| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
From the Department of Surgery, Division of Urology, University of Massachusetts Medical School, Worcester, Massachusetts
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
-glutamyl transpeptidase, and glutathione synthetase. Testosterone replacement in castrated rats partially reduced expression of Noxs but restored expression of superoxide dismutase 2, glutathione peroxidase 1, thioredoxin, and peroxiredoxin 5 to complete normalcy and induced a compensatory increase in expression of catalase, glutathione reductase, -glutamyl transpeptidase, and glutathione synthetase in the regenerating VP. Expression of superoxide dismutase 1, glutathione S-transferase-
, and glucose-6-phosphate dehydrogenase was unaffected by castration and testosterone replacement. These findings indicate androgen-deprivation induces OS in the rat VP through elevation of ROS anabolism and diminution of antioxidant detoxification. Androgen replacement partially reduces OS in rat VP to precastration levels. Expression of Noxs remained high amid a broad-based recovery of antioxidant defense mechanism(s). These data might have implications on the use of androgen blockade for prostate cancer prevention and androgen therapy for andropause treatment in elderly men.
B, and HIF-1
) and intermediate signaling molecules (eg, protein kinase C, ERK, and JNK).1,4
In contrast, induction of high levels of ROS subjects the cell to a state of oxidative stress (OS), which may damage cellular DNA, proteins, and lipids and result in cell-cycle arrest, cellular senescence, and cell death.1
Chronic OS has been implicated in neoplastic transformation5
and promotion of tumorigenesis.6
Apropos to the development of diseases in the prostate, increased oxidative damage to cellular macromolecules has been observed with aging7
and the development of malignancy.8,9
The degree of OS in a cell is dependent on the balance between ROS anabolism and catabolism (Figure 1)
. Although mitochondrial respiration is a major source of ROS, superoxide is also produced by a family of membrane-bound enzymes known as NAD(P)H oxidases (Noxs). Human Nox1, gp91phox (the catalytic subunit of Nox2), Nox3, Nox4, Nox5, and rat Nox1, gp91phox, and Nox4 have been cloned.5,10
Different Noxs are expressed in different cell types, including phagocytes and nonphagocytic cells.11,12
Little is known about expression of Noxs and their functions in the prostate. Ectopic expression of Nox1 in prostate cancer cells enhances growth, tumorigenicity, and angiogenicity,6
whereas down-regulation of Nox5 causes growth arrest and apoptosis.12
No information is available on Nox expression in the prostate gland in vivo.
|
-glutamyl transpeptidase (GGTP), and glutathione synthetase (GS) play major interactive roles in the replenishment of cellular reducing power. Other antioxidant molecules include thioredoxin (Txn), a protein thiol that directly detoxifies ROS and serves as an electron donor to other antioxidant enzymes, including Prdx5.15 Androgens are required for maintaining the homeostasis of cell proliferation and apoptosis in the prostate gland. Castration-induced androgen ablation results in regression of the prostate with a rapid rate of decline in fresh weight, RNA, DNA, and protein content mainly via apoptosis of epithelial cells, cell shrinkage, reduced protein synthesis, and secretion.16-19 Androgen replacement to castrated animals stimulates regeneration of the prostatic epithelium and other cellular and structural components.19 Although prostatic involution and regrowth have long been regarded as direct responses to circulating androgen levels, recent reports suggest that some cellular events are mediated by indirect processes, such as hypoxia17,20 or alterations in the redox state.21 Furthermore, as men age, the incidence of prostatic diseases rises dramatically.22 Related issues such as the risk associated with androgen replacement therapy for andropause treatment23,24 and the risk/benefit ratio of the use of Finasteride for prevention of prostate cancer25 continue to grow in magnitude in response to demographic changes. All in all, these unresolved issues related to the prostate and prostate diseases have prompted us to hypothesize that androgen may play a significant role in regulating the cellular redox state, ie, the balance in ROS levels, in the prostate, which in turn influences cell growth, apoptosis, and disease development in the gland.
