Isolation and characterization of a basic carboxypeptidase from human seminal plasma

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 267, No. 2, December, pp. 660-667,1988

Isolation and Characterization of a Basic Carboxypeptidase from Human Seminal Plasma’ RANDAL A. SKIDGEL,*+’

PETER A. DEDDISH,*

AND

RICHARD M. DAVIS+

Departments of *Anesthesiology and ~Pharmacolcgy, University of Illinois College of Medicine, Chicago, Illinois 60612; and *Department of Neurology, University of Texas Health Science Center, Dallas, Texas 7’5235 Received April

11,1988, and in revised form August 5,1988

A carboxypeptidase which cleaves the C-terminal arginine or lysine from peptides was purified by a two-step procedure; gel filtration on Sephacryl S-300 and affinity chromatography on arginine-Sepharose. The activity increased 280% after the first step, indicating the removal of an inhibitor from the crude starting material. The activity in the crude seminal plasma eluted from the Sephacryl S-300 column with an apparent M, 98,000 and after purification with an il!l, 67,000, indicating that it binds to another protein in the crude seminal plasma. When analyzed by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate, a single band at M,. 53,000 was seen which was converted to two smaller bands (M, 32,000 and/or 26,000) after reduction. The seminal plasma carboxypeptidase has a neutral pH optimum, is inhibited by o-phenanthroline and by the inhibitor of carboxypeptidase B-type enzymes, 2-mercaptomethyl-3-guanidinoethylthiopropanoic acid, and can be activated by cobalt. The purified enzyme has a high specific activity (6’7.8 pmol/min/mg) with the ester substrate benzoyl (Bz)-Glyargininic acid and readily cleaves Bz-Ala-Lys, Bz-Gly-Arg, and Bz-Gly-Lys. It also hydrolyzes biologically active peptides such as bradykinin (Km = 6 PM, kcat = 43 min-l), Arg’-Met5-enkephalin (Km = 103 PM, kcat = 438 min-‘), and Lys’-Met5-enkephalin (Km = 848 PM, kcat = 449 min-l). The seminal plasma carboxypeptidase did not cross-react with antiserum to human plasma carboxypeptidase N; other properties distinguish it from the blood plasma enzyme as well as from pancreatic carboxypeptidase B and granular, acid carboxypeptidase H (enkephalin convertase). The carboxypeptidase could be involved in the control of fertility by activating or inactivating peptide hormones in the seminal plasma. In addition it could contribute to the degradation of basic proteins during semen liquefaction. 0 1988Academic press, I~~.

Human seminal plasma is rich in proteases and peptidases, including seminin (l3), kallikrein (4-6), collagenase-like and elastase-like metallopeptidases (3, ‘7), neutral endopeptidase (“enkephalinase”) (8), and angiotensin I converting enzyme (kininase II) (8,9). The kallikrein-kinin system may be involved in the regulation of male fertility, as both kallikrein and bradykinin were reported to enhance sperm motility 1 This work was supported by Grants HL36081 and HL364’73 from the National Institutes of Health. ’ To whom correspondence should be addressed. 0003-9861/88 $3.00 Copyright All rights

Q 1988 by Academic Press, Inc. of reproduction in any form reserved.

660

(10-12). Significant levels of peptides such as Met-enkephalin (13, 14), substance P (14), ,&endorphin (13, 15), and calcitonin (13) are present in human seminal plasma and these peptides may have important functions in the male reproductive tract. Thus, enzymes which can process or inactivate peptide hormones may affect fertility and sperm motility. We detected the presence of a carboxypeptidase in human seminal plasma which cleaves C-terminal arginine or lysine residues of peptides (16). We report here the purification and characterization of this

BASIC

CARBOXYPEPTIDASE

enzyme and show that it is distinct from other known human carboxypeptidases.3 MATERIALS

