| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Laboratory of Cell Regulation and Carcinogenesis (K.V.D., J.E.G.), Biometrics Research Branch (A.M.M., J.H.S.), EPN/8132, Center for Cancer Research, National Cancer Institute, Bethesda, Maryland 20892; Department of Molecular Reproduction Development and Genetics (P.K.), Indian Institute of Science, Bangalore 560012, India; and Comparative Medicine Branch (J.M.W.), National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland 20892
Address all correspondence and requests for reprints to: Jeffrey E. Green, M.D., Laboratory of Cell Regulation and Carcinogenesis, 41 Medlar’s Drive, Room C619, National Institutes of Health, Bethesda, Maryland 20892. E-mail: jegreen{at}nih.gov.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
, and phosphatase and tensin homolog deleted on chromosome 10 (PTEN) genes, previously implicated in apoptosis. We show that PTEN protein was increased only in the luminal epithelial cells of the VP, suggesting that it may be a key mediator of VP apoptosis in the absence of androgens. The castration-induced immune/inflammatory gene cluster observed specifically in the VP included IL-15 and IL-18. Immunostaining of the VP, but not the DLP, showed an influx of T cells, macrophages, and mast cells, suggesting that these cells may be the source of the immune signature genes. Interestingly, IL-18 was localized mainly to the basal epithelial cells and the infiltrating macrophages in the regressing VP, whereas IL-15 was induced in the luminal epithelium. The VP castration model exhibits immune cell infiltration and loss of PTEN that is often observed in progressive PCa, thereby making this model useful for further delineation of androgen-regulated gene expression with relevance to PCa. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-reductase deficiency who have low to absent levels of dihydrotestosterone (3). Cumulative lifetime exposure to androgens or specific polymorphisms of the androgen receptor are associated with increased PCa risk (4). The role of androgens in prostate carcinogenesis is also supported experimentally. Castration in early life greatly reduces the risk of developing PCa in rodent and xenograft models, whereas androgens promote carcinogenesis in chemically induced rat cancer models (5, 6, 7).
Current therapy for androgen-responsive (AR) PCa thus involves a blockade of androgen signaling by the androgen receptor antagonist flutamide or the 5-reductase inhibitor finasteride thus resulting in the regression of PCa by epithelial cell death (4). However, decreased apoptosis of PCa cells after such treatment often results in the resurgence of a more aggressive androgen-independent (AI) cancer (8). These observations have prompted the evaluation of apoptotic genes as targets for PCa therapy (9). Although treatment options for AR disease exist, combating the AI form of PCa remains problematic. An in-depth understanding of how androgens control prostate cell growth and the identification of pathways that overcome androgen dependence of early-stage PCa are key to developing new strategies for treating the disease. However, studies on androgen-regulated gene expression in human PCa have been restricted to in vitro cell culture assays and the CWR22 xenograft model (10, 11, 12).
The rat castration model has been used to study androgen-regulated gene expression in VP using subtractive hybridization, differential display PCR, and microarray analysis (13, 14). The apoptosis of the rat VP after androgen withdrawal has been a valuable in vivo model system to explore genes involved in this process (15). The rat prostate is comprised of paired lobes divided into the anterior, dorsal and lateral (DLP) and ventral (VP) lobes. Although these lobes are known to vary in their response to androgen blockade by castration and/or by antagonist treatment, the molecular aspects of these differences remain poorly characterized. The VP responds to androgen withdrawal by massive apoptosis, whereas the DLP lobes show negligible cell death (16). However, both the lobes decrease in wet weight, lose RNA and protein content, and regress after castration (17). Thus, a parallel comparison of gene expression changes that occur in the VP and the DLP after androgen depletion may help delineate the metabolic vs. cell survival/apoptosis pathways governed by androgens.
