Androgen-Independent Growth of Lncap Prostate Cancer Cells Is Mediated by Gain-Of-Function Mutant P531

[CANCER RESEARCH 63, 2228–2233, May 1, 2003] Androgen-independent Growth of LNCaP Prostate Cancer Cells Is Mediated by Gain-of-Function Mutant p531

Nancy J. Nesslinger, Xu-Bao Shi, and Ralph W. deVere White2 Department of Urology, University of California, Davis, School of Medicine, Sacramento, California 95817

ABSTRACT versial, most agree that mutations of p53 are common in advanced CaP. Navone et al. (6) reported elevated p53 protein expression levels Mutations of p53 are common in hormone-refractory prostate cancer in 94% (16 of 17) of hormone-refractory specimens and 22% (6 of 27) (CaP), suggesting the possibility that these mutations may be involved in of primary untreated tumors. Heidenberg et al. (7) confirmed p53 the progression of CaP to androgen-independent (AI) growth. However, at present no direct evidence has been presented linking p53 mutations with gene alterations in 82% (9 of 11) of representative specimens. These AI growth of CaP. We established five stably transfected LNCaP cell lines: authors found a clear progression in the number of cases having p53 four containing gain-of-function (GOF) mutant p53 alleles (G245S, alterations from untreated primary to hormone-refractory disease (7). R248W, R273H, and R273C) and one containing a non-GOF p53 mutant In our laboratory, we have reported previously the frequency of p53 allele (P151S). The four GOF p53 sublines were able to grow under alterations to be ϳ39% in surgically resected primary CaPs (8) and androgen-depleted conditions, whereas the LNCaP parental line, vector- 71% in CaP metastases to bone (9). Thus, investigations of clinical only line, and the non-GOF line were unable to grow. To investigate the samples are in agreement that there is an increased rate of p53 mechanism of the AI growth displayed by the GOF p53 mutants, Western mutation in advanced CaP. These observations support the hypothesis blotting or ELISA were used to examine the expression of the androgen that p53 mutations may be involved in the progression of CaP. receptor (AR), the AR-regulated prostate-specific antigen (PSA), as well However, to date this hypothesis has not been tested by direct exper- as Akt and Bcl-2 under androgen-depleted conditions. On androgen ablation, the levels of AR decreased in the four GOF p53 sublines compared imentation. with the control lines. This decreased AR expression was accompanied by We have published previously an extensive functional analysis of attenuated receptor activity, because a decrease in prostate-specific anti- 16 mutant p53 alleles derived from CaP (10). We observed that some gen levels compared with parental LNCaP cells was also observed. Levels of the p53 mutants that can dominate over wt p53 also possessed GOF of phosphorylated Akt increased in both the GOF p53 sublines and the properties, defined as the induction of genomic instability (11), tu- control lines. Bcl-2 remains unchanged or showed reduced expression in morigenicity (12, 13), or transactivation of the promoters of certain all of the cell lines in the absence of androgen compared to the presence of genes, such as MDR-1 and proliferating cell nuclear antigen that are androgen. These observations suggest that GOF p53 mutants mediate the not regulated by wt p53 (14). To study whether these GOF p53 AI growth of LNCaP cells in an AR-independent fashion, and that both mutants mediate progression of CaP to androgen independence, we Akt and Bcl-2 are not involved in this process. selected four GOF alleles (G245S, R248W, R273H, and R273C), which are frequent in CaP (15), and a non-GOF mutant allele (P151S). INTRODUCTION The four GOF alleles possess different GOF properties including the