AND

METHODS

Benzoyl (Bz)4-Ala-Lys, Bz-Gly-argininic acid, and guanidinoethylmercaptosuccinic acid (GEMSA) were provided by Dr. Yehuda Levin of the Weizmann Institute of Science (Rehovot, Israel). Human plasma carboxypeptidase N was purified to homogeneity from outdated human plasma and the subunits were isolated after treatment with 3 M guanidine, as described (17). Frozen human semen was obtained from the Family Planning Clinic of the University of Texas Health Science Center at Dallas. L-Arginine-Sepharose was prepared from epichlorohydrin-activated Sepharose 6B (17). Lys’-Met’-enkephalin was purchased from Peninsula Laboratories (Belmont, CA), Met’-enkephalin was from Boehringer-Mannheim Biochemicals (Indianapolis, IN), and Arg’-Met5-enkephalin and des-Argg-bradykinin were obtained from Bachem (Torrance, CA). Bradykinin, Bz-GlyArg, Bz-Gly-Lys, Bz-Gly-phenyllactic acid, and furylacryloyl (FA)-Phe-Phe were purchased from Sigma Chemical Co. (St. Louis, MO). DL-Z-Mercaptomethyl3-guanidinoethylthiopropanoic acid (MGTA) was from Calbiochem-Behring (La Jolla, CA). Glutaryl -Ala-Ala - Phe - 4 - methoxy - 2 - naphthylamine (glut-Ala-Ala-Phe-MNA) was from Enzyme Systems Products (Livermore, CA). All other chemicals were of reagent grade or better. Enzyme assays. Carboxypeptidase activity with BzGly-argininic acid, Bz-Ala-Lys, FA-Phe-Phe, or BzGly-phenyllactic acid was measured in a continuous spectrophotometric assay using a Varian-Cary 118 or 219 instrument (18). Briefly, enzyme (lo-200 ~1) was mixed with substrate (1 mM final) and buffer (0.1 M Hepes, pH 7.0, for Bz-Ala-Lys or FA-Phe-Phe; 0.1 M Tris-HCl, pH 8.5, for Bz-Gly-argininic acid or Bz-Glyphenyllactic acid) in a final volume of 1 ml and the reaction was conducted at 37°C in a thermostatted cuvette in the recording spectrophotometer. The change in absorbance over time was monitored at 254 nm for the benzoyl substrates and 336 nm for FAPhe-Phe. Hydrolysis of glut-Ala-Ala-Phe-MNA was

a These results were reported in preliminary form at a meeting of the Federation of American Societies for Experimental Biology, Anaheim, CA, 1985. 4 Abbreviations used: Bz, benzoyl; GEMSA, guanidinoethylmercaptosuccinic acid; FA, furylacryloyl; MGTA, DL-2-mercaptomethyl-3-guanidinoethylthiopropanoic acid; glut, glutaryl; MNA, 4-methoxy-2naphthylamine; SDS, sodium dodecyl sulfate; Hepes, 4-(2hydroxyethyl)-l-piperazineethanesulfonic acid; Chaps, 3-[(3-cholamidopropyl)dimethylammonio]propanesulfonic acid.

