- Email: [email protected]
Gelatinolytic And Caseinolytic Proteinase Activities In Human Prostatic Secretions
0022-534 7/93/1493-0653$03.00/0 THE JOURNAL OF UROLOGY Copyright© 1993 by AMERICAN UROLOGICAL ASSOCIATION, INC.
Vol. 149, 653-658, March 1993
Printed in U.S.A.
GELATINOLYTIC AND CASEINOLYTIC PROTEINASE ACTIVITIES IN HUMAN PROSTATIC SECRETIONS MICHAEL J. WILSON*, HOLLY NORRIS, DEEPAK KAPOOR, MILDRED WOODSON, CATHERINE LIMAS AND AKHOURI A. SINHA From the VA Medical Center and Departments of Laboratory Medicine and Pathology, Urologic Surgery and Genetics and Cell Biology, University of Minnesota, Minneapolis, Minnesota
ABSTRACT
Seminal fluid contains a number of proteinase activities, many of which are secreted by the prostate gland. Our objective was to determine proteinase activities in human prostatic secretions which can degrade gelatin and/or casein. Prostatic secretions were collected by prostate massage from men with benign prostatic hyperplasia prior to surgery to relieve obstruction. Significant proteinase activities towards gelatin of about 81, 86, 94, 111, 115 and 163 Kd as well as less active forms of 23, 36, 38, 132, 137, and 148 Kd were detected using protein substrate-polyacrylamide gel zymography. In addition, Ca2+ stimulated activities of approximately 64, 66, 71 and 76 Kd; however, EDTA and EGTA inhibited all activities but the 23, 36 and 38 Kd forms (these were inhibited by benzamidine and epsilon-amino caproic acid). This suggests that the gelatinolytic activities of 64 Kd and greater were metalloproteinases and those of 23, 36, and 38 Kd were serine proteinases. Significant caseinolytic activities of 22, 25, 35, 37, 57, 90, 96, 102 and 116 Kd were found as well as several less active forms and a 12 Kd activity stimulated by Ca2+. Caseinolytic activities of 12, 14, 16, 96, 102, 116, and 126 Kd were inhibited by EDTA and EGTA indicating they are metalloproteinases. The 35, 37, 57 and 58 Kd caseinolytic activities were inhibited by benzamidine, and the 57 and 58 Kd forms by epsilon-aminocaproic acid suggesting they were serine proteinases. There was considerable variability among individuals in the molecular forms of proteinase activity expressed as well as the level of their activity. A significant decrease in the frequency of expression of the 132 Kd gelatinolytic activity was found in secretions from men with atypia or adenocarcinoma, as compared with men with benign prostatic hyperplasia alone. Our results show that human prostatic secretion contains a variety of proteinase activities. The expression of the 132 Kd gelatinolytic activity could prove useful in further evaluation of neoplastic prostatic disease. KEY WORDS: metalloproteinases, serine proteinases, prostate
The secretion of the human prostate gland, as obtained by rectal prostate massage, contains about 2% protein. 1 However, the identity and function of individual proteins, including a wide variety of enzymes, in this complex mixture remains mostly unknown. Among the enzymes present in prostatic secretion are a number of proteolytic activities which imbue seminal plasma with one of its most characteristic attributes.1 Proteinases that have been described in seminal plasma include plasminogen activator, pepsinogen II, Pz Peptidase, seminin/ prostatic specific antigen (PSA), and gelatinases. 2 ' 3 An important function of proteolytic enzymes in semen is the liquefaction of the postejaculatory seminal coagulum and the subsequent degradation of seminal proteins to proteases, peptides, and amino acids. 1 The proteinase that appears to have a central role in semen liquefaction is the neutral enzyme, seminin/PSA, which can cleave the major basic proteins of seminal vesicle secretion that are responsible for forming the coagulum. 2 • 4 ' 5 The measurement of serum PSA levels has also become an important tool in the diagnosis and evaluation of progression of prostatic cancer.6-8 Since the role of most prostatic secretory proteinases in semen remains unknown and in view of the possible use of secretory proteins as markers of prostatic function and/or pathology, we undertook a study of gelatinolytic and caseinolytic proteinase activities in human prostatic secretions. Accepted for publication September 4, 1992. • Requests for reprints: VA Medical Center, One Veterans Drive, Minneapolis, Minnesota 55417.
653
MATERIALS AND METHODS
Prostatic secretions were collected by prostate massage from 57 men who were about to undergo a transurethral resection of the prostate to relieve symptomatic bladder outlet obstruction. The prostate massage was done after obtaining an informed consent under protocols approved by the Human Studies ComCasein
Gelatin Kd -162 =-137 -132 -115 -111 -94 -88 -a1
Kd
10680-
Kd
Kd
=--132 -111 -94 -86 -81 -76 _71 -66 -54
49.5-
Kd
Kd
27.5-
:'-126 =102 =- 80 -60 -57 _37 -35 -27 =-22
18.5-
-16
1068049.532.5-
Kd
Kd =-145 =--116 a,-102 =--so -60 -57 _37 -35
-
-25 -22 -16 =12
32.5-
-CaCl 2
+CaCl2
-CaCl2
+CaCl2
FIG. 1. Multiple forms of gelatinolytic and caseinolytic proteinase activities in human prostatic secretion. Prostatic secretions (0.5 µl. per lane) of two men with BPH were electrophoresed in gelatin-containing and casein-containing polyacrylamide gels and prepared as zymograms. Molecular mass of protein standards electrophoresed in gel are given in column to left of gel and approximate molecular mass of proteolytic activities detected are given in column to right of each gel. Samples in this figure were chosen to depict most (but do not show all) molecular forms of proteinase activities detected in human prostatic secretions. Note that 64, 66, 71, and 76 Kd forms of gelatinolytic activity and 12 Kd form of caseinolytic activity are stimulated by addition of 5 mM. CaC12 to incubation buffer.
654
Kd
WILSON AND ASSOCIATES
None
EGTA
Benz.
Kd
None
EGTA
Benzam.
10680-
10680-
49.549.532.527.5-
32.527.5-
18.5-
10680-
1068049.5-
49.5-
+CaCl2
+CaCl2
32.527.5-
FIG. 2. Effect of EG TA and benzamidine on gelatinolytic proteinase activities of human prostatic secretions. Replicate aliquots (0.5 µI. per lane) of two samples of prostatic secretions from men with BPH were electrophoresed into gelatin-containing polyacrylamide gels. These gels were sliced into pieces that were incubated with 50 mM. Tris-HCl buffer (pH 8.4) alone, buffer + 5 mM. EGTA, buffer + 10 mM. benzamidine, buffer + 5 mM. CaCl 2 , buffer + 5 mM. CaC12 + 5 mM. EGTA, and buffer + 5 mM. CaC12 + 10 mM. benzamidine. EGTA inhibited gelatinolytic proteinase activities of 64 Kd and greater (EDTA inhibited these activities also, data not shown). EGTA did not affect 23, 36, and 38 Kd activities which were inhibited by benzamidine (also by epsilon-aminocaproic acid, data not shown).
