HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 17, 14877-14883, April 26, 2002

This Article Abstract Full Text (PDF) Supplemental Data All Versions of this Article: 277/17/14877    most recent M200373200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Schwarze, S. R. Articles by Jarrard, D. F. Search for Related Content PubMed PubMed Citation Articles by Schwarze, S. R. Articles by Jarrard, D. F. Social Bookmarking                      What's this?

Novel Pathways Associated with Bypassing Cellular Senescence in Human Prostate Epithelial Cells^*,

Steven R. Schwarze, Samuel E. DePrimo^§¶, Lisa M. Grabert, Vivian X. Fu, James D. Brooks^§, and David F. Jarrard

From the  Department of Surgery, Division of Urology, University of Wisconsin Medical School, Molecular and Environmental Toxicology and the University of Wisconsin Comprehensive Cancer Center, Madison, Wisconsin 53972 and the ^§ Department of Urology, Stanford University, Stanford, California 94305

Received for publication, January 14, 2002, and in revised form, February 6, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Cellular senescence forms a barrier that inhibits the acquisition of an immortal phenotype, a critical feature in tumorigenesis. The inactivation of multiple pathways that positively regulate senescence are required for immortalization. To identify these pathways in an unbiased manner, we performed DNA microarray analyses to assess the expression of 20,000 genes in human prostate epithelial cells (HPECs) passaged to senescence. These gene expression patterns were then compared with those of HPECs immortalized with the human Papillomavirus 16 E7 oncoprotein. Senescent cells display gene expression patterns that reflect their nonproliferative, differentiated phenotype and express secretory proteases and extracellular matrix components. A comparison of genes transcriptionally up-regulated in senescence to those in which expression is significantly down-regulated in immortalized HPECs identified three genes: the chemokine BRAK, DOC1, and a member of the insulin-like growth factor axis, IGFBP-3. Expression of these genes is found to be uniformly lost in human prostate cancer cell lines and xenografts, and previously, their inactivation was documented in tumor samples. Thus, these genes may function in novel pathways that regulate senescence and are inactivated during immortalization. These changes may be critical not only in allowing cells to bypass senescence in vitro but in the progression of prostate cancer in vivo.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Normal primary cells proliferate in culture for a limited number of population doublings prior to undergoing terminal growth arrest and acquiring a senescent phenotype (1). Senescent cells are resistant to mitogen-induced proliferation, express senescence-associated -galactosidase (SA--gal),¹ and assume a characteristic enlarged, flattened morphology (reviewed in Ref. 2). With progressive cell divisions, errors in DNA replication, oxidative metabolism, and environmental insults generate somatic mutations and place cells at risk for oncogenic transformation. Accumulating data suggest that the terminal arrest associated with cellular senescence represents a major barrier, analogous to apoptosis, that cells must circumvent to become malignant (3). Research into the pathways that positively regulate senescence and ways cells bypass senescence is therefore critical in understanding carcinogenesis.

In part, senescence is mediated by the loss of telomere length, which occurs with progressive divisions of mortal cells (4). Cellular senescence also can be induced prematurely by a variety of stimuli including DNA damage, perturbations of chromatin structure, and overexpression of mitogenic signals including E2F1, oncogenic Ha-ras, Raf, or MEK (mitogen-activated protein kinase/extracellular signal-regulated kinase kinase) (5-8). Two genes implicated in and critical for the induction and maintenance of senescence are pRb and p53 (3). Overexpression of downstream members of these pathways including p16, the p53 effector molecule p21, and p14ARF can elicit cell cycle arrest and senescence (5, 9, 10). These gain of function experiments demonstrating that senescence can be prematurely induced, however, do not directly address the complex pathways regulating senescence in unmanipulated, serially passaged, normal primary cells. These mechanisms may significantly differ. For example in fibroblasts, the overexpression of E2F1 leads to senescence; however, E2F1 expression is actually repressed as unmanipulated cells senesce (11).