3 ability to transactivate the promoter of the MDR-1 gene, promotion of CaP is a hormone-dependent tumor that requires androgen to growth in soft agar, and/or an increased S phase fraction (10). We develop, grow, and differentiate (1). Clinically, CaP is treated differ- found that the LNCaP cells transfected with each of four GOF mutant ently depending on the stage of the disease. In men with localized alleles were able to grow in the absence of androgens, whereas the disease, radical prostatectomy and definitive radiotherapy are the non-GOF P151S-transfected LNCaP cells and the LNCaP parental ϳ main curative treatments, achieving clinical cures in 80% of cases. cells could not, indicating that GOF p53 mutations can lead to AI However, in men with metastatic disease, androgen blockade is the growth of LNCaP cells. In addition, we determined the expression sole treatment available. Androgen ablation usually decreases the levels of several proteins that are known to be involved in AI growth volume of the primary and metastatic lesions by inducing apoptosis of CaP tumors, to better understand their role in the AI growth of the (2). However, after this initial response, the tumors recur in an AI GOF p53 sublines. To our knowledge, this is the first report demon- form that is unresponsive to additional androgen withdrawal (3) and strating the direct involvement of GOF p53 mutations in the progres- are resistant to cure by chemotherapy (4). The mechanism of progression of CaP to AI growth. sion to hormone independence remains unclear. The p53 tumor suppressor gene is a transcription factor that regu- MATERIALS AND METHODS lates the expression of genes involved in cell-cycle inhibition, apoptosis, genetic stability, and angiogenesis (5). Mutation of p53 is one Cell Culture. LNCaP cells were obtained from the American Type Culture of the most common genetic events occurring in human cancer. In Collection (Rockville, MD). Cells were routinely maintained in androgen- CaP, p53 alterations have been investigated widely. Although the containing medium, RPMI 1640 supplemented with 10% FBS (Omega Scien- reported frequencies of p53 mutations in primary CaP remain contro- tific, Tarzana, CA). The androgenic activity of 10% FBS has been shown to be equivalent to the activity of 0.05 nM R1881 (16). For the androgen-depletion experiments, cells were grown in androgen-depleted medium, phenol red-free Received 8/21/02; accepted 3/5/03. The costs of publication of this article were defrayed in part by the payment of page RPMI 1640 supplemented with 5% charcoal/dextran-treated FBS (HyClone, charges. This article must therefore be hereby marked advertisement in accordance with Logan, UT). 18 U.S.C. Section 1734 solely to indicate this fact. Stable Transfections. The G245S, R248W, R273H, R273C, and P151S 1 Supported by NIH Grants NCI RO1 CA 77662 and NCI RO1 92069. alleles were individually cloned into the pCR 3.1 vector (Invitrogen, Carlsbad, 2 To whom requests for reprints should be addressed, at Department of Urology, University of California, Davis, School of Medicine, 4860 Y Street, Suite 3500, Sacra- CA) downstream of the cytomegalovirus promoter. Each allele was sequenced mento, CA, 95817. Phone: (916) 734-2824; Fax: (916) 734-8094; E-mail: rwdeverewhite@ to ensure the presence of the mutation of interest. The plasmids containing the ucdavis.edu. different p53 mutant alleles or the empty pCR3.1 plasmid (vector-only) were 3 The abbreviations used are: CaP, prostate cancer; AI, androgen-independent; wt, stably transfected into LNCaP cells using 9 ␮g plasmid DNA and calcium wild-type; GOF, gain-of-function; MDR, multidrug resistance; FBS, fetal bovine serum; PSA, prostate-specific antigen; RT-PCR, reverse transcription-PCR; AR, androgen recep- phosphate precipitation with a glycerol shock at 5 h post-transfection. After tor; p-Akt, phosphorylated Akt; BAG-1L, long BAG-1. 48 h, cells were grown under G418 (500 ␮g/ml) selection for 2–3 weeks until 2228

