(10):850–7. 36. Chin SP, Dickinson JL, Holloway AF. Epigenetic regulation of 53. Zhao JC, Fong KW, Jin HJ, Yang YA, Kim J, Yu J. FOXA1 acts prostate cancer. Clin Epigenetics. 2011;2(2):151–69. upstream of GATA2 and AR in hormonal regulation of gene ex- pression. Oncogene. 2016;35(33):4335–44. 37. Lin PC, Giannopoulou EG, Park K, Mosquera JM, Sboner A, Tewari AK, et al. Epigenomic alterations in localized and ad- 54. Lupien M, Eeckhoute J, Meyer CA, Wang Q, Zhang Y, Li W, et al. vanced prostate cancer. Neoplasia. 2013;15(4):373–83. Fox A1 translates epigenetic signatures into enhancer-driven line- age-specific transcription. Cell. 2008;132(6):958–70. 38. Kang GH, Lee S, Lee HJ, Hwang KS. Aberrant CpG island hyper- 55. Wang Q, Li W, Liu XS, Carroll JS, Jänne OA, Keeton EK, et al. A methylation of multiple genes in prostate cancer and prostatic hierarchical network of transcription factors governs androgen intraepithelial neoplasia. J Pathol. 2004;202(2):233–40. receptor-dependent prostate cancer growth. Mol Cell. 39. Mahon KL, Qu W, Devaney J, Paul C, Castillo L, Wykes RJ, et al. 2007;27(3):380–92. Methylated glutathione S-transferase 1 (mGSTP1) is a potential plasma free DNA epigenetic marker of prognosis and response to 56. Wong OG, Huo Z, Siu MK, Zhang H, Jiang L, Wong ES, et al. chemotherapy in castrate-resistant prostate cancer. Br J Cancer. Hypermethylation of SOX2 promoter in endometrial carcinogen- 2014;111(9):1802–9. esis. Obstet Gynecol Int. 2010;10(682504):9. 40. Fu M, Wang C, Reutens AT, Wang J, Angeletti RH, Siconolfi- 57. Lin F, Lin P, Zhao D, Chen Y, Xiao L, Qin W, et al. Sox 2 targets Baez L, et al. p300 and p300/cAMP-response element-binding cyclinE, p 27 and survivin to regulate androgen-independent hu- protein-associated factor acetylate the androgen receptor at sites man prostate cancer cell proliferation and apoptosis. Cell Prolif. governing hormone-dependent transactivation. J Biol Chem. 2012;45(3):207–16. 2000;275(27):20853–60. 58. Kregel S, Kiriluk KJ, Rosen AM, Cai Y, Reyes EE, Otto KB, et al. 41. Jin L, Garcia J, Chan E, de la Cruz C, Segal E, Merchant M, et al. Sox2 is an androgen receptor-repressed gene that promotes Therapeutic targeting of the CBP/p300 bromodomain blocks the castration-resistant prostate cancer. PLoS One. 2013;8(1):11. growth of castration-resistant prostate cancer. Cancer Res. 59. Li X, Xu Y, Chen Y, Chen S, Jia X, Sun T, et al. SOX2 promotes 2017;77(20):5564–75. tumor metastasis by stimulating epithelial-to-mesenchymal transi- 42. Fu M, Liu M, Sauve AA, Jiao X, Zhang X, Wu X, et al. Hormonal tion via regulation of WNT/beta-catenin signal network. Cancer control of androgen receptor function through SIRT1. Mol Cell Lett. 2013;336(2):379–89. Biol. 2006;26(21):8122–35. 60. Russo MV, Esposito S, Tupone MG, Manzoli L, Airoldi I, Pompa 43. Lavery DN, Bevan CL. Androgen receptor signalling in prostate P, et al. SOX2 boosts major tumor progression genes in prostate cancer: the functional consequences of acetylation. J Biomed cancer and is a functional biomarker of lymph node metastasis. Biotechnol. 2011;862125(10):28. Oncotarget. 2016;7(11):12372–85. Curr Mol Bio Rep (2018) 4:101–115 113 61. Esposito S, Russo MV, Airoldi I, Tupone MG, Sorrentino C, 76. Suzuki H, Ueda T, Ichikawa T, Ito H. Androgen receptor involve- ment in the progression of prostate cancer. Endocr Relat Cancer. Barbarito G, et al. SNAI2/Slug gene is silenced in prostate cancer and regulates neuroendocrine differentiation, metastasis- 2003;10(2):209–16. suppressor and pluripotency gene expression. Oncotarget. 