Inactivation of the p16/pRb and p53 pathways is necessary to bypass senescence; however, this is insufficient for immortalization. Telomerase activation is also required, although the overexpression of telomerase alone is not sufficient to confer immortalization in epithelial cells (12). In addition, genetic analyses of immortalized prostate epithelial and other cell types reveal that a number of consistent genetic alterations are required, including the amplification of chromosome 20q and 8q (13, 14). Based on these findings and other microcell transfer experiments, it has been proposed that alterations in multiple pathways are required for the acquisition of the immortal phenotype (15). In a series of classical experiments the fusion of different immortalized cell lines were shown to restore senescence, and these cell lines could be sorted into four complementation groups. This work implies that there are a limited number of pathways critical to senescence, that these pathways must be activated to initiate senescence, and that they can override other transforming genetic events to yield a senescent phenotype (16).

Genes that have been implicated in senescence are also targeted in the development and progression of human prostate and bladder cancer. The alteration of the p16/pRb pathway in 85% of primary prostate cancers is a predictor of adverse clinical outcome.² The tumor suppressor p53 is one of the most commonly mutated genes in human cancer and is altered frequently in advanced prostate and bladder cancers (18, 19). Several lines of evidence suggest that bypassing senescence may also be important in cancer progression in vivo. Immortalized colon cancer cell lines can only be established from large, pathologically advanced lesions (20). When noninvasive papillary bladder neoplasms are grown in culture, they uniformly senesce and maintain an intact p16/pRb pathway (19, 21). In contrast, cells cultured from myoinvasive, aggressive transitional cell bladder carcinomas do not senesce, have alterations of the p53 pathway, and inactivate either p16 or pRb. Furthermore, prostate tumors that grow as xenografts have been generated solely from advanced, metastatic cancers (22). These results suggest that bypassing senescence is important in cancer progression and occurs late in the progression of cancers in vivo.

To gain insight into novel pathways involved in bypassing senescence, we developed a model in which transcript levels for more than 20,000 unique sequences were tracked in replicate cultures as human prostate epithelial cells senesced or were immortalized. A major advantage of this approach is that it avoids exogenous gene expression for the induction of premature senescence that can result in biased or aberrant results. Using this strategy, we have identified several genes, BRAK, DOC1, and IGFBP-3, that are up-regulated in senescence and inactivated during immortalization. We found, utilizing quantitative RT-PCR, that these genes are significantly down-regulated in human prostate cancer cell lines and xenografts. Interestingly, all of these genes have been previously implicated in the tumorigenic phenotype (23-25). These data support the concept that pathways activated at senescence and selectively inactivated with immortalization are important in carcinogenesis and tumor progression. Furthermore, these results provide an important effort in cataloging genes altered in senescence, a biologically relevant model of aging.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Cell Culture-- Prostate tissue was obtained under an approved Internal Review Board protocol from men (ages 44-66) undergoing cystoprostatectomy for bladder cancer at the University of Wisconsin Hospitals and Clinics. Histology confirmed that no bladder or prostate cancer was present in the tissues harvested for our studies. Prostate epithelial cultures were established as described previously (13). Prostate tissues were minced with a scalpel, digested in a solution containing 500 units/ml collagenase (Sigma), and plated on collagen-coated plates. Cells were maintained in Ham's F-12 medium (Invitrogen) supplemented with regular insulin, 0.25 units/ml; hydrocortisone, 1 µg/ml; human transferrin, 5 µg/ml; dextrose, 2.7 mg/ml; nonessential amino acids, 0.1 mM; penicillin, 100 units/ml; streptomycin,100 µg/ml; L-glutamine, 2 mM; cholera toxin, 10 ng/ml; bovine pituitary extract, 25 µg/ml; and 1% fetal bovine serum (26). Cells were passaged when confluent after incubation with trypsin-EDTA. Proliferating cells were harvested 2 days following the first passage. Pre-senescent cells were harvested when cell growth ceased, 1-5% of all cells appeared morphologically circular, flattened and enlarged, and SA--gal expression was negative. Terminally senescent cells were harvested when >60% appeared morphologically senescent and expressed high levels of SA--gal.

