Associate editor: J.L. Turgeon
Androgen receptors in hormone-dependent and castration-resistant prostate cancer

https://doi.org/10.1016/j.pharmthera.2013.07.003Get rights and content

Abstract

In the United States, prostate cancer (PCa) is the most commonly diagnosed non-cutaneous cancer in males and the second leading cause of cancer-related death for men. The prostate is an androgen-dependent organ and PCa is an androgen-dependent disease. Androgen action is mediated by the androgen receptor (AR), a hormone activated transcription factor. The primary treatment for metastatic PCa is androgen deprivation therapy (ADT). For the most part, tumors respond to ADT, but most become resistant to therapy within two years. There is persuasive evidence that castration resistant (also termed castration recurrent) PCa (CRPC) remains AR dependent. Recent studies have shown that there are numerous factors that contribute to AR reactivation despite castrate serum levels of androgens. These include changes in AR expression and structure through gene amplification, mutation, and alternative splicing. Changes in steroid metabolism, cell signaling, and coregulator proteins are also important contributors to AR reactivation in CRPC. Most AR targeted therapies have been directed at the hormone binding domain. The finding that constitutively active AR splice variants that lack the hormone binding domain are frequently expressed in CRPC highlights the need to develop therapies that target other portions of AR. In this review, the role of AR in normal prostate, in PCa, and particularly the mechanisms for its reactivation subsequent to ADT are summarized. In addition, recent clinical trials and novel approaches to target AR are discussed.

Introduction

In the United States, prostate cancer (PCa) is the most commonly diagnosed non-cutaneous cancer in males and the second leading cause of cancer-related death for men (Siegel et al., 2013). The American Cancer Society estimates that there will be approximately 238,590 men diagnosed with PCa and that 29,720 will die from PCa related causes in 2013 (Siegel et al., 2013). The disease typically appears in men 65 and older. According to the American Cancer Society, PCa is most common in black men, followed by Caucasians, and least common in Asians (Potosky et al., 1995). Some, but not all, of these differences may be due to disparities in screening for PCa (Homma et al., 1997). In 1941, long before the androgen receptor was identified, Huggins and Hodges found that there is a relationship between androgens and prostate or PCa growth. They found that administration of testosterone increased prostate growth in animal models. Moreover, castration of patients greatly reduced levels of testosterone resulting in regression of advanced PCa. Huggins' work establishing that PCa was androgen dependent and could be treated by castration earned him the Nobel Prize in 1966 (Toledo-Pereyra, 2001). Androgen action is dependent on the androgen receptor (AR), a hormone activated transcription factor (Agoulnik et al., 2006). This review will discuss the role of AR in normal prostate, in PCa, and the mechanisms for its reactivation subsequent to androgen deprivation therapy (ADT) resulting in progression to a castration resistant form of PCa. Finally, potential novel treatments and clinical trials that target AR will be discussed.

AR is a member of the nuclear steroid receptor superfamily of transcription factors and is classified as NR3C4 (nuclear receptor subfamily 3, group C, member 4) (Lu et al., 2006). The AR gene is located on Xq11-12; thus, males have a single copy of the gene and inactivating mutations result in androgen insensitivity syndrome (AIS) (Hughes et al., 2012). AR contains eight exons that encode a protein of ~919 amino acids (Fig. 1) (Gelmann, 2002, Koochekpour, 2010). The variation in length is a result of a variable length polyglutamine repeat (19–25 for most men) (Giovannucci et al., 1997) and a variable polyglycine repeat. Both repeats are in the amino terminal domain encoded by exon 1 and thus amino acid numbering in the literature can be inconsistent. The reference length of 919 has been used by the AR gene mutations database at McGill (Gottlieb et al., 2004) and was based on 21 glutamines and a polyglycine tract of 24. In 2012, the database switched to NCBI reference sequence NM_000044.2, which has 23 glutamines and a shorter polyglycine tract (23) for a total of 920 amino acids (Gottlieb et al., 2012). In this review, the original 919 nomenclature is used to be consistent with the cited papers. Shorter glutamine repeats typically result in higher levels of transcriptional activity in multiple cell types (Beilin et al., 2000). There is some evidence that there is a higher risk of PCa in men that have AR with shorter CAG repeats (Giovannucci et al., 1997). In contrast, expansion of this repeat (typically >40 Gln) is associated with SBMA (spinal and bulbar muscular atrophy) (Fischbeck et al., 1999). Like most nuclear receptors, AR is composed of distinct functional motifs. These include the amino-terminal domain (NTD encoded by exon 1), DNA-binding domain (DBD encoded by exons 2 and 3), a hinge region (H encoded by the 5′ portion of exon 4), and a ligand-binding domain (LBD encoded by the remainder of exon 4 through exon 8) (Fig. 1) (Gelmann, 2002). The AF-1 and AF-2 transactivation domains required for optimal transactivation are located in the NTD and the LBD respectively (Gelmann, 2002).

