|Accueil | Aperçu | Revues | Soumettre | Nous Contacter | English|
During progression to an androgen-independent state following androgen ablation therapy, prostate cancer cells continue to express the androgen receptor (AR) and androgen-regulated genes, indicating that AR is critical for the proliferation of hormone-refractory prostate cancer cells. Multiple mechanisms have been proposed for the development of AR-dependent hormone-refractory disease including changes in expression of AR coregulatory proteins, AR mutation, growth factor-mediated activation of AR, and AR protein upregulation. The most prominent of these progressive changes is the upregulation of AR that occurs in greater than 90% of prostate cancers. A common feature of the most aggressive hormone-refractory prostate cancers is the accumulation of cells with neuroendocrine characteristics that produce paracrine factors, and may provide a novel mechanism for the regulation of AR during advanced stages of the disease. In this study, we demonstrate that neuroendocrine-derived PTHrP-mediated signaling through the EGF receptor (EGFR) and Src pathways contributes to the phenotype of advanced prostate cancer by reducing AR protein turnover. PTHrP-induced accumulation of AR depended on the activity of Src and EGFR and consequent phosphorylation of the AR on Tyrosine 534. PTHrP-induced tyrosine-phosphorylation of AR resulted in reduced AR ubiquitination and interaction with the ubiquitin ligase carboxy-terminus of HSP70-interacting protein (CHIP). These events result in increased accumulation of AR and thus enhanced growth of prostate cancer cells at low levels of androgen.
Expression of the Androgen Receptor (AR) plays an integral role in the progression of prostatic adenocarcinoma to hormone-refractory disease (1, 2). Several mechanisms have been described to account for the development of AR-dependent, androgen-refractory prostate cancer including AR mutational activation, coactivator overexpression, and activation via crosstalk with growth factor signaling pathways. The most frequent mechanism however, involves increases in AR protein levels.
Increases in AR are both necessary and sufficient to convert prostate cancer cells from a hormone-sensitive to a hormone-refractory state (3). This finding is consistent with reports demonstrating that high levels of AR are associated with aggressive clinicopathologic features and with decreased recurrence-free survival in prostate cancer patients treated with radical prostatectomy (4). In addition, immunohistochemical studies have shown that the AR is expressed in essentially all human prostate cancers, including those that have become hormone-refractory following hormone ablation therapy, and that the level of AR expression is often increased relative to untreated tumors (5, 6).
One explanation for the elevated levels of AR is gene amplification, reported to occur in approximately 22% of advanced prostate tumors (7). However, increased levels of AR observed in the majority of prostate cancers may also result from increased AR protein stability. As seen in recurrent CWR22 and LNCAP xenograft tumors (8, 9), increases in AR are sufficient to allow continued androgen signaling under conditions of low circulating ligand. Chen and colleagues (3) have shown that increasing the concentration of AR in prostate cancer cells using an AR-expressing lentivirus reduces the latency period for the development of LNCaP and LAPC4 xenograft tumors in castrate mice, supporting the hypothesis that increased AR protein promotes growth and survival of prostate tumors in low levels of androgen.
Another mechanism postulated to promote the progression of androgen-refractory prostate cancer is the appearance of cells with a neuroendocrine phenotype. Neuroendocrine-like cells are more prevalent in androgen-refractory disease, occurring in 30-100% of tumors studied (10, 11). The low proliferative capacity of neuroendocrine cells allows them to resist treatment with most chemotherapeutic agents, as well as endocrine and radiation treatments (12, 13). Neuroendocrine-like cells are thought to provide growth and survival signals to surrounding tumor cells promoting the progression of hormone-refractory prostate cancer (14, 15). This hypothesis is supported by work demonstrating that neuroendocrine-like cells enhance the growth of LNCaP xenografts, with the greatest effects seen under conditions of androgen deprivation (16, 17).
Parathyroid hormone-related protein (PTHrP) is one of the secreted products of prostatic neuroendocrine cells, and its expression in the human prostate is a manifestation of abnormal growth regulation. In culture, PTHrP cooperates with low levels of androgen to promote growth of androgen-dependent prostate cancer cells, but the molecular mechanisms responsible for these effects remain unclear (18). In particular, it is not known whether the effects of PTHrP on growth are dependent on the AR.