77. Suzuki H, Freije D, Nusskern DR, Okami K, Cairns P, Sidransky 2015;6(19):17121–34. D, et al. Interfocal heterogeneity of PTEN/MMAC1 gene alter- ations in multiple metastatic prostate cancer tissues. Cancer Res. 62.�� Mu P, Zhang Z, Benelli M, Karthaus WR, Hoover E, Chen CC, 1998;58(2):204–9. et al. SOX2 promotes lineage plasticity and antiandrogen resis- 78. Jarrard DF, Bova GS, Ewing CM, Pin SS, Nguyen SH, Baylin SB, tance in TP53- and RB1-deficient prostate cancer. Science. et al. Deletional, mutational, and methylation analyses of CDKN2 2017;355(6320):84–8. This work demonstrates that lineage (p16/MTS1) in primary and metastatic prostate cancer. Genes plasticity and neuroendocrine differentiation in androgen in- Chromosomes Cancer. 1997;19(2):90–6. dependence is partly driven by Sox2 in prostate cancer mouse 79. Pakneshan P, Xing RH, Rabbani SA. Methylation status of uPA models carrying loss of function alleles for p53 and Rb tumor promoter as a molecular mechanism regulating prostate cancer suppressors. invasion and growth in vitro and in vivo. FASEB J. 2003;17(9): 63.�� Ku SY, Rosario S, Wang Y, Mu P, Seshadri M, Goodrich ZW, et al. 1081–8. Rb1 and Trp53 cooperate to suppress prostate cancer lineage plas- 80. Ashour N, Angulo JC, Andres G, Alelu R, Gonzalez-Corpas A, ticity, metastasis, and antiandrogen resistance. Science. Toledo MV, et al. A DNA hypermethylation profile reveals new 2017;355(6320):78–83. This work shows that Sox2 and Ezh2 potential biomarkers for prostate cancer diagnosis and prognosis. mediates suppression of prostate cancer lineage plasticity in Prostate. 2014;74(12):1171–82. Rb1 and p53 null tumours. Ezh2 inhibition restores 81. Coffey K, Rogerson L, Ryan-Munden C, Alkharaif D, Stockley J, enzalutamide sensitivity in NEPC variants and recurrent Heer R, et al. The lysine demethylase, KDM4B, is a key molecule prostate cancer cells by opposing lineage transformation. in androgen receptor signalling and turnover. Nucleic Acids Res. 64. Dardenne E, Beltran H, Benelli M, Gayvert K, Berger A, Puca L, 2013;41(8):4433–46. et al. N-Myc induces an EZH2-mediated transcriptional program 82. Hayami S, Kelly JD, Cho HS, Yoshimatsu M, Unoki M, Tsunoda driving neuroendocrine prostate Cancer. Cancer Cell. 2016;30(4): T, et al. Overexpression of LSD1 contributes to human carcino- 563–77. genesis through chromatin regulation in various cancers. Int J 65. Barfeld SJ, Urbanucci A, Itkonen HM, Fazli L, Hicks JL, Thiede Cancer. 2011;128(3):574–86. B, et al. C-Myc Antagonises the transcriptional activity of the 83. Kahl P, Gullotti L, Heukamp LC, Wolf S, Friedrichs N, Vorreuther androgen receptor in prostate cancer affecting key gene networks. R, et al. Androgen receptor coactivators lysine-specific histone EBioMedicine. 