Using immunohistological detection of the specific OS biomarkers,26 8-hydroxy-2'-deoxyguanosine (8-OHdG) and 4-hydroxynonenal protein adducts (4-HNE), we localized a marked elevation in oxidative damages to epithelial cells of the rat ventral prostate (VP) after castration and a partial recovery after androgen replacement. These findings support the hypothesis that ROS levels may mediate involution and regeneration of the prostatic epithelium of the rat. Alternatively, changes in ROS levels in the rat prostate after castration and testosterone replacement may simply be part of the cellular events associated with prostatic involution and regeneration. Nevertheless, quantitative polymerase chain reaction (PCR) analyses revealed marked changes in expression levels of Noxs and antioxidant defense enzymes, which may explain the drastic alterations in ROS levels in rat VP after castration and testosterone replacement. These findings provide direct evidence that androgens regulate ROS balance in the prostate, which is now known to have broad ramifications on prostatic aging and disease development in the gland.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Male NBL rats (5 to 6 weeks old) were purchased from Charles River Laboratories (Wilmington, MA). Rats were housed at the university’s animal facility on a 12-hour light/12-hour dark cycle and allowed access to food and water ad libitum. At 11 to 12 weeks of age, rats weighing 280 to 300 g were surgically castrated via the scrotal route under light isoflurane (Abbott Laboratories, North Chicago, IL) anesthesia. Castrated rats (n = 5) were sacrificed 3 days after castration. For the androgen replacement study, one group of rats (n = 5) that had been castrated 7 days previously before they were implanted subcutaneously at the subscapular region with two hormone-filled capsules, each 2 cm long, that were made of Silastic tubing (1.0-mm inner diameter x 2.2-mm outer diameter; Dow-Corning Corp., Midland, MI) and packed with testosterone (Sigma Chemical Co., St. Louis, MO).27
Each tube contained 
14.4 ± 2.1 mg of testosterone. Castrated rats were treated with testosterone for 7 additional days before sacrifice. Sham-operated age-matched intact control rats (n = 5) were used as controls. VPs were carefully excised from the prostate-urethra-bladder complex and cut in half. One half was formalin-fixed and paraffin-embedded for histological and immunohistochemical studies. The other half was snap-frozen in liquid nitrogen and kept at -70°C until RNA isolation. Protocols of animal usage were approved previously by the University of Massachusetts Medical School Animal Care and Usage Committee.
RNA Isolation and cDNA Synthesis via Reverse Transcription
Total RNA was isolated using TRI reagent (Sigma) according to the manufacturer’s protocol. RNA integrity was checked on denatured agarose gels. Two µg of total RNA was reverse-transcribed using the SuperScript II RNase H- reverse transcriptase system (Invitrogen, Life Technologies, Carlsbad, CA) according to the manufacturer’s instructions. After reverse transcription, the resulting cDNA was brought up to a volume of 40 µl with nuclease-free water (Ambion, Austin, TX).
Real-Time Quantitative PCR
Quantitative real-time PCR was performed with the iCycler IQ real-time PCR detection system (Bio-Rad Laboratories, Hercules, CA) as previously described.28
Before the real-time PCR was performed, cDNA obtained from the VPs of intact control rats, 3-day castrated rats, and 7-day castrated rats treated with testosterone replacement for 7 days were further diluted 10-fold with nuclease-free water. The diluted cDNA (2.5 µl) was amplified in a 25-µl reaction mix containing 1x iQ SYBR Green Supermix (Bio-Rad Laboratories) and 150 nmol/L of each primer. After a 15-minute Taq activation step at 95°C (hot start), reactions were subjected to 40 cycles of 30 seconds of denaturation at 94°C, 30 seconds of annealing at 60°C, and 30 seconds of extension at 72°C. Primers were purchased from MWG Biotech (High Point, NC). Intron-spanning primer pairs were designed to minimize primer dimerization and to generate an amplicon between 150 and 350 bp. The sequences of oligonucleotide primers used in this study are listed in Table 1
. Optical data were collected during the annealing step of each cycle. Amplification specificity was verified by performing melting-curve analysis by melting PCR products for 3 minutes at 90°C and then lowering the temperature to 55°C in 0.2°C increments, with 1 second per increment. Optical data were collected throughout the duration of the temperature drop, with a dramatic increase in fluorescence seen when the strands reannealed. Immediately after completion of melting-curve analysis, products were checked with agarose gel electrophoresis. Relative expression of the real-time reverse transcriptase-PCR products was determined by the 
Ct method. This method calculates relative expression using the equation: fold induction = 2-[Ct], where Ct = the threshold cycle, ie, the cycle number at which the sample’s relative fluorescence rises above the background fluorescence, and Ct = [Ct gene of interest (unknown sample) - Ct housekeeping gene (unknown sample)] - [Ct gene of interest (calibrator sample) - Ct housekeeping gene (calibrator sample)]. One of the control samples was chosen as the calibrator sample and used in each PCR. Each sample was run in triplicate, and the mean Ct was used in the Ct equation. The relative abundance of the mRNAs of interest in the VPs of castrated rats and castrates with testosterone replacement was calculated by arbitrarily assigning the abundance in the VPs of untreated intact controls a value of 1. RLP19 was chosen for normalization because this gene showed consistent expression relative to other housekeeping genes (GAPDH and ß-actin) among the three treatment groups in our initial experiments.