In the present study, we describe a novel mechanism by which neuroendocrine-derived factors, such as PTHrP, provide a proliferative advantage to prostate cancer cells. Our studies reveal that LNCaP cells at low androgen concentration proliferated more rapidly when stimulated with PTHrP. Under these conditions steady state levels of AR protein were increased and this increase, like PTHrP-induced proliferation, required the activation of epidermal growth factor receptor (EGFR) or Src and the subsequent phosphorylation of the AR on Tyr 534. We found that targeting of AR for proteasomal-dependent degradation via the chaperone-associated ubiquitin ligase carboxy-terminus of Hsp70-interacting protein (CHIP) was decreased, resulting in the accumulation of AR in response to PTHrP. The phosphorylation of AR on Tyr534 reduced the interaction of AR with CHIP. Together these findings demonstrate that neuroendocrine-derived PTHrP stimulation of cancer cells results in the upregulation of AR protein, thereby promoting growth under low androgen conditions and progression of hormone-refractory disease.
LNCaP cells were maintained in T-medium with 5% (v/v) fetal bovine serum (FBS; Invitrogen, Carlsbad, CA) in a humidified chamber at 37 °C with 5% CO2. Transfections were performed using TransIT-prostate transfection kit (Mirus, Madison, WI) according to manufacturer's instructions. The plasmid encoding FLAG-tagged AR has been previously described (19). The FLAG-AR-Y534F expression vector was constructed by sub-cloning AR-Y534F (gift of Dr. Young E. Whang, University of North Carolina, Chapel Hill, NC) into pCDNA3.1-FLAG plasmid (20). For experiments with cycloheximide (Calbiochem), cells were treated with cycloheximide (10μg/ml) throughout the period of R1881 and PTHrP stimulation.
Antibodies and reagents were obtained from the following sources: anti-AR PG-21 (Upstate, Lake Placid, NY); anti-FLAG antibody, M2, (Sigma, St. Louis, MO); anti-phosphorylated Src (Tyr418), anti-Src, anti-phosphorylated EGFR (Tyr992), and anti-EGFR (Biosource, Camarillo, CA, USA); anti-PTH/PTHrP receptor, anti-ubiquitin, and anti-CHIP (Santa Cruz, CA, USA); anti-tubulin (Calbiochem); anti-Hsp90 (StressGen, Ann Arbor, MI); goat-anti-rabbit and goat-anti-mouse antibodies (GE Healthcare, Buckinghamshire, England); Alamar Blue (Biosource); PTHrP (1-34aa), PP2, AG1478 (Calbiochem); and PTHrP receptor antagonist (Nle8,18, Tyr34)-pTH(3-34)amide (Bachem, Torrance, CA)
The anti-phospho-Tyr534 antibody was raised against a peptide spanning the Tyr534 phosphorylation site in human AR by standard methods. The peptide (DSYSGPpYGDMRLETC) was synthesized with phospho-tyrosine and an N-terminal cysteine (Anaspec, San Jose, CA), it was then coupled to keyhole limpet hemocyanin, and used for antibody production in rabbits (Cocalico Biologicals, Reamstown, PA). Antibody titers were monitored by immunoblotting, and the terminal bleeds were affinity purified using Sepharose-immobilized peptide. The specificity of the antibody was verified by immunoblot analysis against wt and Y543C mutant AR. The increase in immunoreactivity observed with the phospho-Tyr534 antibody in response to EGF was similar to that reported in published studies (21).
Growth of cells in a 96-well plate format (2500 cells/well), maintained for 2 days in various concentrations of R1881 and 1% charcoal-stripped serum (CSS) with or without PTHrP, was monitored with the Alamar blue assay according to the manufacturer instructions. Cell numbers were determind by seeding 5×104 cells/mL in triplicate onto 6-well plates in RPMI-1640 without phenol red, supplemented with 0.1nM R1881 and 1% CSS. The cells were treated with PTHrP alone or together with PTHrP receptor antagonist (10nM), PP2 (10uM), AG1478 (5uM), or DMSO control, and counted using a hemacytometer.
LNCaP cells were serum starved overnight in RPMI-1640 and stimulated with R1881 (0.1nM) or PTHrP (10nM) for the indicated times. The cells were lysed in radioimmunoprecipitation assay buffer [137mM NaCL, 20mM Tris (pH 7.5), 10% glycerol (v/v), 1% Triton-X100, 0.5% (w/s) deoxycholate, 0.1% (w/s) SDS, 2mM EDTA, 1mM phenylmethylsulfonyl fluoride, 1mM Na3VO4, and protease inhibitors]. The lysate protein concentration was determined using the BCA-containing assay (Pierce, Rockford, IL).
