Prostatic Secretory Protein-PSP94, Probasin, and Seminal

0013-7227/00/$03.00/0 Endocrinology Copyright © 2000 by The Endocrine Society

Vol. 141, No. 12 Printed in U.S.A.

A Comparative Study of Hormonal Regulation of Three Secretory Proteins (Prostatic Secretory Protein-PSP94, Probasin, and Seminal Vesicle Secretion II) in Rat Lateral Prostate* JOSEPH KWONG, J. W. XUAN, PETER S. F. CHAN, SHUK-MEI HO, FRANKY L. CHAN

AND

Departments of Anatomy (J.K., F.L.C.) and Surgery (P.S.F.C.), Chinese University of Hong Kong, Hong Kong, China; Department of Surgery, University of Western Ontario (J.W.X.), London, Ontario, Canada N6A 4G5; and Department of Surgery, Division of Urology, University of Massachusetts Medical School, Worcester, Massachusetts 01655 ABSTRACT The rat dorsolateral prostate secretes several major known proteins, although their physiological and reproductive functions are largely undefined. In the present study we examined and compared the in vivo hormonal regulation of the messenger RNA (mRNA) expression of three major secretory proteins, including prostatic secretory protein of 94 amino acids (PSP94 or ␤-microseminoprotein), probasin, and seminal vesicle secretion II (SVSII), in long-term castrated lateral prostates (LP) by in situ hybridization and semiquantitative RT-PCR. The protein levels of PSP94 in the castrated LPs were also examined by Western blotting. PSP94 is a small protein newly isolated from the rat prostate gland and demonstrates highly specific expression in the LP. The results of in situ hybridization showed that PSP94, probasin, and SVSII were highly expressed in the intact LP. The hybridization signals of probasin and PSP94 disappeared in the 60-day postcastrated LPs, whereas the signals of SVSII dropped sharply in the 14-day postcastrated LPs. Similar patterns of decreasing mRNA levels of the three proteins in the castrated LPs were

T

HE RAT PROSTATE gland is composed of paired ventral, lateral, and dorsal lobes. Although the lateral and dorsal lobes are distinct in their morphology (1–3), secretions (4, 5), secretory manner (6), glycoconjugates (7), and responsiveness to castration and hormones (8, 9), these two lobes are usually described and studied together as the dorsolateral lobe. We are interested in the dorsolateral lobe, as pathological or malignant changes are usually developed in these two lobes. Dunning prostatic tumor, a well studied animal model of prostate cancer, is derived originally from the dorsal prostate of a Copenhagen rat (10). Long-term treatment with both androgen and estrogen can induce a premalignant lesion, dysplasia, in the dorsolateral prostate of Noble rat (11, 12). This lesion is similar to the prostatic intraepithelial neoReceived April 26, 2000. Address all correspondence and requests for reprints to: Dr. Franky L. Chan, Department of Anatomy, Chinese University of Hong Kong, Shatin, New Territories, Hong Kong, China. E-mail: franky-chan@ cuhk.edu.hk. * This work was supported by an RGC Earmarked Research Grant (CUHK 4131/00M) from the Hong Kong Research Grant Council (to F.L.C.) and a grant from the Medical Research Council of Canada (to J.W.X.).

observed by RT-PCR analysis. Their mRNA transcripts were restored to normal levels after replacement with testosterone. The results indicate that these secretory proteins are all under androgen regulation in the rat LP. Interestingly, we also observed that their degrees of sensitivity or responsiveness to androgen withdrawal are different. Their mRNA levels dropped in response to duration of castration in the following decreasing order: SVSII, PSP94, and probasin. Besides androgen [dihydrotestosterone (DHT)], we also examined the effects of glucocorticoid [dexamethasone (DEX)], progestin [medroxyprogesterone acetate (MPA)], and zinc on their gene expressions in castrated LPs. We observed that the mRNA transcripts of both PSP94 and probasin were increased after treatments with DHT, DEX, and MPA, suggesting that these two proteins could also be regulated by glucocorticoid and progestin. In contrast with probasin, PSP94 and SVSII were not induced by ZnSO4 treatment. On the other hand, SVSII expression was only increased significantly by DHT and moderately by MPA, but not by DEX, suggesting that SVSII is under strict control by androgen. (Endocrinology 141: 4543– 4551, 2000)

plasia in human prostate. A few estrogen-dependent and androgen-independent prostate tumor lines are derived from the dorsolateral prostate of estrogen-treated Noble rats (13, 14). Prostatitis is also developed at high incidence in the lateral prostate (LP) of aged rats (15), probably caused by the accumulation of secretions in the glandular acini (16). Several secretory proteins have been isolated and characterized from the rat dorsolateral prostate. For example, the dorsal prostate secretes two major proteins, dorsal proteins I and II (DPI and DPII) (17, 18), whereas the LP secretes DPI (19), probasin (19 –21), and seminal vesicle secretion II (SVSII) (22). Syntheses of these proteins are shown to be regulated mainly by androgens (17, 19, 21, 23). Recently, another small secretory protein, prostatic secretory protein of 94 amino acids (PSP94 or ␤-microseminoprotein), has been isolated and cloned from the rat and mouse prostates (24, 25). This small protein is shown to be highly expressed and synthesized in the LP (26). This protein has a human homolog that was originally isolated from seminal plasma. It is one of the major secretory proteins of the human prostate gland, in addition to prostatic acid phosphatase and prostatespecific antigen (27, 28). Research interest in this protein is related to its potential use as a prognostic marker for prostate

4543

4544

PROBASIN, PSP94, AND SVSII IN RAT PROSTATE

cancer. However, there is little information on its regulation by androgen and other hormones in human or rodent prostate glands. In the present study we investigated the in vivo androgen regulation of the messenger RNA (mRNA) expressions of three major secretory proteins (PSP94, probasin, and SVSII) in long-term castrated LP. Besides androgen, the effects of other steroid hormones (glucocorticoid and progestin) and zinc on their gene expressions in the castrated prostates were also studied.