|
Statistical significance of differences in expression level between treatment groups was determined using Systat software (Student version 6.0.1; SPSS, Chicago, IL) to perform one-way analysis of variance with Tukey post hoc analyses. A P value of <0.05 was taken as a statistically significant difference between the two groups.
Immunohistochemical Analysis of OS Biomarkers
Mouse monoclonal antibodies to 8-OHdG (N45.1) and 4-hydroxynonenal protein adducts (HNEJ-2) were obtained from the Japan Institute for the Control of Aging (Fukuroi, Shizuoka, Japan). Immunohistochemical staining was performed with the Vestastain Elite ABC kit (Vector Laboratories, Burlingame, CA). The sections (5 µm) were dewaxed and then rehydrated in graded alcohols and distilled water. After deparaffinization, sections for 8-OHdG staining were antigen-retrieved by autoclaving at 121°C for 15 minutes in a 10% zinc sulfate solution and then treated with 2 mol/L of HCl at 37°C for 30 minutes;29 no antigen retrieval was needed for immunostaining of 4-HNE protein adducts. Sections were then incubated with 10% normal horse serum for 30 minutes at 37°C before overnight incubation with the primary antibody to 8-OHdG (1:40) and 4-HNE protein adducts (1:40) at 4°C. The slides were then incubated with the biotinylated anti-mouse IgG at a dilution of 1:200 for 30 minutes at 37°C followed by quenching of endogenous peroxidase activity by 3% H2O2 in methanol for 30 minutes at room temperature. The rationale for treating sections with H2O2 after primary antibody incubation was to ensure that all of the detected oxidative damages were not because of oxidation by H2O2 in the quenching step. The immunostaining was developed by peroxidase-conjugated avidin-biotin complexes and diaminobenzidine. For controls, the primary antibodies were preincubated with either excess antigens or 1.5 mg/ml 8-OHdG (Sigma) or were replaced with the corresponding normal isotype sera (Zymed, S. San Francisco, CA). Human seminal vesicle was used as a positive control tissue.8
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
OS in prostate was demonstrated by immunohistochemical detection of 8-OHdG and 4-HNE protein adducts, both of which are biomarkers for OS.26
8-OHdG Immunostaining
Positive staining was found exclusively in nuclei of the prostate tissues in the present study. In the acinar region of VPs from the intact control group (Figure 2A)
, positive nuclear staining was found in glandular epithelial cells but was not often found in periacinar smooth muscle and interstitial stromal cells. Negatively stained nuclei were occasionally seen interspersed with positive nuclei in the glandular epithelium (Figure 2A
and insets). In the ductal region (Figure 2D)
, positive nuclear staining of 8-OHdG was observed in both glandular epithelial cells and periductal smooth muscle cells. The intensity of 8-OHdG staining was conspicuously higher in the ductal region than in acinar region. Three days after castration, the immunoreactivity of 8-OHdG was markedly increased in glandular epithelial cells with no noticeable increase in periacinar smooth muscle cells and interstitial stromal cells such as fibroblasts, inflammatory cells, and endothelial cells (Figure 2B)
. Interestingly, strong 8-OHdG staining often was detected in the condensed nuclei of apoptotic cells in the glandular epithelium. In the ductal region of castrated animals (Figure 2E)
, strong immunoreactivity of 8-OHdG was found in both epithelial cells and periductal smooth muscle cells. In castrated animals with testosterone replacement, most acinar epithelia exhibited a general decline in 8-OHdG staining, with foci of strongly stained nuclei often localized adjacent to weakly or negatively stained nuclei within the same epithelium (Figure 2C
and inset). Periacinar smooth cells and stromal cells showed a weak or negative signal for 8-OHdG immunoreactivity. In the ductal region (Figure 2F)
, the immunostaining of 8-OHdG remained strong in the nuclei of epithelial cells and periductal smooth muscle cells. As a positive control tissue, human seminal vesicle exhibited intense nuclear staining in the pseudostratified columnar epithelium (data not shown), as previously documented.8
A signal was absent or undetectable in tissue samples with 8-OHdG antibody preabsorbed with excessive amounts of 8-OHdG (data not shown), demonstrating the specificity of the antibody.
