Hypoxia and Hormone-Mediated Pathways Converge at the Histone Demethylase KDM4B in Cancer

Hormones play an important role in pathophysiology. The hormone receptors, such as estrogen receptor alpha and androgen receptor in breast cancer and prostate cancer, are critical to cancer cell proliferation and tumor growth. In this review we focused on the cross-talk between hormone and hypoxia pathways, particularly in breast cancer. We delineated a novel signaling pathway from estrogen receptor to hypoxia-inducible factor 1, and discussed the role of this pathway in endocrine therapy resistance. Further, we discussed the estrogen and hypoxia pathways converging at histone demethylase KDM4B, an important epigenetic modifier in cancer.

The exact nature of the relationship between hypoxia and estrogen pathways was a puzzle until our recent findings showing that the HIF-1α gene is a direct target of ERα [59]. In this study, we analyzed the global gene expression profile in response to hypoxia and the ERα antagonist fulvestrant and found a subgroup of genes that were dually responsive to the hormone and to oxygen. These genes were upregulated by hypoxia but the ERα antagonist fulvestrant significantly reduced their expression. These data were consistent with previous studies that showed some genes, such as KDM4B, STC2, and VEGF, bear both a hypoxia response element and estrogen response element [60][61][62][63][64][65][66]. Most interestingly, we found that ERα signaling directly regulates HIF-1α expression. When MCF7 cells were grown without estrogen and then placed in hypoxia or treated with the hypoxia mimetic deferoximine, estradiol greatly enhanced HIF-1α expression and this was reversed by fulvestrant and ERα depletion. By analyzing the HIF-1α genomic sequence that bears 15 exons and 14 introns, we identified a canonical estrogen response element (ERE) located in the first intron ( Figure 1A). Interestingly, there is also a FOXA1 binding site that is 64 nucleotides downstream of ERE, further supporting it as a bona fide ERα binding element, because FOXA1 is a pioneer factor that facilitates ERα recruitment [67]. Actually, one study has shown that overexpression of FOXA1 in ER-positive breast cancer cell lines promotes resistance to tamoxifen and to estrogen deprivation [68]. We further validated our findings by chromatin immunoprecipitation-PCR and a luciferase reporter assay, showing that ERα directly binds to this locus, driving HIF-1α gene expression. This finding not only explains the early findings that estrogen and hypoxia pathways crosstalk, but also indicates that overactive HIF-1α function may partially compensate for estrogen signaling when ERα function is compromised, such as in the circumstance of hormone therapy, leading to hormone therapy resistance ( Figure 1B,C). ERα is bound by its ligand it drives the expression of HIF-1α. However, ERα antagonists block the expression of HIF-1α; (C) The pathways mediated by hypoxia, estrogen, metabolites, and cancer genes converge on HIF-1α, which drives a plethora of genes that are involved in multiple biological processes, cancer progression, and therapeutic resistance.