IN HUMAN

SEMINAL

PLASMA

661

determined fluorometrically in a two-step assay (8). Inhibition of the carboxypeptidase was assayed using Bz-Gly-Lys substrate (18). Hydrolysis of Bz-Gly-Lys, Bz-Gly-Arg, Bz-Gly-Phe, Arg’or Lys’-Met5-enkephalin, or bradykinin was measured by incubating the enzyme (38-250 ng purified seminal plasma carboxypeptidase in a 0.1 M Hepes, pH 7.0, buffer containing 0.1% Chaps detergent) with the substrate (2 mM of Bz-Gly-amino acid substrates or 0.1 mM of the hexapeptide enkephalins or bradykinin) in a final volume of 100 ~1 at 37°C for O-180 min. At selected time intervals, the reaction was stopped with 20 pl of 5% trifluoroacetic acid and 40-~1 aliquots were analyzed in an HPLC system as described earlier (19). Peptide products were separated on a Waters FBondapak Cl8 column with an isocratic mixture or increasing gradient of CH,CN/0.05% trifluoroacetic acid (solvent B) in Hz0/0.05% trifluoroacetic acid (solvent A) as follows: For products of Bz-Gly-Arg or Bz-Gly-Lys hydrolysis, an isocratic mixture of 87% A and 13% B; for Bz-Gly-Phe, a gradient of 13% B to 48% B in 15 min; for Arg’-Met5-enkephalin or Ly&Met’-enkephalin, a gradient of 20% B to 40% B in 15 min; for bradykinin, a gradient of 25% to 35% B in 10 min. Peptides were detected by absorbance at 214 nm and were quantitated by comparison of the peak area to that of a known quantity of authentic standard. For kinetic studies, reactions were conducted in duplicate at each of 6-10 substrate concentrations, ranging (depending on the substrate) from 20 to 500 pM. The initial rates of product formation (v) were measured at each substrate concentration ([S]) and kinetic constants were determined by plotting [S] vs [S]/v. The data were fit to the best straight line by linear regression and correlation coefficients of T = 0.98 or better were obtained. Puri,tcation of human seminal plasma carboxypep tidase. All purification steps were performed at 4°C. Human seminal fluid was centrifuged at 15,OOOgfor 5 min and to the supernatant seminal plasma (12 ml) was added 13,200 units of aprotinin. The seminal plasma was applied to a column of Sephacryl S-300 (2.6 X 74.7 cm) preequilibrated with 0.05 M Tris-HCl, pH 7.2, with 0.1 M NaCl (buffer A) and eluted at 25 ml/h with buffer A. Fractions (5.6 ml) were collected and assayed for protein and activity. The active fractions were pooled (120 ml), 13,200 units aprotinin was added, and then the pooled fractions were applied (60 ml/h) to an arginine-Sepharose affinity column (1.5 X 14 cm) preequilibrated with buffer A. Unbound protein was eluted with 174 ml of buffer A (5 ml/h) and the bound carboxypeptidase was then eluted with 36 ml of buffer A containing 1 mM GEMSA. The active fractions were pooled (24 ml), dialyzed overnight against 2 X 5 liters of 0.1 M NH4HC03, pH 7.8, and concentrated to a volume of 1.2 ml in an Amicon stirred cell concentrator with a YM-10 membrane. An

662

SKIDGEL,

DEDDISH,

AND

DAVIS

Fraction FIG. 1. Gel filtration of human seminal plasma on a Sephacryl S-300 column. The column was eluted at 25 ml/h with 0.05 M Tris-HCl, pH 7.2, with 0.1 M NaCl. The elution positions of molecular weight standards are shown by the arrows. K = M, X 1000. Protein (w) and carboxypeptidase activity (A) were measured as described under Materials and Methods.

equal volume of glycerol was added and the preparation was stored at -20°C. Molecular weight detemnination. A column of Sephacryl S-300 was calibrated with the following standards: blue dextran (void volume), thyroglobulin (M, 669,000), ferritin (Mr 440,000), aldolase (M, 158,000), bovine serum albumin (A& 67,000), ovalbumin (Mr 43,000), chymotrypsinogen A (M, 25,000), and ribonuclease A (M, 13,700). The K,” was calculated for each standard using the equation: K,, = (V, - V,) + (V, - I’,) where VO= void volume, V, = elution volume, and V, = total volume of the packed bed. K,, was plotted vs the log of the molecular weight to yield a straight line. The molecular weight of seminal plasma carboxypeptidase was determined by calculating its K,, and interpolating off the standard curve.