Transformation of HPECs with HPV16 E6 and/or E7-- Infections and characterization of HPEC cell lines was carried out as described (13). We used a retrovirus construct carrying either the HPV16 E6 and/or E7 gene(s) as well as a gentamicin resistance cassette (received from Dr. D. Galloway, Seattle, WA) (27). Subconfluent proliferating HPECs were infected with 10³-10⁵ infectious viral units at pre-passage (~5 × 10⁵ cells/100-mm dish) in 3 ml of 1% fetal bovine serum, F-12+ containing 4 µg/ml polybrene (Sigma). After 6 h, the virus-containing medium was exchanged, and infected cells were selected with 50 µg/ml G418 (Invitrogen) for a minimum of 7 days. Immortalized cell lines were screened for HPV16 E6 and/or E7 protein expression and loss of p53 and pRB by Western blot analysis. All immortalized lines were passaged in 1% fetal bovine serum F-12+ for well over 20 times to confirm their immortality.

Microarray Hybridizations and Data Analysis-- Microarray manufacture and hybridizations were carried out in accordance with previously published methods available at (28). For this set of experiments we used DNA microarrays with either 41,000 or 47,000 spots representing over 20,000 unique human genes and ESTs. For each hybridization, 2 µg of poly(A) mRNA from each sample was reverse-transcribed and labeled with fluorescence-tagged nucleotides (Cy3 for the reference sample, Cy5 for the experimental sample). Samples were hybridized against a common reference pool of mRNA derived from a panel of human cell lines as described previously (29). Hybridizations were carried out for 16 to 18 h at 65 °C, and the arrays were washed. After drying, the microarrays were scanned with a confocal laser GenePix microarray scanner (Axon Instruments) and analyzed with GenePix software. After visual inspection, spots of poor quality were flagged and excluded from analysis. Data files containing fluorescence ratios were entered into the Stanford Microarray Data Base, and compiled experiments were further analyzed with hierarchical clustering software and visualized with Treeview software (30). Prior to cluster analysis, gene entries were filtered to select those with >2.5-fold changes in signal intensity when comparing proliferating cells to pre-senescent, terminally senescent, and immortalized cell lines on at least one-third of the array samples, with fluorescent intensity in each channel that was greater than 1.6 times the local background. The identities of spotted cDNAs are based on information available in UniGene cluster Build 137 (June, 2001).³

Statistical Analysis of Microarray Data-- Genes with significant expression changes in response to senescence and immortalization in normal prostate epithelial cells were identified using the significance analysis of microarrays (SAM) procedure (31). The changes in gene expression for any single gene as measured in several array experiments provide a statistically testable measure of robustness, regardless of the magnitude of change in expression. The SAM procedure computes a two-sample T-statistic (e.g. proliferating versus senescent HPECs) for the normalized log ratios of gene expression levels for each gene. It thresholds the T-statistics to provide a "significant" gene list and provides an estimate of the false discovery rate (the percent of genes identified by chance alone) from randomly permuted data.

For the experiments described in our study, the raw expression ratio data set was filtered, using the program Cluster, for genes in which transcript levels differ from their median value by at least 1.5-fold in androgen-treated cells compared with controls in at least two experiments (with not more than 30% of measurements discarded because of poor data quality for each entry). We selected a data set in which 0.5 is the median number of likely false positive genes (false detection rate of 0.088%) for further analysis. We selected this value because it provided statistical assurance that most genes in the set were significantly altered but was not so stringent as to exclude genes of biological interest.