isolated colonies appeared. Colonies were selected and expanded in 24-well plates before being transferred to culture flasks. Cell Counts. Cells from the LNCaP parental and mutant p53-transfected sublines were plated in triplicate in androgen-containing medium on 60-mm2 Petri dishes using 1 ϫ 105 cells/dish. Cells were allowed to attach overnight before switching to androgen-depleted medium. At days 0, 1, 3, and 5, the cells were trypsinized, pelleted, and resuspended in 1 ml of medium. A 10-␮l aliquot was combined with an equal volume of trypan blue, and the cells were counted using a hemocytometer. Western Blotting. Pelleted cells were lysed in lysis buffer [150 mM NaCl, 50 mM Tris-HCl (pH 8.0), 0.5% sodium deoxycholate, 1% Triton X-100, and Fig. 1. Expression of p53 and p21waf1 protein in the mutant p53-transfected LNCaP 0.1% SDS] supplemented with 10 mg/ml leupeptin, 0.1 M aprotinin, 0.1 M cells. Top, Western blotting demonstrates the overexpression of p53 protein in each ␮ phenylmethylsulfonyl fluoride, 0.1 M NaVO4, and 0.05 units/ l RNase-free subline compared with the parental LNCaP, confirming the presence of exogenous p53 in DNase. The cell lysates were spun at 12,000 rpm at 4°C, and the supernatants these sublines. Middle, Western blotting analysis of p21waf1 expression in mutant p53- transfected LNCaP sublines 4 h after 10-Gy radiation. The parental LNCaP cells, after were used for protein analysis. Protein concentrations were determined using waf1 ␮ induction by radiation, express high levels of p21 , whereas the four GOF p53 sublines the BCA Protein Assay Reagent (Pierce, Rockford, IL). Protein (25–50 g) have comparatively lower p21waf1 levels, suggesting a dominant-negative effect. Bottom, was separated on an 8% (for AR), a 12% (for Akt), or a 15% (for Bcl-2) the ␤-actin, used as a standard to confirm equivalent loading, was derived from the same SDS-PAGE mini-gel and transferred to nitrocellulose membranes. The AR protein preparations as was used for the p53 and p21 Western blotting. Each lane contains ␮ antibody (NeoMarkers, Fremont, CA), p-Akt antibody (Cell Signaling Tech- 50 g protein. nology, Beverly, MA), and Bcl-2 antibody (BD PharMingen, San Diego, CA) were all used at a 1:1,000 dilution, whereas the ␤-actin antibody (Sigma, St. Louis, MO) was used at a 1:10,000 dilution. the four GOF p53 sublines were grown for 5 days in androgen- PSA ELISA. The ELISA reactions were carried out using 1 ϫ 105 cells depleted medium. The parental LNCaP cells and the P151S subline grown in 24-well plates in 500 ␮l of androgen-containing or androgen- underwent a growth arrest (Fig. 2A). Similarly, the vector-only subline depleted medium. The cells were grown for 3 days before plating in 24-well demonstrated growth arrest (data not shown). In contrast, the GOF plates. Cells were grown for additional 2 days, and medium was collected. A p53 sublines continued to grow in androgen-depleted medium (Fig. PSA ELISA kit (Medicorp, Montreal, Quebec, Canada) was used to measure 2A), thus exhibiting an AI phenotype. The growth rates of these lines PSA protein in 20 ␮l of medium in duplicate as per the manufacturer’s in androgen-depleted medium were examined by counting cells on directions. Dilutions were made as necessary to ensure that the sample readings fell within the standard curves. The experiment was performed three times and days 0, 1, 3, and 5. As shown in Fig. 2B, the three control lines (the the results averaged. parental LNCaP, vector-only subline, and the P151S subline) failed to PSA RT-PCR. RNA was extracted from cell pellets using TRIzol Reagent grow in androgen-depleted medium, whereas the four GOF p53 sub- (Life Technologies, Inc., Grand Island, NY). cDNA synthesis was carried out lines were able to proliferate in the medium with a doubling-time of using Moloney murine leukemia virus Reverse Transcriptase (Promega, Mad- ϳ2.5 days for three of the sublines. We additionally compared the ison, WI). The following primers were used to amplify the PSA transcripts: four GOF p53 sublines to determine the effect of different GOF p53 PSA-forward: 5Ј-TGGGAGTGCGAGAAGCATTC and PSA-reverse: 5Ј- mutants on the growth of LNCaP cells. Cells were grown for 5 days GCACACAGCATGAACTTGGTCAC. The following amplification condi- in the presence or absence of androgens, and the fold increases in cell tions were used in a MJ PTC-100 thermal cycler (MJ Research, Incline number were calculated by dividing the cell number in androgen- Village, NV): an initial denaturation for 5 min at 94°C, 30 cycles of 94°C for containing medium by cell number in androgen-depleted medium. We 30 s, 59°C for 30 s, and 72°C for 1 min, followed by a final extension for 10 min at 72°C. observed that the growth of three GOF sublines (G245S, R248W, and R273C) was slightly faster in androgen-containing medium than in RESULTS androgen-depleted medium, whereas the R273H subline had similar growth rates in the two media (Fig. 2C). Overall, these experiments Establishment of Stably Transfected LNCaP Sublines. Four demonstrate that only GOF mutant p53 alleles resulted in the AI GOF p53 mutant alleles (G245S, R248W, R273H, and R273C) and the growth of LNCaP cells. non-GOF P151S mutant allele were selected for the present study. All Expression and Activity of the AR. LNCaP cells express a mu- are dominant negative, with the R248W, R273H, and R273C mutants tant AR that has been identified frequently in CaP patients who have dominant over two wt copies of p53, and the G245S and P151S alleles AI disease (17). Despite this mutation, the LNCaP AR has been dominant over one wt copy of p53. The four mutant alleles were able shown to bind androgens and maintain its ligand-dependent transac- to activate the MDR-1 promoter, to grow in soft agar, and/or to tivational activity (18). To investigate how this mutant receptor is increase S phase, exhibiting GOF activity (10). Each of the five involved in the GOF mutant p53-mediated AI growth, cells were mutant alleles was stably transfected into LNCaP cells to yield four grown in androgen-containing or androgen-depleted medium for 5 GOF and a non-GOF p53 subline. As a control, the empty pCR3.1 days, and AR expression was evaluated by Western blotting. As plasmid (vector-only) was also stably transfected into LNCaP cells. shown in Fig. 3, the expression levels of AR protein were similar in The expression of p53 in these stable sublines was confirmed by the LNCaP parental cells, vector-only, and five mutant p53-trans- Western blotting. In each case, the transfected subline demonstrated fected sublines when grown in androgen-containing medium, indicat- increased p53 protein compared with the LNCaP parental cells (Fig. ing that transfection of these mutant p53 alleles into LNCaP cells did 1), indicating that the exogenous p53 was expressed. To examine not affect the basal levels of receptor expression. When these lines whether these exogenous p53 mutants dominate over the LNCaP were grown in androgen-depleted medium, a slight decrease in the AR endogenous wt p53, the expression level of p53-regulated p21waf1 levels was observed in the parental, vector-only, and P151S lines. protein after 10-Gy radiation was tested using Western blotting anal- However, in the case of the four GOF p53 sublines, the expression of ysis. As shown in Fig. 1, reduced p21waf1 levels were observed in AR was markedly down-regulated after 5-day growth in androgen- these GOF sublines as compared with the parental LNCaP cells, depleted medium. These results suggest that the AR is not required for suggesting a dominant-negative effect. GOF p53 mutant-induced AI growth of LNCaP cells. The reduction of AI Growth of LNCaP Cells Expressing GOF Mutant p53. The AR expression after androgen depletion in the GOF p53 sublines LNCaP parental cells, the vector-only subline, the P151S subline, and prompted us to examine the transactivational activity of the AR. We 2229