In this study, we performed a comprehensive analysis of gene expression profiles of the rat VP and DLP lobes after castration as a function of time. Our results identify common and distinct patterns of gene regulation in the DLP and VP and indicate that many more genes are androgen regulated in the VP as compared with DLP. The data identified phosphatase and tensin homolog deleted on chromosome 10 (PTEN), a tumor suppressor gene deleted in late stage PCa, AI tumors and 30% of metastatic cancer as an androgen-repressed gene and as a possible mediator of castration-induced apoptosis. The most novel finding was the identification of an immune/inflammatory gene signature after the induction of VP apoptosis, including IL-15 and IL-18 as androgen-repressed genes. However, unlike the situation in vivo, IGF binding protein (IGFBP)-3, IGFBP-5, and IL-18 were up-regulated by androgens in the rat prostate NRP-152 cells, whereas the expression of PTEN and IL-15 was unaffected. These data suggest that expression of certain genes may depend upon the cross talk between cells in a whole organ system that does not occur in cell culture systems. These data thus emphasize the relevance of animal models to study hormone-regulated gene expression.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
schematically represents the number of genes modulated in each lobe. By definition, genes that were up-regulated after castration were designated as androgen-repressed genes and those with decreased expression after castration were classified as androgen-induced genes. These assignments were confirmed by testosterone treatment of castrated rats (Figs. 2 and 3, lane T).
|
|
|
Lobe-Specific Responses to Castration
Comparative microarray analysis of the VP and DLP lists generated by F test analyses identified several interesting changes in both the lobes. However, a major aim of this study was to find genes/pathways that distinguished androgen driven differential gene regulation in the VP and DLP. The data described below focus on such differences.
DLP
In DLP, 256 genes (79 named genes and 177 ESTs) changed 2-fold in the androgen deprived DLP and followed a single pattern of expression as shown in Fig. 2
. The DLP-specific genes were not involved in any particular cellular pathway (supplemental Table III). Androgens induced genes involved in cell defense (defensin ß 1) and collagen metabolism and extracellular matrix production. Very few genes were androgen-repressed in this lobe. Among these were the CYP2B21 gene, mineralocorticoid receptor (MlR) and their regulator the hepatic nuclear factor 3 (HNF-3).
VP
A considerable overlap was observed between the gene expression profiles of VP obtained by Pang et al. (14) and the present study, thus providing additional validation of both these studies. In the ventral prostate, 1496 genes differed at least 2-fold in their expression levels between the experimental groups (Fig. 3A). Hierarchical clustering of these genes resulted in 16 distinct clusters (supplemental Table I). The most dominant temporal trend observed was the induction (n = 634) or suppression (n = 690) of genes by the third day after castration (C3) that continued up to C7 (Fig. 3B). Less than 10 genes in each of the remaining 14 clusters exhibited a deviation from this general pattern. Testosterone treatment reverted most of the castration-induced genes to the sham castrated levels (Fig. 3B). Although many interesting pathways were modulated by androgens in a lobe-specific manner in VP, the apoptosis-related genes and novel pathways are discussed.
Apoptosis-Related Genes
The androgen-repressed gene list contained several factors previously associated with apoptosis including IGFBP-3, CCAAT-enhancer binding protein- (C/EBP), and PTEN (18, 19, 20). Semiquantitative RT-PCR analysis confirmed the induction of these genes after castration (Fig. 4). Because previous studies have shown that testosterone-repressed prostatic message-2 (TRPM-2) was induced in the VP but remained unchanged in the DLP, RT-PCR for this gene was used to confirm the lobe-specific regulations observed by microarray analysis (Fig. 4). As shown, IGFBP-3 induction was observed at C1 and C3, whereas IGFBP-5 was induced at later time points after castration (C5 to C7) in the rat VP. C/EBP followed a trend similar to IGFBP-3. PTEN was induced as early as C3 and was maintained from C5 to C7 in the VP. However, both C/EBP and PTEN mRNAs were up-regulated in DLP after castration at C1 to C5, but the IGFBPs were not modulated in DLP. After androgen administration, all genes studied reverted to gene expression levels displayed by the normal prostate (Fig. 4, lane T). 18SrRNA was used to normalize the amounts of cDNA used for each reaction.
|
). A band of lower molecular weight was evident at C5 in the ventral lobe. Caspase-3 has been implicated in the cleavage of PTEN protein, and caspases are induced after androgen ablation in the VP (21). It is possible that the lower band is cleaved PTEN protein (22).
|
). A specific induction of nuclear PTEN was observed as early as C3 in the ventral lobe that subsequently increased at C5 (data not shown) and C7 (Fig. 5B, upper panels, indicated by an arrow). On the other hand, immunoreactivity for PTEN was not detected in the dorsolateral lobe from the sham-castrated (S), castrated (C3 and C7), or T animals (Fig. 5B, lower panels). The intratubular staining (*) observed in all the sections was also observed in sections treated with secondary antibody alone, demonstrating that it was a nonspecific finding.
