Androgen Receptors and Prostate Cancer Therapies
Written by Ben Bunting: BA(Hons), PGCert. Sport & Exercise Nutrition. British Army Physical Training Instructor.
--
A central role of the androgen receptor (AR) in prostate cancer tumorigenesis has led to the development of antiandrogen drugs. Unfortunately, patients often recur with more aggressive tumors known as hormone-refractory PCa (HRPC) or androgen-independent prostate cancer (AIPC).
Several alterations have been identified in ARs that may contribute to this pathologic change. Among them are chromatin transactivation, reactivation of androgen-regulated genes, and phosphorylation to activate the receptor in the absence of ligand.
AR Phosphorylation by PKC
AR is a key regulator of androgen biosynthesis in prostate cancer (PCa) cells. The presence of high levels of androgens promotes the development of prostate tumors, but depletion of these hormones leads to shrinkage and apoptosis. However, PCa often reoccurs as castration-resistant prostate cancer (CRPC) within 18 to 36 months of androgen deprivation therapy, thus providing a therapeutic challenge.
AR phosphorylates upon hormone treatment at multiple sites. This phosphorylation switch may regulate AR transcriptional activity, translocation to the nucleus and stability. Several posttranslational modifications are known to occur in the activation function-1 (AF1) region of AR, which contains the transcriptional activation unit-1 (TAU1) and 5 (TAU5) regions [11-12]. Some of these posttranslational modifications, like acetylation, have been shown to be androgen dependent, but others, such as SUMOylation, are independent of androgens.
The NTD of AR contains two SUMOylation sites (K386 and K520) that are critical for protein stability, nuclear localization and chromatin interactions. The ubiquitin ligase Ubc9 interacts with these SUMOylation sites and subsequently ubiquitinates AR to detach it from the chromatin. During short-term proteotoxic cell stress, such as hyperthermia or heat shock, AR acetylates at the SUMOylation sites, which promotes its detachment from the chromatin and its accumulation in the nuclear matrix compartment.
In addition, a phosphorylation site called S81 is located in the hinge region of AR, which is responsible for its poly-ubiquitination and degradation. The mutation of S81 to alanine significantly reduced the amount of AR acetylated at the acetylation site and altered its cellular localization. These results suggest that S81 plays an important role in the regulation of AR acetylation and could be a potential therapeutic target for patients with CRPC.
Similarly, p65 and S578 are two phosphorylation sites in the AF2 region of AR that are involved in its cellular localization. The phosphorylation of these residues has been observed in xenograft models of prostate cancer and is associated with androgen resistance.
A phosphorylation loop of AR and its coregulators that occurs when ERK is activated has also been identified as a non-genomic mechanism for AR transcriptional activity at very low androgen levels (17,67). This autocrine loop can involve phosphorylation of AR and AR coregulators to facilitate their binding to AR and enhance AR transcriptional activity. The resulting increase in transcriptional activity may be a significant factor in the development of prostate cancer, and is likely to be critical for PCa initiation and progression.
AR Phosphorylation by PIM1
Despite androgen deprivation therapy (ADT), prostate cancer remains a lethal disease that often recurs within 18-36 months, transforming into castration resistant prostate cancer (CRPC). Activation of the AR, which is a key regulator of cell growth and proliferation, occurs in response to androgens. Nevertheless, the mechanisms that determine how androgens stabilize and activate AR in CRPC are unclear.
In the absence of androgen, AR resides in a non-translating state, primarily in the cytoplasm. Its expression is regulated by a variety of transcriptional co-regulators that are differentially expressed in cells with different hormone receptor status and in cells with various other cellular functions. In addition, a number of posttranslational modifications, such as phosphorylation, acetylation, methylation, and ubiquitination, have been identified.
One of the most important posttranslational modification is phosphorylation, which has been shown to increase AR activity and stability. This is likely due to the interaction of phosphorylated AR with phosphatase enzymes and with DNA. Several residues, including Ser-81 and S213 (S213 phosphorylation is dependent on the PI3K pathway), are known to be phosphorylated in AR.
Another important posttranslational modification is acetylation, which also increases AR activity and stability. This is also likely due to the interaction of acetylated AR with phosphatase enzymes. In addition, acetylation of AR may be responsible for the association with heat shock proteins in the cytoplasm and their subsequent interaction with the LBD.
Finally, SUMOylation is another important posttranslational modification that affects the mobility of AR in chromatin. The SUMOylation of AR can be reduced by the inhibitor SUMO-1 or by arginine substitution. In HEK293 cells, SUMOylation of AR significantly decreased the binding of AR to FOXO4 and enhanced the expression of SPOCK1.
Additional residues located in the TAU1 region, Y223, Y307, Y346, and Y357, were phosphorylated when SRC kinase was coexpressed with AR in a hormone-refractory xenograft. These modified tyrosine residues were correlated with the development of PCa from an androgen-dependent to a hormone-refractory state. These results provide evidence that the TAU1 region plays a key role in androgen independent regulation of AR in CRPC. This may help explain why CRPC cells are unable to respond to androgen deprivation therapy.
AR Phosphorylation by MED1
In conventional prostate cancer cells, AR-mediated signaling occurs via a genomic or transcriptional pathway that involves the translocation of AR from the cytoplasm to the nucleus and its interaction with cognate androgen response elements on promoters, leading to repression or activation of target genes. However, when androgen levels are low (in castration-resistant PCa or CRPC), the AR is capable of modulating cell proliferation without the need for genomic signaling.