mittee of the Minneapolis VA Medical Center. The prostatic fluids were centrifuged at 960 X g for 10 minutes at 4C to remove cells; the supernatant was frozen in powdered dry ice and stored frozen at -20C until zymographic analysis. Prostatic tissues removed in the surgery were submitted for pathologic examination. In some cases prostatitis/inflammation, atypia, or adenocarcinoma were determined to be present. Zymography of proteinases. Aliquots (0.5 µl.) of prostatic secretion were subjected to electrophoresis in duplicate gelatinor casein-containing polyacrylamide (9 and 11 % acrylamide respectively) gels in the presence of sodium dodecyl sulfate under nonreducing conditions. 9 • 10 Each protein substrate was present at 0.1 % final concentration in the gel. The gels (0. 75 mm. thick) were subjected to electrophoresis for about 35 minutes at 200 v. in a BioRad Mini-Protean II system (BioRad, Richmond, California). Following electrophoresis, the gels were rinsed with distilled water and then washed with gentle shaking at room temperature with 2.5% Triton X-100 (2 changes) for 1 hour. The gels were again rinsed with distilled water and one gel was incubated in 50 mM. Tris-HCl (pH 8.4) containing 5
32.527.518.5FIG. 3. Effect ofEGTA and benzamidine on caseinolytic proteinase activities of human prostatic secretions. Replicate aliquots (0.5 µI. per lane) of same samples shown in fig. 2 were electrophoresed into caseincontaining acrylamide gels. These gels were sliced into pieces that were incubated under same conditions as detailed in legend to fig. 2; 12, 14, 16, 96, 102, 116, and 126 Kd caseinolytic proteinase activities were inhibited by EGTA (also by EDTA, data not shown). Benzamidine inhibited 35, 37, 57, and 58 Kd caseinolytic activities.
mM. CaCb (2 changes), whereas the other was incubated with buffer alone. Incubation was overnight (18 to 20 hours) at 37C. Following the incubation they were stained with amido black. Areas of proteolysis appear as clear zones against a blue background. Molecular mass determinations were made in reference to prestained protein standards (BioRad) co-electrophoresed in these gels: phosphorylase B, 106 Kd; bovine serum albumin, 80; ovalbumin, 49.5 Kd; carbonic anhydrase, 32.5 Kd; soy bean trypsin inhibitor, 27.5 Kd; and lysozyme, 18.5 Kd. The effect of pH of the incubation media and selected proteinase inhibitors on the proteinase activities in the proteinsubstrate polyacrylamide gels was examined to determine some of the properties of these enzymes. Replicate aliquots (0.5 µl.) of prostatic secretions were electrophoresed in gelatin or casein containing SDS-polyacrylamide gels. The gels were cut into strips, each of which contained a lane(s) of the replicate sample(s). These strips were washed in Triton X-100 as described above and then incubated with and without 5 mM. CaCb in 50 mM. Tris-HCl buffers of varied pH (6.8, 7.6, 8.0, 8.4, 9.0, 9.2) or with 50 mM. Tris-HCl buffers of pH 8.4 containing one of the following inhibitors: 10 mM. epsilon-aminocaproic acid, 10 mM. tranexamic acid, 5 mM. sodium iodoacetate, 5 mM. nethylmaleimide (NEM), 10 mM. benzamidine, 5 mM. EDTA, or 5 mM. EGT A. The proteinase activities were stable to
PROTEINASES IN HUMAN PROSTATIC SECRETION
IA I 1
BPH 2
3
4
PROST
5
6
7
8
IA I
655
CAP
9 10 11 12 13 14 15
106 -
80-
49.532.5FIG. 4. Gelatinolytic proteinase activities of prostatic secretions from men with BPH, inflammation/prostatitis, atypia, or adenocarcinoma. Aliquots (0.5 µI. per lane) of prostatic secretions from men with BPH, inflammation/prostatitis (Prost), Atypia (Atyp), and adenocarcinoma (CAP) were electrophoresed in gelatin-containing SDS polyacrylamide gels and prepared for zymography. Activities observed within individuals with BPH alone indicate variability in gelatinolytic proteinase activities among men, and this variability is observed across the disease groups.
freezing since some samples were frozen and thawed 6 times and the proteinase pattern was not changed. Other procedures. Protein concentrations of secretions were estimated using bicinchoninic acid (Pierce Chemical Co., Rockford, Illinois) with bovine serum albumin as standard.11 Statistical analysis. The statistical evaluation of the frequency of expression of individual proteinase activities in secretions of men with different conditions of the prostate was done by chi-square analysis. 12 RESULTS
Gelatinolytic and caseinolytic proteinase activities in prostatic secretions of men with benign prostatic hyperplasia (BPH). An examination of proteinase activities in human prostatic secretions using zymography showed multiple activities towards gelatin or casein as the protein substrates (fig. 1). Significant gelatinolytic proteinase activities of 81, 86, 94, 111, 115, and 162 Kd and less active forms of 23, 36, 38, 132, 137, and 148 Kd were detected. The addition of 5 mM. CaCb to the reaction buffer stimulated proteinase activities of approximately 64, 66, 71, and 76 Kd. Significant caseinolytic proteinase activities of 22, 25, 35, 37, 57, 90, 96, 102, and 116 Kd as well as less active forms of 11, 14, 16, 20, 27, 39, 54, 58, 60, 72, 80, 86, 96, 126, 133, 137, 145, and 151 were detected. A 12 Kd caseinolytic activity was stimulated by Ca2+. However, there was considerable variability among men in the expression of both activity and number of molecular forms of the gelatinolytic and caseinolytic proteinases. Enzymic properties of the gelatinolytic and caseinolytic proteinase activities. Some of the enzymic properties of the gelatinolytic and caseinolytic proteinase activities were examined by varying the pH of the reaction buffer or by adding different inhibitors of proteinase activities to the reaction media. Both
the gelatinolytic and caseinolytic proteinases demonstrated activity over a broad range of pH values (7.2-9.2) in the neutral to alkaline domain (data not shown). Optimal proteinase activities were found at a reaction pH of 8.4 and this pH was chosen for the incubation of the protein substrate gels in these studies. The chelators EDTA and EGTA (5 mM.) inhibited all the gelatinolytic proteinase activities except 23, 36, and 38 Kd forms (fig. 2). The greater sensitivity of gelatinolytic activities to chelator inhibition versus the lower number of activities (64, 66, 71, and 76 Kd) stimulated by Ca2+ added to the incubation buffer may be due to the endogenous association of Ca2+ with these proteinases since prostatic secretions are rich in Ca2+. 13, 14 The 23, 36, and 38 Kd activities were inhibited by the serine proteinase inhibitors benzamidine (10 mM.) and epsilon-aminocaproic acid (10 mM.). The 12, 14, 16, 96, 102, 116, and 126 Kd forms of caseinolytic activity were inhibited by 5 mM. EDTA or EGTA (fig. 3), whereas the 35, 37, 57, and 58 Kd forms were inhibited by 10 mM. benzamidine (fig. 3). In addition, the 57 and 58 Kd forms of caseinolytic activity were inhibited by 10 mM. epsilon-aminocaproic acid (data not shown). There was no effect of 5 mM. sodium iodoacetate, 0.2% sodium azide, 5 mM. n-ethylmaleimide, 10 mM. hydrocinnamic acid, and 10 mM. tranexamic acid on either group ofproteinase activities. Considerable variability in the number and relative activity of proteinase bands towards either protein substrate was found for prostatic secretions from men with BPH (figs. 4 and 5). In an attempt to quantitate the number of proteinase activities in prostatic secretion, we examined the zymograms for the relative frequency (the presence or absence) of expression of each molecular form of gelatinolytic and caseinolytic proteinase activity in these specimens (tables 1 and 2). The different molecular forms of caseinolytic and gelatinolytic proteinase
656
Kd
WILSON AND ASSOCIATES
I
CAP 2 3
4
I I ATYP 5 6 7
I
PROST BPH 8 9 10 11 12 13 14
I
57, 60, and 102 Kd activities. The caseinolytic activities of 11, 16, 25, 35, 37, 54, and 58 Kd were detected with a low frequency (<40%). Expression of gelatinolytic and caseinolytic proteinase activities in prostatic secretions from men with different prostatic diseases. We examined the proteolytic activities towards gelatin (fig. 4 and table 1) and casein (fig. 5 and table 2) in prostatic secretions from men with a primary pathologic diagnosis of BPH, prostatitis/inflammation, atypia, and adenocarcinoma. When the frequency of a particular proteinase activity in the secretions of men with BPH alone was compared with that of men with other pathologies, we found a pattern that indicated that: 1) fewer samples from men with atypia or adenocarcinoma showed the 76 and 132 Kd gelatinolytic activity; 2) men with prostatitis/inflammation demonstrated less 132 Kd gelatinolytic activity; 3) men with adenocarcinoma showed less expression of 27, 39, and 60 Kd forms of caseinolytic activities; and 4) men with prostatitis/inflammation showed a decreased frequency of the 14 Kd and increased frequency of the 25 Kd caseinolytic activities. However, only the decreased expression of the 132 Kd gelatinolytic proteinase activity in secretions from men with atypia or cancer was statistically significant (table 1).