Endogenous AR was immunoprecipitated from LNCaP cells with 5μg anti-human AR per 100-mm dish, and immunoprecipitates were captured on protein A-conjugated sepharose beads (Roche, Indianapolis, IN). FLAG-AR was immunoprecipitated with M2 anti-FLAG antibody coupled to agarose (Sigma). Precipitates were resuspended in sample buffer [10% (v/v) glycerol, 62.5mM Tris (pH 6.8), 2% (w/v) SDS, 0.01mg/ml bromophenol blue], resolved by SDS-PAGE and transferred to nitrocellulose for immunoblotting.
Silencing of Src was performed with Validated Stealth RNAi DuoPak (Invitrogen) according to manufacturer instructions. Transfection of Src siRNA was done in LNCaP cells using Oligofectamine (Invitrogen) in Opti-MEM medium. A non-specific control duplex-XIII (Dharmacon Research, Lafayette, CO) was used as a siRNA silencing control. Maximal knockdown of Src was observed on day 3 after transfection.
Real-time RT-PCR was performed as previously described (22). Briefly, total RNA was isolated from LNCaP cells treated for 6 hours with PTHrP, R1881, or both using Qiagen Rneasy kit (Qiagen, Chatworth, CA). RNA quantity was determined with RiboGreen RNA assay kit (Invitrogen). First-strand cDNA was produced following the iScript™cDNA synthesis protocol (Bio-Rad, Hercules, CA). cDNA amplification was conducted on an iCycler optical system, in the presence of iQ SYBR Green mastermix (Bio-Rad). The PCR primers used were: AR-forward, 5'-CCTGGCTTCCGCAACTTACAC-3'; AR-reverse, 5'-GGACTTGTGCATGCGGTACTC-3'; PSA-forward, 5'-TGGTGCATTACCGGAAAGTGGATCA-3' ; P S A-reverse 5'-GCTTGAGTCTTGGCCTGGTCATTTC-3' ; F K B P 5 1-f o r w a r d , 5'-AGGAGGGAAGAGTCCCAGTG-3' ; F K B P 5 1-r e v e r s e , 5'-TGGGAAGCTACTGGTTTTGC-3'; NKX3.1-forward, 5'-GCACATATTTGCATGGAAGG-3'; N K X 3 . 1-reverse, 5'-ACAGCGAGTGCATCTTGTTC-3'; DKK1-forward, 5'-CCTTGGATGGGTATTCCAGA-3'; DKK1-reverse, 5'-CAGTCTGATGACCGGAGACA-3'; GUS-forward, 5'-CCGACTTCTCTGACAACCGACG-3'; GUS-r e v e r s e , 5'-AGCCGACAAAATGCCGCAGACG-3'.
All data were expressed as the mean ± standard error of mean from three or more independent experiments. Statistical analysis was performed by Student's t-test. Significance was determined with P<0.05.
The correlation of hormone-refractory prostate cancer with increased neuroendocrine differentiation and up-regulation of neuropeptide signaling, such as PTHrP (23, 24), prompted us to investigate a potential involvement of PTHrP in androgen-dependent prostate cancer cell proliferation. We performed a dose-response study of the effects of the synthetic androgen R1881 alone or R1881 together with recombinant PTHrP (10nM) on LNCaP cells. After 48 hours, maximal proliferation induced by R1881 alone occurred at 1nM. However, a significant growth increase was seen when cells were maintained in media containing 1% CSS supplemented with as little as 0.01nM R1881 and PTHrP (Fig.1A) indicating that PTHrP reduces the androgen requirement of LNCaP prostate cancer cells for growth. As indicated in Figure 1B, the proliferative advantage provided by PTHrP occurs only in the presence of R1881.
It has previously been documented that EGFR and Src pathways are induced upon GPCR activation (25-28). To gain insight into the role of these pathways in PTHrP-induced proliferation, we evaluated effects of EGFR or Src inhibitors on proliferation of the prostate cancer cells. PTHrP-induced proliferation of the cells was completely abolished by treatment with, AG1478, or PP2, the EGFR or Src inhibitors respectively. Similarly, PTHrP-induced LNCaP cell proliferation was abrogated upon treatment with PTHrP receptor antagonist (Fig. 1C). These observations indicated that transactivation of both EGFR and Src by the PTHrP receptor is necessary for the proliferation of LNCaP cells in response to PTHrP.