Fig. 2. Growth of LNCaP parental and mutant p53-transfected sublines in androgen-depleted medium. A, the LNCaP parental cells and mutant p53-transfected cells grow in androgen-containing medium and/or in androgen-depleted medium for 5 days. Growth arrest was seen for the LNCaP cells and the non-GOF P151S subline (LNCaPϩP151S) in androgen-depleted medium. In contrast, the GOF p53 sublines (LNCaPϩG245S, LNCaPϩR248W, LNCaPϩR273H, and LNCaPϩR273C) continued to grow in the androgen-depleted medium. B, the growth curves show that the four GOF p53 sublines continued to proliferate in androgen-depleted medium, whereas the LNCaP parental, vector-only, and P151S sublines do not proliferate in this medium. C, comparison of the effect of different GOF p53 mutants on the growth of LNCaP in the presence or absence of androgens. The Y axis is the fold-change expressed as the cell number in androgen-containing medium divided by the cell number in androgen-depleted medium on day 5. The LNCaP parental and P151S subline are used as controls, and have a large fold increase, reflecting the growth arrest of these cells in androgen-depleted medium. analyzed the expression levels of PSA protein, a downstream gene Expression of Akt in GOF Mutant p53 Sublines. Because Akt product regulated by the AR in the presence of androgen. Consistent has been implicated in promoting CaP cell survival and growth in the with AR expression levels, both the LNCaP parental cells and the absence of androgen (19), we examined the levels of p-Akt. All of the P151S cells expressed high levels of PSA protein in androgen- cell lines were grown for 5 days in either androgen-containing or containing medium, whereas the PSA levels were markedly decreased androgen-depleted medium, and the p-Akt levels were determined in androgen-depleted medium (Fig. 4A). In contrast, introduction of using Western blot analysis of the extracted protein. As shown in Fig. the GOF p53 alleles into LNCaP cells resulted in significant reduction 5, the transfection of GOF mutant p53 alleles into LNCaP cells did not of the levels of PSA protein (P Ͻ 0.05) when these cells were grown significantly alter p-Akt expression when the cells were grown in in androgen-containing medium, although their AR expression levels androgen-containing medium. However, in each of the cell lines, the remained similar to those of the control cell lines. Furthermore, when p-Akt levels were increased markedly when the cells were grown in these four GOF sublines were cultured for 5 days in androgen- androgen-depleted medium. These results indicate that up-regulation depleted medium, their PSA was undetectable using the ELISA ap- proach (Fig. 4A). In one experiment, ELISA was also performed on cells containing empty plasmids (vector-only control). The PSA levels were comparable with those of the parental LNCaP cells (data not shown). We additionally analyzed PSA mRNA expression using RT-PCR amplification of cDNA derived from these lines. As shown in Fig. 4B, there is decreased PSA mRNA expression in three GOF p53 sublines (R248W, R273H, and R273C) when grown in androgen- containing medium, whereas PSA gene expression was undetectable Fig. 3. Western blotting analysis of AR expression in GOF p53 sublines and control lines in the presence or absence of androgens. There is no change in AR expression in the in the four GOF p53 sublines when grown in androgen-depleted LNCaP parental, vector-only, and P151S sublines grown in androgen-containing medium medium. Therefore, the decreased PSA expression at both protein and or slight AR reduction in androgen-depleted medium. However, in the GOF p53 sublines, there is a marked reduction in AR expression after androgen withdrawal. Each lane mRNA levels may be the result of reduced AR expression and/or contains 30 ␮g of protein, and ␤-actin was used to ensure equivalent loading. The four reduced AR activity. blots shown at the right of the figure are from a separate assay. 2230