The unliganded AR is inactive and is bound to cytoplasmic chaperones including Hsp90 (heat shock protein 90) (Fig. 2) (Roy et al., 1999, Kim et al., 2009). The major circulating androgen is testosterone, produced in the testes. In the prostate as well as in a limited number of other tissues, testosterone is converted to dihydrotestosterone (DHT) by 5α-reductase enzymes. Although testosterone and DHT (see Fig. 3 for structures) both bind to and activate AR, DHT has a substantially higher affinity for AR (Lindzey et al., 1994) and is the primary androgen in the prostate (Roy et al., 1999). Hormone binding induces a conformational change resulting in dissociation of cytoplasmic chaperones and revealing the nuclear localization signal (Fig. 2). The hormone-bound AR dimerizes and translocates to the nucleus, where it binds to DNA and interacts with a series of transcriptional coregulators to regulate target gene expression (Ai et al., 2009, Koochekpour, 2010). More than 150 proteins have been identified as AR coregulators (Heemers & Tindall, 2007). These include the p160 (SRC-1, SRC-2/TIF2, SRC-3/AIB1) family, p300/CBP, ARA54, ARA55, and ARA70 and many other proteins (Wolf et al., 2008). Many of the coregulators are enzymes (histone acetyltransferases, methyl transferases, and kinases) that act to open chromatin structure to facilitate transcription (Wolf et al., 2008). In the best characterized mechanism of action, an AR dimer binds to a consensus DNA binding sequence and recruits a series of coregulators to activate transcription (Fig. 2). It can also regulate transcription by interacting with other transcription factors without directly binding to DNA. AR also represses transcription of other target genes by less well characterized mechanisms (Fig. 2). Upon ligand binding AR can also activate kinases (for example Src through direct interaction (Migliaccio et al., 2011) and EGFR through release of EGFR ligand (Sen et al., 2010)) that can alter transcription independent of any requirement for AR to bind to the target genes (Fig. 2). In addition, AR and its coregulators are phosphoproteins. In some cases altered cell signaling leads to activation of AR in androgen-depleted medium (Culig et al., 1994, Nazareth and Weigel, 1996). Although early models of gene regulation by steroid receptors assumed that binding sites are located in the proximal promoter region near the site for initiation of transcription, newer studies using chromatin immunoprecipitation (ChIP) followed by direct analysis for candidate sites, binding to tiled arrays of genomic sequences (ChIP chip) (Wang et al., 2007), or massively parallel sequencing (ChIP-Seq) (Wang et al., 2009a) have shown that AR binds not only to proximal promoters but many kb upstream, in intronic regions, and well as in the 3′ UTR of regulated genes. In many cases, the distal sites can be brought into proximity to the promoter by protein–protein interactions (Wang et al., 2005) to regulate transcription. Other sites may play as yet unidentified roles.

