Compensatory upregulation of tyrosine kinase Etk/BMX in response to androgen deprivation promotes castration-resistant growth of prostate cancer cells

Abstract

We previously showed that targeted expression of non-receptor tyrosine kinase Etk/BMX in mouse prostate induces prostate intraepithelial neoplasia, implying a possible causal role of Etk in prostate cancer development and progression. Here, we report that Etk is upregulated in both human and mouse prostates in response to androgen ablation. Etk expression seems to be differentially regulated by androgen and interleukin 6 (IL-6), which is possibly mediated by the androgen receptor (AR) in prostate cancer cells. Our immunohistochemical analysis of tissue microarrays containing 112 human prostate tumor samples revealed that Etk expression is elevated in hormone-resistant prostate cancer and positively correlated with tyrosine phosphorylation of AR (Pearson correlation coefficient rho = 0.71, P < 0.0001). AR tyrosine phosphorylation is increased in Etk-overexpressing cells, suggesting that Etk may be another tyrosine kinase, in addition to Src and Ack-1, which can phosphorylate AR. We also showed that Etk can directly interact with AR through its Src homology 2 domain, and such interaction may prevent the association of AR with Mdm2, leading to stabilization of AR under androgen-depleted conditions. Overexpression of Etk in androgen-sensitive LNCaP cells promotes tumor growth while knocking down Etk expression in hormone-insensitive prostate cancer cells by a specific shRNA that inhibits tumor growth under androgen-depleted conditions. Taken together, our data suggest that Etk may be a component of the adaptive compensatory mechanism activated by androgen ablation in prostate and may play a role in hormone resistance, at least in part, through direct modulation of the AR signaling pathway.

Figure 1. Etk expression is increased in prostate cells after castration

A, The 16-week-old FVB mice were castrated and sacrificed on Day 1 (D1), Day 2 (D2), Day 4 (D4) and Day8 (D8) post-castration, Etk expression was analyzed by Western Blot (Left) and Immunohistochemical (Center) analysis using anti-Etk antibody. IL6 was used as positive control for castration. Actin was used as a loading control. The relative Etk mRNA level in prostates from intact or castrated mice was determined by Real-time PCR analysis (Right). B, At Day 8 post-castration, mice were supplemented with 50mg/kg DHT per day for 5 days. The total prostate tissue lysates were subjected to Western Blot analysis with anti-Etk antibody. Lane 1, intact prostates; Lane 2, age-matched mice at Day 8 post-castration; Lane 3, the castrated mice supplemented with 50mg/kg DHT per day for 5 days. Actin was used as a loading control. C, Hormone sensitive (HS) and hormone refractory (HR) LAPC4 and LNCaP xenograft tumors were derived as described in Methods. Total protein lysates extracted from these tumors were subjected to immunoblotting with anti-Etk antibody. Actin served as a loading control. D, Binding of AR to the putative ARE sites (ARE-P, 1, 2, 3, 4) of human ETK gene in LNCaP cells in response to DHT and IL6 was analyzed by ChIP assays as described in Methods. PCR products were resolved on agarose gels.

Figure 2. Etk is upregulated in hormone refractory prostate cancer and induces AR tyrosine phosphorylation

A, The mean scores and frequency of the immunoreactivity with anti-Etk or anti-pARY534 antibodies detected in hormone-naïve (HN), and hormone-refractory (HR) human prostate tumor arrays (TMAs). *p< 0.001, compared to HN. B, Representative fields of the TMAs. Anti-AR was used as a control. Low (X20) and high (X400) magnification views of Etk, pARY534, and AR expression in HN and HR prostate tumor were shown. C, Etk is associated with AR and induces AR tyrosine phosphorylation in COS-1 cells. COS-1 cells were co-transfected with the plasmids indicated. At 24h post-transfection, cells were serum starved overnight and then lysed. The association between Etk and AR was determined by immunoprecipitation with anti-AR antibody and followed by immunoblotting with anti-Etk antibody. The effects of Etk on AR tyrosine phosphorylation were examined by immunoprecipitation with anti-AR antibody and followed by immunblotting with anti-phosphotyrosine (pY) antibody. The levels of Etk or AR in the total cell lysates were monitored by immunoblotting with anti-Etk and anti-AR antibody respectively. D, Overexpression of Etk in mouse prostate induces AR tyrosine phosphorylation. Mouse prostates were dissected from the Etk transgenic and their WT littermate mice respectively. One half of the dorsolateral prostate lobes were fixed in 10% Formalin and subjected to immunohistochemical staining with anti-phosphorylated ARY534 antibody (Left). The lysates of the other half gland were subjected to immunoblotting with the antibodies as indicated (Center). The intensity of pARY534 and AR in each lane was quantified by using the software Quantity One and the change of pARY534 was normalized with AR (Right). The value in Lane 1 was set as 1.