Endo • 2000 Vol. 141 • No. 12

of rat seminal vesicle as a template. The 5⬘-primer (5⬘-CGG AAT TCA GTG GAC AGC TGA AAT CTG-3⬘; the underline indicates the linker sequence of EcoRI) and the 3⬘-primer (5⬘-CCA TCG ATT AGG ATT GGG AGC GTT CTT G-3⬘; the underline indicated the linker sequence of ClaI) were used to amplify an 780-bp fragment of rat SVSII (spanning nucleotides 2661–3424 of rat SVSII; GenBank accession no. J05443). The purified PCR product was verified by restriction enzyme digestion with NcoI before subcloning into EcoRI (XbaI) and ClaI (XhoI) sites of the pBluescript SK plasmid. The resulting plasmid (SVSII-780) was linearized with EcoRI (or XbaI; antisense) or ClaI (or XhoI; sense) to generate DIG-labeled RNA probes.

ISH Materials and Methods Castration and hormonal treatments of animals Mature male Noble rats (obtained from the Laboratory Animal Services Center of the Chinese University of Hong Kong), weighing 300 – 400 g, were surgically castrated under anesthesia via the scrotal route (9). Castrated rats were kept in cages with chow and water ad libitum for 3– 80 days. Intact age-matched rats (at least 10 for each experimental group) were used as normal controls. Castrated rats were killed at 3, 5, 14, 30, and 60 days postcastration. For androgen replacement study, rats that had been castrated for 60 days were implanted sc at the subscapular region with two 2-cm-long SILASTIC brand tubings (id, 1.96 mm; od, 3.18 mm; Dow Corning Corp., Midland, MI) filled with testosterone (11, 12). Each tubing contained about 43.9 ⫾ 1.1 mg testosterone. Castrated rats were treated with testosterone for 3 additional weeks before death by an overdose of chloral hydrate (700 mg/kg, ip). About 5.0 ⫾ 0.8 mg testosterone (per two tubings) were released over the 3-week period. For negative controls, castrated rats were implanted with two empty tubings. The animal protocols used in this study were approved by the animal research ethics committee of the Chinese University of Hong Kong. For in situ hybridization (ISH) studies, the excised urethra-prostatic complexes were fixed in 4% paraformaldehyde in PBS for 4 h or overnight at 4 C before being embedded in paraffin as described previously (26). For total RNA and protein extractions, the lateral prostatic lobes (LP) were quickly dissected from the prostate gland-urethra complex over an ice bath and under a dissecting microscope, snap-frozen in liquid nitrogen, and stored at ⫺70 C until use. For study of the effects of steroid hormones and ZnSO4 on the gene expressions of the prostatic secretory proteins, male Noble rats of about 300 g BW were castrated. The castrated rats were randomly divided into five experimental groups. Each group consisted of six age-matched unoperated males as normal controls. Thirty days postcastration, the rats were received daily sc or ip injections of dihydrotestosterone (DHT; 3 mg/kg䡠day, sc; Ref. 43), dexamethasone (DEX; 3 mg/kg䡠day, sc; Ref. 43), medroxyprogesterone acetate (MPA; 1.5 mg/kg䡠day, ip; Ref. 46), or zinc sulfate (ZnSO4; 5 mg/kg䡠day, ip; Ref. 43) for 7 days. Both DHT and DEX were dissolved in corn oil. MPA was dissolved first in 90% ethanol and then diluted in 0.9% NaCl with 1% gelatin before being administered to animals. ZnSO4 was dissolved in 0.9% NaCl. As controls for treatments, the castrated animals were received vehicle only (corn oil or saline). Animals were killed 7 days after drug treatments. The LPs were dissected, immediately snap-frozen in liquid nitrogen, and stored at ⫺70 C until RNA extraction.

Nonradioactive ISH was performed according to the procedure described previously (26). In brief, the prehybridized sections were hybridized with DIG-labeled RNA probes for rat PSP94, probasin, or SVSII (8 ng/␮l in a hybridization buffer containing 4 ⫻ SSC, 2 ⫻ Denhardt’s solution, 10% dextran sulfate, 50% deionized formamide, and 0.2 mg/ml salmon sperm DNA) overnight at 56 C in a humid chamber. After hybridization and washes, the hybridized signals in sections were visualized by anti-DIG-alkaline phosphatase immunohistochemistry. All enzymes, DIG-labeled UTP, and anti-DIG-alkaline phosphatase were obtained from Roche (Mannheim, Germany).

Protein extraction, SDS-PAGE, and Western blotting analysis Proteins were extracted from the homogenized frozen prostatic tissues by TRIzol reagent (Life Technologies, Inc.) according to the manufacturer’s instructions. The protein concentrations were determined by the bicinchoninic acid protein assay with BSA as standard before aliquoting and were kept at ⫺20 C until electrophoretic analysis. Protein samples were separated by SDS-PAGE in a 15% gel using a Minigel apparatus. After electrophoresis, proteins were transblotted onto the 0.2-␮m polyvinylidene difluoride membranes and stained with a rabbit antiserum against the recombinant glutathione-S-transferase-rat PSP94 as described previously (26, 29).

RNA extraction and RT-PCR Total cellular RNA was extracted from the homogenized frozen tissues by TRIzol reagent. Extracted RNA was dissolved in diethylpyrocarbonate (DEPC)-treated water and quantified by measuring its A260. Approximately 1.5 ␮g DNase I-treated RNA samples were reverse transcribed to first strand cDNA using a SuperScript preamplification system (Life Technologies, Inc.) for each RT-PCR reaction. After RT, PCR was performed in a thermal cycler (GeneAmp 9600, Perkin-Elmer Corp.). All reactions were terminated at 4 C. The sequences of oligonucleotide primers used in this study and the conditions of the PCR are listed in Table 1. The specificity of the SVSII PCR product was verified by digestion with NcoI and sequencing. Control reactions for RT-PCR were performed by replacing RNA sample for RT with DEPC-treated water (no RNA control), using RNA sample that was not reverse transcribed to cDNA (no RT control) and replacing RT product with DEPCtreated water (no cDNA template).