and in clinical samples from AI and metastatic tumors (6, 7, 9, 21, 22), this suggests that p53 mutations may play a role in the development of hormone-refractory CaP. This could occur through two mecha- nisms. First, because the primary effect of p53 mutation is loss of tumor suppressor function, inactivating or reducing wt p53 function could result in androgen independence. This theory is supported by several published studies. One showed that CaP tumor in which p53 was inactivated by the large T antigen of SV40 progressed to androgen independence in transgenic mice (23). Burchardt et al. (24) reported recently that a hormone-resistant phenotype could be induced in LNCaP cells by reduction of wt p53 function as the result of transfection of either p53 antisense transcripts or a truncated, dominant-negative form of p53. These studies demonstrated that loss of wt p53 function could contribute to the androgen resistance of CaP cells. Second, it is well established that, in addition to the loss of wt p53 function, p53 mutations can also result in GOF characteristics. In other types of cancer, some GOF p53 mutants have been shown to enhance tumorigenic potential (12) or to up-regulate genes involved in cell proliferation (10, 25). On the basis of these observations, we hypothesized that, beyond their ability to inhibit the activity of endogenous wt p53, GOF p53 mutants may directly contribute to AI growth of CaP cells. To test this hypothesis, five dominant-negative p53 mutant alleles (four having and one lacking GOF properties) were transfected into LNCaP cells, which normally express wt p53. The ability of these p53 mutants to dominate over the endogenous wt p53 Fig. 4. PSA production in the GOF p53 sublines. A, ELISA analysis of PSA expression. waf1 The LNCaP parental and P151S subline grown in androgen-containing medium produce was evidenced by showing that these sublines had reduced p21 high levels of PSA, whereas the four GOF p53 sublines have lower levels of PSA. On expression and increased Bcl-2 expression, because a p53-dependent androgen depletion, PSA levels were much lower in the LNCaP parental and P151S cells, negative response element has been identified in the Bcl-2 gene (26, with levels undetectable in the four GOF mutant p53 sublines. Error bars represent an average of three experiments; bars, ϮSE. B, RT-PCR amplification of PSA. Expression 27). Among these five mutants, only the four GOF p53 mutant of PSA was highest in the LNCaP parental and P151S subline, whereas very little PSA sublines were able to grow in androgen-depleted medium, whereas the expression was seen in the four GOF mutant p53 sublines. N-ras was used as control. non-GOF P151S subline was unable to grow under these conditions. Thus, the data reported here offers strong support to our hypothesis. of p-Akt after androgen depletion is not a specific response of LNCaP In an effort to elucidate the molecular mechanism(s) through which cells to expression of the GOF p53 mutants but rather is a general GOF p53 mutants induce AI growth, we examined whether the response to androgen withdrawal. expression or transactivational activity of the AR was altered by the Bcl-2 Expression in GOF Mutant p53 Sublines. The results transfected p53 mutants. In the GOF p53 LNCaP sublines, we failed described above suggest that the GOF mutant p53-mediated AI to observe an increase in AR expression or in AR-dependent PSA growth may be through a pathway that bypasses the AR. Because expression when grown in androgen-containing medium, whereas we Bcl-2 is a potent inhibitor of apoptosis that has been implicated in the observed a marked decrease in AR and PSA expression when grown development of hormone resistance in CaP (20), we examined these in androgen-depleted medium. This indicates that AI growth of these sublines for their expression of Bcl-2 protein, an obvious bypass GOF p53-transfected cells can occur in an AR-independent fashion. candidate gene. As shown in Fig. 5, transfection of the mutant p53 We observed that these GOF p53 sublines expressed a significantly alleles into LNCaP cells resulted in increased Bcl-2 expression in the low level of PSA in androgen-containing medium when compared G245S, R248W, R273H, and P151S sublines when grown in andro- with the parental LNCaP line and the P151S subline, suggesting that gen-containing medium, as compared with the parental and vector- these p53 mutants repress AR activity. This is consistent with obser- only cells. The up-regulation of Bcl-2 expression may result from a vations reported by other groups. Eastham et al. (28) reported that wt dominant-negative effect of mutant p53 over wt p53. The exception is p53 was able to inhibit the expression of PSA in CaP tumors grown the R273C mutant, which had lower expression than the control lines. in nude mice. Shenk et al. (29) observed recently that in AR-trans- This was a consistent result seen in three independent experiments. fected PC3 cells the R248W p53 mutant weakly repressed AR trans- When these seven lines were grown in androgen-depleted medium for 5 days, their Bcl-2 levels were down-regulated to a greater or lesser extent compared with the corresponding cells grown in androgen- containing medium. These results suggest that down-regulation of Bcl-2 expression is not a specific response of LNCaP cells to the GOF p53 mutants but is a general response to androgen withdrawal.

