|Home | About | Journals | Submit | Contact Us | Français|
The insulin-like growth factor (IGF) system plays an important role in prostate cancer. The BRCA1 gene encodes a transcription factor with tumor suppressor activity. The involvement of BRCA1 in prostate cancer, however, has not yet been elucidated. The purpose of the present study was to examine the functional correlations between BRCA1 and the IGF system in prostate cancer.
An immunohistochemical analysis of BRCA1 was performed on Tissue Microarrays comprising 203 primary prostate cancer specimens. In addition, BRCA1 levels were measured in prostate cancer xenografts and in cell lines representing early stages of the disease (P69 cells) and advanced stages (M12 cells). The ability of BRCA1 to regulate IGF-IR expression was studied by coexpression experiments using a BRCA1 expression vector along with an IGF-IR promoter-luciferase reporter.
We found significantly elevated BRCA1 levels in prostate cancer in comparison to histologically normal prostate tissue (p < 0.001). In addition, an inverse correlation between BRCA1 and IGF-IR levels was observed in the AR-negative P69 and M12 prostate cancer-derived cell lines. Coexpression experiments in M12 cells revealed that BRCA1 was able to suppress IGF-IR promoter activity and endogenous IGF-IR levels. On the other hand, BRCA1 enhanced IGF-IR levels in LnCaP C4-2 cells expressing an endogenous AR.
We provide evidence that BRCA1 differentially regulates IGF-IR expression in AR positive and negative prostate cancer cells. The mechanism of action of BRCA1 involves modulation of IGF-IR gene transcription. In addition, immunohistochemical data is consistent with a potential survival role of BRCA1 in prostate cancer.
The insulin-like growth factors, IGF-I and IGF-II, are a family of mitogenic polypeptides with important roles in normal growth and differentiation as well as in tumor development and progression (1–3). In the specific context of prostate cancer, a significant amount of data has been accumulated over the last twenty years suggesting that the IGF axis plays an important role in the transformation of the prostate epithelium (4–7). The contribution of IGF action to prostate cancer development is further supported by epidemiological studies showing a significant increase in serum IGF-I levels in patients who later developed prostate cancer (8). Acquisition of the malignant phenotype is initially IGF-I receptor (IGF-IR)-dependent, however, the progression of prostate cancer from an organ-confined, androgen-sensitive disease to a metastatic one is associated with dysregulation of androgen receptor (AR)-regulated target genes and with a significant decrease in IGF-IR mRNA and protein levels (9, 10). Likewise, IGF-IR expression is extinguished in a majority of human prostate cancer bone marrow metastases (11). The molecular mechanisms that are responsible for regulation of the IGF-IR gene in prostate cancer, however, remain largely unidentified.
The familial breast and ovarian cancer susceptibility gene (BRCA1) gene encodes a 220-kDa phosphorylated transcription factor with tumor suppressor activity (12). BRCA1 mutation was correlated with the appearance of breast and ovarian cancer at very young ages, although BRCA1 has been also implicated in the etiology of sporadic types of cancer (13–15). BRCA1 is normally targeted to the nucleus via two nuclear localization signals (16). The BRCA1 polypeptide participates in multiple biological pathways, including gene transcription, DNA damage repair, cell growth, and apoptosis (17). Both direct and indirect types of evidence support a tumor suppressor role of BRCA1. Direct evidence was provided by studies showing that transfer of BRCA1 protein arrested growth of breast and ovarian cancer cells, whereas inactivation of the endogenous BRCA1 gene induced cellular transformation (18). On the other hand, indirect evidence was provided by studies showing somatic allelic loss of 17q21 in breast and ovarian tumors (19).
