(Clusterin) among Ventral Prostate, Seminal Vesicle, Testis

BIOLOGY OF REPRODUCTION 49, 233-242 (1993)

Localization and Molecular Heterogeneity of Sulfated Glycoprotein-2 (Clusterin) among Ventral Prostate, Seminal Vesicle, Testis, and Epididymis of Rats1 JULIA A. SENSIBAR, 2 3 YI QIAN,3 MICHAEL D. GRISWOLD, 4 STEVEN R. SYLVESTER, 4 C. WAYNE BARDIN,5 C. YAN CHENG,5 and CHUNG LEE 3 Department of Urology,3 Northwestern Univtersiy Medical School, Chicago, Illinois 60611 Department of Biochemistry,4 Washington State University, Pullman, Washington 99164 Center for Biomedical Research,5 The Population Council, New York, New York 10021 ABSTRACT In the rat reproductive tract, sulfated glycoprotein 2 (SGP-2) is present in the ventral prostate, seminal vesicle, testis, and epididymis. In the ventral prostate, SGP-2 is associated with the paxess of programmed cell death, while in the testis and epididymis a role for SGP-2 in sperm maturation has been proposed. Available information suggests that there are both interand intra-organ variations in SGP-2 localization, molecular forms, and response to androgen ablation. In the present study, localization of SGP-2 within the ventral prostate, seminal vesicle, and epididymis was compared by immunohistochemistry. In the ventral prostate of intact rats, immunoreactive SGP-2 was confined to a discrete population of epithelial cells lining the proximal ducts. Epithelial cells in other regions of the ventral prostate did not stain for SGP-2. A similar staining pattern was observed for the seminal vesicle; a small population of SGP-2-expressing epimelfial cils was found in epithelium that did not stain for SGP-2. The epididymis also demonstrated a non-uniform staining pattern. The caput displayed strong immunoperoxidase reaction over the apical membrane and stereocilia of all principal cells. Principal cells also showed variable degrees of cytoplasmic staining ranging from weak to strongly positive. The corpus and cauda showed a similar staining pattern. After castration, all epithelial cells in the ventral prostate and seminal vesicle became intensely positive for SGP-2 staining. In the caput and cauda epididymis there was an increase in the number of principal cells demonsrati stng intracellular staining for SGP-2. These results suggest that as observed previously in the regressing ventral prostate, increased intracellular SGP-2 staining may also be associated with the regressing epididymis and seminal vesicle. Differences in molar forms of SGP-2 were investigated by two-dimensional Western and lectin blots. Molecular forms of SGP-2 differed between testis and epididymis but were similar between ventral prostate and seminal vesicle. Prostate and seminal vesicle forms of SGP-2 differed from those of both testis and epididymis. Analysis of terminal carbohydrate present on the various SGP-2 molecular forms also confirmed the existence of heterogeneity. These results demonstrate the presence of multiple molecular forms of SGP-2 in various organs of the male reproductive tract in rats and suggest a possible variation in functional activity and/or haffIife of SGP-2 in these organs.

INTRODUCTION Sulphated glycoprotein-2 (SGP-2) was first described as a constitutively expressed secretory protein of Sertoli and epididymal epithelial cells [1]. Recently, comparison of nucleotide and N-terminal amino acid sequences has established that SGP-2 is identical to clusterin, testosterone-repressed prostate message-2, and androgen-repressed message and that it is homologous to SP-40,40, complement cytolysis inhibitor, and apolipoprotein J of human serum [1-7]. SGP2 is now known to be a widely distributed protein with a diverse range of proposed functions. Within the rat reproductive tract, SGP-2 message and protein are present in the ventral prostate, seminal vesicles, coagulating glands, testis, and epididymis [3, 4, 8, 9]. Available information seems to indicate that there are inter- and intra-organ variations in cellular localization [8, 10], level of

expression [4, 9], and molecular forms [11] of SGP-2 synthesized within the male reproductive tract. These variations may have an impact on the functional activity of the protein. By examining the pattern of these variations, we may be able to shed some light on the functional role of SGP-2. For example, within the compartments of the epididymis, variable levels of SGP-2 expression are found in principal cells of distal, proximal, and intermediate regions of the initial segment and between caput, corpus, and cauda [9, 10]. In the normal rat prostate ductal system, SGP-2 is confined to epithelial cells lining the proximal segment in which signs of programmed cell death are apparent, whereas epithelial cells lining intermediate and distal ducts do not contain measurable SGP-2 [8]. Similarly, SGP-2 expression by different organs of the reproductive tract following castration is not uniform. In the prostate and seminal vesicles, SGP-2 message and protein levels increase dramatically upon castration [4, 9,12,13]. In contrast, levels in the coagulating gland and caput epididymidis decline [4, 9]. To derive inferences about the functional role of SGP-2, in the present study we examined the cellular distribution of this protein in various male reproductive organs under normal conditions and after androgen ablation. In addition, we compared variation in molecular forms of SGP-2 in tes-

Accepted April 3, 1993. Received August 18, 1992. 'This work was supported by NIH grants HD-28048, HD10808, HD13541, DK39250, and DK-43541, the Edwin and Lucy Kretschmer Fund of Northwestern University Medical School and the William O. Jeffery III fellowship for Prostate Cancer Research and a grant from the American Foundation for Urologic Disease to JAS. 2Correspondence: Dr. Julia Sensibar, Department of Urology, Tarry 11-714, Northwestern University Medical School, 303 East Chicago Avenue, Chicago, IL 606113008. FAX: (312) 908-7275.