2017;18:83–93. demethylase 1 and four and a half LIM domain protein 2 predict 66. Lee JK, Phillips JW, Smith BA, Park JW, Stoyanova T, McCaffrey risk of prostate cancer recurrence. Cancer Res. 2006;66(23): EF, et al. N-Myc drives neuroendocrine prostate cancer initiated 11341–7. from human prostate epithelial cells. Cancer Cell. 2016;29(4): 84. Battaglia S, Karasik E, Gillard B, Williams J, Winchester T, Moser 536–47. MT, et al. LSD1 dual function in mediating epigenetic corruption 67. Nolan KD, Franco OE, Hance MW, Hayward SW, Isaacs JS. of the vitamin D signaling in prostate cancer. Clin Epigenetics. Tumor-secreted Hsp90 subverts polycomb function to drive pros- 2017;9(82):017–0382. tate tumor growth and invasion. J Biol Chem. 2015;290(13): 85. Li Y, Trojer P, Xu CF, Cheung P, Kuo A, Drury WJ, 3rd, et al. The 8271–82. target of the NSD family of histone lysine methyltransferases de- 68. Bellazzo A, Di Minin G, Collavin L. Block one, unleash a hun- pends on the nature of the substrate. J Biol Chem 2009; 284(49): dred. Mechanisms of DAB2IP inactivation in cancer. Cell Death 34283–34295. Differ. 2017;24(1):15–25. 86. Keats JJ, Maxwell CA, Taylor BJ, Hendzel MJ, Chesi M, 69. Wu K, Liu J, Tseng SF, Gore C, Ning Z, Sharifi N, et al. The role Bergsagel PL, et al. Overexpression of transcripts originating of DAB2IP in androgen receptor activation during prostate cancer from the MMSET locus characterizes all t(4;14)(p16;q32)-posi- progression. Oncogene. 2014;33(15):1954–63. tive multiple myeloma patients. Blood. 2005;105(10):4060–9. 70. Tsai YS, Lai CL, Lai CH, Chang KH, Wu K, Tseng SF, et al. The 87. Yang P, Guo L, Duan ZJ, Tepper CG, Xue L, Chen X, et al. role of homeostatic regulation between tumor suppressor DAB2IP Histone methyltransferase NSD2/MMSET mediates constitutive and oncogenic Skp2 in prostate cancer growth. Oncotarget. NF-kappaB signaling for cancer cell proliferation, survival, and 2014;5(15):6425–36. tumor growth via a feed-forward loop. Mol Cell Biol. 71. Tan HL, Sood A, Rahimi HA, Wang W, Gupta N, Hicks J, et al. 2012;32(15):3121–31. Rb loss is characteristic of prostatic small cell neuroendocrine 88. Kang HB, Choi Y, Lee JM, Choi KC, Kim HC, Yoo JY, et al. The carcinoma. Clin Cancer Res. 2014;20(4):890–903. histone methyltransferase, NSD2, enhances androgen receptor- mediated transcription. FEBS Lett. 2009;583(12):1880–6. 72. Taylor BS, Schultz N, Hieronymus H, Gopalan A, Xiao Y, Carver 89. Nalla AK, Williams TF, Collins CP, Rae DT, Trobridge GD. BS, et al. Integrative genomic profiling of human prostate cancer. Lentiviral vector-mediated insertional mutagenesis screen iden- Cancer Cell. 2010;18(1):11–22. tifies genes that influence androgen independent prostate cancer 73. van der Horst EH, Degenhardt YY, Strelow A, Slavin A, Chinn L, progression and predict clinical outcome. Mol Carcinog. Orf J, et al. Metastatic properties and genomic amplification of the 2016;55(11):1761–71. tyrosine kinase gene ACK1. Proc Natl Acad Sci U S A. 90. Shi Y, Lan F, Matson C, Mulligan P, Whetstine JR, Cole PA, et al. 2005;102(44):15901–6. Histone demethylation mediated by the nuclear amine oxidase 74. Mahajan NP, Liu Y, Majumder S, Warren MR, Parker CE, Mohler homolog LSD1. Cell. 2004;119(7):941–53. JL, et al. Activated Cdc42-associated kinase Ack1 promotes pros- 91. Metzger E, Yin N, Wissmann M, Kunowska N, Fischer K, tate cancer progression via androgen receptor tyrosine phosphor- Friedrichs N, et al. Phosphorylation of histone H3 at threonine ylation. Proc Natl Acad Sci U S A. 2007;104(20):8438–43. 11 establishes a novel chromatin mark for transcriptional regula- 75. Nelson WG, De Marzo AM, Yegnasubramanian S. Epigenetic tion. Nat Cell Biol. 2008;10(1):53–60. alterations in human prostate cancers. Endocrinology. 