The prostate is a gland in the male reproductive system that synthesizes components of the seminal fluid including proteases such as prostate specific antigen (PSA) (Leissner & Tisell, 1979). It develops from the urogenital sinus (UGS) (Cunha et al., 2004), which contains an outer layer of embryonic connective tissue termed the urogenital sinus mesenchyme (UGM) that expresses AR (Cunha et al., 2004). AR regulates gene expression vital to normal prostate development and function (Heinlein & Chang, 2002b). Stromal AR in the UGM regulates ductal morphogenesis through paracrine signaling for the initiation of prostatic development. Tissue recombinant experiments showed that stromal cells from UGM of AR positive mice combined with bladder epithelial cells from androgen insensitive (Tfm) mice produce glandular structures characteristic of a prostate, but do not produce the full complement of androgen-dependent proteins secreted by the dorsolateral mouse prostate (Donjacour and Cunha, 1993, Cunha et al., 2004). However, the reverse combination did not undergo prostatic development in mice even in the presence of androgens (Cunha et al., 2004). This demonstrates a crucial role for mesenchymal AR in prostatic development. Stromal AR functions as a key modulator of epithelial cell proliferation, survival, and differentiation in the normal developing prostate (Cunha et al., 2004). In the normal adult prostate, AR is expressed in all luminal cells and some epithelial basal and intermediate cell types (Mirosevich et al., 1999) as well as in stromal cells. The epithelial cells depend on androgen-induced growth factors secreted from stromal cells. The primary role of epithelial AR appears to be the production of secreted proteins characteristic of the prostate. Interestingly, AR plays both a growth stimulatory and inhibitory role functioning as a survival factor for luminal cells and as a suppressor of basal cell proliferation to tightly regulate normal prostate growth (Niu et al., 2008, Yu et al., 2009).

Although there is good agreement that PCa develops from prostate epithelial cells, there is conflicting evidence regarding whether the tumors arise from basal cells or from the luminal epithelial cells (Wang, Kruithof-de, et al., 2009b, Lawson et al., 2010, Wang et al., 2013). Prostates frequently contain multiple tumor foci as well as prostatic intraepithelial neoplasia (PIN) lesions, the presumed precursor of PCa. As tumors develop, they become progressively less organized with smaller ductal structures and ultimately may lose these structures entirely. With further growth, the tumors invade surrounding tissues and metastasize to the lymph nodes, bladder, and bone (Miller et al., 2003). A number of mutations, deletions, amplifications and, more recently, genomic translocations have been associated with PCa. Some of these alter the activity/structure of AR, some are regulated by AR, and others are independent of AR signaling.

One of the most frequently altered pathways in PCa is the PI3K signaling pathway, which plays a critical role in promoting growth and blocking apoptosis. About 40% of primary PCa and 70% of metastatic PCa exhibit genomic alterations in the PI3K/AKT pathway (El Sheikh et al., 2008, Reid et al., 2010, Taylor et al., 2010). A majority of these are characterized by the loss of PTEN (phosphatase and tensin homolog), a lipid phosphatase that limits AKT activation (El Sheikh et al., 2008, Reid et al., 2010, Taylor et al., 2010). In mouse models, homozygous deletion of PTEN is sufficient to induce tumors that metastasize (Wang et al., 2003).