IGFBP-3 and IGFBP-5 Expression in Prostate Lobes
The castration-induced up-regulation of IGFBP proteins was validated by RT-PCR (Fig. 4). To determine the source of IGFBP production, immunohistochemical analysis was performed. As shown in Fig. 6, after castration both IGFBP-3 and IGFBP-5 proteins were induced in the luminal epithelium of VP (panels B and D). An increase in stromal staining was also evident. However, neither of the isoforms was expressed in an amount detectable by immunohistochemistry in the sham-castrated prostate (Fig. 6, A and C). In contrast, DLP of sham-castrated and castrated animals did not display a significant change in the amounts of IGFBP expression before and after castration (Fig. 6, E–H).
|
). RT-PCR analysis was performed for select genes including IL-15, IL-18, Syk, CX3C, and inhibitor of DNA binding 2 (Fig. 7). As shown in the figure, all the genes tested were induced at C5 to C7 after castration and reverted to sham operated levels in VP upon testosterone supplementation. Our microarray analysis did not show such modulations in DLP, and this was confirmed by RT-PCRs. The mRNA for Syk was undetectable in the DLP.
|
|
, panels A, C, and E, respectively). However, upon castration, at least a 5-fold higher influx of mast cells was observed in the ventral lobe at C7 (Fig. 8B). Also, a marked infiltration of intraepithelial and stromal ED-1 immunoreactive macrophages (Fig. 8D) and the appearance of intraepithelial and connective tissue stromal CD3+ T cells was seen (Fig. 8F). No infiltrating B cells were found in the VP (data not shown). The immune cells were not detected in DLP in the sham-castrated or castrated animals (data not shown).
|
, sham-castrated VP displayed positive staining for IL-18 in the stromal cells previously identified as macrophages (Fig. 8, C and D). Upon castration, the increased macrophages remained positive for IL-18 staining (indicated by *). Interestingly, upon castration, several epithelial cells in the VP acquired IL-18 immunoreactivity (Fig. 9B, indicated by filled arrows). On the other hand, IL-15 was abundantly expressed in luminal epithelial cells in sham-castrated VP and increased further at C7 after androgen withdrawal (Fig. 9, C and D). In the DLP, IL-18 remained unchanged upon castration and was expressed by basal epithelial cells. Similarly, IL-15 was abundant in the luminal epithelial cells of the DLP before and after castration (Fig. 9, E–H).
|
shows RT-PCR levels of these genes at various time points after androgen stimulation (R1881, 5 nM). In contrast to their androgen repression in the rat prostate, IGFBP-3, IGFBP-5, and IL-18 were up-regulated after androgen treatment in these cells. However, PTEN mRNA levels were unchanged in the 24-h period studied in this report. At time points beyond 24 h, the cells showed compromised cell survival. However, PTEN mRNA levels remained unchanged at these later time points (data not shown). The mRNA for IL-15 was undetectable by RT-PCR at all the time points studied.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Relatively few genes (n = 256) were specifically modified in the DLP. HNF-3, a transcription factor that directly binds the promoter of many metabolic enzymes and hormones and increases their expression (23, 24) was androgen repressed in DLP. Of interest was the association HNF-3 induction with two of its targets, the CYP2B21 and MlR in the castrated DLP cluster. These data suggest that HNF-3 maybe an important regulator of DLP in the absence of testosterone. However, this is in contrast to recent data that suggest that HNF-3 influenced the androgen-mediated regulation of the probasin and prostate-specific-antigen gene promoter (25). The second DLP-specific response in androgen-regulated genes appeared upon testosterone administration to castration rats. In this group of animals, a subset of genes including the DNA polymerase and proliferation cell nuclear antigen (PCNA) showed induction beyond the sham-castrated levels. This over-induction of DNA replication genes may contribute to androgen-induced tumor formation in the DLP. These intriguing observations require further study.
