Introduction
Prostate cancer (PCa) is the second most common malignancy in men and the fifth leading cause of death worldwide (Rawla, 2019). Patients with early-stage prostate cancer have a good prognosis (Trewartha and Carter, 2013). However, advanced prostate cancer is associated with high mortality with the development of invasion and metastasis (Zong and Goldstein, 2013). Since Huggins and Hodges discovered that PCa responds to androgen therapy (R, 2002), androgen-deprivation therapy (ADT) has become the first-line treatment use to suppress PCa progression. Despite an initial therapeutic efficiency of 80-90% (Trewartha and Carter, 2013), nearly all patients treated with ADT will ultimately develop castration-resistant prostate cancer (CRPC) (Davies, Conteduca et al., 2019).
Patients with CRPC show a sustained increase in serum prostate-specific antigen (PSA) levels after castration treatment (Saad, Chi et al., 2015), which is generally attributed to the reaction of the androgen receptor (AR), which may be the result of AR gene amplification or other mechanisms (Cai and Balk, 2011 , Tran, Ouk et al., 2009). The androgen biosynthesis inhibitor abiraterone and AR antagonist enzalutamide (Enz) are two novel agents targeting androgen action to treat patients with CRPC and were approved by the Food and Drug Administration (FDA) in 2012 (Nelson and Yegnasubramanian, 2013). Enz is a potent AR signalling inhibitor without agonistic activity. It suppresses the nuclear translocation of activated AR to prevent its localization to androgen response elements and coactivator recruitment, which induces apoptosis while inhibiting proliferation of the CRPC cells (Erdogan, 2018 , Jiang, Chen et al., 2020). A phase III clinical trial reported that patients treated with Enz survived months longer than those in the placebo group (Niu, Guo et al., 2018). Despite the initial therapeutic effect of Enz, CRPC patients will eventually become resistant to it (Erdogan, 2018). Therefore\sout, it is urgent to explore new strategies to resolve resistance to Enz.
In this review, we summarized the emerging information on Enz-resistance mechanisms in CRPC, including AR-related signalling pathways, lineage plasticity, cytokines dysregulation, and gene polymorphisms. Furthermore, new potential therapeutic strategies to treat resistant CRPC which are promising for better therapeutic prognoses are introduced. We hope that these recently discovered mechanisms and emerging solutions provide some inspiration for the treatment of Enz-resistant CRPC.
Mechanisms ofEnz resistance in CRPC
Restoration ofa ndrogen receptor signalling
Androgen receptor reactivation
AR belongs to the steroid hormone receptor family and functions as a transcription factor. It contains three different functional domains: the N-terminal domain (NTD), the DNA-binding domain (DBD), and the carboxy-terminal ligand-binding domain (LBD). The main role of AR is to regulate genes encoding protein transcription needed for prostate function and the promotion of cellular differentiation in the normal prostate (Wong, Ferraldeschi et al., 2014).
Recently, several studies have demonstrated that AR is highly expressed and transcriptionally active in CRPC despite Enz treatment (Yuan, Cai et al., 2014). David Y. et al. (Takeda, Spisák et al., 2018) utilized an integrative analysis of ChIP-seq in primary specimens and identified a somatically acquired AR enhancer located 650 kb centromeric to the AR, which is frequently amplified in CRPC. Under low androgen conditions, additional copies inserted into this region are sufficient to increase cell proliferation and reduce cell sensitivity to Enz. Targeting this AR enhancer may become a potential method for treating resistant CRPC. Galectin-3, a member of the animal lectin family, significantly inhibited the therapeutic effect of Enz by increasing the expression of several AR target genes, such as kallikrein-related peptidase 3 (KLK3) and transmembrane protease serine 2 (TMPRSS2). Such enhancement of AR transcriptional activity and expression of AR-related genes leads to cell resistance to Enz (Dondoo, Fukumori et al., 2017). In contrast to its direct influence on AR expression, BMI1, a polycomb group protein, binds the androgen receptor, preventing ubiquitin E3 ligase-mediated AR protein degradation, which results in sustained AR signalling in CRPC cells (Zhu, Zhao et al., 2018). By constructing an Enz-resistant xenograft model, the authors found that the BMI1 inhibitor PTC209 significantly decreased Enz-resistant CRPC tumour growth. Thus, BMI1 is a novel therapeutic target for CRPC.