Figure 3. Castration promotes association between Etk and AR

A, Association between Etk and AR in xenograft tumors is increased at post-castration. Hormone sensitive (HS) and hormone refractory (HR) LNCaP and LAPC4 xenograft tumors were derived as described in Methods. The total protein lysates extracted from these tumors were subjected to immunoprecipitation with anti-AR or anti-Etk antibody and followed by immunoblotting with anti-Etk or anti-AR antibody. Expression of Etk or AR was monitored by immunoblotting of the total cell lysates. Actin was used as a loading control. The intensity of AR-bound Etk or Etk-bound AR in each lane was quantified by using the software Quantity One and normalized with AR or Etk protein level (Right). The value of HS LNCaP was set as 1. B, COS-1 cells were transfected with the Flag-tagged AR. Cell lysates were incubated with GST-Etk wt or GST-Etk mutants bound to glutathione-Sepharose 4B beads at 4 °C for 1 h. The beads were washed intensively, and the bound proteins were detected by immunoblotting with an anti-Flag antibody. The GST fusion proteins were visualized with Coomassie Blue staining (CBS). C, COS-1 cells were transfected with Flag-tagged AR wt or AR mutants. Cell lysates were incubated with GST-EtkSH2 bound to Glutathione-Sepharose 4B beads at 4 °C for 1 h. The beads were washed intensively, and the bound proteins were detected by immunoblotting with an anti-Flag antibody. The GST fusion proteins were detected with anti-GST. D, The effects of Y534F and Y551/552F mutation on Etk induced AR tyrosine phosphorylation. AR or AR mutants were co-transfected with Etk or the vector control into COS-1 cells. At 48h posttransfection, cell lysates were subjected to immunoprecipitation with anti-AR and followed by immunoblotting with anti-phosphotyrosine (pY). The expression of AR and Etk in total cell lysates was monitored by immunoblotting with the indicated antibody. The intensity of phosphorylated AR and AR in each lane was quantified by using the software Quantity One and the change of tyrosine phosphorylation was normalized with AR (Right). The value of phosphorylated ARWT was set as 100%.

Figure 4. Etk increases AR stability and transcriptional activity

A, LNCaP cells were transfected with control vector or Etk. After serum-starvation, cells were treated with 0.1 ug/ ul Cycloheximide (CHX) for the indicated time. Total cell lysates were immunoblotted with anti-AR, anti-Erk and anti-Etk respectively. Actin was used as a loading control. B, AR wt or ARY551/552F was cotransfected with Etk into COS-1 cells. After serum-starvation, cells were treated with 0.1 ug/ ul Cycloheximide for the time indicated. Total cell lysates were immunoblotted with anti-AR and anti-Etk. C, AR was cotransfected with T7-tagged Etk or EtkKQ into COS-1 cells. Association of AR with endogenous Mdm2 was detected by immunoprecipitation with anti-Mdm2, followed by immunoblotting with anti-AR. D, LNCaP cells were transfected with lenti-virus encoding Etk, shRNA specific for Etk or vector control. PSA, POV1 and AR mRNA level were detected by Quantitative real-time RT-PCR after serum-starvation overnight. **p < 0.01. The PSA, AR and Etk protein levels were detected by Western Blot.

Figure 5. Etk promotes prostate tumor growth under androgen depleted conditions

A–B, The effect of Etk shRNA on prostate cancer cell growth. C-81 (A) or CWR-R1 (B) cells were infected with the lenti-virus encoding shRNA specific for luciferase (control) or Etk. Cells were maintained in the medium containing 5% charcoal-stripped serum for two weeks. Cell colonies were visualized by Coomassie Blue Staining (CBS) (Top). Etk level was detected by immunoblotting at 48h post-infection (Middle). At 48h post-infection, cells were subcutaneously injected into castrated male nude mice respectively and tumor growth was measured as described in Methods (Bottom). **p < 0.01, *p < 0.05. The results were presented as mean tumor volume ± SE (n =5 mice/group). C, Etk promotes prostate tumor growth. LNCaP cells were infected with lentivirus encoding Etk or the vector control. At 48h post-infection, cells were subcutaneously injected into castrated male SCID mice respectively and tumor growth was detected as described in Methods. The results were presented as mean tumor volume ± SE (n = 5 mice/group). **p < 0.01, *p < 0.05. The levels of Etk were detected by Western blot at 48h post-infection.