233

234

SENSIBAR ET AL.

tis, epididymis, seminal vesicle, and ventral prostate to determine whether it is possible that different tissues have different patterns of processing of this protein. MATERIALS AND METHODS Animals Male Sprague-Dawley rats weighing 275-300 g were purchased from Harlan Industries (Cumberland, IN). Rats were housed five per cage in an air-conditioned room with lights-on between 0700 and 1900 h. NIH 07 mouse/rat diet (Harlan Teklad, Bartonville, IL) and water were provided ad libitum. All animal procedures were performed according to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health. Rats were castrated via the scrotal route under metofane anesthesia. To examine the effects of androgen withdrawal on the epididymis, we modified our normal orchiectomy procedure, in which the epididymis and associated fat pads are removed together with the testis. Testes were removed through the tunica albuginea, which was widely incised to minimize risk of damage to epididymal blood supply. The tip of the initial segment of the epididymis along with its blood vessels was ligated before the testis was removed. Ventral prostate ductal systems were obtained by perfusion-fixation of urogenital tissues followed by microdissection as previously described [8,14]. All distal and intermediate ducts belonging to a common proximal duct were carefully removed and considered as a single ductal system. At the indicated times after castration, rats were killed by decapitation and the various tissues were removed for immunohistochemical processing or preparation for two-dimensional gel electrophoresis. Antiserum to SGP-2 (3) Specific antiserum to SGP-2 was a gift of Drs. Steven Sylvester and Michael Griswold. Purified SGP-2 from Sertoli cell-conditioned medium was used as the antigen source. A rabbit immunoglobulin fraction antiserum was used in the present study at a concentration of 25 g/ml for immunohistochemistry and 12.5 pLg/ml for Western blotting. Immunohistochemical Staining Tissues were fixed in neutral buffered formalin for 1224 h and then processed and embedded in paraffin blocks. Five-micrometer sections were placed onto poly-L-lysinecoated slides, deparaffinized in Hemo-De (Fisher Scientific Co., Pittsburgh, PA), and rehydrated through graded alcohols to PBS, pH 7.2. Endogenous peroxidase activity was quenched by incubation in 0.3% hydrogen peroxide for 10 min. Normal goat serum was used to block nonspecific binding sites in the sections. Incubation with SGP-2-specific antiserum was conducted in a humidified chamber at 4°C for 18 h with a 1:200 dilution (25 pLg/ml) of an IgG fraction

antiserum to SGP-2 [1]. Antigen was visualized by subsequent incubations with a 1:2000 dilution of goat anti-rabbit immunoglobulin (Vector Laboratories, Burlingame, CA), avidin-biotin-horseradish peroxidase complex (ABC Kit, Vector Laboratories, Burlingame, CA), and diaminobenzidine tetrahydrochloride before counterstaining with Gills hematoxylin. Negative control sections were processed in an identical manner with normal rabbit IgG substituted for primary antibody. In all negative control sections, no color reaction was observed. Photomicrographs were taken through an Olympus OM2 microscope (Olympus Camera Corp., Woodbury, NY) using Ektachrome 50 film for color slides. Color prints were developed and kept as the permanent record. Preparationof Tissue Extracts and Fluidsfor Electrophoresis Frozen tissues were thawed on ice and homogenized in three volumes of PBS via a Duall glass homogenizer. Aliquots of the tissue homogenates or seminal vesicle fluid were diluted 1:3 in urea solubilization mix (9 M urea, 4% Nonidet P-40, 2% LKB ampholyte range 9-11 [LKB, Rockville, MD], and 2% -mercaptoethanol) and incubated at room temperature for 2 h before centrifugation at 100 000 x g. The resulting supernatant was used for two-dimensional gel electrophoresis. Two-Dimensional Gel Electrophoresis The ISO-DALT system of Anderson and Anderson [15, 16] was used in the present study. First dimension isoelectric focussing was conducted in 5% acrylamide tube gels with 9 M urea and a mixture of ampholytes ranging from pH 3 to pH 10. Catholyte was 0.02 N sodium hydroxide, and anolyte was 0.085% phosphoric acid. Prefocussing was carried out for 200 volt hours; isoelectric focussing was conducted at 800 volts for 14 000 volt hours. After focussing, gels were extruded and equilibrated in 138 mM Tris base, 10% glycerol, 2% SDS, and bromophenol blue, pH 6.8. The second dimension consisted of slab gels with a linear gradient of acrylamide (9-18%) containing 1% SDS. The electrophoresis buffer contained 24 mM Tris base, 0.2 M glycine, and 1% SDS. Electrophoresis was conducted for 18 h at 180 volts. After electrophoresis, gels were fixed in 2.5% sulfosalicylic acid, 5% acetic acid in 20% ethanol; silver staining was then used to detect total protein [17]. Two-Dimensional Western Blot Proteins in unfixed two-dimensional gels were transferred electrophoretically to 0.2-micrometer nitrocellulose at 100 volts for 18 h through use of a transfer buffer consisting of 25 mM Tris, 192 mM glycine in 20% methanol, pH 8.3. The transferred nitrocellulose sheet was then incubated with a blocking solution (5% nonfat dry milk in PBS containing 0.001% thimerosal) for 2 h. Primary anti-