The high sensitivity of the ventral lobe to the loss of androgens by marked epithelial involution was reflected in the number of genes that were modulated (1496 genes) after castration. Multiple pathways have been implicated in the prostate involution (26, 27). Pang et al. (14) suggested that a decrease in GST gene and a concomitant increase in ezrin may lead to oxidative stress in the VP after androgen depletion, leading to cell death. Three additional genes that have previously been associated with PCa cell apoptosis were identified by our microarray analysis. These include PTEN, C/EBP , and IGFBP3. IGFBP-3 has been shown to induce apoptosis in PCa cells and PC3 cells in the absence of IGF-I and IGF-II signaling (18). C/EBP was shown previously to be an androgen-repressed gene in the rat ventral prostate and thus was not evaluated further (19). Interestingly, GST and ezrin, genes implicated in VP apoptosis by Pang et al. (14) were also induced in the DLP after castration. However, in contrast to the VP, the DLP does not undergo significant apoptosis after castration. Thus, changes common to both the DLP and VP are likely not responsible for the lobe-specific apoptosis observed in response to androgen withdrawal.
Because the overexpression of PTEN induced apoptosis of LNCaP cells in vitro (20), the modulation of PTEN by androgens was investigated further. Although PTEN mRNA was induced in both VP and DLP, no change in PTEN protein was evident in DLP after castration (Fig. 5). In the VP, an increased expression of nuclear PTEN protein was observed by both immunohistochemistry and immunoblot analysis. Treatment of rats with the antiandrogen flutamide does not induce cell death in the ventral prostate (28), and PTEN staining was not observed in prostate sections from these animals (data not shown). Therefore, PTEN protein levels correlate with in vivo cellular apoptosis, suggesting that it may be an important regulator of apoptosis in the VP in response to androgen withdrawal.
Both IGFBP-3 and IGFBP-5 were induced in the VP after castration. Detectable signal for either isoform was not observed in the sham-castrated rats by immunohistochemistry. In contrast, the DLP abundantly expressed both IGFBP isoforms in the presence and absence of testosterone, suggesting that these genes may not be regulated by androgens directly in the dorsolateral prostate in the rat castration model. Further studies are required to determine the influence of these isoforms in VP apoptosis.
The most novel cluster of genes specifically regulated in the VP after castration was comprised of genes involved in the process of inflammation and immune response (Table 1). These genes appear to be involved in T-cell and NK-cell activation and mast cell degranulation. To determine the source of these cytokines, we performed immunohistochemistry and demonstrated an influx of mast cells, macrophages, and CD3-positive T cells at C5 in the VP. Increased mast cell numbers and macrophage influx have been previously reported at C7 (29). One function attributed to the immune cell influx was the clearance of debris from dying epithelial cells (16). The androgen repression of IL-15 and IL-18 was validated by RT-PCR and immunohistochemical analysis. Although these cytokines are produced by macrophages, recent reports have localized IL-18 to the basal epithelial cells of the normal human prostate using the same antibody used in our study (30, 31). Furthermore, IL-18 staining was observed in benign prostatic hyperplasia, but advanced AI cancer cells were negative for IL-18 staining. In the rat VP, IL-18 was mostly expressed in macrophages before castration. However, after androgen loss the staining was retained in macrophages but was also evident in basal and some luminal epithelial cells. IL-15, which is induced by IL-18 in several inflammatory pathways, was localized to the luminal cells of the dying VP. However, the exact functions of these cytokines in epithelial cells remain unknown and require further investigation. In the DLP, these immune response pathways were unaffected by the androgen status. Therefore, the induction of apoptosis at C3 in the rat VP appeared to precede the influx of immune cells at C5 in response to androgen ablation, suggesting that apoptosis was prerequisite for immune cell invasion.
To determine whether the genes identified as androgen-repressed in rat prostate were similarly regulated by androgens in vitro, NRP-152 cells were treated with testosterone. Interestingly, the IGFBPs and IL-18 were androgen-induced genes in these cells, whereas PTEN and IL-15 were unresponsive to the change in androgen status. A similar induction of IGFBP-3 was reported in LNCaP cells (32). These results were opposite of the androgen-repression found in vivo. The NRP-152 cells were derived from the normal rat DLP and were shown to be basal epithelial cells (33). In addition, in vitro cell culture lacks cross talk between different cell types as exists in our in vivo study. Moreover, unlike the rat prostate, many prostate cell lines are not dependent on androgens for survival but are responsive to this hormone. These reasons may, in part, account for the differences between the castration model and the in vitro cell culture systems.