DISCUSSION Once CaP becomes unresponsive to androgen withdrawal, no ad- Fig. 5. Western blotting analysis of p-Akt and Bcl-2 expression in GOF p53 sublines and control lines. Increased p-Akt was detected in all of the cell lines in the absence of ditional curative treatment is available for this disease. Understanding androgens, indicating a general response to androgen withdrawal. Bcl-2 expression the molecular changes that occur during progression to androgen increased in G245S, R248W, R273H, and P151S sublines in the presence of androgen, and their expression decreased after androgen withdrawal. Levels of Bcl-2 were similar independence is essential for the development of targeted therapy. before and after androgen withdrawal in the remaining three lines. ␤-Actin was used to Because p53 mutations are found at high frequencies in CaP cell lines, ensure equivalent loading. Each lane contains 30 ␮g of protein. 2231

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 2003 American Association for Cancer Research. MUTANT p53 MEDIATES ANDROGEN-INDEPENDENT GROWTH activation of the PSA promoter in the presence of androgen. This vitro. In the presence of these GOF p53 mutants, the expression and negative activity of p53 is independent of specific DNA binding (30), transactivational activity of the AR are down-regulated, indicating and requires the p53 DNA-binding domain and COOH terminus (29). that an activated AR is not needed for GOF p53 mutant-mediated Thus, the four transfected GOF p53 mutants that contain mutations in proliferation in the absence of androgens. Furthermore, Bcl-2 expres- the DNA-binding domain and possess the COOH terminus may be sion is not involved in this AI growth. Whereas increased Akt ex- able to down-regulate the transactivational activity of the AR, result- pression was not induced directly by GOF p53 mutants, it may be ing in the reduced PSA expression. However, the extent of down- associated with androgen ablation and the development of androgen regulation may be dependent on the cellular context. We additionally independence. Taken together, these observations suggest that there found that the expression of AR protein was markedly repressed when are additional pathways involved in the GOF p53 mutant-induced these GOF mutant p53-transfected LNCaP sublines were grown for 5 androgen independence. Our LNCaP sublines should be valuable days in androgen-depleted medium. This was accompanied by a models for investigating these pathways. marked repression of PSA expression at both the protein and the mRNA levels. It is difficult to attribute this down-regulation of PSA ACKNOWLEDGMENTS expression to specific inhibition of AR expression by GOF p53 mutants in an androgen-depleted environment. More probably, the We thank Dr. Arline Deitch for insightful discussions and editorial assist- decreased AR levels that we observed are attributable to degradation ance. We also thank Dr. Philip C. Mack for assistance with the statistical of AR protein in the absence of androgen. This is in accord with the analysis and helpful discussions, and Dr. Clifford G. Tepper for analysis of the observation that the synthetic androgen R1881 increases AR stabili- microarray data. zation (31). The fact that the GOF mutant p53 sublines that express very low levels of AR protein are nonetheless able to grow in REFERENCES androgen-depleted medium additionally supports our hypothesis that 1. Huggins, C., and Hodge, C. V. Studies on prostatic cancer: The effects of castration, AI growth of these sublines is probably via an AR-independent of estrogen and androgen injection on serum phosphatases in metastatic carcinoma of pathway. the prostate. Cancer Res., 1: 293–297, 1941. We examined several additional factors that have been implicated 2. Kyprianou, N., English, H., and Isaacs, J. T. Programmed cell death during regression of PC-82 human prostate cancer following androgen ablation. Cancer Res., 50: in the AI growth of CaP. Akt is a serine-threonine protein kinase that 3748–3753, 1990. inhibits the transactivational activity of the AR (32). Furthermore, Akt 3. Isaacs, J. T. The biology of hormone refractory prostate cancer. Urol. Clin. North expression has been shown to increase under conditions of androgen Am., 26: 263–273, 1999. 4. Oh, W. K., and Kantoff, P. W. Management of hormone refractory prostate cancer: withdrawal. This increased expression is sustained throughout the current standards and future prospects. J. Urol., 160: 1220–1229, 1998. progression to androgen independence (33). Phosphorylation of Akt 5. Vogelstein, B., Lane, D., and Levine, A. J. Surfing the p53 network. Nature (Lond.), 408: 307–310, 2000. results in full activation of this kinase (34). We found that expression 6. Navone, N. M., Troncoso, P., Pisters, L. L., Goodrow, T. L., Palmer, J. L., Nichols, of the active, phosphorylated form of Akt is considerably higher in all W. W., von Eschenbach, A. C., and Conti, C. J. p53 protein accumulation and gene of the LNCaP cell lines when grown in androgen-depleted conditions. mutation in the progression of human prostate carcinoma. J. Natl. Cancer Inst., 85: 1657–1669, 1993. This appears to be a general response to androgen ablation, because it 7. Heidenberg, H. B., Sesterhenn, I. A., Gaddipati, J. P., Weghorst, C. M., Buzard, G. S., is not specifically induced by GOF p53 mutants. Moul, J. W., and Srivastava, S. Alteration of the tumor suppressor gene p53 in a high Bcl-2 overexpression has been associated with the progression of fraction of hormone refractory prostate cancer. J. Urol., 154: 414–421, 1995. 