235

SGP-2 IN REPRODUCTIVE TRACT

FIG. 1. Immunohistochemical localization of SGP-2 in rat ventral prostate ductal system of intact and castrated rats. A) SGP-2 immunostaining in the intermediate region of the prostate ductal system from an intact rat. Epithelial cells do not contain measurable SGP-2. B) SGP-2 immunostaining in the proximal duct of the ventral prostate from an intact rat. Epithelial cells are positively stained for SGP-2 (arrows). C) SGP-2 immunostaining in the ventral prostate of a 3-day castrated rat. All epithelial cells in the entire ductal system demonstrate staining for SGP-2. Staining is confined to the apical region of epithelial cells. Positive staining is also noted in the prostatic lumens. A-C, x129.

body (1:400 dilution of an IgG fraction antiserum against SGP-2) [1] diluted in blocking buffer was applied for overnight incubation. The nitrocellulose sheets were then washed with blocking buffer and incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG for 4 h. Immunoreactive proteins were detected by incubation in substrate solution containing 3 mg 4-chloro-1 naphthol/ml methanol and 30 Il hydrogen peroxide in 50 ml Tween/Tris-buffered saline (0.05% Tween-20, 50 mM Tris, 200 mM NaCl, pH 7.5). Glycoprotein Staining and Lectin Blotting Proteins in unfixed two-dimensional gels were transferred to nitrocellulose under conditions identical to those reported for Western blots. Total glycoprotein detection was conducted according to the manufacturer's directions for the glycan detection kit (Boehringer-Mannheim Biochemicals, Indianapolis, IN). Briefly, the transferred nitrocellulose sheets were first washed with PBS, pH 6.5. Adjacent hydroxyl groups in sugars of carbohydrate present on glycoproteins were then oxidized by incubation in 10 mM sodium metaperiodate in 0.1 M sodium acetate buffer, pH 5.5 for 30 min. Blots were washed in PBS and then covalently labeled via incubation with digoxigenin-conjugated succinyl-e-aminocaproic acid hydrazide in 0.1 M sodium acetate, pH 5.5 for 1 h. The blots were washed again with Tris-buffered saline (TBS; 50 mM Tris, 200 mM NaCI, pH 6.5) and then blocked with a solution of 0.5% casein overnight. After

the overnight blocking, filters were washed with TBS and incubated with antidigoxigenin-horseradish peroxidase (HRP) conjugate for 1 h. Color development was identical to that described for Western blots. For lectin blotting, the transferred nitrocellulose sheets were first blocked with 0.5% casein in PBS for 2 h. Blots were then washed twice in TBS and once with TBS containing 1 mM MgC12, 1 mM MnC12, 1 mM CaCI2, pH 7.5. Various digoxigenin-conjugated lectins were then incubated with the blots for 1 h. After lectin incubation, blots were washed and then incubated with HRPconjugated antidigoxigenin for 1 h. Color development was identical to that described for Western blots. All chemicals were of reagent quality and were purchased from Sigma Chemical Co. (St. Louis, MO) or from Boehringer-Mannheim Biochemicals (Indianapolis, IN). RESULTS Localization of SGP-2 in the Ventral Prostate, Epididymis, and Seminal Vesicle of Intact and CastratedRats Figure 1 compares the pattern of immunohistochemical staining for SGP-2 in the ventral prostate ductal system of intact and castrated rats. The rat ventral prostate is composed of 7-8 glandular ductal systems, each of which consists of branches and sub-branches originating from a common proximal duct and extending towards the distal tips. According to their relative distance from the urethra, the various duct regions are designated the proximal, inter-

236

SENSIBAR ET AL.

FIG. 2. Immunohistochemical staining of SGP-2 in epididymis of intact and castrated rats. Comparison of immunohistochemical localization of SGP2 in caput (A, D), corpus (B, E), and cauda (C, F) epididymidis of intact (A, B, C) and 3-day castrated rats (D, E, F). Areas with positive staining for SGP-2 are indicated by bright brown color. Luminal spermatozoa are strongly positive for SGP-2. In intact rats, principal cells in caput, corpus, and cauda epididymidis demonstrate weak cytoplasmic staining and strong staining of apical membrane and stereocilia (A, C). Occasional cells were seen to stain throughout their cytoplasm (A, B, C). In castrated rats, SGP-2 staining in the caput epididymidis shows increased intracellular staining of principal cells and appearance of large positively stained vacuoles lining the tubule (D, arrow). SGP-2 staining in corpus epididymidis of castrated rats appears diminished compared to that in corpus epididymidis of intact rats (E, B). Cauda epididymidis of castrated rats shows increased number of cells with intense intracellular staining for SGP-2 (F). A--F, x125.

mediate, and distal segments [8]. In noncastrated rats, SGP2 immunostaining was confined to the epithelial cells lining the proximal ducts of the prostate (Fig. 1B); epithelial cells lining the other regions of the prostate ductal system did not express measurable SGP-2 immunostaining (Fig. 1A). In contrast to the discrete regionalized staining pattern for

SGP-2 in the prostate of intact rats, there was intense immunostaining for SGP-2 in all epithelial cells in the prostate ductal system of 3-day castrated rats (Fig. 1C). These findings are consistent with those from our previous study of SGP-2 immunostaining within the rat ventral prostate ductal system [8].