Quantitative RT-PCR Analysis-- Total RNA was isolated using the RNeasy RNA isolation kit (Qiagen) and treated with DNase, and 1 µg was used to prepare cDNA. Quantitative RT-PCR was performed by monitoring in real time the increase in fluorescence of the SYBR green dye as described using a iCycler detection system (Bio-Rad) (32, 33). For comparison of transcript levels between samples, a standard curve of cycle thresholds for several serial dilutions of a cDNA sample was established. This value was then used to calculate the relative abundance of each gene. These values were then normalized to the relative amounts of 18S cDNA. All PCR reactions were performed in duplicate. The sequences of the primers used for PCR analysis are available upon request.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Analysis of Senescence Pathways in Human Prostate Epithelial Cells-- Genome-wide RNA expression analysis of serially passaged primary human prostate epithelial cells (HPECs) and HPV16 E7 immortalized HPECs was performed with DNA microarrays representing over 20,000 distinct human genes and ESTs. To identify genes induced at senescence and inactivated at immortalization, HPEC cultures derived from three separate patients were analyzed at three defined stages in each patient's life span. Cells harvested in their first passage and growing exponentially were defined as proliferating. Pre-senescent cells were harvested when growth was halted, but the cells had not yet expressed the morphologic characteristics of senescence and were not yet SA--gal positive. We have previously shown that this is a distinct point at which cells transiently elevate the protein levels of selected cell cycle inhibitory genes, and it appears critical in signaling the onset of senescence (34). At terminal senescence, >60% of HPECs expressed SA--gal and had acquired an enlarged, flattened morphology (13, 34).

Three independently proliferating HPEC cultures were then immortalized utilizing HPV16 E7 retrovirus. We have used this molecular tool previously to dissect the genetic events involved in bypassing senescence (13, 35). The E7 oncoprotein acts, in part, by inactivating the pRb/p16 pathway and, via the binding of p21, the p53 pathway, thus recapitulating several genetic alterations commonly observed in prostate tumors (13, 18). HPV E7 immortalized cells are non-tumorigenic (36), making them an applicable model with which to isolate genes associated with the immortalization phenotype. However, because of some of the diverse activities of the HPV oncoproteins, we note that not all targets relevant to bypassing senescence will be identified utilizing this approach. Poly(A)+ mRNA was isolated from each of these experimental samples, reverse-transcribed, labeled with Cy-5, and hybridized in comparison with a common reference pool of cDNA (labeled with Cy-3) (29).

The determination of genes that significantly differed in expression levels between cellular phenotypes was made utilizing the analytical tool SAM (31). A comparison of proliferating to terminally senescent HPECs generated 262 unique entries using SAM, of which 93 were induced at senescence (51 of known function) and 159 were repressed (77 of known function). Complete details and transcript identities for all outputs can be accessed in the supplemental data (Supplemental Figs. 1 and 2). SAM outputs comparing proliferating to pre-senescent cells identified no genes with significant induction in expression in pre-senescent cultures but did find 101 genes in which expression levels were repressed. These 101 genes also displayed repressed transcript levels in terminally senescent cultures. The results generated by SAM were filtered to remove ESTs and genes of unknown function. The remaining genes were then classified based on their cellular function and placed into one or more related groups.

Consistent with a nonproliferative senescent state, 28 of the 77 known genes that were repressed at senescence are positive growth regulators or transcription factors involved in promoting cell cycle progression. An additional 28 repressed genes are directly involved in DNA replication, repair or mitosis. Of greater interest to us were those genes induced as cells progressed to terminal senescence as these genes may lend insights into the induction of senescence. It has been postulated that p16 and p53 function as tumor suppressors, in part, by inducing senescence pathways (37). We have previously reported at the cyclin-dependent kinase (cdk) inhibitor p16 protein level, a known marker of senescent cells, is elevated at senescence (34). In this study, we found an ~50-fold increase in p16 mRNA at senescence, suggesting that p16 protein increases are due, at least in part, to increased transcription. Transcript levels of other negative growth regulators were also increased including the p53-regulated targets BTG2, p21, IGFBP-3, and mdm2 (Fig. 1A), suggesting elevated p53 activity at senescence. Overexpression of BTG2 (38), p21 (39), IGFBP-3 (25), and even mdm2 in normal diploid cells (40) can result in G_0/1 cell cycle arrest. These p53-inducible genes may reflect the participation of p53 in regulating the onset of senescence, lending further insight into senescence-associated p53 targets.

View larger version (47K): [in this window] [in a new window]   Fig. 1.   Identity and cellular role of specific categories of genes induced in terminally senescent HPECs (A) and repressed in immortalized HPECs (B). Transcripts of named genes were selected from the SAM outputs and placed into functional categories. Included in A are genes induced at senescence associated with differentiated cell or neuronal functions and those associated with growth regulation. Genes repressed in immortalized cells that are involved in signaling, transcription, and DNA repair are listed in B. The GenBankTM accession number and fold change from normal proliferating HPEC proliferating values are listed for each entry. A "+" indicates gene induction, and "" indicates repression.