The Hypoxia and Estrogen Signaling Pathways Converge on Histone Demethylases
Genetic abnormalities that drive tumorigenesis are usually coupled with epigenetic alterations that engage multiple important biological processes such as DNA replication, DNA repair, and gene expression [69][70][71][72][73]. One such aberrant chromatin modification is histone lysine methylation [69,70], which was believed to be irreversible until the discovery of lysine-specific demethylase 1 (LSD1) [74].
The KDM4 subfamily of histone demethylases consists of four members. KDM4A, KDM4B, and KDM4C share high sequence homology in their catalytic domains, and they remove methyl groups from H3K9me2/me3 and H3K36me2/me3 [75]. KDM4A-4C members also bear other similar functional domains that include two PHD domains and two Tudor domains. However, KDM4D is less conserved and removes methyl groups only from H3K9me2/me3. KDM4B plays important roles in the self-renewal of embryonic stem cells and the conversion of induced pluripotent stem cells [85,86], and is linked to many forms of cancer [87]. KDM4B is amplified in medulloblastoma [88] and malignant peripheral nerve sheath tumors [89], and is overexpressed in many other cancers [90][91][92]. KDM4B regulates the expression of key oncogenes, such as C-MYC [93][94][95][96] and CDK6 [97], and is involved in cancer invasiveness, metastasis, and therapeutic resistance [98][99][100]. Interestingly, KDM4B is a direct target of p53, exerting its DNA repair function in response DNA damage [101][102][103].
Recently, we showed that KDM4B is involved in neuroblastoma growth and tumor maintenance [104]. The expression of KDM4B was highly correlated with that of the MYCN oncogene in neuroblastoma, and it formed a complex with N-Myc protein, thereby facilitating its function by maintaining low levels of repressive H3K9me2/me3 marks at Myc-binding sites. In breast cancer we have shown that HIF-1α and ERα can coordinate expression of genes, such as KDM4B, whose expression is driven by both ERα and HIF-1α and epigenetically regulates the G2/M phase of cell cycle progression in breast cancer cells [63] and other cancer cell lines (unpublished data), as the expression of several key cell cycle genes is correlated with changes in the KDM4B substrate, H3K9me3 [63,104]. Similar to other dual responsive genes such as VEGF and STC2 that are regulated by both estrogens and hypoxia, the genomic locus of KDM4B bears both HIF-1α and ERα binding elements [60,63] (Figure 2A). The cross-talk between HIF-1α and ERα converges at KDM4B, which is important for cell cycle progression and tumor growth in ER positive breast cancer [63,93]. Importantly, in endocrine therapy-resistant breast cancer cells, the regulation of KDM4B by HIF-1α and ERα is intact and KDM4B is still required for G2/M phase progression [59] ( Figure 2B). In addition, KDM4B is not only required for enhancing androgen receptor (AR) transcriptional activity through histone modification, but it also enhances AR protein stability via inhibition of AR ubiquitination [105], demonstrating the functional connection between AR and KDM4B in prostate cancer. Therefore, HIF-1α plays an important role in modulating anti-androgen responses via KDM4B in prostate cancer.

Future Prospects
Many transcriptional factors, such as Myc, ERα, and AR, exert oncogenic functions to drive cancer cell proliferation. Directly targeting these oncogenic transcription factors is either technically challenging or leads to therapeutic resistance. Therefore, new approaches need to be developed to overcome these obstacles. Transcription factors need to complex with other cofactors to drive gene expression and many of these cofactors are histone modifiers. Thus, development of small molecules to target the histone modifiers, such as KDM4B, may provide an opportunity to enhance the efficacy of standard chemotherapeutics or to overcome drug resistance. Recently, efforts have been made by us and other groups to identify and develop KDM4B inhibitors for cancer treatment [106][107][108][109]. By using a chemoinformatics in combination with high-content imaging approach we identified ciclopirox as a novel histone demethylase inhibitor. Ciclopirox targeted KDM4B, inhibited Myc signaling, resulting in suppression of neuroblastoma cell viability and tumor growth associated with an induction of differentiation [107]. We also found that MCF7 cells (ERα-positive) were much more sensitive to MDA-MB-231 cells (ERα-negative) (Jun Yang, St Jude Children's Research Hospital, Memphis, TN, USA. unpublished data), suggesting ERα-positive breast cancer cells are more addicted to KDM4B. Chu et al. identified a KDM4B inhibitor that significantly blocked the viability of cultured prostate cancer cells, which was accompanied by transcriptional silencing of growth-related genes, a substantial portion of which were AR-responsive [106]. Recently, a more potent and selective KDM4 inhibitor was developed by Cellgene [108,109], which was efficacious in breast and colon cancer models. Although whether these KDM4B inhibitors are able to reverse endocrine therapy resistance needs to be tested, we believe specific and potent KDM4B inhibitors hold a promise for overcoming endocrine therapy resistance to breast cancer and prostate cancer.