Recently, Hwang et al. (Hwang, Seo et al., 2019) indicated that resistant prostate cancer overexpressed or amplified cAMP-response element-binding protein 5 (CREB5), mediating the enhancement of AR activity at a subset of promoters and enhancers, including Myc and genes involved in the cell cycle. Notably, the pioneering factor forkhead box protein A1 (FOXA1) was also bound to CREB5/AR-binding sites and was necessary for CREB5-mediated resistance, which was not eliminated by Enz. The expression of the proto-oncogene c-Myc is positively associated with AR expression (Bai, Cao et al., 2019). Alex Bainbridge et al. (Bainbridge, Walker et al., 2020) observed the significant inhibition of AR-regulated gene expression upon siRNA-mediated I-kappa-B kinase (IKBKE) depletion or pharmacological inhibition of IKBKE in an LNCaP-derived cell line resistant to Enz. Further analysis showed that IKBKE regulated AR levelsvia Hippo pathway inhibition to reduce c-Myc levels at cis-regulatory elements within the AR gene, leading to the formation of a large pool of AR. Another study showed that erythropoietin-producing human hepatocellular (Eph) receptors positively regulated AR by inducing proto-oncogene c-Myc expression, which was critical for Enz resistance. To date, several EphB4 inhibitors have been tested in clinical trials (Li, Lanman et al., 2020). Considering the low AR conditions after AR inhibitor treatments, Liu et al. discovered that deficient AR altered the balance between eukaryotic initiation factor 4E-binding protein 1(4EBP1) and eukaryotic translation initiation factor 4F (eIF4F) complex formation (Liu, Horn et al., 2019). The loss of AR increased eIF4F assembly to drive the translation of a network of pro-proliferation mRNAs that were needed for tumour growth. Thus, the suppressing eIF4F may be a possible strategy for treating Enz-resistant CRPC.
Moreover, inhibitors of signal transducer and activator of transcription 3 (STAT3) and transactivation activity, namely, galiellalactone, (GPA500) (Thaper, Vahid et al., 2018), and triptolide (Han, Huang et al., 2017), have been proven to overcome drug resistance by targeting AR signalling. Tetra-aryl cyclobutanes are novel AR antagonists that are structurally different from Enz and inhibit the growth of Enz-resistant xenograft tumours (Pollock, Wardell et al., 2016). Proteolysis-targeting chimaeras (PROTACs) are used in an emerging approach that has recently attracted considerable attention. They are small molecules that control intracellular protein levels by recruiting targeted proteins into the ubiquitin-proteasome system (UPS) for selective degradation (Toure and Crews, 2016). Jemilat Salami et al. (Salami, Alabi et al., 2018) performed a head-to-head comparison of Enz and the PROTAC derivative ARCC-4 by utilizing different cellular models of PCa drug resistance. The results indicated that ARCC-4 was better at overcoming resistance in CRPC cellular models. Furthermore, at low nanomolar concentrations ARCC-4 inhibited the proliferation of AR-amplified PCa cells. Hence, PROTAC-mediated AR degradation is a very promising method of Enz-resistant CRPC therapy. In addition to protein degradation by PROTACs, the orally bioavailable AR degrader UT-34 and interferon regulatory factor 8 (IRF8)-induced drugs are alternative strategies for Enz-resistant CRPC treatment (Ponnusamy, He et al., 2019 , Wu, You et al., 2020).
Androgen receptor splice variants
AR is composed of NTD, DBD, and LBD. When LBD binds to a ligand, the conformational changes of LBD allow AR to be transported into the nucleus where it binds to DNA. AR can undergo alternative splicing to generate isoforms that contain exons 1 to 3 with a deleted LBD. A truncated AR protein without the LBD fails to respond to AR targeting by Enz (J, Bubley et al., 2017).
AR variant 7 (AR-V7) is the most common variant that detected in circulating tumour cells from patients with CRPC, and it is significantly increased upon Enz therapy (Antonarakis, Lu et al., 2014). The splicing factor hnRNPA1 and the long non-coding RNA (lncRNA) Malat1 have been identified as contributors to Enz resistance by promoting the generation and expression of AR-V7 (Stone, 2017). Furthermore, it has been reported that tyrosine kinase ACK1 (TNK2) phosphorylates histone H4 functioned as an epigenetic modifier mediatingc a feedforward ACK1/pY88-H4/WDR5/MLL2/AR epigenetic circuit. The transcription of AR and AR-V7 was increased, leading to Enz resistance (Mahajan, Malla et al., 2017). Recently, Zhao et al. (Zhao, Peacock et al., 2019) found that arginine vasopressin receptor 1a (AVPR1A) was the most downregulated gene after the depletion of AR-V7 or the versatile coactivator VAV3. Therefore, suppressing AVPR1A in CRPC cells can decrease cell proliferation. Utilizing AVPR1A as a therapeutic target is effective not only for CRPC but also for bone-metastatic CRPC. High AR/AR-V7 expression and a positive correlation of AR-V7 with an E2F score indicated a supraphysiological testosterone (SPT) response (Lam, Nguyen et al., 2020). SPT produced an enduring response as shown by the sustained inhibition of AR-V7, E2F transcriptional output, and the DNA damage response (DDR) transcriptome. Hence, SPT treatment is a potent candidate for treating Enz-resistant CRPC. Moreover, CRPC cells can recruit more natural killer (NK) cells than normal prostate epithelial cells in the PCa microenvironment. NK cells altered microRNA-34 and microRNA-449 to suppress AR-V7 expression and AR-V7-induced enhancer of zeste homologue 2 (EZH2) expression. Utilizing these immune cells provides a novel angle for approaching Enz resistance (Lin, Chou et al., 2017).