Semiquantitative RT-PCR Synthesis of digoxigenin (DIG)-labeled RNA probes Full-length rat PSP94 complementary DNA (cDNA) generated by RT-PCR was subcloned into BamHI and EcoRI sites of a pBluescript KS⫺ plasmid (25). The rat PSP94 plasmid was linearized with BamHI (sense probe; control) or EcoRI (antisense probe) and then used to generate DIG-labeled RNA probes with DIG-labeled UTP by in vitro transcription as described previously (26). A 550-bp probasin cDNA fragment subcloned in pAT153 (pM-40.3, a clone provided by Dr. Robert J. Matusik, Vanderbilt University, Nashville, TN) was cleaved by digestion with PstI and purified by electrophoresis in 1% agarose. The cleaved cDNA fragment was then subcloned into BamHI and EcoRI sites of the pBluescript SK plasmid. The resulting plasmid (pB-550) was linearized with BamHI (antisense) or EcoRI (sense) to generate DIG-labeled RNA probes. The rat SVSII transcripts were amplified by RT-PCR using the first strand cDNA

Semiquantitative RT-PCR of PSP94, probasin, and SVSII mRNAs was performed and compared with that of the ␤-actin mRNA. PCR conditions were same as described above and are summarized in Table 1. Total cellular RNA (1.5 ␮g) was reverse transcribed, and 2 ␮l of the resulting cDNA samples were subsequently used in PCRs for PSP94, probasin, SVSII, and ␤-actin. The optimal PCR conditions were determined so that the amplifications were within the exponential phase (linear range) of PCR with respect to the amount of input RNA. Separate reactions were also performed in parallel using primers for rat ␤-actin cDNA for comparative measurements. PCR cycle numbers for rat PSP94, probasin, and SVSII were 20, whereas the number was 27 for ␤-actin. After PCR, the products were resolved on a 1.5% agarose gel with ethidium bromide. The fluorescent images were captured under UV transillumination with a CCD camera (Fotodyne, Inc., Hartland, WI) and recorded on black and


4545

TABLE 1. Sequences of oligonucleotide primers and conditions of PCR Primers

a. Rat PSP94 5⬘-Primer

3⬘-Primer b. Rat probasin 5⬘-Primer

3⬘-Primer c. Rat SVS II 5⬘ Primer 3⬘ Primer d. Rat ␤-actin 5⬘-Primer 3⬘-Primer

Sequences

5⬘-AGT CCT GCT TTG CCA CCA TG-3⬘

PCR conditions

References

94 C for 1 min, 63 C for 1 min, 72 C for 2 min (for 1–5 cycles); 94 C for 1 min, 57 C for 1 min, 72 C for 2 min (for 6 –20 cycles)

24, 26; GenBank accession no. U65486

94 C for 30 sec, 65 C for 30 sec, 72 C for 30 sec (for 1–5 cycles); 94 C for 30 sec, 60 C for 30 sec, 72 C for 30 sec (for 6 –20 cycles)

GenBank accession no. M27156

5⬘-AGT GGA CAG CTG AAA TCT GGA-3⬘ 5⬘-TAG GAT TGG GAG CGT TCT TG-3⬘ (spanning nucleotides 2661–3424)

94 C for 1 min, 60 C for 1 min, 72 C for 2 min (for 1–20 cycles)

22; GenBank accession nos. J05443, M54795

5⬘-ATG GAT GAC GAT ATC GCT GCG-3⬘

Same conditions as of PSP94, probasin or SVS II but reaction cycles were extended to 27

GenBank accession no. V01217

5⬘-TTT CAA GAC ATC CTG GCT CTG-3⬘ (spanning nucleotides 32–522) 5⬘-CTT GTC AGT GAG GTC CAG ATA CCT ACA GAG-3⬘ 5⬘-GCA AGT GAA ATA GAT TTA TTT AAA CAT TCA-3⬘ (spanning nucleotides 1–776)

5⬘-CTC CAT ATC GTC CCA GTT GGT-3⬘ (spanning nucleotides 1–250 of cds)

white photographs. The intensities of the bands were converted into digitalized signals with an image scanner and quantified by a gel documentation program (ImageQuant, Molecular Dynamics, Inc., Sunnyvale, CA). Signal intensities of the rat PSP94, probasin, and SVSII products were normalized to those of rat ␤-actin products as ratios to produce arbitrary units of relative abundance. The mean values obtained from five separate experiments were plotted as percentages of the day 0 control value (normal intact LP).

Statistical analysis Statistical analyses were performed using the SigmaStat program (SPSS, Inc., Chicago, IL). For studies of androgen regulation, statistical differences between intact control and various castration groups were determined by one-way ANOVA and then Tukey’s post-hoc test. For studies of regulation by steroids and zinc, statistical differences between corn oil-treated control and various treatment groups were determined by the same analysis. Differences were considered significant if P ⬍ 0.05.

Results Effects of castration on the mRNA expressions of PSP94, probasin, and SVSII

The results of ISH of the three secretory proteins demonstrated intense and specific hybridization signals over the secretory epithelium of the lateral lobe of the intact rat prostate gland (Figs. 1a–3a). Seven to 14 days postcastration, the hybridization signals of PSP94 and probasin remained intense in the secretory epithelium (not shown). However, the intensity of SVSII signals dropped significantly after 7 days of castration (Fig. 3b) and disappeared at 14 days postcastration (Fig. 3c). Significant reductions of PSP94 and probasin signals were observed at 30 days postcastration (Figs. 1b and 2b). Their signals became totally lost 60 days after castration (Figs. 1c and 2c). The signals of all three secretory proteins