92. Jia L, Berman BP, Jariwala U, Yan X, Cogan JP, Walters A, et al. 2009;150(9):3991–4002. Genomic androgen receptor-occupied regions with different 114 Curr Mol Bio Rep (2018) 4:101–115 functions, defined by histone acetylation, coregulators and tran- enhancers and superenhancers through BET bromodomain inhibition is a promising therapeutic option in cancer and scriptional capacity. PLoS One. 2008;3(11):10. 93. Jia L, Shen HC, Wantroba M, Khalid O, Liang G, Wang Q, et al. highlight the key role of this epigenetic master regulator in cancer progression Locus-wide chromatin remodeling and enhanced androgen receptor-mediated transcription in recurrent prostate tumor cells. 109. Jang MK, Mochizuki K, Zhou M, Jeong HS, Brady JN, Ozato K. Mol Cell Biol. 2006;26(19):7331–41. The bromodomain protein Brd4 is a positive regulatory compo- 94. Lasko LM, Jakob CG, Edalji RP, Qiu W, Montgomery D, nent of P-TEFb and stimulates RNA polymerase II-dependent Digiammarino EL, et al. Discovery of a selective catalytic p300/ transcription. Mol Cell. 2005;19(4):523–34. CBP inhibitor that targets lineage-specific tumours. Nature. 110. Zuber V, Bettella F, Witoelar A, Andreassen OA, Mills IG, 2017;550(7674):128–32. Urbanucci A. Bromodomain protein 4 discriminates tissue- 95.� Valdes-Mora F, Gould CM, Colino-Sanguino Y, Qu W, Song JZ, specific super-enhancers containing disease-specific susceptibility Taylor KM, et al. Acetylated histone variant H2A.Z is involved in loci in prostate and breast cancer. BMC Genomics. 2017;18(1): the activation of neo-enhancers in prostate cancer. Nat Commun. 2017;8(1):1346. This study highlights the important role of 111. Faivre EJ, Wilcox D, Lin X, Hessler P, Torrent M, He W, et al. histone acetylation in prostate cancer beyond active pro- Exploitation of castration-resistant prostate cancer transcription moters via activation of AR associated enhancers and the in- factor dependencies by the novel BET inhibitor ABBV-075. Mol crease in chromatin accessibility Cancer Res. 2017;15(1):35–44. 96. Ruscetti M, Dadashian EL, Guo W, Quach B, Mulholland DJ, 112. Asangani IA, Wilder-Romans K, Dommeti VL, Krishnamurthy Park JW, et al. HDAC inhibition impedes epithelial- PM, Apel IJ, Escara-Wilke J, et al. BET Bromodomain inhibitors mesenchymal plasticity and suppresses metastatic, castration- enhance efficacy and disrupt resistance to AR antagonists in the resistant prostate cancer. Oncogene. 2016;35(29):3781–95. treatment of prostate cancer. Mol Cancer Res. 2016;14(4):324–31. 97. Di Croce L, Helin K. Transcriptional regulation by polycomb 113. Arora VK, Schenkein E, Murali R, Subudhi SK, Wongvipat J, group proteins. Nat Struct Mol Biol. 2013;20(10):1147–55. Balbas MD, et al. Glucocorticoid receptor confers resistance to 98. Melling N, Thomsen E, Tsourlakis MC, Kluth M, Hube-Magg C, antiandrogens by bypassing androgen receptor blockade. Cell. Minner S, et al. Overexpression of enhancer of zeste homolog 2 2013;155(6):1309–22. (EZH2) characterizes an aggressive subset of prostate cancers and 114. Blee AM, Liu S, Wang L, Huang H. BET bromodomain-mediated predicts patient prognosis independently from pre- and postoper- interaction between ERG and BRD4 promotes prostate cancer cell atively assessed clinicopathological parameters. Carcinogenesis. invasion. Oncotarget. 2016;7(25):38319–32. 2015;36(11):1333–40. 