The formation of fusion proteins through recurrent chromosomal translocations is also a major factor in PCa. The majority of these fusions are not only androgen-regulated, but there is evidence that AR promotes the formation of the fusions. Tomlins and others have found that more than 50% of PCa harbor a gene fusion between the promoter of TMPRSS2 and portions of the coding regions of ETS transcription factors (Tomlins et al., 2005, Tu et al., 2007). TMPRSS2 (transmembrane protease, serine 2) is an androgen responsive gene that is highly expressed in PCa (Yu et al., 2010). The ETS (E-twenty six) family members are transcription factors, which when over-expressed, function in tumorigenesis by regulating cell proliferation, cell migration, cell cycle control, and apoptosis (Gutierrez-Hartmann et al., 2007). Fusions between TMPRSS2 and ERG (Fig. 4) are much more common than fusions with other ETS family members, such as ETV1 or ETV4. Presumably, this is because TMPRSS2 and ERG are located approximately 3 Mb apart on the same chromosome, whereas other ETS factors are located on other chromosomes (Gutierrez-Hartmann et al., 2007). There is evidence that AR facilitates their formation. DHT treatment of normal prostate epithelial cells induces colocalization of TMPRSS2 and ERG measured by FISH (Lin et al., 2009). Using the AR positive LNCaP PCa cell line and FISH, two groups showed that treatment with androgens and a DNA damaging agent (radiation or chemical) resulted in very low frequency genomic rearrangements that mimic the translocations seen in PCa (Lin et al., 2009, Mani et al., 2009). That so many tumors contain these fusions suggests that there is a substantial selective advantage for their expression. The fusion is not detected in normal prostate, but some studies have found it in more than 15% of PIN and in a much higher proportion of primary tumors (Zhang et al., 2010). Despite the frequency of this fusion in PCa, the VCaP cell line is the only commonly studied PCa cell line that expresses the fusion. There is good agreement that expression of the fusion enhances expression of genes associated with invasiveness (Tomlins et al., 2008, Wang et al., 2008). Some investigators have found that depletion of the protein reduces growth in vitro (Sun et al., 2008, Wang et al., 2008) and Ittmann's group has shown that depletion of the fusion diminishes VCaP xenograft growth (Wang et al., 2008). Several groups have reported that artificial expression of the protein in mouse prostate is insufficient to produce tumors (Carver et al., 2009, Zong et al., 2009). However, the combination of elevated PI3K signaling (through reduction in PTEN or constitutively active kinase) in combination with ETS factor expression does produce tumors (Carver et al., 2009, Zong et al., 2009). Epidemiological studies have yielded conflicting results. One confounding factor is that there are many different fusions (Wang et al., 2006) and some of the fusions rely on an alternate translation start site within the fusion rather than the natural TMPRSS2 or ETS factor start site. Thus, the amount of RNA does not necessarily correlate well with protein levels. However, the combination of fusion and PTEN deletion predicts a shorter time to biochemical recurrence (rising serum PSA) in humans (Yoshimoto et al., 2008).

There are a number of other common genomic alterations and additional changes that result in altered expression of proteins. Some of these are oncogenes or tumor suppressors whose expression is altered in many cancers. In prostate, amplification and over-expression of c-myc is common (Taylor et al., 2010). Both p53 and Rb can be lost or inactivated, but this is more common in metastatic tumors rather than in the primary tumors (Taylor et al., 2010). There also are a variety of changes in AR and in proteins, which modulate AR function. These are discussed later in this review.

Section snippets

Androgen receptor and hormone dependent prostate cancer

Increasing levels of serum PSA, an androgen-regulated gene, detected in screening exams is often the first indication of PCa. Serum PSA is an indication that prostate cells are inappropriately releasing PSA into the circulation rather than into the lumen of the prostate. Although PCa often causes this, injury to the prostate can also release PSA. Thus, the presence of cancer typically is confirmed by digital-rectal exam and/or by biopsy. One of the major challenges in PCa is in distinguishing

Changes in androgen receptor that lead to reactivation

Since PCa is androgen dependent, some form of ADT therapy is the primary treatment for metastatic disease (see Section 5 for a discussion of treatments). Although most tumors respond initially, they become refractory to treatment. There is no cure for castration resistant prostate cancer (CRPC), which is ADT resistant but AR dependent (Seruga et al., 2011). A variety of in vitro and pre-clinical studies as well as more recent clinical studies support a role for reactivated AR in CRPC. For

Molecular alterations that enhance androgen receptor activity

In addition to alterations in AR expression or structure, there are many factors that contribute to activation of AR despite castrate levels of serum androgens. These alterations include changes in steroid metabolism, coactivator expression/activity, and cell signaling.