SGP-2 IN REPRODUCTIVE TRACT

237

FIG. 3. Immunohistochemical staining of SGP-2 in the seminal vesicle of intact and castrated rats. A and B) Pattern of SGP-2 immunostaining in the seminal vesicle of intact rats. The majority of epithelial cells do not contain measurable levels of SGP-2. Some epithelial cells stain intensely positive for SGP-2 (arrows, panel A) and demonstrate a particulate staining pattern localized to the apical region of the cell. Seminal vesicle secretory fluid (panel A) and the smooth muscle cell layer underlying the epithelium are also positive for SGP-2 (panel B). C) SGP-2 immunostaining in seminal vesicle of a 3-day castrated rat. All epithelial cells stain intensely positive for SGP-2. A-C, x129.

Figure 2 shows the results of immunohistochemical localization of SGP-2 in three regions of the epididymis of intact and castrated rats. In the intact epididymis, the staining pattern varied both within different cell types and within regions of the epididymis. In the caput epididymidis, intense staining for SGP-2 was localized to the apical membranes and stereocilia of principal cells. Luminal spermatozoa were also strongly positive. Weak cytoplasmic staining was present in all principal cells, and occasional cells were also found to stain strongly for SGP-2 throughout their cytoplasm (Fig. 2A). In the corpus and cauda epididymidis, the staining pattern for SGP-2 was similar to that observed in the caput epididymidis but was less intense (Fig. 2, B and C). Occasional cells lacking stereocilia stained intensely throughout their cytoplasm (Fig. 2C). After castration, SGP-2 staining changed markedly within the epididymis. The caput epididymidis underwent a decrease in tubule diameter and loss of luminal spermatozoa resulting in a net decrease in luminal SGP-2 staining. Concomitantly there was an increase in the number of principal cells showing strong intracellular staining for SGP-2. Large vacuole-like structures staining for SGP-2 appeared within the tubules (Fig. 2D). SGP-2 staining in the corpus epididymidis of castrated rats was diminished compared to that in epididymis from noncastrated rats. There was less evidence of tubule shrinkage and no evident loss of luminal sperm content. The in-

tensity of SGP-2 staining of principal cell stereocilia was slightly diminished (Fig. 2E). In the cauda epididymidis of castrated rats, luminal SGP-2 staining of spermatozoa was unchanged; however, there was a dramatic increase in the number of cells demonstrating strong intracellular SGP-2 staining throughout their cytoplasm (Fig. 2F). Figure 3 shows the localization of SGP-2 in the seminal vesicle of intact and castrated rats. The seminal vesicle epithelium of intact rats also displayed a marked regionalization of SGP-2 immunostaining. While the majority of epithelial cells did not contain immunoreactive SGP-2, an apparently random pattern of intensely positive epithelial cells was apparent (Fig. 3A). The staining appeared particulate and was localized to the apical aspect of the epithelial cells. The seminal vesicle secretory material present in the lumen was also strongly positive for SGP-2. The smooth muscle cell layer supporting the seminal vesicle epithelium also showed strong positive staining that appeared to be confined to the cell membrane of the smooth muscle cells (Fig. 3B). SGP-2 expression in the seminal vesicle epithelium of 3-day castrated rats was similar to that observed in the prostate. All epithelial cells in the seminal vesicle demonstrated intense positive staining for SGP-2 (Fig. 3C). SGP2 staining in the smooth muscle underlying the epithelium appeared diminished compared to that in intact rat seminal vesicle.

238

SENSIBAR ET AL.

FIG. 4. Molecular forms of SGP-2 present in ventral prostate (panel A), seminal vesicle (panel B), testis (panel C), and epididymis (panel D). Proteins were separated by two-dimensional gel electrophoresis followed by Western blot transfer and detection with SGP-2 antiserum. Precursor (P), alpha subunit (A), and beta subunit (B)forms of SGP2. Minor SGP-2 precursor isoforms were present in testis and epididymis that were absent from prostate and seminal vesicle (bold arrows).

Comparison of Molecular Forms of SGP-2 in Prostate, Seminal Vesicle, Epididymis, and Testis The molecular forms of SGP-2 present in ventral prostate, seminal vesicle, testis, and epididymis are shown in Figure 4, A-D, respectively. All tissues contain immunoreactive species corresponding to precursor, alpha, and beta subunits of SGP-2. In all tissues, multiple immunoreactive species with differing isoelectric points and similar molecular masses were observed. This pattern is consistent with the glycosylation of precursor, alpha, and beta subunit forms of SGP-2. Relative molecular masses and isoelectric points of SGP-2 forms present in reproductive tract tissues were determined by reference to a standard overlay of molecular mass and isoelectric point markers prepared by co-electrophoresis of sample together with molecular mass and creatinine kinase isoelectric point charge train standards [17] (Fig. 6). Comparison of major molecular forms of SGP-2 present in the various tissues was conducted by co-electrophoresis followed by Western blotting with SGP-2 antiserum. The tissue-specific forms of SGP-2 could be readily