Growth of castrate-resistant prostate cancer occurs in low levels of circulating androgens but requires AR (2). Figure 1D shows that growth of LNCaP cells in lowandrogen and low-serum medium was significantly increased by PTHrP, and this effect could be blocked by AR siRNA. Together these data suggest that PTHrP is driving AR-dependent cell growth in this experimental model.
We next analyzed effects of the neuropeptide on AR protein levels. Serum deprived LNCaP cells were treated with R1881 (0.1nM), PTHrP (10nM), or both, and AR expression was determined by immunoblotting. Androgen was found to increase the steady-state expression of AR protein as previously reported, but this effect was further enhanced in the presence of PTHrP. Treatment of LNCaP cells with PTHrP alone did not increase the AR protein level, indicating that the effect of PTHrP signaling on AR is dependent on the presence of androgen (Fig. 2A). A dose-response analysis of R1881 alone or together with PTHrP showed that AR protein accumulates to a greater extent at low androgen concentrations (0.01-0.1nM) when PTHrP is present (Fig. 2B). Thus, PTHrP signaling leads to increased AR protein under androgen-limiting conditions, suggesting that PTHrP-induced LNCaP cell growth results from its modulation of AR levels.
Significant accumulation of AR protein was observed following treatment of LNCaP cells with PTHrP and R1881 for six hours. To determine whether transcriptional upregulation of the AR gene occurs under these conditions quantitative real-time PCR was utilized to measure AR mRNA. Induction of AR gene expression was observed upon treatment of LNCaP cells with R1881 alone but no significant change was detected relative to treatment with R1881 and PTHrP (Fig. 2C). To ascertain whether PTHrP-induced increases in AR are a result of changes in protein turnover, LNCaP cells were treated with the protein synthesis inhibitor, cycloheximide (CHX, 10ug/ml). As shown in Figure 2D, treatment with CHX reduced the steady state level of AR but did not inhibit the stabilizing effect of PTHrP on AR protein, indicating that PTHrP-mediated signaling influences AR levels posttranslationally. Together these results suggest that the production of PTHrP by neuroendocrine cells hypersensitizes prostate cancer cells to androgen by decreasing AR turnover and thus increasing the level of AR protein.
It is well documented that upregulation of growth factor signaling correlates with AR activation and progression of prostate cancer (29-31). In addition, stimulation of GPCRs has been demonstrated to result in the transactivation of the EGFR and Src pathways, prompting us to examine whether PTHrP signaling stabilizes AR through the activation of these pathways. To determine whether PTHrP stimulation is capable of activating EGFR in LNCaP cells, phosphorylation of Tyr992 was measured by immunoblotting with phosphospecific antibodies. Phosphorylation of Tyr992 was increased after one hour stimulation with PTHrP in the presence of androgen, but was not induced by PTHrP or R1881 alone. Similarly, phosphorylation of the activating Tyr418 residue of Src was shown to increase in response to PTHrP and R1881 treatment (Fig. 3A).
To determine if the PTHrP-induced activation of EGFR and Src mediates the increase in AR protein, these pathways were blocked pharmacologically. Treatment of LNCaP cells with either AG1478 or PP2 abrogated the AR stabilizing effect of PTHrP (Fig. 3B). Reduction of Src protein in cells transfected with specific siRNA also resulted in a decrease in AR protein level relative to LNCaP cells transfected with control siRNA even in the presence of PTHrP (Fig. 3C), further implicating Src in mediating the PTHrP-induced increase in AR protein. Together, these observations suggest that both EGFR and Src signaling are critical to PTHrP-induced regulation of AR.
AR protein turnover is regulated in part by association with chaperones, such as Hsp70 and Hsp90, and occurs via ubiquitin-dependent proteolysis (32, 33). To examine whether the interaction of AR with chaperones is altered in the presence of PTHrP, AR was immunoprecipitated from LNCaP cells deprived of serum and treated with R1881 alone or in combination with PTHrP. As shown in Figure 4A, less Hsp90 was in complex with AR when the cells were treated with R1881 and PTHrP. Analysis of AR ubiquitination following immunoprecipitation also indicated that PTHrP treatment reduced AR ubiquitinated species (Fig. 4B). This observation indicated that AR is subject to proteasome-dependent degradation under normal physiological conditions and that the presence of PTHrP can stabilize the AR protein by interfering with its ubiquitination.