were restored to normal levels when the castrated animals were replaced with testosterone for 3 weeks (Figs. 1d–3d). The relative abundance of mRNA transcripts of the three secretory proteins (expressed as ratios to ␤-actin) in response to androgen withdrawal by castration and subsequent replacement with testosterone was quantified by a semiquantitative RT-PCR protocol. The analysis demonstrated that there was a significant decrease of about 70% in the PSP94 mRNA level in the LP after 30 days of castration (Fig. 4). Its mRNA transcript continued to decline in castrated LP and dropped to a level of only 10% compared with that in intact LP when the castration was prolonged to 60 days. Its expression rebounded to the normal level after testosterone replacement for 3 weeks. PCR analysis also showed that the probasin mRNA level in the LP dropped after 60 days of castration (Fig. 5). However, such a decrease was not statistically significant. Interestingly, we observed a slight rebound of the probasin mRNA level 14 days after castration. A discrepancy between the results of ISH and RT-PCR of probasin in LP castrated for 60 days was noted and attributed to the different sensitivities of mRNA detection between these two techniques. Its transcripts returned to the normal level after replacement with testosterone for 3 weeks. On the other hand, the decrease in SVSII mRNA levels in castrated LP was much faster than that in PSP94 and probasin (Fig. 6). Its expression started to drop in the early days postcastration (3–5 days). At 14 days postcastration, its expression level became very low in LP and dropped to a level of only 10%. After 60 days of castration, its expression level became almost undetected by PCR. Similarly, its transcripts rebounded to normal levels after testosterone replacement.

4546


FIG. 1. In Figs. 1–3, the results of ISH of PSP94, probasin, and SVSII mRNA in normal, castrated, and androgen-replaced LPs are shown. a, ISH of PSP94 in normal LP. The secretory epithelium is intensely reacted. No hybridization signal is seen in the stromal tissue. b, ISH of PSP94 in LP castrated for 30 days. The secretory acini became atrophied. The intensity of hybridization signals is significantly reduced. c, ISH of PSP94 in LP castrated for 60 days. The hybridization signals become totally lost in the secretory epithelium. d, ISH of PSP94 in androgen-replaced castrated LP. The hybridization signals in the epithelium are as strong as those in the intact LP. All micrographs are at the same magnification (⫻40).

The protein levels of PSP94 in castrated LPs were also examined by Western blottings (Fig. 7). The antibody recognized a strong band or smear of about 14.5 kDa in the intact LP. Its intensities became weakened 3–5 days after castration. At 14 days postcastration, its signal became very weak. No binding signal was detected in the LPs castrated for 30 and 60 days. After testosterone replacement, its binding intensity in the castrated LP was as strong as that in the intact LP.

Endo • 2000 Vol. 141 • No. 12

FIG. 2. a, ISH of probasin in normal LP. The secretory epithelium is strongly reacted. b, ISH of probasin in LP castrated for 30 days. The signals in the epithelium are significantly reduced. c, ISH of probasin in LP castrated for 60 days. The epithelium is negatively stained. d, ISH of probasin in androgen-replaced castrated LP. The epithelium is strongly reacted. All micrographs are at the same magnification (⫻40).

Effects of steroid hormones and zinc on PSP94, probasin, and SVSII expressions in castrated rat prostates

Besides androgen, the effects of other steroid hormones and zinc on the mRNA expressions of the three secretory proteins in castrated LPs were analyzed by the semiquantitative RT-PCR (Figs. 8–10). The analysis showed that there were significant increases in PSP94 mRNA transcripts in the castrated LPs after treatments with DHT, DEX, and MPA compared with the level after corn oil treatment (Fig. 8). When treated with DHT, the mRNA level of PSP94 increased to about 90% of normal. Treatments with DEX and MPA increased PSP94 mRNA levels to about 20% and 35% of


4547

FIG. 4. a and b, Semiquantitative RT-PCR of PSP94 in normal, castrated, and androgen-replaced LPs. a, Upper panel, RT-PCR (20 cycles) of PSP94 mRNA (shown as bands of 490 bp) in the normal LP; LPs castrated for 3, 5, 14, 30, and 60 days; and castrated LP followed by androgen replacement. Its signal is strong in normal LP and becomes very weak after 30 days of castration. After androgen replacement, its signal becomes strong again. Controls for RT-PCR lacking cDNA templates are the absence of reverse transcriptase (no RT), total RNA (no RNA), and cDNA templates (no template). No signal is shown for the controls. Lower panel, RT-PCR (27 cycles) of rat ␤-actin mRNA in the same samples tested for PSP94. A band (249 bp) of similar intensity is seen in all tested samples except the controls. b, Relative abundance of PSP94 mRNA transcripts in castrated (PC3– 60) and androgen-replaced (AR) LPs as determined by semiquantitative RT-PCR is expressed as a percentage of the value for normal LP (intact control). Each data point is the mean ⫾ SE of five independent measurements. Asterisks indicate that levels are significantly different from the intact normal LP (*, P ⬍ 0.05).

of the normal level. MPA treatment slightly increased SVSII mRNA to about 20% of the normal level. However, there was no effect on SVSII mRNA expression after DEX and ZnSO4 treatments. FIG. 3. a, ISH of SVSII in normal LP. The secretory epithelium shows intense reaction. b, ISH of SVSII in LP castrated for 7 days. The hybridization signals become dramatically reduced in the secretory epithelium. c, ISH of SVSII in LP castrated for 14 days. No signals are seen in the secretory epithelium. d, ISH of SVSII in androgenreplaced castrated LP. The secretory epithelium is strongly reacted. All micrographs are at the same magnification (⫻40).

normal, respectively. However, there was no increase in PSP94 mRNA expression in the ZnSO4-treated LP. Significant increases in probasin mRNA expression in castrated LPs were observed after treatments with DHT, DEX, and MPA (Fig. 9). Probasin mRNA transcripts rebounded to the normal level (100%) when the castrated rats received DHT. DEX and MPA treatments also increased probasin mRNA to approximately 75% of the normal level. ZnSO4 treatment slightly increased probasin mRNA to about 45% of the normal level. The expression of SVSII mRNA in castrated LPs was only affected significantly by DHT treatment (Fig. 10). After DHT treatment, SVSII mRNA was increased to approximately 80%