Eflect ofpHon activity. The pH activity profile was determined with Bz-Ala-Lys in the following buffers: 0.1 M sodium acetate (pH 5.0 to 5.5), 0.1 M 2-(N-morpholino)ethanesulfonie acid (pH 6.0 to 6.5), 0.1 M TrisHCl (pH 7.0 to 9.5). Polyacrylamide gel electrophoresis. Polyacrylamide gel electrophoresis in sodium dodecyl sulfate (SDS) was done in 9% gels according to Laemmli (20) with a Mighty Small gel unit (Hoefer, San Francisco, CA). Protein samples were added to sample buffer with or without dithiothreitol (25 mg/ml) and then denatured by boiling for 5 min. Proteins were detected after electrophoresis by silver staining (21). Rocket immunoelectrophoresi Rocket immunoelectrophoresis was performed according to a modification (17) of the method of Laurel1 (22) using spe-

TABLE

I

PURIFICATIONOFHUMANSEMINALPLASMACARBOXYPEPTIDASE

Purification

step

Seminal plasma Sephacryl S-300 Arginine-Sepharose

Volume (ml) 12 120 2.4

Protein (mid 252 127 0.05

’ One unit equals 1 pmol of Bz-Gly-argininic

Activity” (units)

Specific activity (units/mg)

Yield (o/o)

Purification (x-fold)

3.26 9.12 3.39

0.013 0.072 67.8

loo 280 104

1 5.5 5215

acid hydrolyzed

per minute at 37°C.

BASIC

CARBOXYPEPTIDASE

IN HUMAN

30

40

SEMINAL

50

PLASMA

663

60

Fraction FIG. 2. Arginine-Sepharose affinity chromatography of human seminal plasma carboxypeptidase. The active fractions from the Sephacryl S-300 column were pooled and applied to the affinity column and the column was washed with Tris buffer as indicated. The carboxypeptidase was eluted with buffer containing 1 mM GEMSA as shown. For further details, see Fig. 1 and Materials and Methods.

cific antiserum raised to human plasma carboxypeptidase N (17). Protein assays. Protein concentration was determined using the method of Bradford (23) with bovine serum albumin as the standard. RESULTS

Purificaticm of human seminal plasma carboxypeptidase. When crude seminal plasma was passed through a Sephacryl S300 column (Fig. l), the peak of carboxypeptidase activity eluted with an apparent M, 98,000 (average value from two different runs). This step gave a 5.5-fold purification (Table I) with an apparent activation as the recovery was 280%. This suggests the possible presence of an inhibitor of the enzyme in the crude seminal plasma. The carboxypeptidase was then bound to an arginine-Sepharose affinity column and eluted with GEMSA, the competitive inhibitor of Arg/Lys-carboxypeptidases (24) (Fig. 2). The purification factor for the affinity step was 948 with a 37% yield. Overall, the enzyme was purified 5215-fold with a yield of 104% due to the increased activity after the first step. The preparation had a high specific activity of 67.8 pmol/min/mg with the ester substrate, Bz-Gly-argininic acid. When the purified enzyme was rechromatographed on the

Sephacryl S-300 column, the M, was 67,000, lower than that in the crude seminal plasma. Similar results were obtained when the purified enzyme was chromatographed on a TSK-G3000SW HPLC gel filtration column. These data indicate that the apparent higher molecular mass of the enzyme in the crude mixture might be due to binding to another protein, possibly an inhibitor. When the enzyme was run on SDS polyacrylamide gels in the absence of reducing agent, a single band at M,. 53,000 was seen (Fig. 3). When analyzed by SDS-polyacrylamide gel electrophoresis under reducing conditions, two bands at M, 32,000 and 26,000 were detected as well as a trace amount of the M, 53,000 band (Fig. 3). A different preparation of the enzyme, when analyzed by SDS-polyacrylamide gel electrophoresis under reducing conditions, yielded approximately equivalent amounts of the two lower molecular weight bands and the band corresponding to the intact enzyme. Characterization of carboxypeptidase activity. With Bz-Ala-Lys substrate, the pH optimum was relatively broad; from about pH 6 to 8 (Fig. 4). In order to determine the nature of the carboxypeptidase activity, the activation

664

SKIDGEL,

DEDDISH,

AND

DAVIS TABLE

II

ACTIVATIONANDINHIBITIONOFHUMAN SEMINAL PLASMA~ARBOXYPEPTIDASE

Addition”