identified by reference to a companion Western blot of each tissue of the combination run alone. As shown in Figure 5A, co-electrophoresis of seminal vesicle fluid and prostatic proteins followed by Western blotting with SGP-2 antiserum demonstrated that the subunits of SGP-2 present in ventral prostate and seminal vesicle were superimposable (3 isoforms, molecular mass 38 kDa, pI 5.1-5.3 [alpha subunit]; and 2 isoforms, 33 kDa, pI 6.7-7.2 [beta subunit]). However, the precursor form of SGP-2 differed between seminal vesicle and prostate. The seminal vesicle SGP-2 precursor (arrowhead) existed as a higher-molecular-mass variant (3 isoforms, 70 kDa, pI 5.8-6.0) compared to prostatic SGP-2 precursor proteins, which showed a small amount of the higher-molecular-mass variant (3 isoforms) in addition to the lower-molecular variant (arrow; 3 isoforms, 68 kDa, pI 5.8-6.0). Co-electrophoresis of ventral prostate and epididymis proteins followed by Western blotting with SGP2 antiserum demonstrated that distinct alpha and beta subunit forms were associated with prostate (arrowheads) and epididymis (arrows). The major isoforms of SGP-2 precur-

SGP-2 IN REPRODUCTIVE TRACT

239

FIG. 5. Co-electrophoresis of molecular forms of SGP-2. Protein combinations were separated by two-dimensional gel electrophoresis followed by Western blot transfer and detection with SGP-2 antiserum. A) Co-electrophoresis and Western blot of seminal vesicle fluid from an intact rat and prostatic proteins from a 3-day castrated rat. Distinct precursor forms of SGP-2 were present in seminal vesicle (arrowhead, the 70-kDa series) and ventral prostate (arrow, the 68-kDa series). Alpha and beta subunits of SGP-2 in these two tissues migrated to identical positions. B) Co-electrophoresis and Western blot of epididymal proteins from an intact rat and prostatic proteins from a 3-day castrated rat. Distinct alpha and beta subunit forms of SGP-2 were present in epididymis (arrowheads) and ventral prostate (arrows). This panel indicates that the precursor proteins in the ventral prostate contain both the 68-kDa series and the 70-kDa series. C) Co-electrophoresis and Western blot of testicular and epididymal proteins from an intact rat. Distinct alpha and beta subunit forms of SGP-2 were present in testis (arrowheads) and epididymis (arrows). This panel indicates that the precursor proteins in the epididymis and the testis are similar and that they contain only the 68-kDa series.

KC

sor present in prostate and epididymis also differed. The epididymal SGP-2 precursor existed as a lower-molecularmass variant (3 isoforms, 68 kDa, pI 5.8-6.0), while the prostatic forms of SGP-2 existed as a combination of the forms present in epididymis and seminal vesicle (3 isoforms, 70 kDa, pI 5.8-6.0; and 3 isoforms, 68 kDa, pI 5.86.0). Additional minor precursor isoforms were present in testis and epididymis (Fig. 4D, bold arrows) that were absent from prostate and seminal vesicle preparations. These isoforms ranged in molecular mass from 62 to 70 kDa and probably represent microheterogeneity of SGP-2 precursor. Co-electrophoresis of testicular and epididymal proteins (Fig. 5C) demonstrated that SGP-2 precursor forms were superimposable between testis and epididymis. In contrast, SGP-2 subunits in testis and epididymis differed. Testicular subunits (arrowheads) were more acidic and of higher molecular mass (multiple isoforms, 46 kDa, pI 3.8-4.3 [alpha subunit]; and 10 isoforms, 33 kDa, p 4.6-5.8 [beta subunit]) than the corresponding subunits in the epididymis (arrows) (10 isoforms, 38 kDa, pI 4.5-5.1 [alpha subunit]; and 8 isoforms, 30 kDa, pI 5.3-6.5 [beta subunit]). ISOELECTRIC

POINT

FIG. 6. Schematic presentation of the relative two-dimensional Western blot locations of major molecular forms of SGP-2 in the male reproductive tract. Precursor (Pre), alpha, and beta forms of SGP-2 subunits present in ventral prostate (Pa, P: solid symbols), testis (Ta, T: solid symbols), and epididymis (Ea, E: open symbols). Seminal vesicle alpha and beta subunits co-migrated to positions identical to those of prostatic alpha and beta subunits (not shown).

Glycosylation of SGP-2 in the Rat Ventral Prostate

SGP-2 in the testis and epididymis is known to be a glycoprotein. To determine whether the prostatic form of SGP2 is also glycosylated, glycoprotein staining of proteins present in the ventral prostate of a 3-day castrated rat was performed. Precursor, alpha, and beta subunits of SGP-2

SENSIBAR ET AL.