The involvement of BRCA1 in prostate cancer etiology has been the focus of controversial debate. Previous studies have suggested that BRCA1 functions as an AR coregulator and plays a positive role in androgen-induced cell death (20, 21). Consistent with a potential tumor suppressor role, prostate cancer cells DU-145 transfected with a wild-type BRCA1 exhibited a decreased proliferation rate, an increased sensitivity to chemotherapy drugs, increased susceptibility to drug-induced apoptosis, and alterations in expression of key regulatory proteins (22). Furthermore, BRCA1 splice variant BRCA1a was recently shown to display antitumoral activity in triple-negative prostate cancer cells (23). Linkage studies have provided conflicting data regarding a potential correlation between BRCA1/BRCA2 status and a familial history of prostate cancer. Thus, Struewing et al (24) reported that by the age of 70, the estimated risk of prostate cancer in Ashkenazi Jewish men carrying mutations in the BRCA1 or BRCA2 genes was 16 percent. The hypothesis that deleterious mutations in BRCA2 are associated with an increased risk of prostate cancer was further substantiated by studies showing that this type of mutations are more likely to be found in unselected individuals with prostate cancer than in age-matched controls (25). In a recent study, the Icelandic BRCA2 999del5 founder mutation was strongly associated with rapidly progressing lethal prostate cancer (26). Specifically, patients carrying this mutation had a lower mean age of diagnosis, more advanced tumor stages, and shorter median survival times. On the other hand, a study by Vazina el al. (27) concluded that the rate of predominant Jewish BRCA1 and BRCA2 mutations in prostate cancer patients was not significantly different from that in the general population. Likewise, no strong evidence for a role of BRCA1 or BRCA2 mutations in the development of prostate cancer was provided by other reports (28, 29).
In view of the putative role of BRCA1 in prostate cancer, and to expand our previous studies on the interactions between BRCA1 and the IGF system, we evaluated in the present study (i) the potential correlation between BRCA1 expression and tumor status in a collection of prostate cancer specimens, and (ii) the capacity of BRCA1 to control IGF-IR expression in prostate cancer cells with different AR status. Results obtained indicate that BRCA1 is expressed at relatively high levels in prostate cancer compared to a very low BRCA1 immunostaining in normal prostate epithelium. There is a significant negative relationship between IGF-IR and BRCA1 expression levels in AR-negative prostate cancer cell lines, whereas in cancer with an active AR this relationship is positive. In addition, we provide evidence that the IGF-IR gene is differentially regulated by BRCA1 in prostate cancer cells with different AR status.
The tissue samples used in this study were tissue microarrays (TMAs) made from human radical prostatectomy specimens acquired and used in conformity with an IRB-approved protocol at the University of Washington. Median patient age was 58 years old (range: 48 to 74 years old). The prostates ranged in weight from 21 to 123 grams (median weight 42 grams). At presentation, 58% of patients were clinical stage cT1 and 42% cT2. The range of serum PSA was 2.2 to 24 ng/m. (median 5.4 ng/ml).
Two TMAs were used for these studies. All samples in all arrays were provided in duplicate as 0.6 mm diameter tissue cores. These arrays contained 203 prostate carcinomas exhibiting a range of Gleason grades (72% Gleason pattern 3, 27% Gleason pattern 4, 1% Gleason pattern 5) and 80 samples of non-malignant prostate tissue of different biologic states (normal, atrophy, and BPH).
Antibodies recognizing BRCA1 (Cat. number sc-7867, Santa Cruz Biotechnology, Santa Cruz, CA, USA) and IGF-IR α-subunit (Cat. number sc-461, Santa Cruz Biotechnology) were used to stain the TMAs. BRCA1 blocking peptide Cat. number #ab40076 was purchased from Abcam (Cambridge, MA, USA). Specificity of staining was confirmed by omission of the primary antibody, by immunostaining the sections with a primary antibody against an irrelevant antigen, and by preincubating the anti-BRCA1 in a five-fold molar excess concentration of BRCA1 peptide prior to incubating the sections with primary anti-BRCA1 antibody. In addition, specificity was determined by Western blot of a human prostate xenograft, LuCaP 35, which expresses BRCA1 protein. The Western blot was stained with BRCA1 antibody or a 10-fold excess of the BRCA1 blocking peptide plus BRCA1 antibody. Antigen was localized using a three-step avidin-biotin-peroxidase method.