8. Chi, S-G., deVere White, R. W., Meyers, F. J., Siders, D. B., Lee, F., and Gumerlock, CaP to a metastatic phenotype and correlates with the development of P. H. p53 in prostate cancer: frequent expressed transition mutations. J. Natl. Cancer therapeutic resistance (24, 35). Despite these observations, Bcl-2 Inst., 86: 926–933, 1994. 9. Meyers, F. J., Gumerlock, P. H., Chi, S-G., Borchers, H., Deitch, A. D., and deVere expression was somewhat decreased in all of our cell lines maintained White, R. W. Very frequent p53 mutations in metastatic prostate carcinoma and in in androgen-depleted medium. This down-regulation may reflect the matched primary tumors. Cancer (Phila.), 83: 2534–2539, 1998. effect of short-term androgen-withdrawal treatment. 10. Shi, X-B., Nesslinger, N. J., Deitch, A. D., Gumerlock, P. H., and deVere White, R. W. Complex functions of mutant p53 alleles from human prostate cancer. Prostate, A recent study shows that p53 mutants up-regulate the promoter of 51: 59–72, 2002. the BAG-1 gene that expresses a multifunctional antiapoptotic protein 11. Koivisto, P. A., and Rantala, I. Amplification of the androgen receptor gene is (36). The BAG-1L protein that arises from translation initiation at a associated with P53 mutation in hormone-refractory recurrent prostate cancer. J. Pathol., 187: 237–241, 1999. noncanonical CUG codon, has been shown to increase AR transacti- 12. Dittmer, D., Pati, S., Zambetti, G., Chu, S., Teresky, A. K., Moore, M., Finlay, C., and vational activity in the presence of low concentration of androgen Levine, A. J. Gain of function mutations in p53. Nat. Genet., 4: 42–46, 1993. 13. Shaulsky, G., Goldfinger, N., and Rotter, V. Alterations in tumor development in vivo (37). Results from our pilot experiments have demonstrated increased mediated by expression of wild type or mutant p53 proteins. Cancer Res., 51: BAG-1L protein expression in GOF p53 sublines compared with the 5232–5437, 1991. controls, whereas the expression of BAG-1 and its other isoform 14. Roemer, K. Mutant p53: Gain-of-function oncoproteins and wild-type p53 inactiva- tors. Biol. Chem., 380: 879–887, 1999. BAG-1M was not affected in these lines. Whether and how BAG-1L 15. deVere White, R. W., Deitch, A. D., Gumerlock, P. H., and Shi, X-B. Use of a yeast is involved in the GOF p53 mutant-mediated androgen independence assay to detect functional alterations in p53 in prostate cancer: review and future is currently under investigation in our laboratory. directions. Prostate, 41: 134–142, 1999. 16. Kokontis, J., Takakura, K., Hay, N., and Liao, S. Increased androgen receptor activity The contribution of other gene pathways to AI growth in the and altered c-myc expression in prostate cancer cells after long-term androgen presence of GOF p53 mutants is currently under investigation in our deprivation. Cancer Res., 54: 1566–1573, 1994. 17. Taplin, M. E., Bubley, G. J., Ko, Y. J., Small, E. J., Upton, M., Rajeshkumar, B., and laboratory. Microarray assays of these sublines have identified 84 Balk, S. P. Selection for androgen receptor mutations in prostate cancers treated with genes as being differentially expressed in GOF p53 sublines compared androgen antagonist. Cancer Res., 59: 2511–2515, 1999. with vector-only subline. Overall, 61 genes are down-regulated, 18. Veldscholte, J., Ris-Stalpers, C., Kuiper, G. G. J. M., Jenster, G., Berrevoets, C., Claassen, E., vanRooij, H. C. J., Trapman, J., Brinkmann, A. O. and Mulder, E. A whereas 23 are up-regulated. These genes encode proteins having a mutation in the ligand binding domain of the androgen receptor of human LNCaP broad range of functions including metastasis, tumor suppression, cell cells affects steroid binding characteristics and response to anti-androgens. Biochem. cycle control, G-protein signaling, translation, or detoxification. Elu- Biophys. Res. Commun., 173: 534–540, 1990. 19. Wen, Y., Hu, M. C-T., Makino, K., Spohn, B., Bartholomeusz, G., Yan, D-H., and cidation of the relative importance of these gene changes will shed Hung, M-C. HER-2/neu promotes androgen-independent survival and growth of light on GOF p53 mutant-mediated AI growth in CaP. prostate cancer cells through the Akt pathway. Cancer Res., 60: 6841–6845, 2000. 20. Raffo, A. J., Perlman, H., Chen, M-W., Day, M. L., Streitman, J. S., and Buttyan, R. In conclusion, we have developed a cell model and demonstrated Overexpression of bcl-2 protects prostate cancer cells from apoptosis in vitro and that GOF p53 mutants contribute to AI growth of LNCaP cells in confers resistance to androgen depletion in vivo. Cancer Res., 55: 4438–4445, 1995. 2232