Protein homeostasis is another potential mechanism leading to drug resistance. Here, Liu et al. (Liu, Lou et al., 2018) demonstrated that UPS was inhibited in Enz-resistant prostate cancer. The interaction between heat shock protein family member HSP70 and functional E3 ubiquitin ligase STUB1 was required for AR/AR-V7 proteostasis. Suppressing HSP70 significantly inhibited tumour growth and resistance to Enz by decreasing AR-V7 expression. Similarly, aldo-keto reductase family 1 member C3 (AKR1C3) enhanced AR-V7 protein stability viathe alteration of the UPS. AKR1C3-targeted indomethacin greatly reduced AR/AR-V7 protein expression in vitro and in vivo through the activation of the ubiquitin-mediated proteasome pathway (Liu, Yang et al., 2019). A preclinical study indicated that the anti-apoptotic B cell lymphoma-2 (BCL2) protein inhibitor ABT263 increased ROS levels, leading to the inhibition of ubiquitin specific protease 26 (USP26) activity. As a result, the ubiquitination and ubiquitin-proteasome-dependent degradation of AR and AR-V7 were increased (Xu, Sun et al., 2020). Other agents, including fatty acid synthase inhibitor (FASN) IPI-9119, luteolin, triterpenoid antioxidant drug bardoxolone-methyl (CDDO-Me), and GnRH antagonist degarelix (Firmagon), have also been identified as inhibitors of the AR-V7 protein in Enz-resistant CRPC (Zadra, Ribeiro et al., 2019 , Naiki-Ito, Naiki et al., 2019 , Khurana, Chandra et al., 2020 , Cucchiara, Yang et al., 2019).
In prostate cancer cells, c-Myc is one of the most overexpressed genes and is frequently upregulated in CRPC. c-Myc promotes AR gene transcription and enhances the stability of full-length AR (AR-FL) and AR-V proteins without altering AR RNA splicing. The inhibition of c-Myc, such as with 5-Lox inhibitor MK591/quiflapon, can interrupt oncogenic c-Myc signalling to kill Enz-resistant cells (Bai, Cao et al., 2019 , Monga, Subramani et al., 2020). GATA2 is also a critical regulator of AR-Vs, and its cistromes considerably overlap with bromodomain and extra-terminal (BET) proteins (Chaytor, Simcock et al., 2019). Small-molecule inhibitors of GATA2 are potent agents that suppress the expression and transcriptional function of both AR-FL and AR-Vs (He, Lanz et al., 2014). Moreover, the selective inhibitors of the small-molecule BET, namely, JQI and I-BET762, have also been proven to be effective in treating Enz-resistant CRPC by targeting the amino-terminal bromodomains of bromodomain-containing protein 4 (BRD4) (Asangani, Dommeti et al., 2014 , Chan, Selth et al., 2015). In addition, later research demonstrated that the conserved zinc-finger transcription factor (ZFX) interacted with AR-V7 and co-occupied the unique AR-V7-binding sites in target gene promoters in CRPC cells. Targeting this ZFX can provide a more effective Enz-resistant CRPC treatment (Cai, Tsai et al., 2018).
Considering the difficulty in distinguishing the activity of AR-Vs from that of AR-FL, there is urgent demand for AR-V cell models that enable the identification of AR-V-specific functional and phenotypic read-outs that are not influenced by AR-FL. Kounatidou et al. (Kounatidou, Nakjang et al., 2019) established CWR22Rv1-AR-EK cells in which the expression of AR-FL was abrogated and all endogenous AR-Vs were retained through CRISPR/Cas9 editing. By utilizing this model, they validated that AR-Vs regulated a DDR gene network, including an AR-FL-like ‘BRCAness’ signature, which was essential for cell survival after ionizing radiation treatment. Moreover, AR-Vs interacted with poly (ADP-ribose) polymerase (PARP) and depended on PARP activity for transcriptional function. Notably, suppressing PARP influenced AR-V gene expression and reduced the growth of CRPC cells. The successful establishment of this model provided a platform for accurate studies of AR-Vs in CRPC.