1068049.5-
-CaCl2
32.527.518.5-
1068049.532.527.518.5-
DISCUSSION
FIG. 5. Caseinolytic proteinase activities of prostatic secretions from men with BPH, inflammation/prostatitis, atypia, or adenocarcinoma. Aliquots (0.5 µl. per lane) of prostatic secretions from men with BPH, inflammation/prostatitis (Prost), Atypia (Atyp), and adenocarcinoma (CAP) were electrophoresed in casein-containing SDS polyacrylamide gels and prepared for zymography. Activities observed within individuals with BPH alone indicate variability in caseinolytic proteinase activities among men. This variability is observed across disease groups.
activities were evaluated in zymograms in which prostatic secretion samples were electrophore~ed by both constant volume (0.5 µl. per lane) or a constant amount of protein (5 µg. per lane). This was done so that differences in protein content of prostatic secretions would not affect the detection of a particular band of proteinase activity. We interpreted these gels on a qualitative basis and did not attempt to quantitate the activity of individual bands. The reason for this is that there are many proteinase bands and each proteinase band expresses its own enzymic rate of protein substrate hydrolysis. We incubated the gels to maximize the activities of all bands and thus to optimize detection of individual activities. Most of the gelatinolytic proteinase activities were expressed by the majority of men, that is the activities of 64, 71, 81, 86, 111, 115, 132, 148, and 162 Kd were found in more than 70% of secretion samples from patients with BPH. There were, however, quantitative differences in the apparent activity of individual gelatinolytic bands. The 66 and 104 Kd gelatinolytic activities were found in a small number of samples (<40%). Many of the caseinolytic activities were expressed by most of the men with BPH (>70% ); these activities were 12, 22, 72, 80, 86, 90, 96, 116, 126, 137, 145, and 151 Kd. Other activities were found with a moderate frequency (40 to 70% of samples) and included the 14, 27, 39, TABLE
Human prostatic secretion contains multiple gelatinolytic and caseinolytic proteinase activities of varying intensity and molecular size as indicated in protein substrate-SDS-polyacrylamide gel zymography. All of the gelatinolytic activities except the 23, 36, and 38 Kd bands appeared to be metalloproteinases whereas the 23, 36, and 38 Kd activities are attributed to the serine proteinase class. Based on inhibitor studies, some caseinolytic activities could be classified as metallo- or serine proteinases, while most remained unaffected by the agents tested. There are patterns in the distribution of proteinase activities in the zymograms that indicate that some proteolytic activities may be related. For example, whereas most caseinolytic activities were not affected by EDT A or EGT A, those activities grouped as 12, 14, and 16 Kd and 96, 102, 116, and 126 Kd were inhibited by these chelators. Differences in the extent of glycosylation and/or possible proteolytic modification of a proteinase or the presence of proenzyme and active enzyme forms of a proteinase could account for multiple forms of proteinase activities observed in these zymograms. Other examples of proteinase activities in these gels possibly related to a mature proteinase molecule would be the 23, 36, and 38 Kd gelatinolytic activities inhibited by benzamidine and epsilon-aminocaproic acid and the benzamidine-sensitive caseinolytic activities of 35 and 37 and of 57 and 58 Kd. These groupings also suggest that the 36 and 38 Kd gelatinolytic and 35 and 37 Kd caseinolytic activities, which are similar in molecular mass and are sensitive to the same inhibitor, may be the manifestation of the same enzymes. PSA/seminin is a prominent serine proteinase in prostatic secretion which can hydrolyze both casein and gelatin. 15• 16 The gelatinolytic and caseinolytic activities of approximately 35-38
l. The expression of gelatinolytic proteinase activities in prostatic secretions from men with BPH, BPH
+ prostatitis/inflammation, BPH
+ atypia, and BPH + adenocarcinoma Percent of samples expressing proteinase activities of approximate molecular mass (x10- 3 )
BPH lnflamm/prostatitis Atypia Adenocarcinoma*
N
23
36
38
64
66
71
76
81
86
94
104
lll
ll5
132
137
148
162
28 14 6 9
62 57 33 67
24 43 17 33
24 43 17 22
100 93 100 100
4 7 0 ll
100 100 100 100
64 50 33 22
100 93 83 89
100 100 100 100
100 100 100 100
14 36 17
93 100 100 100
100 100 100 100
79 43 Ot llt
71 57 50 33
96 100 100 100
100 100 100 100
11
* Patients with adenocarcinoma included those with a combined Gleason score as follows: 1-2, 1-4, 5-5, and 2-7. t Statistically significant " = 0.01.
PROTElNit,SES TABLE
0
L.
657
The expression of caseinolytic proteinase activities in prostatic secretions frorri men with l3Pl-l, BP}{ and prostatitis/inflanimation, BPH + atypia, and BPH + adenocarcinoma
Condition
Percent of samples expressing proteinase activities of approximate molecular mass (X 10-")
N
11
12
14
16
20
22
25
27
35
37
39
54
57
58
60
72
80
86
90
96
102
116
126 133 137 145
151
25 15
4 0
20 53 17 22
40 47 50 22
20 20 17 33
64 60 67 33
36 27 50
56 40 50 33
24 53 33 33
100 93 100 100
96 100 100 100
76 93 83 100
100 100 100 100
100 100 100 100
56 87 100 89
96 100 100
9
100 93 83 89
20 53 67
Adenocarcinoma *
0 11
32 27 33 11
56
6
48 20 67
24
Atypia
96 100 100 89
92 100 100 100
100 100 100 100
BPH Prostatitis/lnflam
67
47 50 44
11
22
47 50 22
100
32 33 0 11
72
47 50 89
100 100 100 89
* Patients with adenocarcinoma included those with a combined Gleason score as follows: 1-2, 1-4, 5-5, and 2-7.