21. Eastham, J. A., Stapelton, A. M. F., Gousse, A. E., Timme, T. L., Yang, G., 29. Shenk, J. L., Fisher, C. J., Chen, S-Y., Zhou, X-F., Tillman, K., and Shemshedini, L. Slawin, K. M., Wheeler, T. M., Scardino, P. T., and Thompson, T. C. Association p53 represses androgen-induced transactivation of prostate-specific antigen by dis- of p53 mutations with metastatic prostate cancer. Clin. Cancer Res., 1: 1111– rupting hAR amino- to carboxyl-terminal interaction. J. Biol. Chem., 276: 38472– 1118, 1995. 38479, 2001. 22. Isaacs, W. B., Carter, B. S., and Ewing, C. M. Wild-type p53 suppresses growth of 30. Levine, A. J. p53, the cellular gatekeeper for growth and division. Cell, 88: 323–331, human prostate cancer cells containing mutant p53 alleles. Cancer Res., 51: 4716– 1997. 4720, 1991. 31. Manin, M., Baron, S., Goossens, K., Beaudoin, C., Jean, C., Veyssiere, G., 23. Gingrich, J. R., Barrios, R. J., Kattan, M. W., Nahm, H. S., Finegold, M. J., and Verhoeven, G., and Morel, L. Androgen receptor expression is regulated by the PI3- Greenberg, N. M. Androgen-independent prostate cancer progression in the TRAMP kinase/Akt pathway in normal and tumoral epithelial cells. Biochem. J., 366: 729– model. Cancer Res., 57: 4687–4691, 1997. 736, 2002. 24. Burchardt, M., Burchardt, T., Shabsigh, A., Ghafar, M., Chen, M-W., Anastasiadis, 32. Murillo, H., Huang, H., Schmidt, L. J., Smith, D. I., and Tindall, D. J. Role of PI3K signaling in survival and progression of LNCaP prostate cancer cells to the androgen A., de la Taille, A., Kiss, A., and Buttyan, R. Reduction of wild type p53 function refractory state. Endocrinology, 142: 4795–4805, 2001. confers a hormone resistant phenotype on LNCaP prostate cancer cells. Prostate, 48: 33. Lin, H-K., Yeh, S., Kang, H-Y., and Chang, C. Akt suppresses androgen-induced 225–230, 2001. apoptosis by phosphorylating and inhibiting androgen receptor. Proc. Natl. Acad. Sci. 25. van Oijen, M. G. C. T., and Slootweg, P. J. Gain-of-function mutations in the tumor USA, 98: 7200–7205, 2001. suppressor gene p53. Clin. Cancer Res., 6: 2138–2145, 2000. 34. Chan, T. O., Rittenhouse, S. E., and Tsichlis, P. N. AKT/PKB and other D3 26. Miyashita, T., Harigai, M., Hanada, M., and Reed, J. C. Identification of a p53- phosphoinositide-regulated kinases: kinase activation by phosphoinositide-dependent dependent negative response element in the bcl-2 gene. Cancer Res., 54: 3131–3135, phosphorylation. Annu. Rev. Biochem., 68: 965–1014, 1999. 1994. 35. Shi, X-B., Gumerlock, P. H., and DeVere White, R. W. Molecular biology of prostate 27. Miyashita, T., Krajewski, S., Krajewska, M., Wang, H-G., Lin, H-K., Liebermann, cancer. World J. Urol., 14: 318–328, 1996. D. A., Hoffman, B., and Reed, J. C. Tumor suppressor p53 is a regulator of bcl-2 and 36. Yang, X., Pater, A., and Tang, S. C. Cloning and characterization of the human bax gene expression in vitro and in vivo. Oncogene, 9: 1799–1805, 1994. BAG-1 gene promoter: upregulation by tumor-derived p53 mutants. Oncogene, 18: 28. Eastham, J. A., Grafton, W., Martin, C. M., and Williams, B. J. Suppression of 4546–4553, 1999. primary tumor growth and the progression to metastasis with p53 adenovirus in 37. Froesch, B. A., Takayama, S., and Reed, J. C. BAG-1L protein enhances androgen human prostate cancer. J. Urol., 164: 814–819, 2000. receptor function. J. Biol. Chem., 273: 11660–11666, 1998.

2233

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 2003 American Association for Cancer Research. Androgen-independent Growth of LNCaP Prostate Cancer Cells Is Mediated by Gain-of-Function Mutant p53

Nancy J. Nesslinger, Xu-Bao Shi and Ralph W. deVere White

Cancer Res 2003;63:2228-2233.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/63/9/2228

Cited articles This article cites 37 articles, 15 of which you can access for free at: http://cancerres.aacrjournals.org/content/63/9/2228.full#ref-list-1

Citing articles This article has been cited by 16 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/63/9/2228.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/63/9/2228. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.