115. Janouskova H, El Tekle G, Bellini E, Udeshi ND, Rinaldi A, 99. Lee TI, Jenner RG, Boyer LA, Guenther MG, Levine SS, Kumar Ulbricht A, et al. Opposing effects of cancer-type-specific SPOP RM, et al. Control of developmental regulators by Polycomb in mutants on BET protein degradation and sensitivity to BET inhib- human embryonic stem cells. Cell. 2006;125(2):301–13. itors. Nat Med. 2017;23(9):1046–54. 100. Liu L, Xu Z, Zhong L, Wang H, Jiang S, Long Q, et al. Enhancer 116. Zhang P, Wang D, Zhao Y, Ren S, Gao K, Ye Z, et al. Intrinsic of zeste homolog 2 (EZH2) promotes tumour cell migration and BET inhibitor resistance in SPOP-mutated prostate cancer is me- invasion via epigenetic repression of E-cadherin in renal cell car- diated by BET protein stabilization and AKT-mTORC1 activation. cinoma. BJU Int. 2016;117(2):351–62. Nat Med. 2017;23(9):1055–62. 101. Xu H, Xu K, He HH, Zang C, Chen CH, Chen Y, et al. Integrative 117. Gupta-Elera G, Garrett AR, Robison RA, O'Neill KL. The role of analysis reveals the transcriptional collaboration between EZH2 oxidative stress in prostate cancer. Eur J Cancer Prev. 2012;21(2): and E2F1 in the regulation of cancer-related gene expression. Mol 155–62. Cancer Res. 2016;14(2):163–72. 102. Cha TL, Zhou BP, Xia W, Wu Y, Yang CC, Chen CT, et al. Akt- 118. Hussong M, Borno ST, Kerick M, Wunderlich A, Franz A, mediated phosphorylation of EZH2 suppresses methylation of ly- Sultmann H, et al. The bromodomain protein BRD4 regulates sine 27 in histone H3. Science. 2005;310(5746):306–10. the KEAP1/NRF2-dependent oxidative stress response. Cell 103.�� Xu K, Wu ZJ, Groner AC, He HH, Cai C, Lis RT, et al. EZH2 Death Dis. 2014;5:e1195. oncogenic activity in castration-resistant prostate cancer cells is 119. Theurillat JP, Udeshi ND, Errington WJ, Svinkina T, Baca SC, Polycomb-independent. Science. 2012;338(6113):1465–9. This Pop M, et al. Prostate cancer. Ubiquitylome analysis identifies paper provides compelling evidence that the oncogenic role dysregulation of effector substrates in SPOP-mutant prostate can- of Ezh2 in CRPC is partly independent of its repressive func- cer. Science. 2014;346(6205):85–9. tion and acts via coactivator functions with key transcription 120. Kikuchi M, Okumura F, Tsukiyama T, Watanabe M, Miyajima N, factors including AR Tanaka J, et al. TRIM24 mediates ligand-dependent activation of 104. Labbe DP, Sweeney CJ, Brown M, Galbo P, Rosario S, Wadosky androgen receptor and is repressed by a bromodomain-containing KM, et al. TOP2A and EZH2 provide early detection of an ag- protein, BRD7, in prostate cancer cells. Biochim Biophys Acta. gressive prostate cancer subgroup. Clin Cancer Res. 2017;23(22): 2009;1793(12):1828–36. 7072–83. 121. Grasso CS, Wu YM, Robinson DR, Cao X, Dhanasekaran SM, 105. Urbanucci A, Mills IG. Bromodomain-containing proteins in Khan AP, et al. The mutational landscape of lethal castration- prostate cancer. Mol Cell Endocrinol. 2017; resistant prostate cancer. Nature. 2012;487(7406):239–43. 106. Fujisawa T, Filippakopoulos P. Functions of bromodomain- 122. Zhao D, Lu X, Wang G, Lan Z, Liao W, Li J, et al. Synthetic containing proteins and their roles in homeostasis and cancer. essentiality of chromatin remodelling factor CHD1 in PTEN- Nat Rev Mol Cell Biol. 2017;18(4):246–62. deficient cancer. Nature. 2017;542(7642):484–8. 107.�� Loven J, Hoke HA, Lin CY, Lau A, Orlando DA, Vakoc CR, et al. 123. Zaret KS, Carroll JS. Pioneer transcription factors: establishing Selective inhibition of tumor oncogenes by disruption of super- competence for gene expression. Genes Dev. 2011;25(21):2227– enhancers. Cell 2013; 153(2):320–334. 