|
In the intact control group, acinar epithelium exhibited weak or no immunostaining of 4-HNE protein adducts, which if present were localized in the cytoplasm and nuclei of epithelial cells (Figure 2G)
. In ducts, immunostaining of 4-HNE-modified proteins was diffuse and localized mainly in the cytoplasm of epithelial cells (Figure 2J)
. The smooth muscle cells and stromal cells in both acinar and ductal regions showed weak or no staining. Three days after castration, the immunoreactivity of 4-HNE protein adducts was distinctly increased in nuclei and cytoplasm of cells in both glandular and ductal epithelium (Figure 2, H and K
, respectively). The staining was focally distributed within epithelial tubules with foci of strongly stained regions often localized adjacent to weakly or negatively stained areas. Most important, 4-HNE protein adducts were detected in epithelial cells that had undergone apoptosis. In the acinar region, there was no noticeable increase in the staining of 4-HNE protein adducts in smooth muscle cells and stromal cells, including interstitial fibroblasts, inflammatory cells, and endothelial cells (Figure 2H)
. However, increased 4-HNE protein adducts were found in smooth muscle cells and stroma of prostatic ducts (Figure 2K)
. In castrated animals treated with testosterone for 7 days, the amount of 4-HNE protein adducts was significantly reduced, and they presented a patchy distribution in the glandular epithelium with diffuse and heterogeneous cytoplasmic staining (Figure 2I)
. Discretely positive- and negative-stained nuclei were found mingled together within the same epithelial region. Periacinar smooth muscle and stromal cells showed weak or no staining for 4-HNE-modified proteins. In the ductal region (Figure 2L)
, however, 4-HNE protein adducts were uniformly distributed in the epithelium and in periductal smooth muscle cells, with diffuse staining found in nuclei and cytoplasm. As a positive control tissue, human seminal vesicle exhibited intense cytoplasmic staining in the pseudostratified columnar epithelium (data not shown), as previously reported.8
The signal was absent or undetectable in tissue samples with primary antibody replaced by normal isotype antiserum (data not shown), demonstrating the specificity of the antibody.
Induction of NAD(P)H Oxidase Expression by Castration and Its Partial Reversal by Androgen Replacement
Quantitative real-time reverse transcriptase-PCR was used to measure the relative abundance of steady-state mRNA levels of rat gp91phox (the catalytic subunit of Nox2) and two other members of the family, Nox1 and Nox4 (Figure 3A)
. Relative mRNA levels of Nox1, gp91phox, and Nox4 in the VPs of castrated rats were twofold (P < 0.05), eightfold (P < 0.01), and 10-fold (P < 0.001) higher than levels found in VPs of intact control rats, respectively. After testosterone replacement in castrated rats, transcript levels of gp91phox and Nox4 were significantly decreased (P < 0.05 and 0.001, respectively) as compared with levels in castrated rats but were still elevated by twofold (P < 0.05) and 2.5-fold (P < 0.01) over levels found in intact rats, respectively. However, in the castrates with testosterone replacement, Nox1 exhibited only a trend of reduction in expression (P > 0.05) as compared with values in castrates but remained onefold elevated (P < 0.05) over levels in intact controls.
