WO1999028749A1

WO1999028749A1 - USES OF p27 IN PROSTATE CANCER

Info

Publication number: WO1999028749A1
Application number: PCT/US1998/025483
Authority: WO
Inventors: Carlos Cordon-Cardo
Original assignee: Sloan-Kettering Institute For Cancer Research
Priority date: 1997-12-01
Filing date: 1998-12-01
Publication date: 1999-06-10
Also published as: AU1801499A

Abstract

This invention provides a method for determining the aggressiveness of a prostate carcinoma comprising: (a) obtaining a sample of the prostate carcinoma; and (b) detecting the presence of p27 protein in the prostate carcinoma, the absence of p27 indicating that the prostate carcinoma is aggressive. This invention also provides a method for diagnosing a benign prostate hyperplasia comprising: (a) obtaining an appropriate sample of the hyperplasia; and (b) detecting the presence of the p27 RNA, a decrease of the p27 RNA indicating that the hyperplasia is benign. Finally this invention provides various uses of p27 in prostate cancer.

Description

USES OF p27 IN PROSTATE CANCER

This application claims the benefit of U.S. provisional application No. 60/067,190, the contents of which is hereby incorporated by reference.

This invention was made in part with support under United States Government National Cancer Institute grant CA-DK-47650. Accordingly, the United States Government has certain rights in the invention.

Throughout this application, various references are referred to within parentheses. Disclosures of these publications in their entireties are hereby incorporated by reference into this application to more fully describe the state of the art to which this invention pertains. Full bibliographic citation for these references may be found at the end the specification, preceding the claims.

Background of the Invention

It has been postulated that the loss of function of a new family of negative cell cycle regulators, which act as cyclin-dependent kinase inhibitors and have been termed CKI , might lead to tumor development. CKIs fall into two families, Kip and Ink, on the basis of sequence ho ology. p27^Kιpl is implicated in Gl phase arrest by associating with multiple Gl cyclin-dependent kinases, abrogating their activity. However, no tumor-specific p27^Kιpl genomic mutations have been found in a large group of primary human cancers studied. More recently, it has been reported that proteasome-mediated degradation of p27 protein occurs during the cell cycle and that this degradation is increased in a subset of breast and colon carcinomas of poor prognosis. Purpose: The present study was undertaken in order to assess for potential alterations of p27 expression in benign prostatic hyperplasia (BPH) and in a well characterized cohort of patients with prostatic cancer. Inactivation of the p53 and RB tumor suppressor genes has been implicated in the development and progression of a number of different cancers (1,2) . It has also been postulated that the loss of function of a new family of negative cell cycle regulators, which act as cyclin-dependent kinase inhibitors and have been termed

CKI , might also lead to tumor development (3) . CKIs fall into two families, Kip and Ink, on the basis of sequence homology (4) . Kip family members include p21 (also known as WAFl, Cipl, or Sdil) (5-7), p27^Kιpl (8-10 ) and p57 ^Kιp2

(11,12). The Ink group includes four members: _{p l 6} INK4A/MTSl/CDKN2 ^ ^ ^ INKIB/MTSZ ₍ ^ ^ p l^MC _{( 1 5 )} ^ ^ ^ 9 ™

(16) . p27 is a negative regulator implicated in Gl phase arrest by TGFβ, cell-cell contact, agents that elevate cyclic AMP, and the growth inhibitory drug rapamycin (17-21) . p27 associates with multiple Gl cyclin-dependent kinases in non-proliferating cells, abrogating their activity (4, 8-10) .

To assess its role as a potential tumor suppressor, the p27^Kιpl gene was mapped to 12pl2-12pl3.1 and no tumor-specific genomic mutations in a large group of primary human cancers were observed (22-24) . More recently, it has been reported that proteasome-mediated degradation of p27 occurs during the cell cycle and that this degradation is increased in a subset of breast and colon carcinomas of poor prognosis (25-28) . The present study was undertaken in order to assess for potential alterations of p27 expression in BPH and in a well characterized cohort of patients with primary and metastatic prostatic cancer.

74 prostate carcinomas from primary and metastatic sites, representing different hormone sensitivities were analyzed. Normal prostatic tissues and cases of benign prostatic hyperplasia were also studied. In order to evaluate prostatic tissue of p27 null mice, eight 7 month old and six greater than 12 month old littermate pairs of wild-type and p27 knockout animals were used. Levels of expression and microanatomical localization of p27 protein and RNA transcripts were determined by immunohistochemistry and in si tu hybridization with specific antibodies and probes, respectively. Comparative analyses between immunohistochemistry, immunoblotting and immunodepletion assays were also conducted in a subset of cases. Association between alterations in p27 expression and clinicopathological variables were evaluated using the two-tailed Fisher's exact test. Disease relapse-free survivals were evaluated using the Kaplan-Meier method and the Logrank test. Distinct anomalies in the expression of p27 in benign and malignant human prostate tissues are reported. The normal human prostate shows abundant amounts of p27 and high levels of p27 messenger in both epithelial and stroma cells. However, p27 protein and transcripts are almost undetectable in epithelial and stroma cells of BPH lesions. It is also reported that p27-null mice develop hypercellular prostatic glands which histologically resemble human BPH. Based on these findings we postulate that the loss of p27 expression in human prostate may be causally linked to BPH. Prostatic carcinomas can be categorized into two groups: those that contain detectable p27 protein and those that do not. In contrast to BPH, however, both groups of prostatic carcinomas contain abundant p27 transcripts. Moreover, primary prostatic carcinomas displaying the p27 -negative phenotype appear to be biologically more aggressive, based on their association with time to prostate specific antigen (PSA) failure following radical prostatectomy. These results support the postulate that BPH is not a premalignant lesion in the pathway of prostate cancer development . Data also suggest that prostatic carcinoma develops along two different pathways, one involving the loss of p27 and the other using other processes that circumvent the growth suppressive effects of p27. Summary of the Invention

This invention provides a method for determining the aggressiveness of a prostate carcinoma comprising: (a) obtaining a sample of the prostate carcinoma; and (b) detecting the presence of p27 protein in the prostate carcinoma, the absence of p27 indicating that the prostate carcinoma is aggressive.