This non-genomic AR signaling occurs in caveolin-rich lipid rafts and mediates a rapid androgen-induced increase in intracellular Ca2+ concentration, which can be inhibited by pertussis toxin or phospholipase C (PLC) inhibitor. Moreover, in androgen-dependent prostate cancer cells, a truncated version of AR lacking the LBD, AR-Vs, expresses high levels and is constitutively active even in the absence of androgen. Interestingly, increased expression of AR-Vs has also been observed in CRPC, but the functional contribution of the Vs to cell growth is unclear.
One possible mechanism that negatively regulates AR transcriptional activity is through the interaction of stress kinases. The kinase cascades MKK4/p38 and MKK6/JNK converge to negatively regulate AR transcriptional activity by activating a phosphatidyl-inositol 3-kinase (PI3K)/Akt-dependent pathway, which results in inhibition of AR-mediated gene expression. Using a siRNA knockdown strategy, we found that PI3K/Akt is inhibited in a stress kinase-dependent manner in AR-dependent COS-1 cells, but not in androgen-independent C4-2 cells or AR-Vs.
Upon activation of the stress kinase pathways, p38 and JNK phosphorylate Y223, a truncated residue in the TAU1 region, adjacent to serine 308, which is an important negative regulator of AR transcriptional activity. The phosphorylation of this residue is cycle-cycle dependent and is associated with the expression of PIM-1, which is long (L) or short (1S) in both androgen dependent and CRPC cells.
The phosphorylation of this residue is further stimulated by a phosphatidyl-inositol-3 kinase (PI3K)/Akt-dependent, E3 ubiquitin ligase RNF6. Upon phosphorylation of T850, the long (L) isoform of PIM-1 stabilizes AR and is able to bind to androgen-dependent promoters.
These data provide evidence that a phosphorylation switch controls androgen biosynthesis in PCa, allowing AR to reactivate even under castration-resistant conditions. This discovery is a major step in understanding how AR can survive and maintain the alternative gene expression patterns that sustain prostate cancer.
AR Phosphorylation by ESRP2
There is an important phosphorylation switch that controls androgen biosynthesis in prostate cancer. The phosphorylation switch is the interaction between androgen receptor (AR) and its kinases. These kinases are activated by androgen-induced cAMP levels, which are regulated by intracellular Ca2+ concentration, resulting in phosphorylation of the AR.
The interaction between AR and its kinases is also influenced by other factors such as ubiquitination and SUMO tyrosine mutase. The ubiquitination of AR lysine 845 (K845) in the LBD by RNF6 promotes monoubiquitination and polyubiquitination of AR [116-119]. In contrast, SUMO-1 tyrosine mutase inhibits monoubiquitination and SUMO-2 tyrosine mutase enhances polyubiquitination.
We have previously reported that a kinase-inactive form of Akt, dAkt, suppresses transactivation of AR by a mechanism independent of cAMP. Moreover, cAkt markedly reduces AR protein levels in COS-1 cells. The phosphorylation of AR at the two Akt consensus sites (Ser210 and Ser790) was a major factor in this suppression.
In CRPC, the androgen-dependent transcriptional activity of AR shifts from the TAU1 to TAU5 region, which requires S308 phosphorylation. Phosphorylation of S308 is inhibited by the PI3K inhibitor LY294002 and inhibits the expression of AR in a dose-dependent manner. It was also found that S308 phosphorylation was negatively correlated with PSA-Enh-Luc assay indicating a negative role for this site in androgen dependent regulation of AR.
This suggests that ESRP2 is a key regulator of the phosphorylation switch in CRPC, and that it may play a role in determining the transition from androgen dependent to androgen independent activation of AR. This would make ESRP2 an important target for therapy in CRPC and prostate cancer patients.
Interestingly, we have recently found that ESRP2 can be induced by the androgen dihydrotestosterone (DHT). This suggests that ESRP2 could be a direct mediator of the androgen-induced switch in the transcriptional activity of AR.
Conclusion
The androgen receptor (AR) is the ligand-activated transcription factor that mediates prostate cancer cell growth. AR is a serine/threonine kinase that is expressed in most PCa cells and drives gene expression mainly by binding androgens to its ligand-binding domain (LBD) and translocating to the nucleus. However, in a minority of PCa cells, including those that display castration-resistant (CRPC) characteristics, constitutively active AR variants exist that lack the LBD and drive gene expression in a nonandrogen dependent manner.
S96 phosphorylation of AR is a major signaling event that regulates the stability, nuclear localization, and transcriptional activity of AR in response to nutrients or mitogenic stimuli. Moreover, high S96 phosphorylation is a significant survival factor in human HCC and is an independent predictor of overall survival.
BMX is an important regulator of androgen biosynthesis that directly interacts with 3bHSD1 in a complex, which activates BMX kinase activity to promote androgen synthesis. BMX kinase inhibition blocks the formation of this complex and inhibits androgen synthesis in cultured prostate cancer cells and patient tissues.
In addition to its role in regulating androgen synthesis, BMX also mediates the regulation of a diverse set of tyrosine kinases by forming a complex with androgen receptor-binding protein 3bHSD1 that regulates the conversion of DHEA to dihydrotestosterone (DHT). Thus, blocking BMX may be an effective strategy for treating relapsed/recurrent androgen-sensitive PCa.
Despite this, the exact function of ESRP2 is unknown. However, it is known that ESRP2 is an epithelial-specific splicing regulator. It has been shown that ESRP2 upregulation predicts poor prognosis in prostate cancer. It is also known that ESRP2 can alter the alternative splicing patterns of FGFR2 genes.