Concentration (mM) 1.0 0.1 5.0 0.01 1.0

CoClz PCMS Dithiothreitol MGTA o-Phenanthroline

FIG. 3. SDS-polyacrylamide gel electrophoresis of purified human seminal plasma carboxypeptidase with or without reduction. The enzyme was boiled for 5 min in the presence of SDS with (lane 3) or without (lane 2) dithiothreitol and then electrophoresed in a 9% polyacrylamide gel. Lane 1: molecular weight standards (from top to bottom): az-macroglobulin (subunit M, 180,000) and P-galactosidase (subunit M, 116,000) were not resolved; fructose-6-phosphate kinase (subunit M, 84,000); pyruvate kinase (subunit M, 58,000); fumarase (subunit M, 48,500); lactic dehydrogenase (subunit M, 36,500); triosephosphate isomerase (subunit M, 26,600). Protein bands were detected by silver staining.

and inhibition of the enzyme was studied. Cobalt chloride (1 mM) stimulated the activity fourfold while 1 mM o-phenanthroline inhibited it by 85% (Table II), indicating that the seminal plasma carboxypeptidase is a metalloenzyme. MGTA (10 PM), the potent inhibitor of carboxypeptidase B-type enzymes (24), also inhibited the

t

FIG. 4. pH activity profile of human seminal plasma carboxypeptidase. Peptidase activity was measured at various pH values with Bz-Ala-Lys substrate as described under Materials and Methods.

404 94 77 9 15

“Preincubated with purified seminal plasma carboxypeptidase for 2 h on ice. PCMS, pchloromercuriphenylsulfonate; MGTA, 2-mercaptomethyl-3-guanidinoethylthiopropanoic acid. b Activity with Bz-Gly-Lys in the absence of CoCl, = 100%.

seminal plasma enzyme (91%). However, dithiothreitol(5 mM) did not stimulate the activity and p-chloromercuriphenylsulfonate (0.1 mM) did not inhibit it, indicating the lack of a reactive sulfhydryl group in or close to the active center. When tested with synthetic peptide substrates, the seminal plasma carboxypeptidase had the highest activity with Bz-Glyargininic acid (67.8 pmol/min/mg; Table III). The high specific activity with Bz-Glyargininic acid was similar to values obtained in our laboratory with homogeTABLE

III

HYDROLYSISOFSUBSTRATESBYHUMANSEMINAL PLASMA~ARBOXYPEPTIDASE

Substrate”

20

Activity* (%I

Bz-Gly-argininic acid Bz-Ala-Lys Bz-Gly-Lys Bz-Gly-Arg Bz-Gly-Phe FA-Phe-Phe Bz-Gly-phenyllactic acid Glut-Ala-Ala-Phe-MNA

Concentration (m@ 1.0 1.0 2.0 2.0 2.0 0.1 1.0 0.04

Activity (pmol/ min/mg) 67.8 15.3 0.11 0.52 0.01 Not cleaved Not cleaved 0.01

’ Abbreviations: Bz, benzoyl; FA, furylacryloyl; Glut, glutaryl; MNA, 4-methoxy-2-naphthylamine.