Discussion

In the present study we showed that the mRNA expression levels of three secretory proteins decreased or disappeared in LP castrated for over 30 days, and they could be restored to normal levels by replacement with androgen. We also observed that their degrees of sensitivities or responsiveness to androgen withdrawal by castration were different. Their mRNA levels dropped in the following decreasing order: SVSII, PSP94, and probasin. The results suggest that the mRNA expressions of the three prostatic secretory proteins are all under regulation of androgen. The decrease in SVSII mRNAs in response to androgen withdrawal is more rapid than those in PSP94 and probasin. Regulation of probasin and SVSII expressions by androgen has been studied previously in short-term castrated dorsolateral prostates. Using RNA hybridization, Matusik and colleagues (19, 21, 22, 30) reported that mRNA levels of probasin (clone M-40) and SVSII (clone RWB) decrease 6 days

4548


FIG. 5. a and b, Semiquantitative RT-PCR of probasin in normal, castrated, and androgen-replaced LPs. a, Upper panel, RT-PCR (20 cycles) of probasin mRNA (shown as bands of 776 bp) in the normal LP; LPs castrated for 3, 5, 14, 30, and 60 days; and castrated LP followed by androgen replacement. Its signal is strong in normal LP and becomes slightly weakened after 30-day castration. After androgen replacement, its signal becomes intense again. Controls for RTPCR are same as those for PSP94. No signal is shown for the controls. Lower panel, RT-PCR (27 cycles) of rat ␤-actin mRNA in the same samples tested for probasin. A band (249 bp) of similar intensity is seen in all tested samples except in the controls. b, Relative abundance of probasin mRNA transcripts is expressed as a percentage of the value in normal LP (intact control). Each data point is the mean ⫾ SE of five measurements. Statistical analysis shows that mRNA levels of probasin in LPs castrated for 60 days do not differ significantly from those in the intact LP control (P ⬎ 0.05).

after castration, and their levels are restored to normal after testosterone administration, whereas SVSII mRNA decreases more rapidly than probasin in response to androgen removal and remains low in the castrated prostates but can be restored to normal 48 h after androgen replacement. In contrast with SVSII, rebound of the probasin mRNA level is observed 6 days after castration and reaches a normal level 12 days after castration (21, 30). Such a rebound after castration was also detected by RT-PCR in the present study. However, regulation of the rebound in castrated LP remains unclear. Matusik et al. suggest that androgen is not the only inducible factor for the probasin gene expression. The authors have reconfirmed by ISH that probasin, but not SVSII, is still detected in the LP after castration for 12 days (30). A similar study using immunohistochemistry has shown that probasin is not detected in the acinar lumen, but its staining is still detected in the prostatic epithelium after castration for 7 days (20). Further analysis of the probasin gene by deoxyribonuclease I footprinting has identified two different cis-acting DNA elements in the 5⬘-flanking region, and both DNA elements bind more strongly to androgen receptor than to glucocorticoid receptor (31). Kasper and colleagues (32) further characterized these two androgen receptor-binding sites in the probasin gene and found that the binding of androgen receptor to these DNA elements occurs in a cooperative and

Endo • 2000 Vol. 141 • No. 12

FIG. 6. a and b, Semiquantitative RT-PCR of SVSII in normal, castrated, and androgen-replaced LPs. a, Upper panel, RT-PCR (20 cycles) of SVSII mRNA (shown as bands of 764 bp) in the normal LP; LPs castrated for 3, 5, 14, 30, and 60 days; and castrated LPs followed by androgen replacement. Its signal is strong in normal LP and LP castrated for 3–5 days. Its signal is reduced significantly 14 days after castration and is undetectable 30 days after castration. After androgen replacement, its signal becomes strong again. Controls for RTPCR are same as those for PSP94. No signal is shown for the controls. Lower panel, RT-PCR (27 cycles) of rat ␤-actin mRNA in the same samples tested for SVSII. b, Relative abundance of SVSII mRNA transcripts is expressed as a percentage of the value in normal LP (intact control). Each data point is the mean ⫾ SE of five measurements. Asterisks indicate that levels are significantly different from the intact LP (*, P ⬍ 0.05).

mutually dependently manner. Sequence analysis of the SVSII gene has identified one DNA element that binds androgen receptor, and another DNA element that binds glucocorticoid receptor, in the 5⬘-flanking region (22). However, the present study shows that the synthetic glucocorticoid (DEX) has no effect on SVSII mRNA expression. In contrast with probasin and SVSII, little is known about androgen regulation of PSP94 expression in both rodent and human prostate glands. Sequence analyses have identified more than one steroid hormone-responsive element in the promoter region of three primate PSP94 genes, including human, and one estrogen-responsive element in the first intron, suggesting that the expression of PSP94 in these primate species could be regulated by steroid hormones (33, 34). In addition, a DNA element that binds glucocorticoid receptor is found in the first intron of the human PSP94 gene (35). Besides steroid hormone-responsive elements, a cAMP response element was demonstrated in the promoter of the human PSP94 (36). However, the significance of these DNA elements is still unclear, as detailed functional assays have not been performed for this gene. Using a transient transfection assay, Ochiai et al. (35) demonstrated that the human PSP94 promoter contains two regions, from ⫺275 to ⫺207


FIG. 7. Western blot analysis of PSP94 protein in normal LP, castrated LP, and castrated LP followed by androgen replacement. The antiserum (1:2500 dilution) recognizes a strong band or smear of about 14.5 kDa in normal LP. Its intensity becomes significantly reduced in LP castrated for 14 days and disappears after castration for 30 and 60 days. After androgen replacement, its signal becomes as strong as in the normal LP.