In brief, deparaffinized sections were rehydrated in phosphate-buffered saline (PBS) and subjected to antigen retrieval using a microwave (15 minutes in citrate buffer solution). Sections were then incubated sequentially in solutions of 5% albumin in PBS, 10% hydrogen peroxide in water, primary antibody, secondary antibody [biotinylated anti-rabbit IgG (BA-1000, Vector Labs, Burlingame, CA, USA)], and avidin-biotin-peroxidase solution (Vector Labs) with interval washes in PBS. Reaction product was detected by incubating the sections in an aqueous solution of 0.05% diaminobenzidine and 0.3% hydrogen peroxide. The sections were counterstained with hematoxylin. Nuclear BRCA1 localization was assessed by staining the TMA without hematoxylin counterstain to more clearly show only BRCA1 staining in the nucleus.
The immunohistochemical stains were evaluated in a blinded fashion by two independent pathology reviewers using the following scale: 0 equals no expression, 1 equals faint/focal/equivocal staining, 2 equals < 50% of the cells express the antigen, 3 equals > 50% of cells express the antigen. The following cell types were evaluated: secretory and basal epithelial cells, high-grade prostate intraepithelial neoplasia, and Gleason pattern 3, Gleason pattern 4, and Gleason pattern 5 tumor cells. Statistical analysis was performed using two-way ANOVA and Bonferonni correction for multiple comparisons. Statistics were performed using the Statview Statistical Program™ (Calabasas, CA, USA).
Derivation of the P69 and M12 cell lines has been previously described (30, 31). Briefly, the P69 cell line was obtained by immortalization of prostate epithelial cells with simian virus 40 T antigen, and the M12 cell line was derived by injection of P69 cells into athymic nude mice and serial reimplantation of tumor nodules into nude mice. P69 and M12 cells were cultured in RPMI 1640 medium supplemented with 10 ng/ml EGF, 0.1 nM dexamethasone, 5 μg/ml insulin, 5 μg/ml transferrin, and 5 ng/ml selenium. P69 cells are responsive to IGF-I and are rarely tumorigenic whereas M12 cells are highly tumorigenic and metastatic, and exhibit a reduced IGF-I responsiveness (32). P69 and M12 cells express extremely low levels of AR. The LnCaP C4-2 cell line was maintained in T-Medium (InVitrogen) containing 5% fetal bovine serum.
Cells were harvested with ice-cold PBS containing 5 mM EDTA, and lysed in a buffer containing 150 mM NaCl, 20 mM Hepes, pH 7.5, 1% Triton X-100, 2 mM EDTA, 2 mM EGTA, 1 mM PMSF, 2 μg/ml aprotinin, 1 mM leupeptin, 1 mM pyrophosphate, 1 mM vanadate, and 1 mM DTT. Protein content was determined using the Bradford reagent. Samples were electrophoresed through 10% SDS-PAGE, followed by blotting of the proteins onto nitrocellulose membranes. After blocking with 5% skim milk in T-TBS (20 mM Tris-HCl, pH 7.5, 135 mM NaCl, and 0.1 % Tween-20), blots were incubated with a rabbit polyclonal anti-human IGF-IR β-subunit antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), washed with T-TBS, and incubated with an Horseradish peroxidase-conjugated secondary antibody. In addition, blots were incubated with antibodies against BRCA1 (C20, Santa Cruz Biotechnology), tubulin (T-5168, Sigma-Aldrich Co., Saint Louis, MO, USA), Akt and phospho-Akt (Ser473) (#9272 and #9271, respectively, Cell Signaling, Beverly, MA, USA), Erk1 and phospho-Erk1/2 (Thr202/Tyr204) (SC-94, Santa Cruz Biotechnology, and #9101, Cell Signaling, respectively), and actin (A-5060, Sigma-Aldrich Co.). Proteins were detected using the SuperSignalR West Pico Chemiluminescent Substrate (Pierce, Rockford, IL, USA).