Kd, which are sensitive to the serine proteinase inhibitors benzamidine and epsilon-aminocaproic acid, probably do not represent PSA/seminin since PSA/seminin separated from seminal plasma by gel filtration chromatography does not demonstrate activity in gelatin-SDS-polyacrylamide gel zymograms,3 indicating that PSA/seminin is not reactivated after exposure to SDS. Both the human and rat prostate show zone/lobe specific secretion of proteinases. The central zone of the human prostate secretes pepsinogen Il 17 and tissue-type plasminogen activator 18 preferentially, whereas seminin/PSA is found in the peripheral zone. 19 In the rat model, plasminogen activator activities are secreted only by the ventral lobe. 10 Similarly, gelatinolytic and caseinolytic proteinase activities of varying number, molecular size, and relative activity are secreted individually by the ventral, lateral, and dorsal lobes of the rat prostatic complex 20 • 21 (Wilson and associates, unpublished results). The reason for lobe/zone specific origins of secretory proteinases in the prostate is not known, nor is the zone of origin within the human prostate of the gelatinolytic and caseinolytic proteinase activities we describe here. However, like pepsinogen II, 17 some gelatinolytic proteinase activities appear to arise from both the prostate and seminal vesicle in men. 3 The analysis of proteins in prostatic secretion has been studied, not only to better understand the biology of prostatic secretions, but to appraise prostate disease. Altered electrophoretic protein pattems 22 • 23 and diminished acid phosphatase, !eucine aminopeptidase and protein kinase activities 24 · 25 have been reported for prostatic secretions from men with prostatitis. Similarly, changes in protein patterns, 23 LDH V/LDH I ratio, and activities/levels of prostatic acid phosphatase, complement components C3 and C4, and transferrin have been noted for prostatic secretions of men with cancer of the prostate. 26 fo study, we detected changes in the presence of a gelatinolytic proteinase activity with prostatic disease beyond that of BPH. Since altered vc,"u.w," patterns in prostatic secretions could arise due to changes in proteinases 1-'"Juuv::,u tumor cells or ,m111oee1 ""'"'~" secreted leucocytes in inflam28 the development of antibodies or enzyme assays to the of specific proteinases could prove useful in the evaluation of ~-'"""·"·- with prostate disease. Acknowledgments: We gratefully acknowledge the excellent assistance of Mr. Roger Paul in the photography of the zymograms. This research was supported by the General Research Fund of the Department of Veterans Affairs.
our
REFERENCES
L Mann, T. and Lutwak-Mann, C.: Male Reproductive Function and Semen. New York: Springer-Verlag, 198L 2. Tauber, P. F., Propping, D., Schumacher, D. F. R and Zaneveld, L. F. D.: Biochemical aspects of the coagulation and liquefaction of human semen. J. AndroL, 1: 280, 1980. 3. Yin, H. Z., Vogel, M. M., Schneider, M., Ercole, E., Zhang, G., Sinha, A. A. and Wilson, M. J.: Gelatinolytic protease activities in human seminal plasma. J. Reprod. FertiL, 88: 491, 1990. 4. Lilja, H.: A kallikrein-like serine protease in prostatic fluid cleaves the predominant seminal vesicle protein. J. Clin. Invest., 76: 1899, 1985.
5. Lee, C., Keefer, l\lL, Zhao, Z. W., Kroes, R, Berg, L., Liu, X. and Sensibar, J.: Demonstration of the role of prostate-specific antigen in semen liquefaction by two-dimensional electrophoresis. J. AndroL, 10: 432, 1989. 6. Stamey, T. A, Yang, N., Hay, A. R, McNeal, J. E, Freiha, F. S. and Redwine, K: Prostatic-specific antigen as a serum marker for adenocarcinoma of the prostate. New EngL J. Med., 317: 909, 1987. 7. Oesterling, J. K, Chan, D. W., Epstein, J. L, Kimball, A W., Brusek, D. J., Rock, R C., Brendler, C. R and Walsh, P. C.: Prostatic-specific antigen in the preoperative and postoperative evaluation of localized prostatic cancer treated with radical prostatectomy. J. UroL, 139: 766, 1988. 8. Lange, P. H., Ercole, C. J., Lightner, D. J., Fraley, K K and Vessella, R: The value of serum PSA determinations before and after radical prostatectomy. J. UroL, 141: 873, 1989. 9. Heusen, C. and Dowdle, KR: Electrophoretic analysis ofplasminogen activators in polyacrylamide gels containing sodium dodecylsulfate and copolymerized substrates. AnaL Biochem., 102: 186, 1980. 10. Wilson, M. J., Sinha, A, A., Powell, J. KR and Estensen, RA: Plasminogen activator activities in the developing rat prostate. BioL Reprod., 38: 723, 1988. IL Smith, P. K, Krohn, R L, Hermanson, G. T., et al.: Measurement of protein using bicinchoninic acid. AnaL Biochem., 150: 76, 1985. 12. Duncan, R C., Knapp, R G. and Miller, M. C. III: Introductory Biostatistics for the Heaith Sciences. New York: John Wiley and Sons, ed. 2, 1983. 13. Zaneveld, L J. D. and Tauber, P. F.: Contribution of prostatic fluid component to the ejaculate. Prog. Clin. BioL Res., 75A: 265, 198L 14. Kavanagh, J. P.: Sodium, potassium, calcium, magnesium, zinc, citrate and chloride content of human prostatic and seminal fluid. J. Reprod. Fert., 75: 35, 1985. 15. Tauber, P. F. and Zaneveld, L. J. D.: Coagulation and Liquefaction of Human Semen. In: Human Semen and Fertility Regulation in Men, edited by E. S. K Hafez. St. Louis: Mosby, pp. 153-166, 1976. 16. Watt, KW. K, Lee, P.-J., M'Timkulu, T., Chan, W.-P. and Loo, R: Human prostatic-specific antigen: structural and functional similarity with serine proteases. Proc. NatL Acad. Sci. U.S.A, 83: 3166, 1986. 17. Reese, J. H., McNeal, J, K, Redwine, K A., Samloff, L NL and Stamey, T. A.: Differential distribution ofpepsinogen II between the zones of the human prostate and the seminal vesicle. J. UroL, 136: 1148, 1986. 18. Reese, J. H., McNeal, J. K, Redwine, E. A., Stamey, T. A. and Freiha, F. S.: Tissue-type plasminogen activator as a marker for functional zones, within the human prostate gland. Prostate, 12: 47, 1988. 19. Reese, J. H., McNeal, J. K, Goldenberg, S. L., Redwine, KA. and Sellers, R G.: Distribution of lactoferrin in the normal and inflamed human prostate: an immunohistochemical study. Prostate, 20: 73, 1992. 20. Wilson, M. J., Strasser, M., Vogel, M. M. and Sinha, A. A: Calciumdependent and independent gelatinolytic proteinase activities of the rat ventral prostate and its secretion: characterization and effects of castration and testosterone treatment. BioL Reprod., 44: 776, 1991. 21. Lee, C., Sensibar, J. A., Sinha, A. A and Wilson, M. J.: Induction of specific gelatinolytic proteinases in the lateral prostate of rats by ectopic pituitary grafts. Biol. Reprod., 46: 671, 1992. 22. Balerna, M., Colpi, G. M., Campana, A., Roveda, L., Tommasini-
658
WILSON AND ASSOCIATES
Degna, A. and Zanollo, A.: High-resolution protein patterns of human expressed prostatic secretion: a new tool for the diagnosis of prostatitis. Arch. Androl., 8: 97, 1982. 23. Tsai, Y. C., Harrison, H. H., Lee, C., Daufeldt, J. A., Oliver, L. and Grayhack, J. T.: Systematic characterization of human prostatic fluid proteins with two-dimensional electrophoresis. Clin. Chem., 30: 2026, 1984. 24. Kavanagh, J. P., Darby, C. and Costello, C.B.: The response of seven prostatic fluid components to prostatic disease. Int. J. Androl., 5: 487, 1982. 25. Wilson, M. J., Steer, R. C. and Kaye, K. W.: Protein kinase activities in human prostatic secretion: biochemical character-
ization and effect of prostatitis. J. Lab. Clin. Med., 103: 905, 1984. 26. Grayhack, J. T., Wendel, E. F., Oliver, L. and Lee, C.: Analysis of specific proteins in prostatic fluid for detecting prostatic malignancy. J. Urol., 121: 295, 1979. 27. Werb, Z.: Proteinases and matrix degradation. In: Textbook of Rheumatology, edited by Kelly, W. N., Harris, E. D., Jr., Ruddy, S. and Sledge, C. B. Philadelphia: W.B. Saunders, pp. 300-321, 1989. 28. Khokha, R. and Denhardt, D. T.: Matrix metalloproteinases and tissue inhibitor of metalloproteinases: a review of their role in tumorigenesis and tissue invasion. Invasion Metastasis 9: 391, 1989.