BASIC

CARBOXYPEPTIDASE

IN HUMAN

SEMINAL

PLASMA

665

its low K, (6 PM), bradykinin had the highest specificity constant (kcat/K,) of the HYDROLYSIS OF NATURALLY OCCURRING PEPTIDE~ peptides tested, but the kcat values for Arg’-Met’-enkephalin and Lys’-Met5-enKm kc,, k,,t/Km kephalin were about lo-fold higher (Table (@M-l mm’) Peptide (FM) (min-‘) IV). As with the synthetic peptide sub6 Bradykinin 43 7.2 strates, the enzyme cleaved C-terminal arArg’-Met5-enkephalin 103 438 4.3 ginine faster than lysine. The K, for Arg’Lys’-Met’-enkephalin 848 449 0.5 Met’-enkephalin was about eightfold lower than that for Lys’-Met’-enkephalin, a Reactions were done in duplicate at each of the 6yielding an eightfold higher specificity 10 substrate concentrations used. Initial rates were constant for the arginine derivative. calculated and the data were fit to the best straight Rocket immunoelectrophoresis. To establine (plotting [S] vs [S]/V) by linear regression. The turnover number (k,,,) was calculated assuming Y&f, lish whether seminal plasma carboxypeptidase contained antigenic determinants 67,000. Results shown are the average values from which would cross-react with antiserum to two separate determinations. human plasma carboxypeptidase N, it was subjected to rocket immunoelectrophoresis. As shown in Fig. 5, the seminal plasma neous preparations of human plasma carboxypeptidase N, 38-58 pmol/min/mg (17, enzyme did not cross-react with antiserum to carboxypeptidase N while carboxypepti25), or human urinary carboxypeptidase, 48.7 pmol/min/mg (26). This ester sub- dase N and its isolated, purified M, 83,000 inactive subunit and M, 48,000 active substrate was hydrolyzed over loo-fold faster unit gave the characteristic “rocket” prethan the corresponding peptide substrate, cipitin lines indicative of cross-reactivity. Bz-Gly-Arg. The enzyme preferred arginine over lysine as Bz-Gly-Arg was cleaved five times faster than Bz-Gly-Lys DISCUSSION at 2 mM (Table III). As has been shown We report here the presence of a carwith other basic carboxypeptidases (19, boxypeptidase in human seminal plasma 25), the penultimate amino acid also influwhich cleaves basic C-terminal amino ences the rate of hydrolysis; Bz-Ala-Lys was cleaved much faster than Bz-Gly-Lys (Table III). The carboxypeptidase A substrate, Bz-Gly-Phe, was hydrolyzed very slowly, at about 2% of the rate obtained with Bz-Gly-Arg. However, the hydrolysis of two other carboxypeptidase A substrates, FA-Phe-Phe and Bz-Gly-phenyllactic acid (an ester substrate), was not detected (Table III). The preparation may still have a slight endopeptidase contamination undetectable by polyacrylamide gel electrophoresis, as glut-Ala-Ala-PheMNA was slowly hydrolyzed (Table III). This was not due to the basic carboxypeptiFIG. 5. Rocket immunoelectrophoresis of human dase, as MGTA did not inhibit the activity. seminal plasma carboxypeptidase, human plasma Hydrolysis of active peptides. Biologicarboxypeptidase N, and its isolated subunits in agacally active peptides, which can be potenrose containing antiserum to carboxypeptidase N. tial substrates for the enzyme in vivo, were The wells contained the following: (1) J4,48,000 active tested at varying substrate concentrasubunit of carboxypeptidase N (0.7 fig). (2) M, 83,000 tions. The products were separated and inactive subunit of carboxypeptidase N (0.7 pg). (3) quantitated by HPLC and the kinetic con- Seminal plasma carboxypeptidase (0.8 pg). (4) Intact stants were calculated (Table IV). Due to carboxypeptidase N (0.8 fig). TABLE