Although the VP and DLP have similar embryonic origins, this study reinforces the idea that a single hormone, testosterone, regulates extremely different molecular pathways in closely related tissues. Embryologically, the DLP is thought to more closely represent the human prostate than the VP. Thus, studies on the VP have been considered by some investigators to be marginally relevant to human PCa. The results from this study demonstrate that the VP is an important model system to identify novel apoptotic targets. For example, we have identified several ESTs that cluster closely with PTEN and may be therefore involved in prostate apoptosis. Additionally, genes important in PCa, like the IGFBPs, ILs, are modulated by androgens in the VP. An influx of immune cells was demonstrated in PCa patients that responded to androgen blockade therapy (34). Therefore, in at least two principle molecular responses to androgen blockade, apoptosis and immune cell influx, the rat VP appears similar to human PCa. Finally, a comparison of our data sets with those from human and rodent models of PCa may help identify relevant androgen-dependent pathways involved in disease development and progression. The data from this study may serve as an important resource for unraveling such molecular changes involved in androgen-related cancer.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
RNA Extraction and Microarray Analysis
Total RNA was extracted as described previously (35) from pooled tissue of three rats per time point to minimize animal variation in response to castration. At each time point, four pooled RNA samples were prepared for each of the three conditions; castrated, sham castrated and testosterone administered. CodeLink UniSet Rat I Bioarrays, purchased from Amersham-Pharmacia Life Sciences (Piscataway, NJ), were used for microarray analysis. The CodeLink platform is a glass slide covered with a three-dimensional polyacrylamide gel matrix, in which 5'-amine-terminated oligonucleotides are covalently attached. All the DNA oligonucleotide probes are 30 bases long. The array elements consist of 9911 named and unknown rat ESTs (together referred to as discovery probes by the manufacturer) and a set of bacterial genes designed for the assessment of array performance (50 negative and 18 positive control probes). Negative hybridization controls were developed to estimate the lower limits of true signal detection. Positive control probes were used as an internal control for the process of target preparation and to estimate the sensitivity of the signal detection. A total of 10 µg total RNA plus six bacterial mRNA spikes were used to generate target solution for each array. The labeling and hybridization procedures were performed as described by the manufacturer. The signal was obtained from hybridization of biotinylated cRNA probe, followed by fluorophore staining (Cy5-Streptavidin) and slide scanning. Images were produced with a GenePix 4000B laser scanner (Axon Instruments, Foster City, CA) and processed using CodeLink Expression Analysis software (Amersham-Pharmacia Life Sciences). The characteristics of the platform performance were described by Ramakrishnan et al. (36).
Data Analysis
Normalization.
Each array was globally normalized by dividing the signal values by the median intensity calculated from all discovery probes on an array. Array elements flagged as being of poor quality during image analysis or signals lower than the estimated negative control threshold were calculated as the mean + 3 SD from the middle 80% of the 300 negative control signals and were excluded from analysis. Nine percent of the array elements were excluded in this way. The final data sets contained 8733 and 8372 discovery probes from dorsolateral and ventral lobe arrays, respectively. The signal values transformed to log base 2 were used in the analysis. Statistical analyses were performed with the Insightful S-Plus package (Insightful Corp., Seattle, WA).
Reproducibility.
Additional replicate experiments were used to evaluate microarray data reproducibility. Five samples of rat testis and rat kidney tissue were pooled and five parallel target preparations, and hybridizations per tissue were performed. Data were preprocessed as described above and only probes with qualified signals across all slides were included, yielding 8074 discovery probes. The within-tissue pair-wise correlation was high (the median correlation coefficients are 0.997 and 0.998 in kidney and testis replicates, respectively) compared with the between-tissue correlation in the range of 0.55 and 0.57. For the majority of probes, the biological variability between testis and kidney specimens was much larger than the technical variation that represents measurement error arising from repeated arrays of the same specimen. The median ratio of biological variance to technical variance was as large as 50.9, indicating the microarray data generated from CodeLink UniSet Rat I Bioarrays platform were highly reproducible.
Statistical Analysis
Gene expression profiles of the VP and DLP lobes were analyzed separately. Within each lobe, the identification of probes that were differentially expressed between S, C, (C1, C3, C5, and C7), and T prostate glands was done using univariate F tests. Differentially expressed genes were identified as those that were significant at the 0.001 level (P < 0.001) and that were at least 2-fold different in the geometric means of the signals between these experimental groups. These identified genes were further grouped according to the trajectory of gene expression change over time using average linkage hierarchical clustering with 1 minus Pearson correlation as the distance metric.