Vol. 149, 653-658, March 1993
Printed in U.S.A.
GELATINOLYTIC AND CASEINOLYTIC PROTEINASE ACTIVITIES IN HUMAN PROSTATIC SECRETIONS MICHAEL J. WILSON*, HOLLY NORRIS, DEEPAK KAPOOR, MILDRED WOODSON, CATHERINE LIMAS AND AKHOURI A. SINHA From the VA Medical Center and Departments of Laboratory Medicine and Pathology, Urologic Surgery and Genetics and Cell Biology, University of Minnesota, Minneapolis, Minnesota
ABSTRACT
Seminal fluid contains a number of proteinase activities, many of which are secreted by the prostate gland. Our objective was to determine proteinase activities in human prostatic secretions which can degrade gelatin and/or casein. Prostatic secretions were collected by prostate massage from men with benign prostatic hyperplasia prior to surgery to relieve obstruction. Significant proteinase activities towards gelatin of about 81, 86, 94, 111, 115 and 163 Kd as well as less active forms of 23, 36, 38, 132, 137, and 148 Kd were detected using protein substrate-polyacrylamide gel zymography. In addition, Ca2+ stimulated activities of approximately 64, 66, 71 and 76 Kd; however, EDTA and EGTA inhibited all activities but the 23, 36 and 38 Kd forms (these were inhibited by benzamidine and epsilon-amino caproic acid). This suggests that the gelatinolytic activities of 64 Kd and greater were metalloproteinases and those of 23, 36, and 38 Kd were serine proteinases. Significant caseinolytic activities of 22, 25, 35, 37, 57, 90, 96, 102 and 116 Kd were found as well as several less active forms and a 12 Kd activity stimulated by Ca2+. Caseinolytic activities of 12, 14, 16, 96, 102, 116, and 126 Kd were inhibited by EDTA and EGTA indicating they are metalloproteinases. The 35, 37, 57 and 58 Kd caseinolytic activities were inhibited by benzamidine, and the 57 and 58 Kd forms by epsilon-aminocaproic acid suggesting they were serine proteinases. There was considerable variability among individuals in the molecular forms of proteinase activity expressed as well as the level of their activity. A significant decrease in the frequency of expression of the 132 Kd gelatinolytic activity was found in secretions from men with atypia or adenocarcinoma, as compared with men with benign prostatic hyperplasia alone. Our results show that human prostatic secretion contains a variety of proteinase activities. The expression of the 132 Kd gelatinolytic activity could prove useful in further evaluation of neoplastic prostatic disease. KEY WORDS: metalloproteinases, serine proteinases, prostate
The secretion of the human prostate gland, as obtained by rectal prostate massage, contains about 2% protein. 1 However, the identity and function of individual proteins, including a wide variety of enzymes, in this complex mixture remains mostly unknown. Among the enzymes present in prostatic secretion are a number of proteolytic activities which imbue seminal plasma with one of its most characteristic attributes.1 Proteinases that have been described in seminal plasma include plasminogen activator, pepsinogen II, Pz Peptidase, seminin/ prostatic specific antigen (PSA), and gelatinases. 2 ' 3 An important function of proteolytic enzymes in semen is the liquefaction of the postejaculatory seminal coagulum and the subsequent degradation of seminal proteins to proteases, peptides, and amino acids. 1 The proteinase that appears to have a central role in semen liquefaction is the neutral enzyme, seminin/PSA, which can cleave the major basic proteins of seminal vesicle secretion that are responsible for forming the coagulum. 2 • 4 ' 5 The measurement of serum PSA levels has also become an important tool in the diagnosis and evaluation of progression of prostatic cancer.6-8 Since the role of most prostatic secretory proteinases in semen remains unknown and in view of the possible use of secretory proteins as markers of prostatic function and/or pathology, we undertook a study of gelatinolytic and caseinolytic proteinase activities in human prostatic secretions. Accepted for publication September 4, 1992. • Requests for reprints: VA Medical Center, One Veterans Drive, Minneapolis, Minnesota 55417.
653
MATERIALS AND METHODS
Prostatic secretions were collected by prostate massage from 57 men who were about to undergo a transurethral resection of the prostate to relieve symptomatic bladder outlet obstruction. The prostate massage was done after obtaining an informed consent under protocols approved by the Human Studies ComCasein
Gelatin Kd -162 =-137 -132 -115 -111 -94 -88 -a1
Kd
10680-
Kd
Kd
=--132 -111 -94 -86 -81 -76 _71 -66 -54
49.5-
Kd
Kd
27.5-
:'-126 =102 =- 80 -60 -57 _37 -35 -27 =-22
18.5-
-16
1068049.532.5-
Kd
Kd =-145 =--116 a,-102 =--so -60 -57 _37 -35
-
-25 -22 -16 =12
32.5-
-CaCl 2
+CaCl2
-CaCl2
+CaCl2
FIG. 1. Multiple forms of gelatinolytic and caseinolytic proteinase activities in human prostatic secretion. Prostatic secretions (0.5 µl. per lane) of two men with BPH were electrophoresed in gelatin-containing and casein-containing polyacrylamide gels and prepared as zymograms. Molecular mass of protein standards electrophoresed in gel are given in column to left of gel and approximate molecular mass of proteolytic activities detected are given in column to right of each gel. Samples in this figure were chosen to depict most (but do not show all) molecular forms of proteinase activities detected in human prostatic secretions. Note that 64, 66, 71, and 76 Kd forms of gelatinolytic activity and 12 Kd form of caseinolytic activity are stimulated by addition of 5 mM. CaC12 to incubation buffer.
654
Kd
WILSON AND ASSOCIATES
None
EGTA
Benz.
Kd
None
EGTA
Benzam.
10680-
10680-
49.549.532.527.5-
32.527.5-
18.5-
10680-
1068049.5-
49.5-
+CaCl2
+CaCl2
32.527.5-
FIG. 2. Effect of EG TA and benzamidine on gelatinolytic proteinase activities of human prostatic secretions. Replicate aliquots (0.5 µI. per lane) of two samples of prostatic secretions from men with BPH were electrophoresed into gelatin-containing polyacrylamide gels. These gels were sliced into pieces that were incubated with 50 mM. Tris-HCl buffer (pH 8.4) alone, buffer + 5 mM. EGTA, buffer + 10 mM. benzamidine, buffer + 5 mM. CaCl 2 , buffer + 5 mM. CaC12 + 5 mM. EGTA, and buffer + 5 mM. CaC12 + 10 mM. benzamidine. EGTA inhibited gelatinolytic proteinase activities of 64 Kd and greater (EDTA inhibited these activities also, data not shown). EGTA did not affect 23, 36, and 38 Kd activities which were inhibited by benzamidine (also by epsilon-aminocaproic acid, data not shown).