AR is a target of the ubiquitin ligases Mdm2- or CHIP, leading to degradation by the 26S proteasome (34, 35). Based on the observation that PTHrP reduces the interaction of AR with the Hsp90 chaperone complex, we hypothesized that the reduction in AR ubiquitination may be due to decreased interaction with the CHIP E3-ligase. Analysis of the AR immunoprecipitate showed reduced CHIP protein upon treatment with PTHrP (Fig. 4B). These data suggest that upon PTHrP stimulation the interaction of AR with chaperone-associated ubiquitin ligases and subsequent proteasomal-directed degradation is reduced.
In previous work, we showed that EGF signaling results in the tyrosine phosphorylation of AR and modulates its interaction with Src (19). To determine whether PTHrP treatment influences the tyrosine phosphorylation of AR, LNCaP cells were deprived of serum and stimulated with R1881 or R1881 and PTHrP for 6 hours. Cell extracts were subsequently prepared and immunoblotted with phosphospecific antibodies to Tyr534, the primary site of AR tyrosine phosphorylation (21). As shown in figure 5A, phosphorylation of AR was increased upon treatment with R1881 and PTHrP relative to R1881 treatment alone. Similar phosphorylation levels were observed when AR was immunoprecipitated and immunoblotting with anti-phospho-tyrosine antibodies was performed (data not shown).
To assess whether the CHIP/AR interaction is modulated by AR tyrosine phosphorylation, LNCaP cells were transiently transfected with either FLAG-AR or with FLAG-AR-Y534F. Immunoprecipitation of exogenously expressed FLAG-AR complexes revealed that, similar to endogenous AR, the CHIP interaction is reduced in the presence of PTHrP. However, the phosphorylation-deficient AR-Y534F mutant showed enhanced association with CHIP in response to treatment with R1881 alone or in combination with PTHrP (Fig. 5B). This observation suggests that tyrosine phosphorylation of AR is required for decreasing the interaction between CHIP and AR, thus reducing its degradation.
To confirm that the phosphorylation of AR Tyr534 plays a role in the PTHrP-induced accumulation of the protein, LNCaP cells ectopically expressing FLAG-AR or FLAG-AR-Y534F were deprived of serum and stimulated with R1881 alone or PTHrP and R1881. Immunoblotting with anti-FLAG antibodies revealed that FLAG-AR, like endogenous AR, accumulated in the presence of PTHrP (Fig. 5C). FLAG-AR-Y534F failed to accumulate in response to PTHrP, indicating that the phosphorylation of this site is important for PTHrP-induced stabilization of AR.
We examined AR transactivation in LNCaP cells stimulated with PTHrP, R1881 or both. Quantitative real-time PCR analysis of androgen-regulated genes demonstrated that PTHrP does not significantly modulate NKX3.1 expression in the presence of R1881, but antagonizes R1881 induction of PSA transcription and enhances mRNA levels of FKBP51. Downregulation of DKK was observed in the presence of R1881 or PTHrP (Fig. 5D).
Increases in the neuroendocrine status of prostate tumors have been correlated with tumor size and progression to hormone-refractory stages. Existing evidence suggests that neuroendocrine-like cells arise within the tumor upon exposure to various differentiation factors, including β-adrenergic agonists, cytokines, or long-term androgen ablation (36-38). Although neuroendocrine-like cells are non-mitotic, they produce neuropeptides, such as PTHrP, neurotensin, and bombesin, that are thought to induce proliferative or survival signals to adjacent cancer cells (16, 39). Our data show that PTHrP-mediated signaling enhances the proliferation of prostate cancer cells by an alternative mechanism involving the regulation of AR protein stability.
Androgen ablation therapy reduces the circulating testosterone concentration from 4.5ng/ml (21nM) to 0.28-1.3ng/ml (1.3-6nM), resulting in apoptosis of androgen-dependent tissues and regression of prostate cancer (40). In comparing the growth properties of LNCaP cells, an androgen-dependent human prostate cancer cell line, in the presence of PTHrP a disparity was observed in the levels of androgen required for cell proliferation. LNCaP cells treated with PTHrP required androgen concentrations two orders of magnitude lower for growth stimulation (Fig. 1), suggesting that in the setting of low levels of androgen following pharmacological or surgical castration PTHrP may enhance the proliferation of prostate cancer cells. The proliferative advantage provided by PTHrP appears to be associated with a complex pattern of androgen-regulated genes, as indicated by the small sampling of AR targets we examined. Further analysis will be required to determine how the global pattern of gene expression contributes to enhanced prostate cancer cell proliferation in the presence of PTHrP and low androgen (Fig. 5D).