|
Transcript levels of a total of 11 genes critically involved in the intracellular antioxidant defense or ROS scavenging were quantified in the rat VP after castration and administration of testosterone to castrated rats. Based on their patterns of response to castration and testosterone replacement, the genes were grouped into three clusters. The first cluster of genes (Figure 3B)
was regarded as androgen-inducible genes whose expression was decreased by castration and restored to complete normalcy on subsequent testosterone replacement. Castration caused a significant reduction in relative mRNA abundance of SOD2 (threefold, P < 0.05), Gpx1 (twofold, P < 0.05), Txn (sixfold, P < 0.001), and Prdx5 (sixfold, P < 0.01) in the VPs. On testosterone replacement, mRNA levels of genes in this cluster were restored to the levels in intact rats. The second cluster of genes (Figure 3C)
demonstrated a more complex relationship with in vivo androgen status. Relative mRNA abundance of catalase, GR, GGTP, and GS were insensitive to castration-induced androgen deprivation but were markedly up-regulated by testosterone replacement to 220% (P < 0.05), 260% (P < 0.001), 400% (P < 0.001), and 200% (P < 0.05), respectively, of levels in intact controls. Finally, expression of the third cluster (Figure 3D)
of genes, including SOD1, GST-, and G6PDH, appeared to be insensitive to castration and testosterone replacement.
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
To elucidate the mechanism(s) underlying androgenic regulation of redox state in rat VP, we focused on the metabolic pathways responsible for ROS anabolism and catabolism. Specifically, quantitative real-time PCR was used to measure transcript levels of three NAD(P)H oxidases (superoxide-generating enzymes), a broad spectrum of key antioxidant enzymes, and key ROS scavengers. Steady-state mRNA levels of all three NAD(P)H oxidases, Nox1, gp91phox (Nox2), and Nox4, were significantly up-regulated in the rat VP after castration. Concomitantly, expression of several key ROS detoxification enzymes/scavengers, SOD2, Gpx1, Txn, and Prdx5 declined whereas transcripts levels of catalase, GR, GGTP, and GS remained unchanged in the VPs of castrated rats. Because Noxs are responsible for generating superoxide and SOD2 responsible for its removal, whereas Gpx1, Txn, and Prdx5 are intimately linked to the removal of H2O2, alterations in the expression of these genes, collectively, are expected to result in accumulation of ROS, leading to a higher pro-oxidant state in the VPs of castrated rats. In contrast, testosterone replacement in castrated rats caused a complete restoration of expression of SOD2, Gpx1, Txn, and Prdx5 mRNA to levels found in intact rats. However, expression of the Nox genes, although much reduced when compared with levels in castrates, were still significantly higher than those found in intact rats. Interestingly, although levels of mRNA expression of catalase, GR, GGTP, and GS were not changed by castration, they were significantly elevated when androgen was replaced in castrated rats. Collectively, these changes in expression of ROS anabolic and catabolic enzymes and ROS scavengers may explain why evidence of OS continued to exist in the VPs of testosterone-treated castrates. Our findings are in general agreement with recent findings demonstrating altered expression of antioxidant enzymes and specific genes related to ROS scavenging in rat VP after castration.21,31
A large body of evidence has demonstrated that ROS evokes apoptosis1
and proliferation,2
two extremely contrasting cellular events. In culture, proliferating cells exhibit a broad spectrum of responses to graded levels of oxidants.3
A very low level of H2O2 (3 to 15 µmol/L) causes a significant mitogenic response. Higher levels of H2O2 (120 to 400 µmol/L) results in a growth-arrested state, whereas a further increase of H2O2 to 1 mmol/L triggers apoptosis. However, the pro-apoptotic or pro-proliferation role of ROS in the prostate gland has yet to be demonstrated. In this study, we observed high levels of OS in the regressing VP and moderate levels of OS in the regenerating gland. Extrapolated from cell culture data, these findings thus support the hypothesis that different levels of ROS may contribute to cell death, as well as to cell growth, in vivo. Specifically, 8-OHdG and 4-HNE protein adducts were found to be highly expressed in the regressing VP epithelia, with the frequent association of these markers with apoptotic cells, suggesting a possible causal relationship between OS and apoptosis. OS-mediated apoptosis has been demonstrated in a variety of biological systems, including brain-derived neurotrophic factor- or zinc-induced neuronal cell death,32,33
anti-cancer drug-induced apoptosis,34
and hyperoxia- or vanadium-induced pulmonary apoptosis.35,36
On the other hand, it has been shown that ROS can elicit a mitogenic response via redox-sensitive transcription factors, such as AP-1 and nuclear factor-B,37
as well as through other signaling mediators such as protein kinase C4
and MAP kinases.38
Previous reports have revealed that low concentrations of oxidants are effective in stimulating the growth of fibroblasts in cultures.2,3
Treatment of hepatoma cells with low concentrations of ROS in vitro induces proliferation through the phosphatidylinositol 3-kinase/Akt pathway.39
After the testosterone-treatment of castrated rats, we observed evidence of continued elevation of OS at moderate levels in the regenerating VP, suggesting a plausible role for low levels of ROS in the regulation of cell proliferation, differentiation, and/or tissue remodeling in the gland.