An expression vector encoding wild-type BRCA1 was constructed by cloning the BRCA1 cDNA into artificially engineered HindIII and NotI sites in the pcDNA3 vector (Invitrogen, Carlsbad, CA, USA) (33). The BRCA1 vector was kindly provided by Dr. Lawrence Brody (National Institutes of Health, Bethesda, MD, USA). For transient cotransfection experiments an IGF-IR promoter luciferase reporter construct was employed that includes 476 bp of 5′-flanking and 640 bp of 5′-untranslated regions of the IGF-IR gene [p(−476/+640)LUC]. The promoter activity of this genomic fragment has been previously described (34). P69 and M12 cells were transfected with 1 μg of the p(−476/+640)LUC reporter construct, along with 1 μg of the BRCA1 expression vector and 0.3μg of a β-galactosidase expression plasmid (pCMV-β, Clontech, Palo Alto, CA, USA) using the Jet-PEI™ (Polyplus, Illkirch, France) transfection reagent. Promoter activities were expressed as luciferase values normalized for β-galactosidase activity.
For stable transfections, parental P69 and M12 cells were plated in 6-well plates and transfected with a wild-type BRCA1 expression vector (or pcDNA3 as a control) using the Jet-PEI™ reagent. After 24 h, selection by 500 μg/ml of G418 (Geneticin, A.G. Scientific, Inc., San Diego, CA, USA) was started. Following 2 weeks of G418 selection, independent colonies were picked up and BRCA1 expression was assessed by quantitative RT-PCR, as described below. Stable-transfected clones employed in this study expressed at least 50% more BRCA1 mRNA than control cells.
Quantitative Real-time PCR was performed using TaqMan® Universal PCR MasterMix and Assay-on-Demand Gene Expression primers and probes (Applied Biosystems, Foster City, CA, USA). An ABI Prism 7000 Sequence Detection System was employed. The GAPDH mRNA levels were analyzed as an internal control and used to normalize the BRCA1 mRNA values. Amplification was carried out after an incubation of 2 min at 50°C and 10 min at 95°C, followed by 20 cycles at 95°C for 15 sec, 1 min at 55°C, and 30 sec at 72°C. The number of PCR cycles to reach the fluorescence threshold was the cycle threshold (Ct). Each cDNA sample was tested in triplicate and mean Ct values are reported. Furthermore, for each reaction, a “no template” sample was included as a negative control. Fold differences were calculated using the 2ΔΔCt method.
Cells were plated in 6-well plates (2 × 105 cells/well) in complete medium. After 24 h the medium was changed to fresh, serum-containing medium. Cells were counted daily by trypsin treatment, followed by Trypan Blue staining, and manual counting with a hemocytometer. At least four fields were counted at each time point and ligand dose. Proliferation experiments were replicated using XTT staining (Biological Industries, Beit Haemek, Israel), with similar results.
The potential involvement of tumor suppressor BRCA1 in the etiology of prostate cancer has been the topic of controversial research. To investigate the expression of BRCA1 in primary prostate tumors, immunoreactive BRCA1 was measured in two TMAs, which contained 203 prostate cancer specimens (Figure 1). Only specimens including both tumor and normal prostate epithelium were included in our analysis. In general, no to very faint BRCA1 immunoreactivity was observed in benign glands, while variably intense staining was seen in prostate cancer. Statistical analysis of the data indicates a highly significant difference (p < 0.001) between BRCA1 expression levels in prostate cancer compared to normal adjacent tissue (Figure 1C). IGF-IR immunostaining revealed no correlation between BRCA1 and IGF-IR staining in the benign luminal cells of the 203 specimens on the TMAs that stained positively for both proteins (r= −0.11, p > 0.05). In contrast, in the malignant tissue from the same TMAs, there was a significant positive correlation between IGF-IR and BRCA1 (r = 0.21, p < 0.02). IGF-IR levels were significantly higher in the malignant epithelium compared to normal luminal cells (p < 0.01). With respect to AR expression in the tissue samples by benign prostate luminal cells, 6% lacked AR immunoreactivity, 18% expressed AR at a level of 1, 31% at 2, and 45% at 3. All cancers expressed AR: 24% at 1, 34% at 2, and 42% at 3. BRCA1 was mainly localized to the nucleus, as demonstrated by staining the TMA without hematoxylin counterstain (Figure 2A). Furthermore, the specificity of the BRCA1 staining was addressed by preincubating the BRCA1 antibody in a five-fold molar excess concentration of BRCA1 peptide prior to immunostaining. As shown in Figure 2B, the intensity of the BRCA1 staining was significantly reduced in the present of the peptide. Likewise, the intensity of the ~220-kDa BRCA1 band in a Western blot of a prostate cancer xenograft was largely reduced in the presence of a ten-fold molar excess of the BRCA1 blocking peptide (Figure 2C).