240

FIG. 7. Glycoprotein staining of SGP-2 from the ventral prostate and testis. A) Prostatic proteins from a 3-day castrated rat were separated by twodimensional gel electrophoresis, then transferred to nitrocellulose and stained for total glycoproteins. Glycosylated forms of SGP-2 are indicated by arrows. B) Testicular proteins separated by two-dimensional gel electrophoresis followed by transfer to nitrocellulose and staining for total glycoproteins. Glycosylated forms of SGP-2 are indicated by arrows.

were identified (Fig. 7A, arrows) and found to be glycosylated by comparison to a companion Western blot of an identical sample probed with antiserum to SGP-2. As a comparison, glycoprotein staining was also conducted for tes-

TABLE 1. Lectin reactivities of SGP-2 in epididymis, testis, ventral prostate, and seminal vesicle.* LECTINt GNA

SNA

MAA

DSA

+ +

-

-

-

+++

-

-

+

++

-

-

++ + -

-

-

+ +

-

+ + -

+ +++ +

PNA

Epididymis precusor a subunit p3 subunit

Testis precusor a subunit /3 subunit

D3VP precusor a subunit p subunit

SVF precusor a subunit t subunit

-

-

++

*Tissue extracts (epididymis, testis, and ventral prostate [D3VP] of 3-day castrated rat) and seminal vesicle fluid (SVF) were separated by two-dimensional gel electrophoresis and transferred to nitrocellulose for reaction with lectins. tGNA, Galanthus nivalis agglutinin; SNA, Sambucus nigra agglutinin; MAA, Maackia amurensis agglutinin; DSA, Datura stramonium agglutinin; PNA, Peanut agglutinin.

ticular proteins. Testicular forms of SGP-2 were identified (Fig. 7B, arrows) and also confirmed to be glycosylated by comparison to a companion Western blot of rat testis probed with antiserum to SGP-2. Comparison of Lectin Reactivities of SGP-2 in Prostate, Seminal Vesicle, Epididymis, and Testis Lectin reactive forms of SGP-2 were identified on lectin blots by comparison to a companion Western blot of an identical sample probed with SGP-2 antiserum. A summary of the lectin reactivities of SGP-2 molecular forms in ventral prostate, seminal vesicle, epididymis, and testis is shown in Table 1. Lectin reactivities were found to differ between the molecular forms of SGP-2 within a tissue and between the same molecular forms of SGP-2 present in different tissues. Epididymal precursor and alpha subunit, prostate precursor and alpha subunit, and testicular alpha subunit forms of SGP-2 were bound by the lectin Galanthus nivalis agglutinin (GNA). The specificity of this lectin is toward terminal mannose residues linked a (1-3), u- (1-6), or t (12) to mannose [18]. These terminal linkages are unique to the high-mannose type of carbohydrate chain. In contrast, all forms of SGP-2 in the seminal vesicle fluid failed to react with GNA. The lectin Sambucus nigra agglutinin (SNA) recognizes terminal sialic acid residues linked xt(2-6) to galactose or N-acetyl glucosamine [19]. Such linkages are common to complex, sialylated carbohydrate chains. Epididymal and testicular forms of SGP-2 alpha subunit reacted

SGP-2 IN REPRODUCTIVE TRACT strongly with this lectin, suggesting that these forms of SGP2 have terminal sialic acid residues. The seminal vesicle and prostatic forms of SGP-2 were bound preferentially by Datura stramonium agglutinin (DSA). This lectin recognizes terminal galactose 1 (1-4) N-acetylglucosamine linkages in complex and hybrid chains that lack terminal sialic acid residues [20]. The strong reactivity toward DSA in contrast to the lack of reaction (prostate) or weak reactivity (seminal vesicle fluid) with lectins MAA and SNA, which identify terminal sialylated carbohydrate chains, suggests that prostatic and seminal vesicle forms of SGP-2 are largely nonsialylated [19,21]. This is consistent with the more basic isoelectric points for prostatic and seminal vesicle SGP-2 subunits as compared to the more acidic, sialylated forms of SGP-2 subunit present in testis and epididymis (Fig. 6). No reaction of SGP-2 molecular forms with peanut agglutinin (PHA) was observed. This lectin recognizes the core disaccharide galactose 13(1-3)-N-acetylgalactosamine typical of O-linked glycoproteins [22]. This result was as expected, as SGP-2 is known to be an N-linked glycoprotein [23]. DISCUSSION In the ventral prostate, increased SGP-2 protein and message expression has been associated with the programmed cell death of androgen-dependent epithelial cells following castration-induced regression [3,9, 24]. The low level of expression of SGP-2 by the ventral prostate of intact rats has been attributed to a discrete population of epithelial cells that line the proximal ducts and are undergoing programmed cell death to maintain organ homeostasis [8,25]. Epithelial cells lining the distal and intermediate ducts of the prostate of intact rats do not contain measurable levels of SGP-2 [3, 8]. In contrast to this, the significantly higher levels of SGP-2 expression and secretion by the testis, epididymis, and seminal vesicles of intact rats have been attributed to a normal physiological role for SGP-2 distinct from that associated with cell death. Previous studies documenting the expression of SGP-2 by several androgen-sensitive tissues during castration-induced regression have noted differential responses of SGP2 expression to androgen ablation. In the ventral prostate and seminal vesicles, SGP-2 expression increases dramatically upon castration [4, 9, 12, 13]. In the present study, our comparison of the cellular expression of SGP-2 in ventral prostate, seminal vesicle, and epididymis under conditions of normal androgen and androgen depletion has allowed us to investigate the cellular basis for these differences. In intact rats, SGP-2 expression in epithelial cells of the ventral prostate, seminal vesicle, and epididymis demonstrated marked regionalization; these findings provide further evidence that cellular SGP-2 expression is not directly linked to circulating androgen levels [3, 8]. The observation of significant SGP-2 staining by the smooth muscle cells underlying the seminal vesicle of intact rats suggests a normal