This invention also provides a method for diagnosing a beign prostate hyperplasia comprising: (a) obtaining an appropriate sample of the hyperplasia; and (b) detecting the presence of the p27 RNA, a decrease of the p27 RNA indicating that the hyperplasia is beign. In an embodiment, the above method further detects the protein expression of p27 wherein this additional step may be performed before or after the detection of the presence of the p27 RNA.

This invention provides a method for predicting the life- span of patient with prostate carcinoma comprising: (a) obtaining a sample of the prostate carcinoma; and (b) detecting the presence of p27 protein in the prostate carcinoma, the presence of the p27 protein indicating that the patient can live longer than the patient who are undetectable p27 protein.

This invention also provides a method for increasing the life-span of patient with prostate carcinoma comprising inducing the expression of p27 protein in the prostate carcinoma.

This invention provides a method for prolong life-span of patient with prostate carcinoma which comprises introducing a nucleic acid molecule having sequence encoding a p27 protein into the carcinoma cell under conditions permitting expression of said gene so as to prolong the life-span of the patient with said prostate carcinoma. In an embodiment, the nucleic acid molecule comprises a vector. The vector includes, but is not limited to, an adenovirus vector, adeno-associated virus vector, Epstein-Barr virus vector, Herpes virus vector, attenuated HIV virus, retrovirus vector and vaccinia virus vector.

This invention provides a method for prolong life-span of patient with prostate carcinoma which comprises introducing an effective amount of p27 protein into the carcinoma cell so as to thereby prolong the life-span of the patient with said prostate carcinoma.

This invention provides a method for prolong life-span of patient with prostate carcinoma which comprises introducing an effective amount of a substance capable of stabilizing the p27 protein into the carcinoma cell so as to thereby prolong the life-span of the patient with said prostate carcinoma .

This invention provides a composition for prolong life-span of patient with prostate carcinoma which comprises an effective amount of a nucleic acid molecule having sequence encoding a p27 protein and a suitable carrier.

This invention provides a composition for prolong life-span of patient with prostate carcinoma which comprises an effective amount of the p27 protein and a suitable carrier.

This invention provides a composition for prolong life-span of patient with prostate carcinoma which comprises an effective amount a substance capable of stabilizing the p27 protein and a suitable carrier. Brief Description of the Figures

Figure 1. Histological analysis, immunohistochemistry, and in si tu hybridization of human primary and metastatic prostatic carcinomas. (A-C) Photomicrographs of primary prostatic carcinomas processed as follows: (A) Immunohistochemical staining against p27 of a prostatic intra-epithelial neoplastic (PIN) lesion; note the intense positive immunoreactivities observed in the nuclei of the tumor cells growing into the lumen. (B) Immunohistochemical staining against p27 of another PIN lesion showing dysplastic changes; note the intense positive immunostaining in the nuclei of normal epithelial cell and the low-to-undetectable staining of the tumor cells dissecting the gland and growing into the lumen. (C) Undetectable levels of p27 protein in an invasive primary prostatic carcinoma; note the staining of a normal gland trapped into the tumor.

(D-F) Photomicrographs of metastatic prostatic carcinomas processed as follows: (D) Immunohistochemical staining against p27 of a metastatic prostate carcinoma to lymph node; note the intense nuclear staining of both tumor cells and lymphocytes (cells in the germinal center display low p27 levels) . (E) Immunohistochemical staining against p27 of another metastatic prostate carcinoma to lymph node; note the intense positive immunostaining in the nuclei of lymphocytes and the undetectable levels of p27 staining on the tumor cells. (F) Immunohistochemical staining against p27 of a metastatic prostate carcinoma to bone; note the positive immunoreactivities in the nuclei of osteoblasts and the lack of staining of tumor cells.

(F and G) Photomicrographs of a primary invasive prostatic carcinoma processed as follows: (F) Low-to undetectable immunohistochemical staining against p27 in the tumor cells; note the staining of a normal gland trapped into the tumor. (G) In si tu hybridization on a consecutive section from the case illustrated in panel (F) showing high mRNA levels of p27^Kιpl even in p27 -negative tumor cells utilizing the anti-sense probe to p27^ipl. Original magnification (A) trough ( F) 400x .

Figure 2. In certain prostatic carcinomas p27 protein is a functional cyclin-dependent kinase inhibitor. (A) Immunohistochemical staining correlates with the presence of p27 by immunoblotting . Tumors #1 and #2 were negative and tumor #3 positive for p27 protein expression, paralleling their IHC patterns. (B) Immunodepletion of p27 extracts. Extracts obtained from tumors #2 and #3 were subjected to sequential depletion with antibodies specific to p27 or a non-specific rabbit anti-mouse (RaM) . Following depletion, the proteins in the supernatants were resolved and the presence of p27 determined by immunoblotting. (C) Depletion of p27 depletes heat stable cyclin-dependent kinase inhibitory activity. The supernatant shown in panel B was boiled and following clarification the soluble fraction was incubated with different amounts of recombinant cyclin E/CDK2 kinase and the degree of inhibition of cylin E/CDK2 activity on histone HI substrate was measured. The amount of each kinase used is shown in the panel and the bars are representative activities on an arbitrary scale. Depletion with either RaM or p27 specific antibodies did not affect the inhibitory activity of the p27 negative tumor; however, depletion of p27 from the positive tumor extract completely ablated the heat stable inhibitor activity.