and from ⫺186 to ⫺128, that might function in a prostatic cell-specific manner. As the receptors for androgen (37), glucocorticoid (38), and progesterone (39, 40) are expressed in the rat prostate, it is believed that these steroid hormones may regulate the mRNA expression of PSP94 in the rat LP through their specific receptors. The present study demonstrated for the first time that the expression of rat PSP94 could be up-regulated by androgen, progestin, and glucocorticoid. The in vivo regulation of PSP94, probasin, and SVSII by glucocorticoid, progesterone, and ZnSO4 has not been demonstrated previously in rat prostate. In a cotransfection experiment on a prostatic cancer cell line (PC-3) with both the recombinant probasin promoter-chloramphenicol acetyltransferease (CAT) plasmid and either glucocorticoid or progesterone receptor expression vector, the probasin gene was shown to be inducible by both glucocorticoid and progesterone (31). A deletion mapping study has demonstrated that the steroid hormone-responsive elements are localized to the region between ⫺244 and ⫺158 of the probasin promoter (31). Recently, a DNA element (PB-ARE-2) in the promoter region of probasin gene has been studied by gel retardation analysis and shown to be induced only by androgens, not by glucocorticoids (41). However, our present observation of the in vivo effect of glucocorticoid on the native probasin gene expression in castrated prostate is in contrast to what has been observed in a transgenic mouse model generated by different lengths of DNA fragments of the probasin promoter (⫺426 to ⫹28 bp or ⫺11.5 kb to ⫹28 bp) and simian virus 40-Tag (42, 43). Transgene (CAT) expression in these castrated transgenic mice either is not induced or is weakly induced by DEX. The difference in response of probasin expression to glucocorticoid between transgenic mice and rats suggests that the hormonal regulation of this transgene in the transgenic mouse ventral prostate is different from that of its native gene in the rat lateral prostate, and some other rat prostate lobe-specific factors may be involved in its regulation. On the other hand, our present study demonstrated for the first time that the in vivo expression of probasin

4549

FIG. 8. a and b, Semiquantitative RT-PCR of PSP94 in normal LP and castrated LPs treated with corn oil, steroid hormones, and ZnSO4. a, Upper panel, RT-PCR (20 cycles) of PSP94 mRNA (shown as 490-bp bands) in normal LP and castrated LPs treated with corn oil, DHT, DEX, MPA, and ZnSO4. Controls for RT-PCR are the absence of RT, total RNA, and cDNA templates. No signal is shown for the controls. Lower panel, RT-PCR (27 cycles) of rat ␤-actin mRNA in the same samples as those tested for PSP94. b, Relative abundance of PSP94 mRNA transcripts is expressed as a percentage of the value in normal LP (intact control). Each data point is the mean ⫾ SE of five measurements. Asterisks indicate that levels are significantly different from the castrated LP receiving corn oil only (*, P ⬍ 0.05).

mRNA was also induced by glucocorticoid. It is unclear whether its induction is mediated via the same consensus or different steroid receptor-responsive element in its promoter region. To date there is no information on the regulation of the transcriptional activities of probasin, PSP94, and SVSII by progestins. Besides progesterone receptor, the induction on these proteins by MPA in castrated LPs could be mediated by androgen receptor, as MPA is also a weak androgen agonist that can bind to the androgen receptor, particularly at higher concentrations (44 – 46). It is estimated that a much higher dose of MPA (1000 times that of testosterone) is required to increase ventral prostate weight in castrated rats (47). This is confirmed by its negative action in an androgeninsensitive testicular feminized mouse mutant with defective androgen receptor (44). It is also demonstrated that the binding of MPA to androgen receptor does not involve the interaction between the NH2- and carboxyl-terminals of the receptor (48). MPA can significantly up-regulate the mRNA expression of a secretory protein, prostatic binding protein, in the castrated rat ventral prostate (46, 49). Therefore, the up-regulation of mRNA expression of the three secretory proteins, particularly probasin, by MPA in rat prostate could be mediated through the androgen or progesterone receptor, or both. The regulation of probasin expression by zinc in the rat prostate has been documented, although its mechanism of action is still unclear. This observation is correlated to the

4550


FIG. 9. a and b, Semiquantitative RT-PCR of probasin in normal LP and castrated LPs treated with corn oil, steroid hormones, and ZnSO4. a, Upper panel, RT-PCR (20 cycles) of probasin mRNA (shown as 776-bp bands) in normal LP and castrated LPs treated with corn oil, DHT, DEX, MPA, and ZnSO4. Controls for RT-PCR are same as those for PSP94. No signal is shown for the controls. Lower panel, RT-PCR (27 cycles) of rat ␤-actin mRNA in the same samples as those tested for probasin. b, Relative abundance of probasin mRNA transcripts is expressed as a percentage of the value for normal LP (intact control). Each data point is the mean ⫾ SE of five measurements. Asterisks indicate that levels are significantly different from those for castrated LP receiving corn oil only (*, P ⬍ 0.05).

unusually high zinc content in the LP. Matusik et al. (21) first reported that the probasin mRNA level can be elevated in the castrated dorsolateral prostate by zinc. Similar positive probasin promoter-CAT induction by zinc was observed in the castrated prostates of a transgenic mouse (43). In contrast with probasin, zinc demonstrated no effect on the expression of PSP94 and SVSII in castrated LPs in the present study. The difference in androgen and zinc regulation of probasin and two other proteins may be due to the fact that probasin appears as both a secreted protein and a nuclear protein in prostatic epithelial cells and is shown to be regulated by translational initiation at different AUG codons of the same mRNA (50). In summary, the present study shows that the gene expressions of PSP94, probasin, and SVSII are all under androgen regulation in the rat LP, as their mRNA levels are all decreased significantly or are completely lost in the longterm castrated LPs. Also noted is that their sensitivities toward androgen withdrawal by castration are different. Among the three proteins, the mRNA levels of SVSII dropped more rapidly than those of PSP94 and probasin after castration. Besides androgen, glucocorticoid and progestin can regulate the gene expressions of PSP94 and probasin in castrated LPs. In contrast with probasin, PSP94 and SVSII are not induced by zinc. SVSII expression was increased significantly by DHT and moderately by progestin, but not by glucocorticoid, suggesting that SVSII is under strict control by androgen.