241

physiological role for SGP-2 in smooth muscle cells. SGP2 expression by rat and porcine smooth muscle cells in vitro [26, 27] and atrial myocytes [28] has been recently reported. The significance of these findings is not understood at present. Upon castration, the staining pattern for SGP-2 changed dramatically. The increased number of cells demonstrating positive intracellular SGP-2 staining in the prostate and seminal vesicles is consistent with the previously reported increases in SGP-2 message levels. In the caput epididymidis, we observed a loss of luminal spermatozoa and reductions in tubule diameter and principal cell height consistent with the preferential androgen sensitivity of this segment [29]. The loss of luminal spermatozoa resulted in the marked reduction of luminal SGP-2 staining. In contrast, intracellular SGP-2 staining of principal cells increased upon castration. The combination of a loss of SGP2 staining in the luminal spermatozoa and an increase in intracellular staining in principal cells could account for an overall decrease in total SGP-2 content in the caput epididymis of castrated rats as reported by earlier investigators using a radioimmunoassay [9]. SGP-2 staining became associated with large rounded vacuoles. Morphologically, these SGP-2-positive, intracellular vacuoles fit the typical description of autophagic vacuoles and/or secondary lysosomes seen to form in other tissues undergoing castration-induced regression [30,31]. Increased autophagic and heterophagic vacuole formation by caput principal cells in response to orchiectomy has been reported, supporting our concept that the SGP-2-positive vacuoles could represent autophagic vacuoles and/or secondary lysosomes [29]. In the cauda epididymidis, castration resulted in an increasing number of cells demonstrating intense intracellular staining for SGP-2. These cells fit the characteristic description of the clear cells predominant in the cauda epididymidis [29, 32]. These cells are also rich with lysosome inclusions. Our two-dimensional electrophoretic and lectin-binding studies of SGP-2 in the male reproductive tract have demonstrated that there is a high degree of variation of molecular forms of SGP-2 among prostate, seminal vesicle, epididymis, and testis. The lectin reactivities are consistent with the presence of complex-type high-mannose, sialyated carbohydrate chains on testicular and epididymal forms of SGP2. Alternative means of detecting sialic acid residues have also supported this finding [23]. Prostate and seminal vesicle forms of SGP-2 appear to be either largely nonsialyated (prostate) or lacking in terminal sialic acid residues (seminal vesicle). Seminal vesicle SGP-2 also appears to differ from that of prostatic, testicular, and epididymal forms of SGP-2 in that it was unreactive with a lectin specific for terminal mannose-mannose linkages. In conclusion, results of the present study indicate that, as observed in the regressing ventral prostate, SGP-2 may also be associated with programmed cell death in the regressing epididymis and seminal vesicle. SGP-2 may have two distinct biological functions: a secretory, extracellular

242

SENSIBAR ET AL.

function associated with epididymal sperm maturation [10] and seminal vesicle physiology and an intracellular function associated with programmed cell death. In addition, the differences in molecular forms and lectin reactivities for SGP-2 forms from prostate, seminal vesicle, epididymis, and testis may also be linked to differences in functional activities and/or half-lives of SGP-2 molecular forms. REFERENCES 1. Sylvester SR, Skinner MK, Griswold MD. A sulphated glycoprotein synthesized by Sertoli cells and by epididymal cells is a component of the sperm membrane. Biol Reprod 1984; 31:1087-1101. 2. Cheng CY,Chen CL, Feng ZM, Marshall A, Bardin CW. Rat clusterin isolated from primary Sertoli cell-enriched culture medium is sulfated glycoprotein-2 (SGP-2). Biochem Biophys Res Commun 1988; 155:398-404. 3. Buttyan R, Olsson CA, Pintar J, Chang C, Bandky M, Ng P-Y, Sawczuk 1. Induction of the TRPM-2 gene in cells undergoing programmed death. Mol Cell Biol 1989; 9:3473-3481. 4. Bettuzzi S, Hiipakka RA, Gilna P, Liao S. Identification of an androgen-repressed mRNA in rat ventral prostate as coding for sulphated glycoprotein 2 by cDNA cloning and sequence analysis. Biochem J 1989; 257:293-296. 5. Kirsbaum L, Sharpe JA, Classon B, Hudson P, Walker ID. Molecular cloning and characterization of the novel, human complement-associated protein, SP-40,40: a link between the complement and reproductive systems. EMBOJ 1989; 8:711718. 6. Jenne DE, Tschopp J. Molecular structure and functional characterization of a human complement cytolysis inhibitor found in blood and seminal plasma: identity to sulfated glycoprotein 2, a constituent of rat testis fluid. Proc Nat Acad Sci USA 1989; 86:7123-7127. 7. de Silva HV, HarmonyJAK, Stuart WD, Gil CW, Robbins J. Apolipoprotein J: structure and tissue distribution. Biochemistry 1990; 29:5380-5389. 8. Sensibar JA, Griswold MD, Sylvester SR, Buttyan R, Bardin CW, Cheng CY, Dudek S, Lee C. Prostatic ductal system in rats: regional variation in localization of an androgen repressed gene product, sulphated glycoprotein 2. Endocrinology 1991; 128:2091-2102. 9. Grima J, Zwain I, Lockshin RA, Bardin CW, Cheng CY. Diverse secretory patterns of clusterin by epididymis and prostate/seminal vesicles undergoing cell regression after orchiectomy. Endocrinology 1990; 126:2989-2997. 10. Hermo L, Wright J, Oko R, Morales CR. Role of epithelial cells of the male excurrent duct system of the rat in the endocytosis or secretion of sulfated glycoprotein-2 (clusterin). Biol Reprod 1991; 44:1113-1131. 11. Sylvester SR, Morales C, Oko R, Griswold MD. Localization of sulfated glycoprotein-2 (clusterin) on spermatozoa and in the reproductive tract of the male rat. Biol Reprod 1991; 45:195-207. 12. Montpetit ML, Lawless KR, Tenniswood M. Androgen repressed messages in the rat ventral prostate. Prostate 1986; 8:25-36. 13. Lee C, Sensibar JA Proteins of the rat prostate. I. Synthesis of new proteins in the ventral lobe during castration-induced regression. J Urol 1987; 138:903-908. 14. Lee C, Sensibar JA, Dudek SM, Hiipakka RA, Liao S. Prostatic ductal system in rats: regional variation in morphological and functional activities. Biol Reprod 1991; 43:1079-1086.