Primary prostate cancer

The median age of men at diagnosis of PCa in the US is 67, but PCa may be diagnosed as early as the fourth decade of life in some high-risk groups. Early PCa can be detected by rising serum PSA levels or by digital rectal exam, but difficulties with using these two assessments as large-population screening tools led to the contentious repeal by the United States Preventive Services Task Force (USPSTF) of its previous recommendation for their use in PCa screening (Moyer & Preventive Services

Conclusions

AR is a key regulator of the development and function of normal prostate. AR continues to play a critical role in PCa although its role in the epithelial cell derived tumors shifts to that of a growth promoting rather than a differentiating factor. ADT is the principal treatment for locally advanced and metastatic PCa. Most tumors respond initially, but become resistant to ADT with two years. The recurrent tumors secrete PSA and a wealth of additional studies summarized in this review show that

Conflict of interest statement

The authors declare that there are no conflicts of interest.

References (223)

  • S.M. Green et al.

    Androgen action and metabolism in prostate cancer

    Mol Cell Endocrinol

    (2012)
  • Z. Guo et al.

    Regulation of androgen receptor activity by tyrosine phosphorylation

    Cancer Cell

    (2006)
  • A. Gutierrez-Hartmann et al.

    ETS transcription factors in endocrine systems

    Trends Endocrinol Metab

    (2007)
  • I.A. Hughes et al.

    Androgen insensitivity syndrome

    Lancet

    (2012)
  • I. Ianculescu et al.

    Selective roles for cAMP response element-binding protein binding protein and p300 protein as coregulators for androgen-regulated gene expression in advanced prostate cancer cells

    J Biol Chem

    (2012)
  • C.B. Kallen et al.

    Steroidogenic acute regulatory protein (StAR) is a sterol transfer protein

    J Biol Chem

    (1998)
  • J.M. Kaminski et al.

    Effect of sequencing of androgen deprivation and radiotherapy on prostate cancer growth

    Int J Radiat Oncol Biol Phys

    (2003)
  • I.U. Agoulnik et al.

    Target gene specific regulation of androgen receptor activity by p42/p44 MAPK

    Mol Endocrinol

    (2008)
  • I.U. Agoulnik et al.

    Role of SRC-1 in the promotion of prostate cancer cell growth and tumor progression

    Cancer Res

    (2005)
  • I.U. Agoulnik et al.

    Androgens modulate expression of transcription intermediary factor 2, an androgen receptor coactivator whose expression level correlates with early biochemical recurrence in prostate cancer

    Cancer Res

    (2006)
  • I.U. Agoulnik et al.

    Androgen receptor action in hormone-dependent and recurrent prostate cancer

    J Cell Biochem

    (2006)
  • I.U. Agoulnik et al.

    Androgen receptor coactivators and prostate cancer

    Adv Exp Med Biol

    (2008)
  • J. Ai et al.

    HDAC6 regulates androgen receptor hypersensitivity and nuclear localization via modulating Hsp90 acetylation in castration-resistant prostate cancer

    Mol Endocrinol

    (2009)
  • H. Akaza et al.

    Study group for the combined androgen blockade therapy of prostate cancer. Combined androgen blockade with bicalutamide for advanced prostate cancer: long-term follow-up of a phase 3, double-blind, randomized study for survival

    Cancer

    (2009)
  • J. Anderson

    The role of antiandrogen monotherapy in the treatment of prostate cancer

    BJU Int

    (2003)
  • G. Attard et al.

    Phase I clinical trial of a selective inhibitor of CYP17, abiralerone acetate, confirms that castration-resistant prostate cancer commonly remains hormone driven

    J Clin Oncol

    (2008)
  • M.D. Balbas et al.

    Overcoming mutation-based resistance to antiandrogens with rational drug design

    Elife

    (2013)
  • A.H. Banham et al.

    Expression of the forkhead transcription factor FOXP1 is associated both with hypoxia inducible factors (HIFs) and the androgen receptor in prostate cancer but is not directly regulated by androgens or hypoxia

    Prostate

    (2007)
  • C.E. Barbieri et al.