In situ detection of oxidative damages by the immunohistochemical approach allows examination of OS at the cellular level. Oxidative damages, as revealed by the intensity of immunopositive 8OHdG and 4-HNE protein adducts, were always higher in the ducts of the VP than in the acini. Castration and androgen replacement had minimal effects on the levels of OS markers in the ducts but distinctly regulated those in the acini. These findings indicate differential regulation of OS in ducts versus that in acini by circulating androgen. Previous studies have reported that prostatic ducts are markedly different from acini with regard to morphology, hormonal responsiveness, proliferation, apoptosis, stromal/epithelial cell ratio, and composition of the extracellular matrix.40-42 Thus, constitutively high levels of OS found in the ductal regions may explain the reported high susceptibility of tumor development from prostatic ducts in hormone-treated Noble rats.43,44 It is currently unknown why ductal epithelia express such a high level of OS constitutively. One possible explanation may be related to previously reported retrograde transport of genotoxic substances or pathogens from the urethra to prostatic ducts, a mechanism implicated in prostatitis of the human gland.45
NAD(P)H oxidases are directly involved in the production of superoxide, a first step in the oxidant anabolism pathway.46
The exact mechanisms regulating expression of these genes in nonphagocytic cells are presently unknown and may be cell- or tissue-specific. It has been demonstrated that protein kinase C mediates the regulation of gene expression for NAD(P)H oxidases by angiotensin II in vascular smooth muscle cells47
and by zinc in neuronal cells.48
Serum, platelet-derived growth factor, and prostaglandin F2 cause alterations in gene expression of NAD(P)H oxidases in vascular smooth muscle cells.11,49
Our results clearly demonstrate that expression of NAD(P)H oxidases in rat VP is dependent on the androgen status of the animal. We had examined the promoter sequences (at 1 to 2 kb upstream of the transcription initiation site) of rat and human NAD(P)H oxidases in silico (unpublished observations) and did not find any androgen-responsive elements. A previous study failed to show regulation of superoxide production by testosterone in macrophages.50
Taken together, these findings suggest that androgenic regulation of NAD(P)H oxidases may be indirect. Expression of these enzymes may be secondary to physiological changes that occur during castration-induced involution and androgen-induced regeneration of the rat VP. The sharp increase in the expression of gp91phox in the VPs of castrated rats raises the possibility that invasion of the immune cell may contribute to the elevated OS status in these glands. It has been documented51
that migrating macrophages phagocytosing apoptotic bodies appeared in rat VPs 2 days after castration. Leukocyte infiltration has been reported in the prostates of castrated mice52
and in localized prostate cancer after androgen ablation therapy in humans.53
However, we did not observe any noticeable increase in the number of lymphocytes or neutrophils in the VP of castrated rats. These discrepancies may be a result of species- and/or disease state-governed variations. Alternatively, it is possible that all three NAD(P)H oxidases are also expressed by epithelial cells. Future investigations involving immunolocalization of the enzymes or transcript analyses in microdissected samples are needed to answer this question. Evidence is emerging to suggest that different Nox enzymes are regulated differentially and have different biological outcomes in a cell or a tissue. For example, ectopic expression of Nox1 in fibroblasts induces proliferation and transformation5
but that expression of Nox4 induces senescence.54
In human prostate cancer cells, overexpression of Nox1 increases tumorigenicity6
and anti-sense inhibition of Nox5 suppresses proliferation and triggers apoptosis,12
whereas down-regulation of gp91phox does not have any impacts on cell growth and survival/apoptosis.12
In vascular smooth muscle cells, Nox1 and Nox4 exhibited opposing responses to the same inducing agents.11
In the present study, androgenic regulation of gp91phox and Nox4 stand in contrast to that of Nox1, suggesting that they may be regulated by dissimilar mechanisms and may play nonredundant roles in the control of ROS anabolism in the rat VP.