Figure 3. Recurrence-free proportion analysis of patients with primary prostate carcinoma (n=42) as assessed by time to detectable PSA. Patients who had PSA relapse were classified as failures, and patients with PSA relapse, or those who were still alive or died from other disease or lost to follow-up during the study period, were coded as censored. Time to relapse was defined as the time from date of surgery to the endpoint (relapse or censoring) . Disease relapse- free survivals were evaluated using the Kaplan-Meier method and the Logrank test . A trend was observed between a p27 negative phenotype and early relapse (p=0 . 08 ) .

Figure 4. Histological analysis, immunohistochemistry, and in si tu hybridization of human normal prostate and benign prostatic hyperplasia.

(A-C) Photomicrographs of consecutive sections of normal prostate tissue processed as follows: (A) Immunohistochemical staining against p27; intense positive immunoreactivities are observed in the nuclei of epithelial cells in the luminal side of the acinus, with decreased reactivities in the nuclei of basal and stroma cells. (B) In si tu hybridization showing high mRNA levels of p27^Kιpl in both epithelial and stroma cells utilizing the anti-sense probe. (C ) In si tu hybridization utilizing the sense probe to p27^Kιpl showing lack of signals in both epithelial and stroma cells.

(D-F) Photomicrographs of consecutive tissue sections of a benign prostatic hyperplastic nodule processed as follows:

(D) Immunohistochemical staining against p27; note the lack or almost undetectable levels of immunoreactivity observed in the nuclei of both epithelial and stroma cells in the luminal side of the acinus, with decreased reactivities in the nuclei of basal and stroma cells. (E) In si tu hybridization showing low-to-undetectable p27^Kιpl transcripts also in both epithelial and stroma cells utilizing the anti-sense probe; note the strong signal of the cellular inflammatory infiltrates that serve as an internal positive control. (F) In si tu hybridization utilizing the sense probe to p27^Kιpi showing lack of signals in epithelial and stroma cells, as well as cellular inflammatory elements. Original magnifications: (A), (B) and (C ) lOOOx; (D) , (E) and (F) 400x.

Figure 5. Histopathological analysis of the prostatic tissues of 12 month old p27+/+ (A) and p27-/- (B-D) mice.

Photomicrographs of tissue sections of normal prostate samples processed as follows: (A) Hematoxylin and eosin staining of a prostate gland of a p27+/+ mouse showing well defined acini of epithelial cells surrounded by a stroma containing few fibroblasts and poor in supportive connective tissue components. (B) Hematoxylin and eosin staining of a prostate gland of a p27-/- mouse showing multiple and complex glands and hypercellular acini of epithelial cells surrounded by fibromuscular stroma cells in a connective tissue displaying abundant supportive components. (C and D) Hematoxylin and eosin stainings of a prostate gland of a p27-/- mouse, high power details, illustrating the complexity of the glands and abundant fibromuscular stroma elements (C) , as well as the hypercellularity of the acini (D) . Original magnifications: (A) and (B) 200x; (C) and (D) 400x.

Detailed Description of the Invention

Methods to introduce a nucleic acid molecule into cells have been well known in the art. Naked nucleic acid molecule may be introduced into the cell by direct transformation. Alternatively, the nucleic acid molecule may be embedded in liposomes. Accordingly, this invention provides the above methods wherein the nucleic acid is introduced into the cells by naked DNA technology, adenovirus vector, adeno-associated virus vector, Epstein- Barr virus vector, Herpes virus vector, attenuated HIV vector, retroviral vectors, vaccinia virus vector, liposomes, antibody-coated liposomes, mechanical or electrical means . The above recited methods are merely served as examples for feasible means of introduction of the nucleic acid into cells. Other methods known may be also be used in this invention.

This invention provides a method for prolong life-span of patient with prostate carcinoma which comprises introducing an effective amount of a substance capable of stabilizing the p27 protein into the carcinoma cell so as to thereby prolong the life-span of the patient with said prostate carcinoma. Such substance may be either inhibiting the protease which degrade the p27 protein or it may interact with p27 in such a way that the protein will be resistant to degradation. By administering such substance into the cell, the effective amount of p27 protein will be increased .

As used herein, the term "suitable carrier" encompasses any of the standard carriers. The composition may be constituted into any form suitable for the mode of administration selected. Compositions suitable for oral administration include solid forms, such as pills, capsules, granules, tablets, and powders, and liquid forms, such as solutions, syrups, elixirs, and suspensions. Forms useful for parenteral administration include sterile solutions, emulsions, and suspensions.

This invention provides a composition for prolong life-span of patient with prostate carcinoma which comprises an effective amount a substance capable of stabilizing the p27 protein and a suitable carrier.