mittee of the Minneapolis VA Medical Center. The prostatic fluids were centrifuged at 960 X g for 10 minutes at 4C to remove cells; the supernatant was frozen in powdered dry ice and stored frozen at -20C until zymographic analysis. Prostatic tissues removed in the surgery were submitted for pathologic examination. In some cases prostatitis/inflammation, atypia, or adenocarcinoma were determined to be present. Zymography of proteinases. Aliquots (0.5 µl.) of prostatic secretion were subjected to electrophoresis in duplicate gelatinor casein-containing polyacrylamide (9 and 11 % acrylamide respectively) gels in the presence of sodium dodecyl sulfate under nonreducing conditions. 9 • 10 Each protein substrate was present at 0.1 % final concentration in the gel. The gels (0. 75 mm. thick) were subjected to electrophoresis for about 35 minutes at 200 v. in a BioRad Mini-Protean II system (BioRad, Richmond, California). Following electrophoresis, the gels were rinsed with distilled water and then washed with gentle shaking at room temperature with 2.5% Triton X-100 (2 changes) for 1 hour. The gels were again rinsed with distilled water and one gel was incubated in 50 mM. Tris-HCl (pH 8.4) containing 5
32.527.518.5FIG. 3. Effect ofEGTA and benzamidine on caseinolytic proteinase activities of human prostatic secretions. Replicate aliquots (0.5 µI. per lane) of same samples shown in fig. 2 were electrophoresed into caseincontaining acrylamide gels. These gels were sliced into pieces that were incubated under same conditions as detailed in legend to fig. 2; 12, 14, 16, 96, 102, 116, and 126 Kd caseinolytic proteinase activities were inhibited by EGTA (also by EDTA, data not shown). Benzamidine inhibited 35, 37, 57, and 58 Kd caseinolytic activities.
mM. CaCb (2 changes), whereas the other was incubated with buffer alone. Incubation was overnight (18 to 20 hours) at 37C. Following the incubation they were stained with amido black. Areas of proteolysis appear as clear zones against a blue background. Molecular mass determinations were made in reference to prestained protein standards (BioRad) co-electrophoresed in these gels: phosphorylase B, 106 Kd; bovine serum albumin, 80; ovalbumin, 49.5 Kd; carbonic anhydrase, 32.5 Kd; soy bean trypsin inhibitor, 27.5 Kd; and lysozyme, 18.5 Kd. The effect of pH of the incubation media and selected proteinase inhibitors on the proteinase activities in the proteinsubstrate polyacrylamide gels was examined to determine some of the properties of these enzymes. Replicate aliquots (0.5 µl.) of prostatic secretions were electrophoresed in gelatin or casein containing SDS-polyacrylamide gels. The gels were cut into strips, each of which contained a lane(s) of the replicate sample(s). These strips were washed in Triton X-100 as described above and then incubated with and without 5 mM. CaCb in 50 mM. Tris-HCl buffers of varied pH (6.8, 7.6, 8.0, 8.4, 9.0, 9.2) or with 50 mM. Tris-HCl buffers of pH 8.4 containing one of the following inhibitors: 10 mM. epsilon-aminocaproic acid, 10 mM. tranexamic acid, 5 mM. sodium iodoacetate, 5 mM. nethylmaleimide (NEM), 10 mM. benzamidine, 5 mM. EDTA, or 5 mM. EGT A. The proteinase activities were stable to
PROTEINASES IN HUMAN PROSTATIC SECRETION
IA I 1
BPH 2
3
4
PROST
5
6
7
8
IA I
655
CAP
9 10 11 12 13 14 15
106 -
80-
49.532.5FIG. 4. Gelatinolytic proteinase activities of prostatic secretions from men with BPH, inflammation/prostatitis, atypia, or adenocarcinoma. Aliquots (0.5 µI. per lane) of prostatic secretions from men with BPH, inflammation/prostatitis (Prost), Atypia (Atyp), and adenocarcinoma (CAP) were electrophoresed in gelatin-containing SDS polyacrylamide gels and prepared for zymography. Activities observed within individuals with BPH alone indicate variability in gelatinolytic proteinase activities among men, and this variability is observed across the disease groups.
freezing since some samples were frozen and thawed 6 times and the proteinase pattern was not changed. Other procedures. Protein concentrations of secretions were estimated using bicinchoninic acid (Pierce Chemical Co., Rockford, Illinois) with bovine serum albumin as standard.11 Statistical analysis. The statistical evaluation of the frequency of expression of individual proteinase activities in secretions of men with different conditions of the prostate was done by chi-square analysis. 12 RESULTS
Gelatinolytic and caseinolytic proteinase activities in prostatic secretions of men with benign prostatic hyperplasia (BPH). An examination of proteinase activities in human prostatic secretions using zymography showed multiple activities towards gelatin or casein as the protein substrates (fig. 1). Significant gelatinolytic proteinase activities of 81, 86, 94, 111, 115, and 162 Kd and less active forms of 23, 36, 38, 132, 137, and 148 Kd were detected. The addition of 5 mM. CaCb to the reaction buffer stimulated proteinase activities of approximately 64, 66, 71, and 76 Kd. Significant caseinolytic proteinase activities of 22, 25, 35, 37, 57, 90, 96, 102, and 116 Kd as well as less active forms of 11, 14, 16, 20, 27, 39, 54, 58, 60, 72, 80, 86, 96, 126, 133, 137, 145, and 151 were detected. A 12 Kd caseinolytic activity was stimulated by Ca2+. However, there was considerable variability among men in the expression of both activity and number of molecular forms of the gelatinolytic and caseinolytic proteinases. Enzymic properties of the gelatinolytic and caseinolytic proteinase activities. Some of the enzymic properties of the gelatinolytic and caseinolytic proteinase activities were examined by varying the pH of the reaction buffer or by adding different inhibitors of proteinase activities to the reaction media. Both
the gelatinolytic and caseinolytic proteinases demonstrated activity over a broad range of pH values (7.2-9.2) in the neutral to alkaline domain (data not shown). Optimal proteinase activities were found at a reaction pH of 8.4 and this pH was chosen for the incubation of the protein substrate gels in these studies. The chelators EDTA and EGTA (5 mM.) inhibited all the gelatinolytic proteinase activities except 23, 36, and 38 Kd forms (fig. 2). The greater sensitivity of gelatinolytic activities to chelator inhibition versus the lower number of activities (64, 66, 71, and 76 Kd) stimulated by Ca2+ added to the incubation buffer may be due to the endogenous association of Ca2+ with these proteinases since prostatic secretions are rich in Ca2+. 13, 14 The 23, 36, and 38 Kd activities were inhibited by the serine proteinase inhibitors benzamidine (10 mM.) and epsilon-aminocaproic acid (10 mM.). The 12, 14, 16, 96, 102, 116, and 126 Kd forms of caseinolytic activity were inhibited by 5 mM. EDTA or EGTA (fig. 3), whereas the 35, 37, 57, and 58 Kd forms were inhibited by 10 mM. benzamidine (fig. 3). In addition, the 57 and 58 Kd forms of caseinolytic activity were inhibited by 10 mM. epsilon-aminocaproic acid (data not shown). There was no effect of 5 mM. sodium iodoacetate, 0.2% sodium azide, 5 mM. n-ethylmaleimide, 10 mM. hydrocinnamic acid, and 10 mM. tranexamic acid on either group ofproteinase activities. Considerable variability in the number and relative activity of proteinase bands towards either protein substrate was found for prostatic secretions from men with BPH (figs. 4 and 5). In an attempt to quantitate the number of proteinase activities in prostatic secretion, we examined the zymograms for the relative frequency (the presence or absence) of expression of each molecular form of gelatinolytic and caseinolytic proteinase activity in these specimens (tables 1 and 2). The different molecular forms of caseinolytic and gelatinolytic proteinase
656
Kd
WILSON AND ASSOCIATES
I
CAP 2 3
4
I I ATYP 5 6 7
I
PROST BPH 8 9 10 11 12 13 14
I
57, 60, and 102 Kd activities. The caseinolytic activities of 11, 16, 25, 35, 37, 54, and 58 Kd were detected with a low frequency (<40%). Expression of gelatinolytic and caseinolytic proteinase activities in prostatic secretions from men with different prostatic diseases. We examined the proteolytic activities towards gelatin (fig. 4 and table 1) and casein (fig. 5 and table 2) in prostatic secretions from men with a primary pathologic diagnosis of BPH, prostatitis/inflammation, atypia, and adenocarcinoma. When the frequency of a particular proteinase activity in the secretions of men with BPH alone was compared with that of men with other pathologies, we found a pattern that indicated that: 1) fewer samples from men with atypia or adenocarcinoma showed the 76 and 132 Kd gelatinolytic activity; 2) men with prostatitis/inflammation demonstrated less 132 Kd gelatinolytic activity; 3) men with adenocarcinoma showed less expression of 27, 39, and 60 Kd forms of caseinolytic activities; and 4) men with prostatitis/inflammation showed a decreased frequency of the 14 Kd and increased frequency of the 25 Kd caseinolytic activities. However, only the decreased expression of the 132 Kd gelatinolytic proteinase activity in secretions from men with atypia or cancer was statistically significant (table 1).