We could classify the antioxidant defense enzymes and ROS scavengers that we studied in the rat VP into three categories (Figure 3)
according to their responsiveness to androgen withdrawal and resupplementation. The first group of VP genes includes SOD2, Gpx1, Txn, and Prdx5. These genes are most likely regulated by androgen directly; their expression declines on androgen withdrawal and rebounds to precastration levels on testosterone replacement. SOD2, also termed MnSOD, catalyzes the dismutation of superoxide into H2O2 in the mitochondria, which are considered the main site of ROS production. Gpx1 and Prdx5 convert H2O2 into H2O, thus preventing accumulation of the peroxide and its metabolic activation to hydroxyl radicals. Txn plays an essential role in eliminating ROS by acting as an electron donor to a number of antioxidant enzymes, including those in the peroxiredoxin family.15
The second set of VP genes, including catalase, GR, GGTP, and GS, are insensitive to androgen withdrawal but are significantly up-regulated on testosterone replacement in castrated animals. Intriguingly, this set of genes, except for catalase, are more distantly related to the removal of ROS but closely associated with replenishment of GSH.14
Specifically, GR is involved in the regeneration of GSH from its disulfide (GSSG) in GSH salvage whereas GGTP and GS play a critical role in de novo GSH synthesis. Taken together, our data suggest that replenishment of GSH is an important cellular event during androgen-induced prostatic regeneration. Our data agree with a previous finding demonstrating that androgen-treatment induces GGTP gene expression and enzyme activity in a human prostate cancer cell line.55
The last set of genes, including SOD1 (also termed CuZn SOD), GST-, and G6PDH, showed no response to changes in the androgenic status of the rats. Although studies in the literature have implicated their action of these genes in the maintenance of redox homeostasis, they may not participate directly in the androgen-regulated prostatic involution and regeneration. Our data clearly demonstrate that androgens regulate the expression of mitochondrial SOD2 but not the cytosolic SOD1. This finding suggests that elevated ROS levels expressed in the prostates of castrated rats may, in part, be derived from the mitochondria. Future studies involving localization of key pro- and/or anti-OS mediators both cellularly and subcellularly in the rat prostate are necessary to identify the cell type(s) and the organelle(s) exhibiting most redox changes in response to androgen manipulations.
Recently, vascular endothelial cells were shown to play an important role in mediating the castration-induced regression in the prostate.17,20 Degeneration and constriction of the vascular system in the prostate were found to be early events after castration, leading to reduced oxygen tension in prostate epithelial cells and activation of hypoxia- and stress-signaling pathways. Similarly, oxygen deprivation alone and combined oxygen and glucose deprivation induce ROS production in microvasculature56 and neurons,57 respectively. It is therefore logical to speculate that castration induces a hypoxic environment in the regressing prostate that leads directly or indirectly to elevation of OS in this gland. In a positive-feedback manner, increased OS in prostatic epithelium may, in turn, inflict additional oxidative damages to the vasculature and produce a vicious cycle of hypoxia- and OS-driven events that lead to a rapid involution of rat VP after castration.
In summary, our data suggest that normal physiological levels of androgen maintain a homeostasis between cellular pro-oxidant and antioxidant contents, as well as a balance between cell death and proliferation in the prostate. When normal androgenic status is disrupted, such as under the condition of castration-induced deprivation, OS is induced in the prostate via up-regulation of Nox-dependent ROS anabolism and down-regulation of a number of key antioxidant enzymes/ROS scavengers. The elevated OS state may play a critical role in mediating epithelial apoptosis and involution. When androgen is replaced, epithelial OS levels decline but fail to return to normalcy within a 7-day period, because of primarily high residual levels of Nox expression, although most antioxidant enzymes/ROS scavengers have returned to normalcy and exhibit an overcompensatory rebound to replenish cellular antioxidants and reducing power. This mild state of OS may contribute to cell proliferation, differentiation, and tissue remodeling in the regenerating VP. Collectively, this study provides the first mechanistic evidence that fills a significant data gap regarding the interrelationship between androgen status and redox homeostasis in the prostate. From a broader perspective, understanding androgenic regulation of OS in the prostate is of prime importance because both aging of the gland7 and pathogenesis of prostatic diseases, such as benign prostatic hyperplasia, prostate cancer, and prostatitis, are believed to be caused by oxidative damages of cellular macromolecules in the human prostate.6,8,9,30 Considering the fact that the use of androgen replacement therapy will be increased in our aging male population,23,24 issues surrounding its suspected risk in promoting prostate cancer should be evaluated in light of the present findings. It is possible that long-term use of testosterone may elevate prostatic ROS anabolism and raise the likelihood of tumor initiation and promotion. Furthermore, our data may explain to some degree the results from the Finasteride Prostate Cancer Prevention Trial,25 which demonstrated that androgen-blockade at the cellular level lowers prostate cancer risk but increases the prevalence of high-grade cancers. It is possible that the protective effect of androgen-blockade through Finasteride is mediated via reduction in cell proliferation, while the promotional effects are caused by androgen deprivation-induced chronic oxidative damages at the cellular/tissue level. Therefore, future studies focusing on the androgenic regulation on ROS production and elimination may result in better preventative or therapeutic modalities in controlling prostate diseases.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported in part by the National Institutes of Health (grants AG 13965, DK 61084, and CA 15776 to S. M. H.).