Endo • 2000 Vol. 141 • No. 12

FIG. 10. a and b, Semiquantitative RT-PCR of SVSII in normal LP and castrated LPs treated with corn oil, steroid hormones, and ZnSO4. a, Upper panel, RT-PCR (20 cycles) of SVSII mRNA (shown as 764-bp bands) in normal LP and castrated LPs treated with corn oil, DHT, DEX, MPA, and ZnSO4. Controls for RT-PCR are same as those for PSP94. No signal is shown for the controls. Lower panel, RT-PCR (27 cycles) of rat ␤-actin mRNA in the same samples as those tested for SVSII. b, Relative abundance of SVSII mRNA transcripts is expressed as a percentage of the value for normal LP (intact control). Each data point is the mean ⫾ SE of five measurements. Asterisks indicate that levels are significantly different from those for castrated LP receiving corn oil only (*, P ⬍ 0.05).

Acknowledgment The authors gratefully acknowledge Dr. R. J. Matusik, Vanderbilt University, for providing the pM-40.3 clone of rat probasin.

References 1. Schrodt GR 1961 The fine structure of the lateral lobe of the rat prostate gland. Comparison with the dorsal and other lobes. J Ultrastruct Res 5:485– 496 2. Brandes D 1964 The fine structure and histochemistry of prostatic glands in relation to sex hormones. Int Rev Cytol 20:207–276 3. Hayashi N, Sugimura Y, Kawamura J, Donjacour AA, Cunha GR 1991 Morphological and functional heterogeneity in the rat prostatic gland. Biol Reprod 45:308 –321 4. Mann T 1964 Male accessory organs of reproduction, and their secretory product: the seminal plasma. In: Mann T (ed) The Biochemistry of Semen and of the Male Reproductive Tract. Methuen, London, pp 37–78 5. Fouquet JP 1971 Secretion of free glucose and related carbohydrates in the male accessory organs of rodents. Comp Biochem Physiol [A] 40:305–317 6. Aumu¨ller G, Seitz J 1990 Protein secretion and secretory processes in male accessory sex glands. Int Rev Cytol 121:127–231 7. Chan FL, Ho SM 1999 Comparative study of glycoconjugates of the rat prostatic lobes by lectin histochemistry. Prostate 38:1–16 8. Douglas J, Resnick WI, Grayhack JT 1976 Effect of testosterone, dihydrotestosterone, estradiol and prolactin on the weight and citric acid content of the lateral lobe of the rat prostate. Invest Urol 14:60 – 65 9. Kwong J, Choi HL, Huang Y, Chan FL 1999 Ultrastructural and biochemical observations on the early changes in apoptotic epithelial cells of the rat prostate induced by castration. Cell Tissue Res 298:123–136 10. Dunning WF 1963 Prostate cancer in the rat. Natl Cancer Inst Monogr 12:351–369 11. Leav I, Ho SM, Ofner P, Merk FB, Kwan PWL, Damassa D 1988 Biochemical alterations in sex hormone-induced hyperplasia and dysplasia of the dorsolateral prostates of the Noble rats. J Natl Cancer Inst 80:1045–1053 12. Leav I, Merk FB, Kwan PWL, Ho SM 1989 Androgen-supported estrogenenhanced epithelial proliferation in the prostates of the intact Noble rats. Prostate 15:23– 40 13. Noble RL 1980 Development of androgen-stimulated transplants of Nb rat


14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

26. 27. 28. 29.

30. 31.

32.

carcinoma of the dorsal prostate and their response to sex hormones and tamoxifen. Cancer Res 40:3551–3554 Noble RL 1982 Prostate carcinoma of the Nb rat in relation to hormones. Int Rev Exp Pathol 23:113–159 Muntzing J, Sufrin G, Murphy GP 1979 Prostatitis in the rat. Scand J Urol Nephrol 13:17–22 Aumu¨ller G, Enderle-Schmitt U, Seitz J, Muntzing J, Chandler JA 1987 Ultrastructure and immunohistochemistry of the lateral prostate in aged rats. Prostate 10:245–256 Wilson EM, French FS 1980 Biochemical homology between rat dorsal prostate and coagulating gland. J Biol Chem 255:10946 –10953 Bartlett RJ, French FS, Wilson EM 1984 In vitro synthesis and glycosylation of androgen-dependent secretory proteins of rat dorsal prostate and coagulating gland. Prostate 5:75–91 Dodd JG, Sheppard PC, Matusik RJ 1983 Characterisation and cloning of rat dorsal prostate mRNAs. Androgen regulation of two closely related abundant mRNAs. J Biol Chem 258:10731–10737 Matuo Y, Nishi N, Muguruma Y, Yoshitake Y, Kurata N, Wada F 1985 Localisation of prostatic basic protein (“probasin”) in the rat prostates by use of monoclonal antibody. Biochem Biophys Res Commun 130:293–300 Matusik RJ, Kreis C, McNicol P, Sweetland R, Mullin C, Fleming WH, Dodd JG 1986 Regulation of prostatic genes: role of androgens and zinc in gene expression. Biochem Cell Biol 64:601– 607 Harris SE, Harris MA, Johnson CM, Bean MF, Dodd JG, Matusik RJ, Carr SA, Crabb JW 1990 Structural characterisation of the rat seminal vesicle secretion II protein and gene. J Biol Chem 265:9896 –9903 Dodd JG, Kreis C, Sheppard PC, Hamel A, Matusik RJ 1986 Effect of androgens on mRNA for a secretory protein of rat dorsolateral prostate and seminal vesicles. Mol Cell Endocrinol 47:191–200 Fernlund P, Granberg L, Larsson I 1996 Cloning of ␤-microseminoprotein of the rat: a rapidly evolving mucosal surface protein. Arch Biochem Biophys 334:73– 82 Xuan JW, Kwong J, Chan FL, Ricci M, Imasato Y, Sakai H, Fong GH, Panchal C, Chin JL 1999 cDNA, genomic cloning and gene expression analysis of mouse PSP94 (prostate secretory protein of 94 amino acids). DNA Cell Biol 18:11–26 Kwong J, Xuan JW, Choi HL, Chan PSF, Chan FL 2000 PSP94 (or ␤microseminoprotein) is a secretory protein specifically expressed and synthesized in the lateral lobe of the rat prostate. Prostate 42:219 –229 Lilja H, Abrahamsson PA 1988 Three predominant proteins secreted by the human prostate gland. Prostate 12:29 –38 Hara M, Kimura H 1989 Two prostate-specific antigens, ␥-seminoprotein and ␤-microseminoprotein. J Lab Clin Med 113:541–548 Kwong J, Chan FL, Jiang S, Guo YZ, Imasato Y, Sakai H, Koropatnick J, Chin JL, Xuan JW 1999 Differential expression of PSP94 in rat prostate lobes as demonstrated by an antibody against recombinant GST-PSP94. J Cell Biochem 74:406 – 417 Sweetland R, Sheppard PC, Dodd JG, Matusik RJ 1988 Post-castration rebound of an androgen regulated prostatic gene. Mol Cell Biochem 84:3–15 Rennie PS, Bruchovsky N, Leco KJ, Sheppard PC, McQueen SA, Cheng H, Snoek R, Hamel A, Bock ME, MacDonald BS, Nickel BE, Chang C, Liao S, Cattini PA, Matusik RJ 1993 Characterisation of two cis-acting DNA elements involved in the androgen regulation of the probasin gene. Mol Endocrinol 7:23–36 Kasper S, Rennie PS, Bruchovsky N, Sheppard PC, Cheng H, Lin L, Shiu RPC, Snoek R, Matusik RJ 1994 Cooperative binding of androgen receptors to two DNA sequences is required for androgen induction of the probasin gene. J Biol Chem 269:31763–31769