108.�� Whyte WA, Orlando DA, Hnisz D, Abraham BJ, Lin CY, Kagey 124. Lee CS, Friedman JR, Fulmer JT, Kaestner KH. The initiation of MH, et al. Master transcription factors and mediator establish liver development is dependent on Foxa transcription factors. super-enhancers at key cell identity genes. Cell. 2013;153(2): Nature. 2005;435(7044):944–7. 307–19. This two papers provide evidence that disruption of Curr Mol Bio Rep (2018) 4:101–115 115 125. Watt AJ, Zhao R, Li J, Duncan SA. Development of the mamma- 140. Antony L, van der Schoor F, Dalrymple SL, Isaacs JT. Androgen receptor (AR) suppresses normal human prostate epithelial cell lian liver and ventral pancreas is dependent on GATA4. BMC Dev Biol. 2007;7(1):37. proliferation via AR/beta-catenin/TCF-4 complex inhibition of c- 126. Zhao R, Watt AJ, Li J, Luebke-Wheeler J, Morrisey EE, Duncan MYC transcription. Prostate. 2014;74(11):1118–31. SA. GATA6 is essential for embryonic development of the liver 141. Chen H, Tu SW, Hsieh JT. Down-regulation of human DAB2IP but dispensable for early heart formation. Mol Cell Biol. gene expression mediated by polycomb Ezh2 complex and his- 2005;25(7):2622–31. tone deacetylase in prostate cancer. J Biol Chem. 2005;280(23): 127. Holtzinger A, Evans T. Gata4 regulates the formation of multiple 22437–44. organs. Development. 2005;132(17):4005–14. 142. Min J, Zaslavsky A, Fedele G, McLaughlin SK, Reczek EE, De 128. Wang Q, Li W, Zhang Y, Yuan X, Xu K, Yu J, et al. Androgen Raedt T, et al. An oncogene-tumor suppressor cascade drives met- receptor regulates a distinct transcription program in androgen- astatic prostate cancer by coordinately activating Ras and nuclear independent prostate cancer. Cell. 2009;138(2):245–56. factor-kappaB. Nat Med. 2010;16(3):286–94. 129. Wang D, Garcia-Bassets I, Benner C, Li W, Su X, Zhou Y, et al. 143. Aytes A, Mitrofanova A, Lefebvre C, Alvarez MJ, Castillo-Martin Reprogramming transcription by distinct classes of enhancers M, Zheng T, et al. Cross-species regulatory network analysis iden- functionally defined by eRNA. Nature. 2011;474(7351):390–4. tifies a synergistic interaction between FOXM1 and CENPF that 130. Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka T, Tomoda drives prostate cancer malignancy. Cancer Cell. 2014;25(5):638– K, et al. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell. 2007;131(5):861–72. 144. Aytes A, Mitrofanova A, Kinkade CW, Lefebvre C, Lei M, Phelan 131. Rybak AP, Tang D. SOX2 plays a critical role in EGFR-mediated V, et al. ETV4 promotes metastasis in response to activation of self-renewal of human prostate cancer stem-like cells. Cell Signal. PI3-kinase and Ras signaling in a mouse model of advanced pros- 2013;25(12):2734–42. tate cancer. Proc Natl Acad Sci U S A. 2013;110(37):E3506–15. 132.�� Bishop JL, Thaper D, Vahid S, Davies A, Ketola K, Kuruma H, 145. Mitrofanova A, Aytes A, Zou M, Shen MM, Abate-Shen C, et al. The master neural transcription factor BRN2 is an androgen Califano A. Predicting drug response in human prostate cancer receptor-suppressed driver of neuroendocrine differentiation in from preclinical analysis of in vivo mouse models. Cell Rep. prostate cancer. Cancer Discov. 