This invention will be better understood from the Experimental Details which follow. However, one skilled in the art will readily appreciate that the specific methods and results discussed are merely illustrative of the invention as described more fully in the claims which follow thereafter. Experimental Details

MATERIALS AND METHODS

Patient Characteristics and Tissues. A cohort of 74 patients with prostatic carcinoma were evaluated. Tissues were obtained from the Department of Pathology, Memorial Sloan-Kettering Cancer Center, New York. Samples were formalin-fixed, paraffin-embedded tissue specimens. Fourty-two primary prostate adenocarcinoma specimens were evaluated, as well as 9 metastases to lymph node and 23 metastases to bone. Normal prostatic tissue and/or areas of benign prostatic hyperplasia adjacent to tumor were observed in the majority of the primary cases studied. These tissues were also analyzed as part of the study. In addition, 10 pairs of frozen normal and tumor prostate tissues were utilized for antibody titration, as well as comparative analyses between immunohistochemistry, immunoblotting and immunodepletion assays (see below) . Representative hematoxylin-eosin stained sections were examined to evaluate the histopathological characteristics of the lesions to be analyzed, including the ratio of normal -to-tumor content for microdissection techniques.

In order to evaluate prostatic tissue of p27 null mice, eight 7 month old and six greater than 12 month old littermate pairs of wild-type and p27 knockout animals were used. Tissues were dissected, weighted and processed for histology by formalin fixation and paraffin embedding. Tissue sections were cutted and stained with hematoxylin-eosin for histologic analysis. All sections were utilized to count the number of acini per gland, a process that was conducted utilizing magnifications of 200x.

Antibodies and Immunohistochemistry . The following well characterized antibodies and corresponding final working dilutions were used for the present study: monoclonal antibody p27/Kipl (Ab-2, Oncogene Science, Boston, MA - 0.1 ug/ml final concentration) and anti-p27 affinity purified rabbit antiserum (1:500 dilution) . A non-immune rabbit serum and mouse monoclonal antibody MIgS-KpI were used as negative controls at similar working dilutions. Deparaffinized sections were treated with 3% H₂0₂ in order to block endogenous peroxidase activity. Sections were subsequently immersed in boiling 0.01% citric acid (pH 6.0) in a microwave oven for 15 minutes to enhance antigen retrieval, allowed to cool, and incubated with 10% normal horse or normal goat sera to block non-specific tissue immunoreactivities. Primary antibodies were then incubated overnight at 4°C. Biotinylated horse anti -mouse IgG antibodies (Vector Laboratories, Burlingame, CA - 1:500 dilution) or goat anti -rabbit antibodies (Vector Laboratories - 1:800 dilution) were applied for lhour, followed by avidin-biotin peroxidase complexes for 30 minutes (Vector Laboratories - 1:25 dilution) . Diaminobenzidine was used as the final chromogen and hematoxylin was used as the nuclear counterstain. Nuclear immunoreactivities were classified as a continuum data

(undetectable levels or 0% to homogeneous staining or 100%) . Tumors were grouped into two categories defined as follows: negative (0% or undetectable staining to <20% nuclear immunoreactivity in tumor cells) , and positive (neoplasms with >_.20% tumor cells with nuclear staining) (see statistical section) .

Probes and In Situ Hybridization. Digoxigenin-labeled probes were used for in situ hybridization and 1 ug of recombinant plasmid pCR™II (Invitrogen, San Diego, CA) , containing the full length human p27 gene (gift of Dr. M. Pagano, New York University School of Medicine, NY) was linearized by BamHI and Xbal to generate antisense and sense transcripts. Riboprobes were generated with T7 and SP6 polymerase for 2 hours at 37°C in IX transcription buffer (Boehringer Mannheim, Indianapolis, IN) , 20 U of RNAse inhibitor, 1 mmol/L each of ATP, GTP, CTP, 6.5 mmol/L UTP and 3.35 mmol/L digoxigenin-UTP. Deparaffinized tissue sections were rinsed in water and PBS for 10 minutes. The slides were digested with Proteinase K (50ug/ml) for 18 minutes at 37°C in PBS, and post-fixed at 4 °C in a freshly prepared solution of 4% paraformaldehyde in PBS for 5 minutes. Prehybridization was done for 30 minutes at 45 °C in 50% formamide and 2XSSC. The hybridization buffer consisted of 50% deionized formamide (v/v) , 10% dextran sulphate (50% stock solution) , 2XSSC (20X stock solution) , 1% SDS (10% stock solution), and 0.25 mg/ml of herring sperm DNA (10 mg/ml) . Hybridization was peformed overnight at 45 °C applying 10 pmol/L digoxigenin-labeled riboprobe in 50 ul of hybridization buffer per section under a coverslip. The coverslips were removed and the slides were washed in pre-warmed 2XSSC for 20 minutes at 60 °C twice, followed by washes in pre-warmed 0.5XSSC and 0.01XSSC at 60 °C for 20 minutes, respectively. After these washes the slides were incubated in normal sheep serum diluted in buffer pH 7.5 and successively in the same buffer with antibody anti-digoxigenin-AP (Boehringer Mannheim, Indianapolis, IN) at dilution of 1:1500 for 1 hour at room temperature. The visualization was accomplished by nitro-blue tetrazolium 5-bromo-4-chloro-3 -indoylphosphate . The slides were counterstained with methyl green and mounted.

Immunoblotting and Immunodepletion Assays. Proteins were extracted from three OCT-embedded prostatic carcinomas and resolved on polyacrylamide gels for immunoblotting with p27-specific antibodies. Extracts obtained from p27 positive and negative tumors were subjected to sequential depletion with antibodies specific to p27 or a non-specific rabbit anti-mouse (RaM) . Following depletion, the proteins in the supernatants were resolved and the presence of p27 determined by immunoblotting. Aliquots of these supernatants were briefly boiled and following clarification the soluble fraction was incubated with different amounts of recombinant cyclin E/CDK2 kinase and the degree of inhibition of cylin E/CDK2 activity on histone HI substrate was measured.