1068049.5-
-CaCl2
32.527.518.5-
1068049.532.527.518.5-
DISCUSSION
FIG. 5. Caseinolytic proteinase activities of prostatic secretions from men with BPH, inflammation/prostatitis, atypia, or adenocarcinoma. Aliquots (0.5 µl. per lane) of prostatic secretions from men with BPH, inflammation/prostatitis (Prost), Atypia (Atyp), and adenocarcinoma (CAP) were electrophoresed in casein-containing SDS polyacrylamide gels and prepared for zymography. Activities observed within individuals with BPH alone indicate variability in caseinolytic proteinase activities among men. This variability is observed across disease groups.
activities were evaluated in zymograms in which prostatic secretion samples were electrophore~ed by both constant volume (0.5 µl. per lane) or a constant amount of protein (5 µg. per lane). This was done so that differences in protein content of prostatic secretions would not affect the detection of a particular band of proteinase activity. We interpreted these gels on a qualitative basis and did not attempt to quantitate the activity of individual bands. The reason for this is that there are many proteinase bands and each proteinase band expresses its own enzymic rate of protein substrate hydrolysis. We incubated the gels to maximize the activities of all bands and thus to optimize detection of individual activities. Most of the gelatinolytic proteinase activities were expressed by the majority of men, that is the activities of 64, 71, 81, 86, 111, 115, 132, 148, and 162 Kd were found in more than 70% of secretion samples from patients with BPH. There were, however, quantitative differences in the apparent activity of individual gelatinolytic bands. The 66 and 104 Kd gelatinolytic activities were found in a small number of samples (<40%). Many of the caseinolytic activities were expressed by most of the men with BPH (>70% ); these activities were 12, 22, 72, 80, 86, 90, 96, 116, 126, 137, 145, and 151 Kd. Other activities were found with a moderate frequency (40 to 70% of samples) and included the 14, 27, 39, TABLE
Human prostatic secretion contains multiple gelatinolytic and caseinolytic proteinase activities of varying intensity and molecular size as indicated in protein substrate-SDS-polyacrylamide gel zymography. All of the gelatinolytic activities except the 23, 36, and 38 Kd bands appeared to be metalloproteinases whereas the 23, 36, and 38 Kd activities are attributed to the serine proteinase class. Based on inhibitor studies, some caseinolytic activities could be classified as metallo- or serine proteinases, while most remained unaffected by the agents tested. There are patterns in the distribution of proteinase activities in the zymograms that indicate that some proteolytic activities may be related. For example, whereas most caseinolytic activities were not affected by EDT A or EGT A, those activities grouped as 12, 14, and 16 Kd and 96, 102, 116, and 126 Kd were inhibited by these chelators. Differences in the extent of glycosylation and/or possible proteolytic modification of a proteinase or the presence of proenzyme and active enzyme forms of a proteinase could account for multiple forms of proteinase activities observed in these zymograms. Other examples of proteinase activities in these gels possibly related to a mature proteinase molecule would be the 23, 36, and 38 Kd gelatinolytic activities inhibited by benzamidine and epsilon-aminocaproic acid and the benzamidine-sensitive caseinolytic activities of 35 and 37 and of 57 and 58 Kd. These groupings also suggest that the 36 and 38 Kd gelatinolytic and 35 and 37 Kd caseinolytic activities, which are similar in molecular mass and are sensitive to the same inhibitor, may be the manifestation of the same enzymes. PSA/seminin is a prominent serine proteinase in prostatic secretion which can hydrolyze both casein and gelatin. 15• 16 The gelatinolytic and caseinolytic activities of approximately 35-38
l. The expression of gelatinolytic proteinase activities in prostatic secretions from men with BPH, BPH
+ prostatitis/inflammation, BPH
+ atypia, and BPH + adenocarcinoma Percent of samples expressing proteinase activities of approximate molecular mass (x10- 3 )
BPH lnflamm/prostatitis Atypia Adenocarcinoma*
N
23
36
38
64
66
71
76
81
86
94
104
lll
ll5
132
137
148
162
28 14 6 9
62 57 33 67
24 43 17 33
24 43 17 22
100 93 100 100
4 7 0 ll
100 100 100 100
64 50 33 22
100 93 83 89
100 100 100 100
100 100 100 100
14 36 17
93 100 100 100
100 100 100 100
79 43 Ot llt
71 57 50 33
96 100 100 100
100 100 100 100
11
* Patients with adenocarcinoma included those with a combined Gleason score as follows: 1-2, 1-4, 5-5, and 2-7. t Statistically significant " = 0.01.
PROTElNit,SES TABLE
0
L.
657
The expression of caseinolytic proteinase activities in prostatic secretions frorri men with l3Pl-l, BP}{ and prostatitis/inflanimation, BPH + atypia, and BPH + adenocarcinoma
Condition
Percent of samples expressing proteinase activities of approximate molecular mass (X 10-")
N
11
12
14
16
20
22
25
27
35
37
39
54
57
58
60
72
80
86
90
96
102
116
126 133 137 145
151
25 15
4 0
20 53 17 22
40 47 50 22
20 20 17 33
64 60 67 33
36 27 50
56 40 50 33
24 53 33 33
100 93 100 100
96 100 100 100
76 93 83 100
100 100 100 100
100 100 100 100
56 87 100 89
96 100 100
9
100 93 83 89
20 53 67
Adenocarcinoma *
0 11
32 27 33 11
56
6
48 20 67
24
Atypia
96 100 100 89
92 100 100 100
100 100 100 100
BPH Prostatitis/lnflam
67
47 50 44
11
22
47 50 22
100
32 33 0 11
72
47 50 89
100 100 100 89
* Patients with adenocarcinoma included those with a combined Gleason score as follows: 1-2, 1-4, 5-5, and 2-7.
Kd, which are sensitive to the serine proteinase inhibitors benzamidine and epsilon-aminocaproic acid, probably do not represent PSA/seminin since PSA/seminin separated from seminal plasma by gel filtration chromatography does not demonstrate activity in gelatin-SDS-polyacrylamide gel zymograms,3 indicating that PSA/seminin is not reactivated after exposure to SDS. Both the human and rat prostate show zone/lobe specific secretion of proteinases. The central zone of the human prostate secretes pepsinogen Il 17 and tissue-type plasminogen activator 18 preferentially, whereas seminin/PSA is found in the peripheral zone. 19 In the rat model, plasminogen activator activities are secreted only by the ventral lobe. 10 Similarly, gelatinolytic and caseinolytic proteinase activities of varying number, molecular size, and relative activity are secreted individually by the ventral, lateral, and dorsal lobes of the rat prostatic complex 20 • 21 (Wilson and associates, unpublished results). The reason for lobe/zone specific origins of secretory proteinases in the prostate is not known, nor is the zone of origin within the human prostate of the gelatinolytic and caseinolytic proteinase activities we describe here. However, like pepsinogen II, 17 some gelatinolytic proteinase activities appear to arise from both the prostate and seminal vesicle in men. 3 The analysis of proteins in prostatic secretion has been studied, not only to better understand the biology of prostatic secretions, but to appraise prostate disease. Altered electrophoretic protein pattems 22 • 23 and diminished acid phosphatase, !eucine aminopeptidase and protein kinase activities 24 · 25 have been reported for prostatic secretions from men with prostatitis. Similarly, changes in protein patterns, 23 LDH V/LDH I ratio, and activities/levels of prostatic acid phosphatase, complement components C3 and C4, and transferrin have been noted for prostatic secretions of men with cancer of the prostate. 26 fo study, we detected changes in the presence of a gelatinolytic proteinase activity with prostatic disease beyond that of BPH. Since altered vc,"u.w," patterns in prostatic secretions could arise due to changes in proteinases 1-'"Juuv::,u tumor cells or ,m111oee1 ""'"'~" secreted leucocytes in inflam28 the development of antibodies or enzyme assays to the of specific proteinases could prove useful in the evaluation of ~-'"""·"·- with prostate disease. Acknowledgments: We gratefully acknowledge the excellent assistance of Mr. Roger Paul in the photography of the zymograms. This research was supported by the General Research Fund of the Department of Veterans Affairs.