RT-PCR
To validate the microarray results, we performed semiquantitative RT-PCRs for several representative genes. Total RNA (2 µg) was reverse transcribed with random hexamers using Superscript RT II (Invitrogen Life Technologies, Carlsbad CA). After the first-strand synthesis, the RNA was digested with 0.1 N sodium hydroxide for 10 min at 65 C and the sample was neutralized by 0.1 M Tris·Cl (pH 7.5). The cDNA was purified with the QIAGEN PCR purification kit and eluted in a volume of 50 µl (QIAGEN Corp., Rockville, MD). Two microliters of the cDNA was used for each PCR except for the 18S PCR, where 1 µl sample was used. The primers for TRPM-2, IGFBP-5, phosphatase and tensin homolog (PTEN), IL-15, IL-18 were as previously described (37, 38, 39, 40, 41). Primers for Syk (GenBank accession no. NM_012758) sense 5'-CTGGAACATGTCCTCCCTGT-3'; antisense 5'-AACACTGTGTGCTGCTCTGG-3', inhibitor of DNA binding 2 sense 5'-GGAATTGCCCAATGTAAGCA-3'; antisense 5'-AGGCCATTTCTGACCAAAGA-3' and the chemokine receptor CX3C, sense 5'-CGCTCTGAATAGCTCCAACC-3'; antisense 5'-GCTTCCTCACTCTGGGACAG-3' were designed using the Primer3 software. The PCRs were performed using the Promega PCR kit according to the manufacturer’s instructions (Promega, Madison, WI). The cycling parameters for all PCRs were 94 C for 45 sec denaturation, 55 C for 45 sec annealing, and 72 C extension for 45 sec. All the PCR products were cloned in the pCRII vector using the TOPO TA cloning kit (Invitrogen Life Technologies). A serial dilution of the plasmids (100 pg to 1 pg) was used for determining the linear range of the PCR (data not shown). Initial results suggested that each set of primers required 25 cycles of PCR to obtain products from experimental samples within the linear range of amplification. The final PCRs were performed using the PCR mix as described above with 1 µCi 32P-deoxy-CTP (3000 Ci/mmol; ICN Pharmaceuticals, Costa Mesa, CA) per reaction. The resultant products were analyzed on a 4–20% nondenaturing Tris-borate EDTA gel. The gels were dried for 2 h at 80 C under vacuum and exposed to XO-MAT AR autoradiographic film (Kodak, Rochester, NY).
Histochemistry and Immunohistochemistry
VP and DLP were excised from the animals, fixed in 4% paraformaldehyde and embedded in paraffin by standard procedures. Sections were stained with hematoxylin and eosin. Mast cells were identified with Giemsa stain. Antibodies to CD3 (rabbit polyclonal at 1:400, A0452, DAKO, Carpinteria, CA); rat ED-1 (mouse monoclonal at 1:2000, kindly supplied by Dr. C. D. Dijkstra, Vrije University, Amsterdam, The Netherlands, and now available from Chemicon International, Temecula, CA); and mouse antirat B cells (1:200, Serotec, Oxford, UK) were used. Sections were treated with IL-18 (1:800, R & D Systems, Minneapolis, MN), IL-15 (1:800, R & D Systems), IGFBP-5 (1:500, Upstate Biotechnology, Lake Placid, NY) and IGFBP-3 (1:300, Diagnostic System Laboratories, Webster, TX) antibodies overnight at 4 C to detect the respective proteins. The monoclonal antibody 6H2.1, specific for PTEN was used at 1:100 dilution and was a kind gift from Dr. Charis Eng (Ohio State University, Columbus, OH). Antigen retrieval was achieved by boiling the sections in citrate buffer for PTEN and CD3 immunostaining. Sections treated with secondary antibody alone were used as controls. The ABC Vectastain rabbit or mouse Elite kits (Vector Laboratories, Burlingame, CA) were used for immunohistochemistry according to the manufacturer’s instructions. 3,3'-Diaminobenzidine was used as the chromagen to detect antibody-specific brown staining. Sections were counter stained with hematoxylin.