Under low androgen conditions, augmentation of LNCaP proliferation or AR protein concentration in response to PTHrP is mediated through EGFR and Src kinase activation (Figs. 1 & 3). Transactivation of this pathway in response to neuropeptide stimulation has been demonstrated by Ullrich and colleagues (41), and is consistent with our work in PC3 cells (25). In the current study, phosphorylation of EGFR was detected after one hour of PTHrP treatment corresponding with increased AR protein levels. Activation of Src appears to be an important step in the regulation of AR protein levels in the presence of PTHrP, as its inhibition or siRNA knockdown abrogates PTHrP-induced AR accumulation (Fig. 3). Src association with AR occurs either directly or in a ternary complex with the scaffold protein RACK1 (19, 42), resulting in AR phosphorylation, increased AR activity, and enhanced LNCaP cell proliferation (21). Our data agree with these findings showing that AR tyrosine phosphorylation is increased in response to PTHrP (Fig. 4), and that PTHrP-mediated cell proliferation requires AR as well as Src activation (Fig. 1). Src can also interact directly with Hsp90 chaperone complexes, and its ability to modulate client protein turnover further supports its involvement in the modulation of AR protein levels (43). Chaperone proteins maintain AR in a partially unfolded ligand-free state to prevent aggregation of the protein, but may also modulate AR protein degradation, nuclear translocation and transactivation (32). It is possible that Src could modify or recruit various components of the complex to influence multiple aspects of AR downstream signaling. Our work favors the hypothesis that phosphorylation of AR modulates its ability to associate with the chaperone complex. We observed that interaction of AR with Hsp90 as well as Hsp70 is decreased in the presence of PTHrP (Fig. 4 and data not shown).
Molecular chaperones regulate the proteasome-dependent degradation of several classes of proteins including steroid receptors, membrane proteins, and kinases. One link between chaperones and the proteasome is CHIP. CHIP contains tetratricopeptide repeats for interaction with Hsp70 and Hsp90 and a modified form of the Ring finger domain found in some ubiquitin ligases (44). In conjunction with the UBCh5 group of E2 enzymes CHIP promotes ubiquitination of glucocorticoid receptor, CFTR, and Raf kinase (45-47). Overexpression studies have revealed that CHIP prevents the accumulation of AR and stimulates its ubiquitination in LNCaP or HeLa cells (34, 48). Our observations in LNCaP cells with endogenous CHIP levels are consistent with these studies and show that AR polyubiquitination is reduced in response to PTHrP stimulation and correlates with reduced CHIP binding.
The interaction of CHIP with AR has been characterized in vitro and was shown to depend on both unphosphorylated and serine/threonine phosphorylated peptides of AR (49). In this study we find that the interaction of CHIP with the phosphorylation-deficient AR-Y534F is enhanced, suggesting that unlike serine/threonine phosphorylation, tyrosine phosphorylation may disrupt the AR/CHIP complex. Serine phosphorylation of AR has also been described to enhance its proteasomal degradation. Serine phosphorylation by Akt is postulated to increase AR ubiquitination by Mdm2 and lead to its degradation (49). Further analysis will be required to elucidate the relationship between these apparently antagonistic phosphorylation events in the regulation of AR/CHIP interaction and their role in the regulation of AR protein turnover.
In summary, PTHrP enhances the proliferation of LNCaP cells at low concentrations of R1881 by modulating androgen-induced stabilization of AR. Under low androgen PTHrP functions as a compensatory signal to modulate the AR protein life cycle through pathways that involve the activation of EGFR and Src. We show that PTHrP-induced transactivation of EGFR, Src and consequent AR tyrosine phosphorylation sequesters a subpopulation of AR protein from chaperone/CHIP complexes that normally regulate the proteasomal-dependent degradation of the protein (Fig. 6). This study provides one mechanistic explanation whereby increased neuroendocrine differentiation promotes the progression of prostate cancer through the production of factors that stabilize AR protein. The findings reveal potential new sites of therapeutic intervention that can be combined with androgen ablation.
We thank Vicki Gordon for preparation of the anti-AR phospho-Tyr534 antibody and Dr. Young E. Whang for providing the AR-Y534F construct. This work was supported by the Department of Defense grant W81XWH-07-1-0070 (DaSilva, J), National Institutes of Health-National Cancer Institute grant PO1 CA104106 (Parsons, S & Weber, M), National Institutes of Health grants T32 DK007646 and CA105402 (Weber, M).