our
REFERENCES
L Mann, T. and Lutwak-Mann, C.: Male Reproductive Function and Semen. New York: Springer-Verlag, 198L 2. Tauber, P. F., Propping, D., Schumacher, D. F. R and Zaneveld, L. F. D.: Biochemical aspects of the coagulation and liquefaction of human semen. J. AndroL, 1: 280, 1980. 3. Yin, H. Z., Vogel, M. M., Schneider, M., Ercole, E., Zhang, G., Sinha, A. A. and Wilson, M. J.: Gelatinolytic protease activities in human seminal plasma. J. Reprod. FertiL, 88: 491, 1990. 4. Lilja, H.: A kallikrein-like serine protease in prostatic fluid cleaves the predominant seminal vesicle protein. J. Clin. Invest., 76: 1899, 1985.
5. Lee, C., Keefer, l\lL, Zhao, Z. W., Kroes, R, Berg, L., Liu, X. and Sensibar, J.: Demonstration of the role of prostate-specific antigen in semen liquefaction by two-dimensional electrophoresis. J. AndroL, 10: 432, 1989. 6. Stamey, T. A, Yang, N., Hay, A. R, McNeal, J. E, Freiha, F. S. and Redwine, K: Prostatic-specific antigen as a serum marker for adenocarcinoma of the prostate. New EngL J. Med., 317: 909, 1987. 7. Oesterling, J. K, Chan, D. W., Epstein, J. L, Kimball, A W., Brusek, D. J., Rock, R C., Brendler, C. R and Walsh, P. C.: Prostatic-specific antigen in the preoperative and postoperative evaluation of localized prostatic cancer treated with radical prostatectomy. J. UroL, 139: 766, 1988. 8. Lange, P. H., Ercole, C. J., Lightner, D. J., Fraley, K K and Vessella, R: The value of serum PSA determinations before and after radical prostatectomy. J. UroL, 141: 873, 1989. 9. Heusen, C. and Dowdle, KR: Electrophoretic analysis ofplasminogen activators in polyacrylamide gels containing sodium dodecylsulfate and copolymerized substrates. AnaL Biochem., 102: 186, 1980. 10. Wilson, M. J., Sinha, A, A., Powell, J. KR and Estensen, RA: Plasminogen activator activities in the developing rat prostate. BioL Reprod., 38: 723, 1988. IL Smith, P. K, Krohn, R L, Hermanson, G. T., et al.: Measurement of protein using bicinchoninic acid. AnaL Biochem., 150: 76, 1985. 12. Duncan, R C., Knapp, R G. and Miller, M. C. III: Introductory Biostatistics for the Heaith Sciences. New York: John Wiley and Sons, ed. 2, 1983. 13. Zaneveld, L J. D. and Tauber, P. F.: Contribution of prostatic fluid component to the ejaculate. Prog. Clin. BioL Res., 75A: 265, 198L 14. Kavanagh, J. P.: Sodium, potassium, calcium, magnesium, zinc, citrate and chloride content of human prostatic and seminal fluid. J. Reprod. Fert., 75: 35, 1985. 15. Tauber, P. F. and Zaneveld, L. J. D.: Coagulation and Liquefaction of Human Semen. In: Human Semen and Fertility Regulation in Men, edited by E. S. K Hafez. St. Louis: Mosby, pp. 153-166, 1976. 16. Watt, KW. K, Lee, P.-J., M'Timkulu, T., Chan, W.-P. and Loo, R: Human prostatic-specific antigen: structural and functional similarity with serine proteases. Proc. NatL Acad. Sci. U.S.A, 83: 3166, 1986. 17. Reese, J. H., McNeal, J, K, Redwine, K A., Samloff, L NL and Stamey, T. A.: Differential distribution ofpepsinogen II between the zones of the human prostate and the seminal vesicle. J. UroL, 136: 1148, 1986. 18. Reese, J. H., McNeal, J. K, Redwine, E. A., Stamey, T. A. and Freiha, F. S.: Tissue-type plasminogen activator as a marker for functional zones, within the human prostate gland. Prostate, 12: 47, 1988. 19. Reese, J. H., McNeal, J. K, Goldenberg, S. L., Redwine, KA. and Sellers, R G.: Distribution of lactoferrin in the normal and inflamed human prostate: an immunohistochemical study. Prostate, 20: 73, 1992. 20. Wilson, M. J., Strasser, M., Vogel, M. M. and Sinha, A. A: Calciumdependent and independent gelatinolytic proteinase activities of the rat ventral prostate and its secretion: characterization and effects of castration and testosterone treatment. BioL Reprod., 44: 776, 1991. 21. Lee, C., Sensibar, J. A., Sinha, A. A and Wilson, M. J.: Induction of specific gelatinolytic proteinases in the lateral prostate of rats by ectopic pituitary grafts. Biol. Reprod., 46: 671, 1992. 22. Balerna, M., Colpi, G. M., Campana, A., Roveda, L., Tommasini-
658
WILSON AND ASSOCIATES
Degna, A. and Zanollo, A.: High-resolution protein patterns of human expressed prostatic secretion: a new tool for the diagnosis of prostatitis. Arch. Androl., 8: 97, 1982. 23. Tsai, Y. C., Harrison, H. H., Lee, C., Daufeldt, J. A., Oliver, L. and Grayhack, J. T.: Systematic characterization of human prostatic fluid proteins with two-dimensional electrophoresis. Clin. Chem., 30: 2026, 1984. 24. Kavanagh, J. P., Darby, C. and Costello, C.B.: The response of seven prostatic fluid components to prostatic disease. Int. J. Androl., 5: 487, 1982. 25. Wilson, M. J., Steer, R. C. and Kaye, K. W.: Protein kinase activities in human prostatic secretion: biochemical character-
ization and effect of prostatitis. J. Lab. Clin. Med., 103: 905, 1984. 26. Grayhack, J. T., Wendel, E. F., Oliver, L. and Lee, C.: Analysis of specific proteins in prostatic fluid for detecting prostatic malignancy. J. Urol., 121: 295, 1979. 27. Werb, Z.: Proteinases and matrix degradation. In: Textbook of Rheumatology, edited by Kelly, W. N., Harris, E. D., Jr., Ruddy, S. and Sledge, C. B. Philadelphia: W.B. Saunders, pp. 300-321, 1989. 28. Khokha, R. and Denhardt, D. T.: Matrix metalloproteinases and tissue inhibitor of metalloproteinases: a review of their role in tumorigenesis and tissue invasion. Invasion Metastasis 9: 391, 1989.