Statistical Methods. The statistical analyses were conducted as follows. For alterations of the p27, we divided patients into two groups: p27 negative (0% or no immunohistochemical staining to <20% tumor cells displaying nuclear reactivities) or p27 positive (>20% tumor cells with nuclear immunostaining with IHC) . The data analyses were conducted to explore the relationship between p27 alterations and clinicopathological variables such as presentation (primary, lymph node metastases, and bone metastases) , clinical stage (B, C, D) , total Gleason score

(6 or less versus 7 or more) , and hormonal status (naive versus androgen-independent ) in a total 74 patients. For 42 patients with primary prostate cancer who underwent radical prostatectomy, further analysis was conducted to evaluate the relationship between p27 alterations and clinical variables, including those described above and PSA relapse (yes and no) . Two-tail Fisher's exact test was utilized to assess these associations and two tailed p-values were employed as a significant level (29) . The FREQ procedure in SAS was used in this study (30) . In the analysis of disease relapse-free survival, patients who had PSA relapse were classified as lost failures, and patients with PSA relapse, or those who were still alive or died from other disease or to follow-up during the study period, were coded as censored. Disease relapse- free survivals were evaluated using the Kaplan-Meier method (31) and the Logrank test (32) . The LIFETEST procedure in SAS was used (30) . Proportional hazards analysis was used to obtain maximum likelihood estimates of relative risks and their 95% confidence intervals (33,34). EXPERIMENTAL RESULTS AND DISCUSSION

To determine whether loss of p27 expression was a common feature in prostate cancer, we analyzed 74 prostate carcinomas from primary and metastatic sites, representing different hormone sensitivities. Included were 42 hormone-naive primary tumors, some with associated prostatic intraepithelial neoplastic (PIN) lesions, and 32 metastatic carcinomas from lymph node tumors (n=9) and bone metastases (n=23) . Thirteen of these metastatic lesions were from hormone-naive cases, while the remaining 19 metastases were obtained after hormonal treatment . PIN lesions displaying a cribiform or pseudopapillary pattern expressed high levels of p27 protein (Figure 1A) and were associated with p27-positive invasive prostatic carcinomas.

In contrast, PIN lesions displaying a flat growth pattern had low to undetectable p27 levels (Figure IB) and were associated with p27 -negative invasive tumors. Of the invasive primary prostatic carcinomas studied, 12 of 42 (28.5%) cases had an intense nuclear immunoreactive p27 pattern in the malignant cells (data not shown) . The remaining 30 (71.5%) primary neoplasms displayed altered patterns of expression: 12 cases had undetectable p27 levels (Figure 1C) , while 18 cases had a heterogeneous pattern of expression (data not shown) . In metastatic lesions, 7 of 32 (21.9%) showed intense p27 nuclear immunostaining in most tumor cells (Figure ID) . The remaining 25 (78.1%) metastatic lesions had either heterogeneous (data not shown) or undetectable nuclear expression of p27 (Figures IE and IF) . Interestingly, all but one of the nine patients with hormone-independent bone lesions displayed altered p27 expression. Four of these 9 cases had undetectable p27 protein expression (Figure IF) , 4 cases had heterogeneous patterns of p27 expression ranging from 30% to 40% tumor cells with weak positive staining, and one case displayed 80% positive tumor cells. However, high levels of p27^Kιpl mRNA, as determined by in situ hybridization to a p27 cDNA probe, were found in all tumors even when the lesions displayed undetectable levels of p27 protein (Figures 1G and 1H) .

In the group of tumors that expressed p27, we next determined if the p27 protein was inactivated. To accomplish this we extracted protein from fresh frozen samples and measured the heat stable Cdk inhibitory activity, using cyclin E/CDK2 as a substrate, remaining in extracts following depletion with p27-specific antibodies as described previously (17) (Figure 2) . Depletion of p27 protein was confirmed by immunoblotting. As expected, the depletion of extracts derived from p27 negative tumors did not affect the heat stable inhibitory activity, nor did depletion of p27 positive tumor extract with a non-specific rabbit-anti-mouse immunoglobulin. However, depletion of extracts derived from p27 positive tumors with the p27 -specific antibody completely removed the inhibitory activity, indicating that p27 was functional as a Cdk inhibitor in these samples.

Taken together, these data suggest that prostatic carcinomas develop along two different pathways, one involving the loss of p27 and the other using alternative processes that may circumvent the growth suppressive effects of p27. In order to determine if these distinct pathways of prostate tumorigenesis correlate with clinical parameters, as reported for other tumor types (25-28) , associations between p27 immunostaining, stage, total Gleason score, and hormonal status of the tumor were assessed. No associations between detectable versus undetectable p27 protein, Gleason score (6 or less versus 7 or more) , or hormonal status (naive versus androgen-independent) were observed. To assess disease aggressiveness, we evaluated the time to PSA failure, the most sensitive indicator of success or failure following radical prostatectomy, in patients treated for localized disease. Only patients who had an undetectable PSA level after surgery, an indication that the resection was complete, were considered. A trend toward an association was observed between a p27 negative phenotype and early relapse (p=0.08) (Figure 3) . This difference did not reach statistical significance due to the limited sample size of the cohort analyzed. Supporting this concept is the fact that in a multivariate proportional hazards analysis, after controlling for stage and Gleason score, p27 status still was the strongest factor in predicting PSA relapse (p=0.07) .