Senescent cells show permanent growth arrest and express markers associated with terminal differentiation. Consistent with this phenotype, 15 of 51 genes induced in senescent cells are associated with differentiation, 10 of which are commonly expressed in neuronal tissue (Fig. 1A). These genes may be indicative of the terminally differentiated phenotype rather than the induction of a neuronal phenotype as senescent HPECs do not resemble the elongated, spindle-shaped cells characteristic of neuronal cells in culture (13). Senescent HPECs also acquire a phenotype in which extracellular matrix (ECM)-degrading proteases, growth factors, and other inflammatory cytokines are produced (41). 16 of the 51 genes induced at senescence code for ECM proteins and matrix proteases (Supplemental Fig. 1). This group includes transcripts for chondroitin sulfate proteoglycan 2, matrix metalloproteinase 2, and microfibrillar-associated protein 2, and their respective increases (5.3-, 4.9-, and 4.3-fold increase by cDNA array, respectively) are among the highest we observed. Several other transcripts encoding ECM proteins (CSPG3, COL1A1, and fibronectin) and proteases (PAPPA and cathepsin F) are also induced (2.3- and 4.3-fold by cDNA array, respectively). Senescent cells produce extracellular proteases that promote matrix and basement membrane degradation, changes that are thought to contribute to the aging process in vivo (42). Our data suggest that senescence is associated with significant alterations of the ECM and provides some of the molecular underpinnings for these matrix changes.

The finding that the ras homolog gene family member B (rhoB) was induced at senescence is novel. rhoB mRNA levels were induced 3.6-fold by cDNA array and confirmed with real-time qPCR (17-fold increase, Fig. 2A). It has been demonstrated that the overexpression of oncogenic H-ras promotes senescence in primary, mortal epithelial cells and fibroblasts (6); however we did not observe changes in H-ras expression during senescence in our HPEC expression profiles. RhoB appears to have several functions in the cell including negative growth regulation, the induction of cytoskeletal changes, and sensitizing cells to apoptosis (43, 44). In aging mice, rhoB expression was found to be induced in skeletal muscle, and this alteration was attenuated with caloric restriction, the only known mechanism for retarding the aging process (45). Collectively, these data suggest that RhoB may be integral to both senescence and the aging process, potentially contributing to both the growth arrest and the characteristic morphology observed in senescent cells.

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Quantitative RT-PCR validation of cDNA microarray results. Real-time PCR following the incorporation of SYBR green into DNA was performed on cDNA derived from three independent HPEC cultures at in the proliferating, pre-senescent, and terminally senescent stages and in HPV16 E7 immortalized HPECs. Relative expression levels were normalized to the 18S content, also determined by qRT-PCR, and plotted on a log scale as fold change compared with proliferating cells. Expression levels from three separate microarray experiments were averaged and plotted with the qRT-PCR data for comparison. Good correlation was observed for all transcript levels, although the DNA microarrays tended to underestimate changes in transcript levels at higher magnitudes. The results are separated into three panels (A, B, and C) for clarity.

To validate the alterations in gene expression observed with senescence and immortalization, we performed real-time quantitative PCR (qRT-PCR) in additional HPEC cultures on 20 genes that had high (>3-fold) or low (<3-fold) level expression changes (Fig. 2). The pattern of gene induction or repression observed on the microarrays correlated with that seen by qRT-PCR, consistent with other validation experiments performed using this array system (46). However, microarrays tended to underestimate changes in transcript levels when genes were induced or repressed by more than ~3-fold.

Many of the gene expression changes we observed are also observed in primary fibroblasts undergoing senescence (Table I). However, in contrast to senescent fibroblasts, HPECs show increased transcript levels of the retinoic acid receptor , interleukin 1, interleukin 6, and IGF-I at terminal senescence. These differences in expression may be cell type-specific or may reflect differences between epithelial cells and fibroblasts (29). Comparisons of limited expression profiles from fibroblasts with our data derived from HPECs are extraordinarily consistent, suggesting that pathways that regulate senescence are conserved among cells derived from different tissues.