Immunoblot Analysis
Protein lysates from frozen rat VP and DLP samples at S, C3, C5, and T experimental time points were made as described previously (42) and proteins were quantified by the bicinchoninic acid-protein assay kit (Pierce, Rockford, IL). At C7, the protein lysates were partially degraded and thus this time point was not investigated. Fifty micrograms of protein from each time point from both the DLP and VP was analyzed by SDS-PAGE on a 4–20% denaturing gel (Invitrogen Life Technologies, Carlsbad, CA). Immunoblot analysis was performed using the PTEN monoclonal antibody (1:1000, overnight at 4 C) using standard protocols. The reacted antibodies were detected by antimouse secondary antibody (Santa Cruz Biotechnology Inc., Santa Cruz, CA) at 1:10,000 dilution, visualized using Western Chemiluminescent Reagent Plus kit (PerkinElmer, Boston, MA) and autoradiography using XO-MAT AR autoradiographic film (Kodak).
Cell Culture
NRP-152 cells derived from rat dorsolateral prostate were a kind gift from Dr. David Danielpour (33). Cells were cultured in DMEM F-12 supplemented with 5% fetal bovine serum for routine cell maintenance. For androgen treatment, 1 x 106 cells were plated in phenol red-free DMEM-F12 containing 5% charcoal-stripped fetal bovine serum in 10-cm dishes. The cells were maintained in serum-free medium overnight and 1 nM, 5 nM, or 10 nM synthetic androgen R1881 was added. Cells were harvested at 0, 0.5, 2, 6, 12, and 24 h after treatment. RNA isolation and RT-PCR analysis was performed as described in Materials and Methods.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
, CCAAT-enhancer binding protein-; DLP, dorsolateral prostate; EST, expressed sequence tag; GST, glutathione-S-transferase; HNF, hepatic nuclear factor; IGFBP, IGF binding protein; PCa, prostate cancer; PTEN, phosphatase and tensin homolog deleted on chromosome 10; S, sham castrated; T, testosterone; TRPM-2, testosterone-repressed prostatic message-2; VP, ventral prostate. Received for publication January 28, 2004. Accepted for publication August 30, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
(c/EBP) in rat androgen-dependent tissues and human prostate cancer. J Androl 22:471–480[Abstract]
(Forkhead Box A1) and androgen receptor in transcriptional regulation of prostatic genes. Mol Endocrinol 17:1484–1507 and H2O2 induce IL-18 and IL-18R ß expression in cardiomyocytes via NF-
B activation. Biochem Biophys Res Commun 303:1152–1158[CrossRef][Medline]
expression during tumor progression. Cancer Res 60:6901–6910This article has been cited by other articles:
 |
|
 |
|
  D. Fairweather, S. Frisancho-Kiss, and N. R. Rose Sex Differences in Autoimmune Disease from a Pathological Perspective Am. J. Pathol., September 1, 2008; 173(3): 600 - 609. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. A. Quintar, F. D. Roth, A. L. D. Paul, A. Aoki, and C. A. Maldonado Toll-Like Receptor 4 in Rat Prostate: Modulation by Testosterone and Acute Bacterial Infection in Epithelial and Stromal Cells Biol Reprod, November 1, 2006; 75(5): 664 - 672. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. V. Heemers, G. Verhoeven, and J. V. Swinnen Androgen Activation of the Sterol Regulatory Element-Binding Protein Pathway: Current Insights Mol. Endocrinol., October 1, 2006; 20(10): 2265 - 2277. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Asirvatham, M. Schmidt, B. Gao, and J. Chaudhary Androgens Regulate the Immune/Inflammatory Response and Cell Survival Pathways in Rat Ventral Prostate Epithelial Cells Endocrinology, January 1, 2006; 147(1): 257 - 271. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Saghizadeh, A. A. Kramerov, J. Tajbakhsh, A. M. Aoki, C. Wang, N.-N. Chai, J. Y. Ljubimova, T. Sasaki, G. Sosne, M. R. J. Carlson, et al. Proteinase and Growth Factor Alterations Revealed by Gene Microarray Analysis of Human Diabetic Corneas Invest. Ophthalmol. Vis. Sci., October 1, 2005; 46(10): 3604 - 3615. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||
, panels A and B, magnification, x400), ED1 staining for macrophages (
, panels C and D, magnification x600) and CD3 staining for T cells (*, panels E and F, magnification, x600) revealed a castration-dependent increase in their numbers in the VP. The influx of these cells was not observed in the DLP (data not shown).