These data suggest extending the characterization of p27 expression to normal prostate and benign prostatic hyperplasia. In the normal human prostate, abundant amounts of p27 protein were detected in the ductal and acinar cells, mainly luminal elements, as well as stroma cells using immunohistochemistry. Epithelial cells displayed a strong nuclear immunostaining signal (Figure 4A) . Likewise, both epithelial and stroma cells expressed abundant p27 transcripts (Figures 4B and 4C) , as detected by in situ hybridization. Strikingly, in 12 cases of BPH p27 expression was low to undetectable in epithelial and stroma cells in the hyperplastic nodules. Immunohistochemical staining revealed low to undetectable immunoreactivities in both epithelial and fibromuscular cells in the hyperplastic nodules (Figure 4D) . This contrasts with the strong p27 nuclear immunostaining phenotype observed in the normal prostate. Likewise, p27 mRNA transcript levels were low to undetectable on consecutive sections of BPH by in situ hybridization (Figure 4E and 4F) . In some of these BPH tissue samples we found areas of basal cell hyperplasia. These cellular elements also had low to undetectable amounts of p27 protein and transcripts (data not shown) . Nevertheless, in the non-hyperplastic regions of these same BPH samples, normal ductal and acinar epithelial cells, as well as stroma elements, showed high levels of p27 expression. These results indicate that in the development of BPH, p27 transcription may be down-regulated. This finding was quite unexpected as this gene product is generally regulated at post-transcriptional levels (35-37) , although members of the nuclear hormone receptor superfamily are suggested to regulate p27^κipl mRNA levels (38) .

The targeted deletion of the p27 locus in a murine model was recently reported (39-41) . p27 deficient mice are viable and display organomegaly, increased body size and female infertility. These anomalies could not be attributed to a defect of the growth hormone/lGF-1 axis, rather, they resulted from excess proliferation prior to withdrawal of cells into a terminally differentiated state

(39) . No increased incidence of spontaneous tumors was observed; however, many p27-null mice developed a pituitary hyperplasia reminiscent of adenoma in the intermediate lobe. These data suggest that p27 deficiency leads to hyperplasia in many tissues and organs. The high frequency of benign prostatic hyperplasia (BPH) in men and the alterations on p27 expression in that condition suggested a parallel to p27 deficiency. Previous reports of histopathological analyses of p27 null mice did not include the prostate (39-41) . We next set up to determine the morphologic characteristics of the prostate gland in p27 deficient animals. Comparing the total mean prostate weights of 7 month old age-matched p27+/+ (n=8) and p27-/- (n=8) mice, the differences were not significant [mean +/- SD: 80.6 mg (+/- 8.6 mg) and 90.1 mg ( +/- 13.3 mg) , respectively (p=0.1)] . However, the mean acini counts of the total gland in these groups were significantly different [mean +/- SD: 50.4 (+/- 8.5) and 74.9 (+/- 8.9), respectively (p<0.01)]. A similar relationship was observed in the mean total prostate weights of the old

(greater than 12 months) p27+/+ (n=6) and p27-/- (n=6) mice

[mean +/- SD : 114.0 mg (+/- 18.5 mg) and 119.0 mg (+/- 26.8 mg) , respectively (p=0.7)], and the mean acini counts

[mean +/- SD: 54.7 (+/- 6.5) and 73.8 (+/- 5.3), respectively (p<0.01)] . The significant increase in the number of acini in both young and old p27 deficient mice was associated with histopathological differences that became more accentuated in the elderly group. The hyperplastic prostate of the older p27-/- mice showed enlarged glands, development of hypercellular acini of epithelial cells, and an increase in fibromuscular stroma cells (Figure 5) . These histological changes are reminiscent of BPH in humans and support the hypothesis that the loss of p27 expression in human prostate may be causally linked to BPH.

It has been suggested that BPH and malignant prostate growth share a common pathway because they commonly coexist and demonstrate androgen dependency (42-44) . However, this relationship remains unclear since BPH tends to develop in the transition zone, while the majority of carcinomas develop in the peripheral zone (45-48) . Results from the present study reveal that, unlike in the BPH lesions, prostatic carcinoma cells regulate p27 expression at the post-transcriptional level. Taken together these data support the postulate that BPH is not a premalignant lesion in prostate cancer development .

Coordinate inactivation of the pathways involving the p53 and RB genes appears to be an essential requirement for the genesis of most human cancers. However, both p53 mutations and RB alterations are reported to be late and uncommon events in prostate tumor progression (49-52) . Contrary to these results, data from this study indicate that inactivation of p27 is a frequent and early event in some prostate cancers. It is thus our working hypothesis that p27 represents another pathway of tumor suppression in certain human tumors, prostate cancer being a paradigm in which this concept could be further tested.

In summary, data from this study suggest that p27^Kιpl gene ablation in the mouse causes a pronounced prostatic hyperplasia, and that the loss of p27 expression in human prostate may be causally linked to BPH. In addition, data from this study suggest that prostatic carcinoma develops along two different pathways, one involving the loss of p27 and the other using alternative processes that circumvent the growth suppressive effects of p27. These phenotypes can be identified as early as in the PIN stage. Moreover, primary prostatic carcinomas displaying the p27-negative phenotype appear to be biologically more aggressive, based on their association with time to PSA failure following radical prostatectomy while controlling for other variables. The consistent alteration of p27 expression observed in all androgen- independent metastatic lesions suggests an association with tumor progression, which may be the result of the metastatic process itself. Alternatively, it may be postulated that p27 positive tumors are more sensitive to androgen ablation, the primary treatment of metastatic disease. Finally, two dissimilar mechanisms appear to be involved in the loss of p27 expression in BPH versus a subset of prostatic carcinomas. p27^ιpl mRNA levels are extensively reduced in BPH, whereas p27 proteins are diminished to undetectable levels in some prostatic carcinomas despite detectable p27 mRNA as the result of a post-transcriptional event. These results support the postulate that BPH is not a premalignant lesion in the pathway of prostate cancer development .