Identification of Pathways and Genes Altered following HPEC Immortalization-- To aid in defining the pathways important in bypassing senescence, we immortalized HPEC cell cultures with the HPV16 E7 oncoprotein. A comparison of genes expressed by these cells with proliferating non-immortalized cells revealed 229 significant alterations in expression of which 92 were induced (40 of known function) and 137 repressed (70 of known function). (For a complete list see Supplemental Figs. 3 and 4.) In bypassing senescence, HPECs appear to lose many specialized functions associated with proliferating, mortal prostate epithelial cells. Of 70 genes down-regulated upon immortalization, ~29 are associated with an epithelial cell function. For example, HPV16 E7 immortalized HPECs lose expression of prostate epithelial cell-specific kallikreins 10 and 11, two secreted proteases potentially important in reproduction (47). Not surprisingly, immortalized HPECs activate genes that participate in pathways that promote proliferation, including transcription factors and signaling proteins (Fig. 1B). In addition, genes involved in DNA damage response were also up-regulated during immortalization (Fig. 1B). It is unclear whether induction of these repair genes reflects increased DNA surveillance or an enhanced DNA repair capacity.

Identification of Genes Associated with Bypassing Senescence in Prostate Cancer Cells-- We hypothesized that a gene whose expression is induced during senescence (activation) and repressed following immortalization (inactivation) could be targeted during prostate cancer progression. Hierarchical clustering analysis was performed on genes in which expression changed by more than 1.5-fold as cells progressed to senescence or following immortalization to identify a group of genes with elevated transcript levels at senescence that returned to levels below or comparable with normal proliferating cells upon immortalization (Fig. 3). To further identify genes that may be inactivated during carcinogenesis, we focused on transcripts induced in senescent HPECs and down-regulated in the immortalized HPECs below levels found in HPECs (determined using SAM). This comparison revealed that only BRAK, DOC1 (down-regulated in ovarian cancer), and IGFBP-3 were significantly induced with senescence and repressed in immortalized cells markedly below (>10-fold) levels in proliferating HPECs.

View larger version (72K): [in this window] [in a new window]   Fig. 3.   Cluster analysis of transcripts from senescent and HPV16 E7 immortalized HPECs identified by the SAM procedure. Experiments are grouped to show transcripts that are induced at pre- and terminal senescence and at or below levels of proliferating cells in the immortalized HPECs. The red squares indicate expression levels higher than that of proliferating HPECs, green squares indicate expression levels lower than that of proliferating HPECs, black squares indicate transcript levels approximately equal to those in proliferating HPECs, and gray squares indicate data of insufficient quality. NP, normal prostate; HPVE7, HPV16 E7 immortalized HPECs; Prol, proliferating; Pre-sen., pre-senescence; T-sen., terminally senescent.

To extend these findings, we analyzed the expression of BRAK, DOC1, and IGFBP-3 levels in the human prostate cancer cell lines LNCaP, Du145, Tsu, PPC-1, PC-3, and DuPro using qRT-PCR. In six of six tumorigenic, immortalized prostate cancer cell lines we found DOC-1 to be down-regulated roughly 23-fold and BRAK and IGFBP-3 expression to be undetectable (Fig. 4). We analyzed the expression levels of these three genes found in two additional human prostate cancer xenografts, LAPC4 and LAPC9, to evaluate whether their expression is also lost in vivo. Similar to the prostate cancer cells, expression of BRAK, DOC1, and IGFBP-3 were all reduced markedly in these models compared with normal HPECs (Fig. 4). These prostate cancer xenografts have not been passaged in vitro, suggesting that the decreased expression of BRAK, DOC1, and IGFBP-3 is not due to artifacts introduced by growth in vitro (22). Our results indicate that genes and pathways activated at senescence and repressed during HPV16 E7 immortalization are also down-regulated in human prostate cancer cell lines and xenografts and may have a role in human prostate cancer.