Accepted for publication September 10, 2003.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  B. Kumar, S. Koul, L. Khandrika, R. B. Meacham, and H. K. Koul Oxidative Stress Is Inherent in Prostate Cancer Cells and Is Required for Aggressive Phenotype Cancer Res., March 15, 2008; 68(6): 1777 - 1785. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. S. Arnold, J. He, A. Remo, D. Ritsick, Q. Yin-Goen, J. D. Lambeth, M. W. Datta, A. N. Young, and J. A. Petros Nox1 Expression Determines Cellular Reactive Oxygen and Modulates c-fos-Induced Growth Factor, Interleukin-8, and Cav-1 Am. J. Pathol., December 1, 2007; 171(6): 2021 - 2032. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. N.C. Tam, I. Leav, and S.-M. Ho Sex Hormones Induce Direct Epithelial and Inflammation-Mediated Oxidative/Nitrosative Stress That Favors Prostatic Carcinogenesis in the Noble Rat Am. J. Pathol., October 1, 2007; 171(4): 1334 - 1341. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Peters, M. F. Leitzmann, N. Chatterjee, Y. Wang, D. Albanes, E. P. Gelmann, M. D. Friesen, E. Riboli, and R. B. Hayes Serum Lycopene, Other Carotenoids, and Prostate Cancer Risk: a Nested Case-Control Study in the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial Cancer Epidemiol. Biomarkers Prev., May 1, 2007; 16(5): 962 - 968. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. C. Sullivan, J. M. Sasser, and J. S. Pollock Sexual dimorphism in oxidant status in spontaneously hypertensive rats Am J Physiol Regulatory Integrative Comp Physiol, February 1, 2007; 292(2): R764 - R768. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Peters, C. B Foster, N. Chatterjee, A. Schatzkin, D. Reding, G. L Andriole, E D. Crawford, S. Sturup, S. J Chanock, and R. B Hayes Serum selenium and risk of prostate cancer--a nested case-control study Am. J. Clinical Nutrition, January 1, 2007; 85(1): 209 - 217. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Bedard and K.-H. Krause The NOX Family of ROS-Generating NADPH Oxidases: Physiology and Pathophysiology Physiol Rev, January 1, 2007; 87(1): 245 - 313. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Cayatte, C. Pons, J.-M. Guigonis, J. Pizzol, L. Elies, P. Kennel, D. Rouquie, R. Bars, B. Rossi, and M. Samson Protein Profiling of Rat Ventral Prostate following Chronic Finasteride Administration: Identification and Localization of a Novel Putative Androgen-regulated Protein Mol. Cell. Proteomics, November 1, 2006; 5(11): 2031 - 2043. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S Morimoto, C A Mendoza-Rodriguez, M Hiriart, M E Larrieta, P Vital, and M A Cerbon Protective effect of testosterone on early apoptotic damage induced by streptozotocin in rat pancreas J. Endocrinol., November 1, 2005; 187(2): 217 - 224. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. K. Maranchie and Y. Zhan Nox4 Is Critical for Hypoxia-Inducible Factor 2-{alpha} Transcriptional Activity in von Hippel-Lindau-Deficient Renal Cell Carcinoma Cancer Res., October 15, 2005; 65(20): 9190 - 9193. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Ouyang, T. L. DeWeese, W. G. Nelson, and C. Abate-Shen Loss-of-Function of Nkx3.1 Promotes Increased Oxidative Damage in Prostate Carcinogenesis Cancer Res., August 1, 2005; 65(15): 6773 - 6779. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||