4551

33. Nolet S, Mbikay M, Chre´tien M 1991 Prostatic secretory protein PSP94: gene organisation and promoter sequence in Rhesus monkey and human. Biochim Biophys Acta 1089:247–249 34. Xuan JW, Wu D, Guo Y, Garde S, Shum DT, Mbikay M, Zhong R, Chin JL 1997 Molecular cloning and gene expression analysis of PSP94 (prostate secretory protein of 94 amino acids) in primates. DNA Cell Biol 16:627– 638 35. Ochiai Y, Inazawa J, Ueyama H, Ohkubo I 1995 Human gene for ␤microseminoprotein: its promoter structure and chromosomal localisation. J Biochem 117:346 –352 36. Ueyama H, Ohkubo I 1998 The expression of ␤-microseminoprotein gene is regulated by cAMP. Biochem Biophys Res Commun 248:852– 857 37. Prins GS, Birch L, Greene GL 1991 Androgen receptor localization in different cell types of the adult rat prostate. Endocrinology 129:3187–3199 38. Schultz R, Isola J, Parvinen M, Honkaniemi J, Wikstro¨m AC, Gustafsson JÅ, Pelto-Huikko M 1993 Localisation of the glucocorticoid receptor in testis and accessory sexual organs of male rat. Mol Cell Endocrinol 95:115–120 39. Belis JA, Lizza EF, Tarry WF 1984 Progesterone receptors in the prostate. Prog Clin Biol Res 145:345–361 40. Lau KM, Leav I, Ho SM 1998 Rat estrogen receptor-␣ and -␤, and progesterone receptor mRNA expression in various prostatic lobes and microdissected normal and dysplastic epithelial tissues of the Noble rats. Endocrinology 139:424 – 427 41. Claessens F, Alen P, Devos A, Peeters B, Verhoeven G, Rombauts W 1996 The androgen-specific probasin response element 2 interacts differentially with androgen and glucocorticoid receptors. J Biol Chem 271:19013–19016 42. Greenberg NM, DeMayo FJ, Sheppard PC, Barrios R, Lebovitz R, Finegold M, Angelopoulou R, Dodd JG, Duckworth ML, Rosen JM, Matusik RJ 1994 The rat probasin gene promoter directs hormonally and developmentally regulated expression of a heterologous gene specifically to the prostate in transgenic mice. Mol Endocrinol 8:230 –239 43. Yan Y, Sheppard PC, Kasper S, Lin L, Hoare S, Kapoor A, Dodd JG, Duckworth ML, Matusik RJ 1997 Large fragment of the probasin promoter targets high levels of transgene expression to the prostate of transgenic mice. Prostate 32:129 –139 44. Mowszowicz I, Bieber DE, Chung KW, Bullock LP, Bardin CW 1974 Synandrogenic and antiandrogenic effect of progestins: comparison with nonprogestational antiandrogens. Endocrinology 95:1589 –1599 45. Bardin CW, Brown T, Isomaa VV, Janne OA 1984 Progestins can mimic, inhibit and potentiate the actions of androgens. Pharmacol Ther 23:443– 459 46. Labrie C, Simard J, Zhao HF, Pelletier G, Labrie F 1990 Synthetic progestins stimulate prostatic binding protein messenger RNAs in the rat ventral prostate. Mol Cell Endocrinol 68:169 –179 47. Raynaud JP, Bouton MM, Moguilewsky M, Ojasoo T, Philibert D, Beck G, Labrie F, Mornon JP 1980 Steroid hormone receptors and pharmacology. J Steroid Biochem 12:143–157 48. Kemppainen JA, Langley E, Wong CI, Bobseine K, Kelce WR, Wilson EM 1999 Distinguishing androgen receptor agonists and antagonists: distinct mechanisms of activation by medroxyprogesterone acetate and dihydrotestosterone. Mol Endocrinol 13:440 – 454 49. Pelletier G, Labrie C, Simard J, Duval M, Martinoli MG, Zhao H, Labrie F 1988 Effects of sex steroids on regulation of the levels of C1 peptide of rat prostatic steroid-binding protein mRNA evaluated by in-situ hybridization. J Mol Endocrinol 1:213–223 50. Spence AM, Sheppard PC, Davie JR, Matuo Y, Nishi N, McKeehan WL, Dodd JG, Matusik RJ 1989 Regulation of a bifunctional mRNA results in synthesis of secreted and nuclear probasin. Proc Natl Acad Sci USA 86:7843– 7847