View larger version (35K): [in this window] [in a new window]   Fig. 4.   Expression of BRAK, DOC1, and IGFBP-3 is very low in immortalized human prostate cancer cell lines. qRT-PCR was performed on six prostate cancer cell lines (LNCaP, Du145, Tsu, PPC-1, PC-3, and DuPro) and two xenografts (LAPC4 and LAPC9) and normalized to proliferating HPEC values. Expression levels in all samples were standardized to 18 S RNA levels. Prol, proliferating; Pre-sen., pre-senescent; T-sen., terminally senescent.

Previous reports suggest that BRAK, DOC1, and IGF family members have roles in the progression of other tumor types in vivo. BRAK is a member of the CXC chemokine family, which is widely expressed in normal tissues but is commonly repressed in head and neck squamous cell carcinomas (23). Its location on 5q harbors a putative susceptibility gene for prostate cancer based on familial linkage studies (48). As its name implies, DOC1 was discovered as a transcript down-regulated in ovarian cancers (24). Its function remains unknown. The IGF axis includes several ligands (IGF-I, IGF-II), cell-surface receptors (IGF-R), and binding proteins (IGFBPs) that regulate cell growth and differentiation (reviewed in Ref. 49). IGFBP-3 modulates interaction of the IGFs to their receptors and also possesses IGF-independent functions (50). Decreased IGFBP-3 expression is associated with prostate cancer progression, demonstrating more frequent loss of expression in advanced disease in both human and mouse models (25, 51, 52). Our data suggest that this may be due to bypassing senescence pathways. Another factor implicating IGFBP-3 in a senescence pathway is that the HPV16 E7 oncoprotein targets IGFBP-3 for inactivation (53). Further study of prostate cancer specimens will be necessary to evaluate the role of BRAK and DOC1 specifically in prostate cancer. These findings underscore the potential importance of inactivation of senescence in carcinogenesis and tumor progression.

In summary, using DNA microarray analysis, we have identified genes participating in normal human prostate epithelial cell senescence. We suspect that these pathways participate in the induction of senescence in many other cell types and that these genes are members of a limited number of pathways that can be inactivated to bypass senescence during tumorigenesis (15). We also have identified novel markers for the senescent phenotype that could assist in an analysis of the role of senescence in normal and pathologic states in vivo (54). Future studies will explore whether reactivation of these pathways in immortalized cells can also re-establish cellular senescence and whether this may represent a novel therapeutic approach to prostate cancer.

	ACKNOWLEDGEMENTS

We thank Dr. Rob Reiter for the xenografts and Dr. John Svaren for the critical reading of this manuscript.

	FOOTNOTES

^* This work was supported by grants from the National Institutes of Health (CA76184-01 to D. F. J.), the University of Wisconsin Comprehensive Cancer Center (to D. F. J.), the Doris Duke Foundation (to J. D. B.), and the Kovitz Foundation (to J. D. B.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The on-line version of this article (available at http://www.jbc.org) contains Supplemental Figs. 1-4.

^¶ Current address: Sugen, Inc., South San Francisco, CA 94080.

To whom correspondence should be addressed: Dept. of Surgery, University of Wisconsin, K6/530, 600 Highland Ave., Madison, WI 53792. Tel.: 608-265-2225; Fax: 608-265-8133; E-mail: jarrard@surgery.wisc.edu.

Published, JBC Papers in Press, February 8, 2002, DOI 10.1074/jbc.M200373200

² D. F. Jarrard, J. Modder, C. Sandefur, Y. Shi, D. Heisey, S. Schwarze, and A. Friedl, submitted for publication.

³ Website is ftp.ncbi.nlm.nih.gov/repository/unigene.