2017;7(1):54–71. This works 2015;12(12):2060–71. provides a mechanistic model for neuroendocrine differentia- 146. Kar S, Sengupta D, Deb M, Pradhan N, Patra SK. SOX2 function tion after AR signaling blockade using patiend-derived models and Hedgehog signaling pathway are co-conspirators in promot- 133. Zeng W, Sun H, Meng F, Liu Z, Xiong J, Zhou S, et al. Nuclear C- ing androgen independent prostate cancer. Biochim Biophys Acta. MYC expression level is associated with disease progression and 2017;1863(1):253–65. potentially predictive of two year overall survival in prostate can- 147. Ding L, Chen S, Liu P, Pan Y, Zhong J, Regan KM, et al. CBP loss cer. Int J Clin Exp Pathol. 2015;8(2):1878–88. cooperates with PTEN haploinsufficiency to drive prostate cancer: 134. Rye M, Bertilsson H, Drabløs F, Angelsen A, Bathen T, Tessem implications for epigenetic therapy. Cancer Res. 2014;74(7): M-B. Gene signatures ESC, MYC and ERG-fusion are early 2050–61. markers of a potentially dangerous subtype of prostate can- 148. Li D, Sun H, Sun WJ, Bao HB, Si SH, Fan JL, et al. Role of cer2014. 50 p. RbBP5 and H3K4me3 in the vicinity of Snail transcription start 135. Kim J, Roh M, Doubinskaia I, Algarroba GN, Eltoum IE, site during epithelial-mesenchymal transition in prostate cancer Abdulkadir SA. A mouse model of heterogeneous, c-MYC- cell. Oncotarget. 2016;7(40):65553–67. initiated prostate cancer with loss of Pten and p53. Oncogene. 149. Dong C, Wu Y, Yao J, Wang Y, Yu Y, Rychahou PG, et al. G9a 2012;31(3):322–32. interacts with Snail and is critical for Snail-mediated E-cadherin 136. Hubbard GK, Mutton LN, Khalili M, McMullin RP, Hicks JL, repression in human breast cancer. J Clin Invest. 2012;122(4): Bianchi-Frias D, et al. Combined MYC activation and Pten loss 1469–86. are sufficient to create genomic instability and lethal metastatic 150. Liu S, Ye D, Guo W, Yu W, He Y, Hu J, et al. G9a is essential for prostate cancer. Cancer Res. 2016;76(2):283–92. EMT-mediated metastasis and maintenance of cancer stem cell- 137. Koh CM, Iwata T, Zheng Q, Bethel C, Yegnasubramanian S, De like characters in head and neck squamous cell carcinoma. Marzo AM. Myc enforces overexpression of EZH2 in early pros- Oncotarget. 2015;6(9):6887–901. tatic neoplasia via transcriptional and post-transcriptional mecha- 151. Liu Y, Zeng S, Jiang X, Lai D, Su Z. SOX4 induces tumor inva- nisms. Oncotarget. 2011;2(9):669–83. sion by targeting EMT-related pathway in prostate cancer. Tumour 138. Fan L, Peng G, Sahgal N, Fazli L, Gleave M, Zhang Y, et al. Biol. 2017;39(5):1010428317694539. Regulation of c-Myc expression by the histone demethylase 152. Bilir B, Osunkoya AO, WGt W, Sannigrahi S, Lefebvre V, JMJD1A is essential for prostate cancer cell growth and survival. Metzger D, et al. SOX4 is essential for prostate tumorigenesis Oncogene. 2016;35(19):2441–52. initiated by PTEN ablation. Cancer Res. 2016;76(5):1112–21. 139. Vander Griend DJ, Litvinov IV, Isaacs JT. Conversion of androgen receptor signaling from a growth suppressor in normal prostate 153. Mahajan K, Malla P, Lawrence HR, Chen Z, Kumar-Sinha C, epithelial cells to an oncogene in prostate cancer cells involves a Malik R, et al. ACK1/TNK2 regulates histone H4 Tyr88- gain of function in c-Myc regulation. Int J Biol Sci. 2014;10(6): phosphorylation and AR gene expression in castration-resistant 627–42. prostate cancer. Cancer Cell. 2017;31(6):790–803.e8.

Journal

Current Molecular Biology ReportsSpringer Journals

Published: Apr 18, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off