15. Anderson NG, Anderson NL. Analytical techniques for cell fractions. XXI. Twodimensional analysis of serum and tissue proteins: multiple isoelectric focussing. Anal Biochem 1978; 85:331-340. 16. Anderson NL, Anderson NG. Analytical techniques for cell fractions. XXII. Twodimensional analysis of serum and tissue proteins: multiple gradient-slab electrophoresis. Anal Biochem 1978; 85:341-354. 17. Harrison HH, Miller KL, Dickinson C, Daufeldt JA Quality assurance and reproducibility of high-resolution two-dimensional electrophoresis and silver staining in polyacrylamide gels. Am J Clin Pathol 1992; 97:97-105. 18. Anderson NL, Hickman BJ. Analytical techniques for cell fractions. XXIV. Isoelectric point standards for two-dimensional gel electrophoresis. Anal Biochem 1979; 93:312-320. 19. Shibuya N, Goldstein IJ, Van Damme EJM, Peumans WJ. Binding properties of a mannose-specific lectin from the snowdrop (Galanthus nialis) bulb. J Biol Chem 1988; 262:728-734. 20. Shibuya N, Goldstein IJ, Broekaert WF, Nsimba-Lubaki M, Peeters B, Peumans WJ. The elderberry (Sambucus nigra L) bark lectin recognizes the NeuSAc(a 2-6)Gal/ GalNac sequence. J Biol Chem 1987; 262:1596-1601. 21. Crowley JF, Goldstein J, Arnarp J, Lonngren J. Carbohydrate binding studies on the lectin from Datura stramonium seeds Arch Biochem Biophys 1984; 231:524533. 22. Wang WC, Cummings RD. The immobilized leukoagglutinin from the seeds of Maackia amurensis binds with high affinity to complex-type asn-linked oligosaccharides containing terminal sialic acid linked t 2,3 to penultimate galactose residues. J Biol Chem 1988; 263:4576-4585. 23. Goldstein IJ,Hayes CE. The lectins: carbohydrate binding proteins of plants and animals. Adv Carbohydr Chem Biochem 1978; 35:127-340. 24. Griswold MD, Roberts K,Bishop P. Purification and characterization of a sulfated glycoprotein secreted by sertoli cells. Biochemistry 1986; 25:7265-7270. 25. Leger JG, Montpetit ML, Tenniswood MP. Characterization and cloning of androgen-repressed mRNAs from rat ventral prostate. Biochem Biophys Res Commun 1987; 147:196-203. 26. Thomas-Salgar SA, Millis AJT. Expression of a glycoprotein (clusterin) in differentiating cultures of porcine smooth muscle cells. In: Program of the 32nd annual meeting of the American Society for Cell Biology; 1992; Denver, CO. Abstract 401. 27. Messmer S, James RW. Co-expression of apolipoprotein E and clusterin (ApoJ) In: Proceedings of the First International Workshop on Clusterin; 1992; Pembrooke College, Cambridge, England. Abstract 16. 28. Harmony JAK, Witte DP, Stuart WD, Aronow BJ, Grupp I, Brown TL, Lund SD. The expression of apolipoprotein J (APOJ) in the mammalian heart. In: Proceedings of the First International Workshop on Clusterin; 1992: Pembrooke College, Cambridge, England. Abstract 7. 29. Moore HDM, Bedford JM. Short term effects of androgen withdrawal on the structure of different epithelial cells in the rat epididymis. Anat Rec 1979; 193:293312. 30. Sensibar JA, Liu X, Patai B, Alger B, Lee C. Characterization of castration-induced cell death in the rat prostate by immunohistochemical localization of cathepsin D. Prostate 1990; 16:263-276. 31. Wilson MJ, Whitaker JN, Sinha AA Immunocytochemical localization of cathepsin D in rat ventral prostate: evidence for castration-induced expression of cathepsin D in basal cells. Anat Rec 1991; 229:321-333. 32. Sun EL, Flickinger CJ. Development of cell types and of regional differences in the postnatal rat epididymis. Am J Anat 1979; 154:27-55.