To further examine BRCA1 levels in prostate cancer, protein expression was measured by Western blot in a collection of 27 human prostate cancer xenografts, kindly provided by Dr. Robert Vessella (University of Washington, Seattle). Results obtained showed that BRCA1 was expressed in most of the xenografts, however the levels varied over a wide range between xenografts. Likewise, large variations were seen in IGF-IR levels between xenografts. Equivalent amounts of protein were loaded in each lane and equal loading was confirmed by ERK loading (Figure 3).
In view of the fact that progression to advanced stage disease is usually associated with a reduction in IGF-IR levels (9), and given that BRCA1 was previously shown to control IGF-IR levels in breast cancer cells in a negative fashion (35–37), we examined the pattern of expression of BRCA1 in two AR-negative prostate cancer-derived cell lines with different IGF-IR levels. As previously shown, the poorly tumorigenic P69 cell line expressed high IGF-IR levels, whereas the tumorigenic and metastatic M12 derivative exhibited significantly reduced IGF-IR values (32). Western blot analysis of BRCA1 revealed a diametrically opposite pattern of expression. Thus, BRCA1 levels were approximately 4-fold lower in P69 than in M12 cells (Figure 4A). To assess whether the increased BRCA1 levels in M12 cells were associated with corresponding changes in mRNA levels, BRCA1 mRNA levels were measured in both prostate cell lines using quantitative Real Time-PCR. Results obtained showed that BRCA1 mRNA levels in M12 cells were approximately 1.5-fold higher than in P69 cells (Figure 4B).
To examine whether the reciprocal pattern of BRCA1 and IGF-IR gene expression in prostate cancer cells could be due to transcriptional repression of the IGF-IR promoter by endogenous BRCA1, cotransfection experiments were performed in M12 cells using a BRCA1 expression vector along with construct p(−476/+640)LUC, which contains most of the proximal region of the IGF-IR promoter fused to a luciferase gene. Forty-eight hours after transfection cells were harvested and luciferase and β-galactosidase activities were measured. As shown in Figure 5A, BRCA1 induced a significant reduction in IGF-IR promoter activity in comparison to pcDNA3-transfected cells (~50% suppression).
Next we studied whether BRCA1 could suppress the expression of the endogenous IGF-IR gene. For this purpose, P69 and M12 cells were stably transfected with a BRCA1 expression vector followed by selection with G418. Total RNA was prepared from individual clones and BRCA1 mRNA levels were assessed by qRT-PCR. Clones employed in this study expressed at least 50% higher BRCA1 mRNA levels than control, pcDNA3-transfected P69 and M12 cells (Figure 5B). Western blot analysis revealed that endogenous IGF-IR levels were largely reduced in BRCA1-overexpressing P69 cells in comparison to P69 control cells (Figure 5C). On the other hand, no effect was seen on the already reduced endogenous IGF-IR levels in BRCA1-overexpressing M12 cells. An inhibitory effect of BRCA1 in M12 cells, however, was observed in cells that were transiently transfected with a BRCA1 vector. As shown in Figure 5D, BRCA1 expression led to a reduction in total- and phospho-IGF-IR, as well as in phospho-Akt, but not in total and phospho-Erk.
Since most prostate cancer cells contain an AR and since BRCA1 is a recognized enhancing co-regulator of the AR, we determined the response to BRCA1 in the AR positive LnCaP C4-2 line, which expresses an endogenous AR, although mutated in the androgen-binding domain. The results of these studies are shown in Figure 6. These studies demonstrate by both an AR reporter assay and measurement of the AR responsive gene TSC-22 and IGF-IR mRNA by quantitative PCR that in the presence of a functional AR, enhancement of AR signaling by BRCA1 results in increased IGF-IR and TSC22 gene expression (Figure 6A). In addition, BRCA1 expression enhanced AR transcriptional activity, as shown by cotransfection experiments using an AR-responsive luciferase reporter plasmid (AAR3) (Figure 6B).