REFERENCES

Hartwell, L.H. & Kastan, M.B. (1994) Sci ence 266, 1821-1828.

Cordon-Cardo, C. (1995) Am. J. Path . 147, 545-560.

Sherr, C.J. (1996) Sci ence 274, 1672-1677.

Sherr, C.J. S. and Roberts, J.M. (1995) Genes Dev 9, 1149-1163.

Harper, J.W., Adami , G.R., Wei, N. , Keyomarsi, K. & Elledge, S.J. (1993) Cell 75, 805-816.

El-Deiry, W.S., Tokino, T., Velculescu, V.E., Levy, D.B., Parsons, R. , Trent, J.M. , Lin, D. , Mercer, W.E., Kinzler, K.W. &

Vogelstein, B. (1993) Cell 75, 817-825.

Xiong, Y. , Hannon, G.J., Zhang, H. Casso, D., Kobayashi , R. &

Beach, D. (1993) Na ture 366, 701-704.

Polyak, K. , Kato, J-Y., Solomon, M.J., Sherr, C.J., Mssague, J., Roberts, J.M. & Koff, A. (1994) Genes Dev 8, 9-22.

Polyak, K. , Lee, M-H. , Erdjument-Bromage, H., Koff, A., Roberts,

J.M., Tempst, P. & Massague, J. (1994) Cell 78, 59-66.

Toyoshima, H. & Hunter, T. (1994) CelJ 78, 67-74.

Lee, M-H., Reynisdόttir, I. and Massague, J. (1995) Genes Dev. 9 , 639 - 649 .

Matsuoka, M., Edwards, M.C., Bai , , C, Parker, S., Zhang, P., Baldini, A., Harper, W. & Elledge, S.J. (1995) Genes Dev. , 9, 650-662.

Serrano, M. , Hannon, G.J. & Beach, D. (1993) Na ture 366, 704-707.

Hannon, G.J. & Beach, D. (1994) Nature 371, 257-261.

Guan, K-L., Jenkins, C.W., Li, Y., Nichols, M.A. , Wu, X., O'Keefe, C.L., Matera, A.G. & Xiong, Y. (1994) Genes Dev 8, 2939-2952.

Chan, F.K.M., Zhang, J. , Cheng, L., Shapiro, D.N. & Winoto, A. (1995) Mol . Cell Bi ol . 15, 2682-2688.

Polyak, K. , Kato, J-Y., Solomon, M.J., Sherr, C.J., Massague, J. ,

Roberts, J.M. & Koff, A. (1994) Genes Dev 8, 9-22.

Kato, J.Y., Matsuoka, M. , Polyak, K. , Massague, J. & Sherr, C.J. (1994) Cell 79, 487-496.

Coats, S., Flanagan, W.M. , Nourse, J. & Roberts, J.M. (1996) Sci ence 272, 877-880.

Luo, Y., Marx, S.O., Kiyokawa, H., Koff, A., Massague, J. & Marks, A.R. (1996) Mol Cell Biol 16, 6744-6751.

MiHard, S.S., Yan, J. , Nguyen, J. , Pagano, M. , Kiyokawa, H. & Koff, A. (1997) J Biol Chem 272, 7093-7098.

Ponce-Castaneda, M.V. , Lee, M-H., Latres, E., Polyak, K. , Lacombe, L., Montgomery, K. , Mathew, S., Krauter, K. , Sheinfeld , J . ,

Massague, J. & Cordon-Cardo, C. (1995) Cancer Res 55, 1211-1214.

Pietenpol, J.A. , Bohlander, S.K., Sato, Y., Rowley, J.D.,

Papadopoulos, N. , Liu, B., Friedman, C, Trask, B.J.,

Roberts, J.M., Kinzler, K.W. , Vogelstein, B. (1995) Cancer Res 55, 1206-1210.

Bullrich, F., MacLachlan, T.K., Sang, N. , Druck, T., Veronese, M.L., Allen, S.L., Chiorazzi, N. , Koff A., Huebner, K. , Croce, CM. δ- Giordano, A. (1995) Cancer Res 55, 1199-1205.

Loda, M., Cukor, B., Tarn, S.W., Lavin, P., Fiorentino, M. , Draetta, G.F., Jessup, J.M., S. Pagano, M. (1997) Nature Med. 3, 231-234.

Porter, P.L., Malone, K.E., Heagerty, P.J., Alexander, G.M., Gatti, L.A., Firpo, E.J., Daling, J.R. & Roberts, J.M. (1997) Na ture Med. 3, 222-225.

Catzavelos, C, Bhattacharya, N. , Ung, Y.C., Wilson, J.A. , Roncari, L., Sandhu, C, Shaw, P., Yeger, H. , Morava-Protzner, I., Kapusta, L., Frannssen, E., Pritchard, K.I. & Slingerland, J.M. (1997) Nature Med. 3, 227-230.

Tan, P., Cady, B., Wanner, M. , Worland, P., Cukor, B., Magi-Galluzzi, C, Lavin, P., Draetta, G., Pagano, M. & Loda, M. (1997) Cancer Res . 57, 1259-1263.

Mehta CR. & Patel, N.R. (1983) J Am Sta t Assoc 78, 427-434.

SAS Institute Inc, SAS/STAT User Guide, version 6 (1990) Kaplan, E.L & Meier, P. (1958) J Am Stat Assoc 53, 457-481.

Peto, R. , Pike, M.C., Armitage, P.. Breslow, N.E., Cox, D.R.,

Howard, S.V., Mantel, N. , McPherson, K. , Peto, J. & Smith, P.G. (1977) Br J Cancer 35, 1-39.