	ABBREVIATIONS

The abbreviations used are: SA--gal, senescence-associated -galactosidase; HPEC, human prostate epithelial cell; EST, expressed sequence tag; qRT-PCR, quantitative reverse transcriptase PCR; HPV, human papillomavirus; SAM, significance analysis of microarrays; IGF, insulin-like growth factor; IGFBP, IGF-binding protein; ECM, extracellular matrix.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				B. Bhatia, M. Jiang, M. Suraneni, L. Patrawala, M. Badeaux, R. Schneider-Broussard, A. S. Multani, C. R. Jeter, T. Calhoun-Davis, L. Hu, et al. Critical and Distinct Roles of p16 and Telomerase in Regulating the Proliferative Life Span of Normal Human Prostate Epithelial Progenitor Cells J. Biol. Chem., October 10, 2008; 283(41): 27957 - 27972. [Abstract] [Full Text] [PDF]

				M. Kwon, E. Hanna, D. Lorang, M. He, J. S. Quick, A. Adem, C. Stevenson, J.-Y. Chung, S. M. Hewitt, E. Zudaire, et al. Functional Characterization of Filamin A Interacting Protein 1-Like, a Novel Candidate for Antivascular Cancer Therapy Cancer Res., September 15, 2008; 68(18): 7332 - 7341. [Abstract] [Full Text] [PDF]

				T. Fukushima, T. Tezuka, T. Shimomura, S. Nakano, and H. Kataoka Silencing of Insulin-like Growth Factor-binding Protein-2 in Human Glioblastoma Cells Reduces Both Invasiveness and Expression of Progression-associated Gene CD24 J. Biol. Chem., June 22, 2007; 282(25): 18634 - 18644. [Abstract] [Full Text] [PDF]

				T. Shimizu, L. Esaki, H. Mizuno, and K. Takeda Granulocyte macrophage colony-stimulating factor enhances retinoic acid-induced gene expression J. Leukoc. Biol., October 1, 2006; 80(4): 889 - 896. [Abstract] [Full Text] [PDF]

				P. Klener, M. Szynal, Y. Cleuter, M. Merimi, H. Duvillier, F. Lallemand, C. Bagnis, P. Griebel, C. Sotiriou, A. Burny, et al. Insights into Gene Expression Changes Impacting B-Cell Transformation: Cross-Species Microarray Analysis of Bovine Leukemia Virus Tax-Responsive Genes in Ovine B Cells J. Virol., February 15, 2006; 80(4): 1922 - 1938. [Abstract] [Full Text] [PDF]

				M. Takaoka, C. E. Smith, M. K. Mashiba, T. Okawa, C. D. Andl, W. S. El-Deiry, and H. Nakagawa EGF-mediated regulation of IGFBP-3 determines esophageal epithelial cellular response to IGF-I Am J Physiol Gastrointest Liver Physiol, February 1, 2006; 290(2): G404 - G416. [Abstract] [Full Text] [PDF]

				G. V. Shurin, R. Ferris, I. L. Tourkova, L. Perez, A. Lokshin, L. Balkir, B. Collins, G. S. Chatta, and M. R. Shurin Loss of New Chemokine CXCL14 in Tumor Tissue Is Associated with Low Infiltration by Dendritic Cells (DC), while Restoration of Human CXCL14 Expression in Tumor Cells Causes Attraction of DC Both In Vitro and In Vivo J. Immunol., May 1, 2005; 174(9): 5490 - 5498. [Abstract] [Full Text] [PDF]

				K.-W. Lee, L. Ma, X. Yan, B. Liu, X.-k. Zhang, and P. Cohen Rapid Apoptosis Induction by IGFBP-3 Involves an Insulin-like Growth Factor-independent Nucleomitochondrial Translocation of RXR{alpha}/Nur77 J. Biol. Chem., April 29, 2005; 280(17): 16942 - 16948. [Abstract] [Full Text] [PDF]

				D M Peehl Primary cell cultures as models of prostate cancer development Endocr. Relat. Cancer, March 1, 2005; 12(1): 19 - 47. [Abstract] [Full Text] [PDF]

				S. B. Jones, S. E. DePrimo, M. L. Whitfield, and J. D. Brooks Resveratrol-Induced Gene Expression Profiles in Human Prostate Cancer Cells Cancer Epidemiol. Biomarkers Prev., March 1, 2005; 14(3): 596 - 604. [Abstract] [Full Text] [PDF]

K. Hardy, L. Mansfield, A. Mackay, S. Benvenuti, S. Ismail, P. Arora, M. J. O'Hare, and P. S. Jat Transcriptional Networks and Cellular Senescence in Human Mammary Fibroblasts Mol.