To assess the potential impact of BRCA1 expression on cell proliferation, BRCA1-overexpressing P69 and M12 cells were plated in 6-well plates at a density of 1×105 cells/well and counted after 24 h, 48 h, and 72 h using a hemocytometer. Results obtained indicate that BRCA1-overexpressing P69 and M12 cells consistently displayed an enhanced proliferation rate in comparison to pcDNA3-transfected cells (approximately 1.5-fold increase at 72 h, p < 0.05, in three independent experiments) (Figure 6C).
Tumor suppressor BRCA1 has been shown to be involved in the regulation of a number of biological processes in various cellular and animal models (17). A potential role for BRCA1 in prostate cancer was suggested by both epidemiological and experimental studies (21, 38), although its mechanisms of action and potential targets have not yet been identified. The IGFs have been recognized as important regulators of prostate epithelial cell growth and differentiation as well as prostate cancer development (39). The IGF-IR, which mediates the mitogenic and antiapoptotic actions of IGF-I and IGF-II, has been identified as a pivotal player in prostate cancer initiation and progression (10). The pattern of expression of the IGF-IR gene through the various stages of the disease, however, remains a controversial subject. Thus, while studies have shown that progression of prostate cancer xenografts to androgen independence is associated with a large increase in IGF-IR mRNA levels (compared to the original androgen-dependent tumors) (40), other reports provided substantial evidence that IGF-IR levels were decreased in human prostate carcinoma compared to benign prostate epithelium (9). Consistent with this later study and, as shown in the present paper, IGF-IR levels are much higher in the non-metastatic prostate epithelial cell line P69, compared to its metastatic derivative, the M12 cell line (32). Furthermore, while IGF-IR mRNA levels were shown to be largely suppressed in bone marrow metastases (11), other studies reported a persistent expression of the IGF-IR gene in prostate metastases (41). Our data, showing a negative correlation between IGF-IR and BRCA1 levels in benign luminal cells and a positive correlation in the malignant tissues, suggest that in the transition from benign to malignant prostate epithelium there is a potential enhancement of the IGF system with up-regulation of IGF-IR by BRCA1. Whether this is a direct interaction cannot be determined by this type of correlation analysis.
The present study identifies BRCA1 as a novel player in prostate cancer and establishes a functional link between BRCA1 and the IGF-IR with potentially relevant physiological and pathological implications in the prostate. Immunohistochemistry revealed that BRCA1 levels were approximately 60% higher in transformed epithelium in comparison to normal tissue (p < 0.001). This paradoxical pattern of expression is consistent with the results of assays showing that BRCA1-overexpressing cells exhibit an enhanced proliferation rate. In addition, data is also consistent with the results of ontogenetic analyses in rodents showing that BRCA1 is highly expressed in rapidly proliferating cells (42). BRCA1 expression is induced by positive growth signals at the cell cycle point where cells become committed to replicate their DNA and undergo cell division (17). Maximal BRCA1 expression was detected during the pre-replicative (G1) phase of the cell cycle (43), and it was proved that BRCA1 is involved in the control of the G1-S and G2-M transition checkpoints (44). Furthermore, we have recently shown that IGF-II, whose levels are largely enhanced in prostate carcinoma, is a potent stimulator of BRCA1 expression (9, 45). On the other hand, BRCA1 overexpression in DU-145 prostate cancer cells was previously shown to cause a very small decrease in proliferation rate, as measured by 3[H]-thymidine uptake (46). However, BRCA1 expression was associated with constitutive activation of STAT3, a transcription factor with crucial roles in cell transformation and tumor formation. Moreover, the fact that reduction of STAT3 levels with antisense oligomers inhibited cell proliferation suggests that BRCA1 expression may elicit a cell survival signal with importance in prostate cancer progression. Further support to the notion that BRCA1 may be involved in early (androgen-dependent) stages of the disease is provided by studies showing that BRCA1 directly interacts with AR and stimulates the activity of androgen response elements in prostate cancer cells (20). Of interest, a recent study has shown that the BRCA1 gene is overexpressed in conjunction with a network of genes related to BRCA1 function in aggressive prostate, breast, and lung cancers in transgenic models associated with integrated SV40 T/t-antigens expression (47). The apparent paradox between the increased BRCA1 levels in prostate cancer and a putative tumor suppressing activity may potentially stem from the multiple and often opposite cellular pathways elicited by BRCA1 (21).