Cox, D.R., (1972) J R Statist Soc 34, 187-220. Cox, D.R. (1975) Biometrika 62, 269-279.

Pagano, M. , Tarn, S.W., Theodoras, A.M., Beer-Romero, P., Del Sal, G. , Chau, V., Yew, P.R., Draetta, G.F. & Rolfe, M. (1995) Science 269, 682-685.

King, R.W., Deshaies, R.J., Peters, J-M. & Kirschner, M. , (1996) Science 274, 1652-1659.

Hengst, L. & Sherr, S.I. (1996) Sci ence 271, 1861-1864.

Liu, M., Lee, M.H., Cohen, M. , Bommakanti, M. & Freedman, L.P. (1996) Genes & Dev 10 , 142-153.

Kiyokawa, H., Kineman, R.D., Manova-Todorova, K.O, Soares, V.C, Hoffman, E.S., Ono, M. , Khanam, D., Hayday, A.C., Frohman, L.A. & Koff, A. (1996) Cell 85, 721-732.

Nakayama, K. , Ishida, N. , Shirane, M., Inomata, A., Inoue, T., Shishido, N. , Horii, I., Loh, D.Y. & Nakayama, K-I. (1996) Cell 85, 707-720.

Fero, M.L., Rivkin, M. , Tasch, M. , Porter, P., Carow, C.E., Firpo, E., Polyak, K. , Tsai, L-H. , Broudy, V., Perlmutter, R.M., Kaushansky, K. & Roberts, J.M. (1996) Cell 85, 733-744.

Scher, H. , Steinieck, G. & Kelly, W.K. (1995) Urology 46, 142 - 148 .

Oesterling, J.E. (1996) Prostate (Suppl) 6, 67-73.

Linehan, W. , Cordon-Cardo, C, Isaacs, W., Eds. DeVita, V.T., Hellman, S, Rosenberg, S.A. (J.B. Lippincott Company, Philadelphia, 1997) .

Bostwick, D.G. (1996) Cancer 78, 330-336.

Parkinson, M.C. (1995) Histopathol 27, 301-307.

Bostwick, D., Pacelli, A., Lopez-Beltran, A. (1996) Prosta te 29, 117-134.

Lipski, B.A., Garcia, R.L. & Brawer, M.K. (1996) Semin Urol Oncol 14, 149-155.

Aprikian, A.G., Sarkis, A.S., Fair, W.R., Zhang, Z-F., Fuks, Z. S Cordon-Cardo, C. (1994) J Urol 151, 1276-1280.

Brooks, J.D., Bova, G.S., Marshall, F.F. & Isaacs, W.B. (1993) J Urol 150, 1278-1283.

Dinjens, W.N., van der Weiden, M.M., Schroeder, F.H., Bosman, , F.T. & Trapman, J. (1994) Int J Cancer 56, 630-633.

Grignon, D.J., Caplan, R. , Sarkar, F.H., Lawton, C.A., Hammond, E.H., Pilepich MV, For an JD, Mesic J, Fu KK, Abrams RA, Pajak TF, Shipley WU & Cox JD . (1997) J Natl Cancer Inst 89, 158-165. What is claimed is:

1. A method for determining the aggressiveness of a prostate carcinoma comprising: (a) obtaining a sample of the prostate carcinoma; and

(b) detecting the presence of p27 protein in the prostate carcinoma, the absence of p27 indicating that the prostate carcinoma is aggressive.

2. A method for diagnosing a beign prostate hyperplasia comprising :

(a) obtaining an appropriate sample of the hyperplasia; and

(b) detecting the presence of the p27 RNA, a decrease of the p27 RNA indicating that the hyperplasia is beign.

3. A method of claim 2, further comprising detecting the protein expression of p27 wherein this additional step may be performed before or after the detection of the presence of the p27 RNA.

4. A method for predicting the life-span of patient with prostate carcinoma comprising: (a) obtaining a sample of the prostate carcinoma; and

(b) detecting the presence of p27 protein in the prostate carcinoma, the presence of the p27 protein indicating that the patient can live longer than the patient who are undetectable p27 protein.

5. A method for increasing the life-span of patient with prostate carcinoma comprising inducing the expression of p27 protein in the prostate carcinoma.

A method for prolong life-span of patient with prostate carcinoma which comprises introducing a nucleic acid molecule having sequence encoding a p27

Claims

protein into the carcinoma cell under conditions permitting expression of said gene so as to prolong the life-span of the patient with said prostate carcinoma .

7. The method of claim 6, wherein the nucleic acid molecule comprises a vector.

8. The method of claim 7, wherein the vector is an adenovirus vector, adenoassociated virus vector,

Epstein-Barr virus vector, retrovirus vector or vaccinia virus vector.

9. A method for prolong life-span of patient with prostate carcinoma which comprises introducing an effective amount of p27 protein into the carcinoma cell so as to thereby prolong the life-span of the patient with said prostate carcinoma.

10. A method for prolong life-span of patient with prostate carcinoma which comprises introducing an effective amount of a substance capable of stabilizing the p27 protein into the carcinoma cell so as to thereby prolong the life-span of the patient with said prostate carcinoma.

11. A composition for prolong life-span of patient with prostate carcinoma which comprises an effective amount of a nucleic acid molecule having sequence encoding a p27 protein and a suitable carrier.

12. A composition for prolong life-span of patient with prostate carcinoma which comprises an effective amount of the p27 protein and a suitable carrier.

13. A composition for prolong life-span of patient with prostate carcinoma which comprises an effective amount a substance capable of stabilizing the p27 protein and a suitable carrier.