    Exome sequencing identifies recurrent SPOP, FOXA1 and MED12 mutations in prostate cancer

    Nat Genet

    (2012)
  • D.A. Barron et al.

    The reactive stroma microenvironment and prostate cancer progression

    Endocr Relat Cancer

    (2012)
  • D.R. Bauman et al.

    Identification of the major oxidative 3alpha-hydroxysteroid dehydrogenase in human prostate that converts 5alpha-androstane-3alpha, 17beta-diol to 5alpha-dihydrotestosterone: A potential therapeutic target for androgen-dependent disease

    Mol Endocrinol

    (2006)
  • J. Beilin et al.

    Effect of the androgen receptor CAG repeat polymorphism on transcriptional activity: Specificity in prostate and non-prostate cell lines

    J Mol Endocrinol

    (2000)
  • D.R. Berthold et al.

    Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer: Updated survival in the TAX 327 study

    J Clin Oncol

    (2008)
  • C.L. Bevan et al.

    The AF1 and AF2 domains of the androgen receptor interact with distinct regions of SRC1

    Mol Cell Biol

    (1999)
  • G.R. Blackledge et al.

    Casodex (bicalutamide): Overview of a new antiandrogen developed for the treatment of prostate cancer

    Eur Urol

    (1997)
  • J.L. Boddy et al.

    The androgen receptor is significantly associated with vascular endothelial growth factor and hypoxia sensing via hypoxia-inducible factors HIF-1a, and the prolyl hydroxylases in human prostate cancer

    Clin Cancer Res

    (2005)
  • C.E. Bohl et al.

    Structural basis for antagonism and resistance of bicalutamide in prostate cancer

    Proc Natl Acad Sci U S A

    (2005)
  • M. Bolla et al.

    Improved survival in patients with locally advanced prostate cancer treated with radiotherapy and goserelin

    N Engl J Med

    (1997)
  • L. Bubendorf et al.

    Survey of gene amplifications during prostate cancer progression by high-throughout fluorescence in situ hybridization on tissue microarrays

    Cancer Res

    (1999)
  • C. Cai et al.

    Intratumoral de novo steroid synthesis activates androgen receptor in castration-resistant prostate cancer and is upregulated by treatment with CYP17A1 inhibitors

    Cancer Res

    (2011)
  • B.S. Carver et al.

    Aberrant ERG expression cooperates with loss of PTEN to promote cancer progression in the prostate

    Nat Genet

    (2009)
  • K.H. Chang et al.

    Dihydrotestosterone synthesis bypasses testosterone to drive castration-resistant prostate cancer

    Proc Natl Acad Sci U S A

    (2011)
  • C.D. Chen et al.

    Molecular determinants of resistance to antiandrogen therapy

    Nat Med

    (2004)
  • A.C. Chung et al.

    Genetic ablation of the amplified-in-breast cancer 1 inhibits spontaneous prostate cancer progression in mice

    Cancer Res

    (2007)
  • N.J. Clegg et al.

    ARN-509: A novel antiandrogen for prostate cancer treatment

    Cancer Res

    (2012)
  • J.M. Crook et al.

    Intermittent androgen suppression for rising PSA level after radiotherapy

    N Engl J Med

    (2012)
  • N. Craft et al.

    A mechanism for hormone-independent prostate cancer through modulation of androgen receptor signaling by the HER-2/neu tyrosine kinase

    Nat Med

    (1999)
  • Z. Culig et al.

    Androgen receptor — an update of mechanisms of action in prostate cancer

    Urol Res

    (2000)
  • Z. Culig et al.

    Androgen receptor activation in prostatic tumor cell lines by insulin-like growth factor-I, keratinocyte growth factor, and epidermal growth factor

    Cancer Res

    (1994)
  • G.R. Cunha et al.

    Role of stromal–epithelial interactions in hormonal responses

    Arch Histol Cytol

    (2004)
  • Cited by (260)

    View all citing articles on Scopus
    View full text