While the biological significance of IGF-IR reduction in prostate cancer is still unclear, the data presented here demonstrates that the IGF-IR gene is a downstream target for BRCA1 action in this organ. In prostate cancer cells not expressing an AR, BRCA1 expression resulted in a ~50% reduction in the activity of a cotransfected IGF-IR promoter construct, probably by a direct effect at the IGF-IR promoter. The physiological relevance of these results is highlighted by the fact that the endogenous IGF-IR gene, as well as IGF-IR and Akt phosphorylation, were reduced in BRCA1-expressing M12 prostate cancer cells. However, in prostate epithelial cells that express an active AR, the effect of BRCA1 on IGF-IR gene expression is mediated through its enhancement of AR transcription and subsequent AR-mediated IGF-IR expression. These results are consistent with studies showing an interplay between BRCA1 and AR in transcriptional regulation (48). In terms of the mechanism of action of BRCA1, we have previously identified a proximal IGF-IR promoter region that mediated the effect of BRCA1 (35, 36). Specifically, this region included a cluster of four GC boxes, which are bona fide binding sites for zinc finger protein Sp1. While we were unable to show direct binding of the in vitro translated BRCA1 protein to this promoter region, we identified a BRCA1 domain involved in Sp1 binding. Physical interaction between BRCA1 and Sp1 prevented binding of the zinc finger to cis-elements in the IGF-IR promoter, with ensuing reduction in promoter activity. A related mechanism of action was recently reported for the von Hippel-Lindau (VHL) tumor suppressor in the specific context of IGF-IR regulation in clear cell renal carcinoma (49). Thus, similar to BRCA1, VHL was shown to reduce IGF-IR promoter activity and mRNA levels via a mechanism that involves functional and physical interactions between VHL and Sp1.
In summary, we have shown that BRCA1 regulates IGF-IR gene expression in prostate cancer cells via a mechanism that involves repression of IGF-IR gene transcription. In addition, immunohistochemical data is consistent with a potential survival role of BRCA1 in prostate cancer. Regulation of IGF-IR expression by BRCA1 may constitute a novel control mechanism that allows the IGF system to engage in both differentiative and proliferative types of actions.
The breast and ovarian cancer susceptibility gene-1 (BRCA1) was originally identified as a protein whose mutated form was associated with familial breast and/or ovarian cancer. However, it is clear that the non-mutated (wild type) form of BRCA1 has distinct cellular functions, including activity as an androgen receptor (AR) co-activator as well as inhibition of insulin-like growth factor-I receptor (IGF-IR) gene expression. In this study, we were interested in determining the role BRCA1 may have in regulation of the IGF-IR gene in prostate cancer. We have demonstrated that BRCA1 protein expression is increased in prostate cancer but rather than suppressing IGF-IR expression, as we demonstrate in AR-negative prostate epithelial cell lines, we show that BRCA1 is positively correlated with IGF-IR. Further we demonstrate that the mechanism responsible for this correlation involves enhancing AR transactivation. These findings are of relevance because they demonstrate a new mechanism for IGF and AR stimulation of prostate cancer and further support the relevance of targeting AR and IGF-IR in prostate cancer with BRCA1 expression as a marker for defining the target activity.
This work was performed in partial fulfillment of the requirements for a Ph.D. degree by Hagit Schayek in the Sackler Faculty of Medicine, Tel Aviv University. The authors wish to thank Dr. L. Brody (National Institutes of Health, Bethesda, MD) for providing the BRCA1 expression vector, Dr. R. Vessella (University of Washington, Seattle, WA) for prostate cancer xenografts, and Ms. Tal Ohayon for help with the manuscript. This research was supported by Grant 2003341 of the United States-Israel Binational Science Foundation (to H.W. and S.R.P.) and CA97186-06 and Veterans Affairs Research Service (to S.R.P.).