IV

666

SKIDGEL,

DEDDISH,

acids. We also describe a relatively simple purification procedure consisting of two steps. The purified seminal plasma carboxypeptidase had apparent molecular weights of 67,000 in gel filtration and 53,000 in SDS-polyacrylamide gel electrophoresis. However, after reduction, SDSpolyacrylamide gel electrophoresis of one preparation yielded two bands of M, 32,000 and 26,000 while in another preparation, about 50% of the enzyme was converted to the two lower molecular weight bands and the rest remained intact. These data indicate that, despite the protease inhibitor added, a susceptible peptide bond in the native carboxypeptidase might have been cleaved by a protease, either in the seminal plasma or during purification, but the enzyme was still held together in an active conformation by disulfide bonds until reducing agent was used in the electrophoresis. This phenomenon has been noted with many enzymes including pancreatic kallikrein (27) and neutral endopeptidase 24.11 (28). Although human seminal plasma contains some blood plasma proteins (29), the carboxypeptidase we isolated is clearly different from human blood plasma carboxypeptidase N. The seminal plasma enzyme differs from carboxypeptidase N (M,. 280,000) in molecular weight (l’i’), lack of cross-reactivity with antiserum to carboxypeptidase N, and very high esterase activity relative to peptidase activity (25, 26). It also has a marked preference for cleaving C-terminal Arg vs Lys, while the opposite is true for carboxypeptidase N (19, 25, 26). In addition, the apparent molecular weight of the seminal plasma carboxypeptidase differs from that of human pancreatic carboxypeptidase B (Mr 34,000) (30). Carboxypeptidase H has an acidic pH optimum and is inhibited by sulfhydryl-directed reagents (31), while the seminal plasma enzyme is not inhibited by these reagents and has a neutral pH optimum. The seminal plasma carboxypeptidase has a molecular weight which is similar to that of carboxypeptidase M, but it is soluble while carboxypeptidase M is bound to cell membranes in placenta, kidney, lung, and blood vessels (19,32). Whether the seminal

AND

DAVIS

plasma carboxypeptidase would be membrane bound in the prostate and then released into seminal fluid in a soluble form will require further experimentation. These data indicate that the seminal plasma carboxypeptidase is a novel enzyme, synthesized in the male reproductive tract where it may have specific functions. Seminal plasma carboxypeptidase could participate in the control of fertility and sperm motility by activating or inactivating peptide hormones by removal of a Cterminal basic amino acid. Substance P, enkephalins, and other peptide hormones are synthesized as large precursor molecules which are processed by proteases, frequently at dibasic residues, to release a peptide with an extra C-terminal Arg or Lys which is then removed by carboxypeptidase B-type enzymes to produce the mature peptide (33). Enkephalins and substance P have been found in high concentrations in human seminal plasma (13,14) and enkephalinergic nerve fibers are present in human prostate and seminal vesicles (34). Since seminal plasma carboxypeptidase readily cleaves the C-terminal Arg or Lys from hexapeptide enkephalins, it could be involved in the processing of these or possibly other peptide hormones. Alternatively, it could inactivate peptides, such as bradykinin, by removing their C-terminal basic amino acid. In fact, the specificity constant with bradykinin is about 2.5-fold greater than that obtained with human plasma carboxypeptidase N (19), which was originally named kininase I because it inactivates bradykinin (35). The carboxypeptidase could also affect sperm motility in another way. Because the seminal plasma is known to contain proteins and peptides rich in basic amino acids (36, 37), the enzyme, by releasing Cterminal arginine from these proteins and peptides, could contribute to the high concentration (5 mM) of free arginine found in human seminal plasma (38). The importance of this was indicated in previous studies showing that free arginine stimulated sperm motility in a dose-dependent fashion (39) and arginine administration to oligospermic men increased sperm count and motility (40).

BASIC

CARBOXYPEPTIDASE

Human semen partially coagulates a few minutes after ejaculation and then liquifies within about 30 min. During the liquefaction process, proteolysis occurs, resulting in the production of peptides and free amino acids (3, 41). Liquefaction has been correlated with the degradation of the major seminal vesicle protein into smaller basic proteins (36,37). Thus, a carboxypeptidase which removes C-terminal basic amino acids may be part of the proteolytic process during semen liquefaction. ACKNOWLEDGMENTS We thank Elizabeth Ohiku for technical assistance and Dr. Ervin G. Erdos for his help and advice. Dr. R. Klauser participated in the initial pilot studies. REFERENCES 1. LUNDQUIST, F., THORSTEINSSON, TH., AND Buus, 0. (1955) Biochem. J. 59,69-79. 2. SYNER, F. N., AND MOGHISSI, K. S. (1972) B&hem. J. 126,1135-1140. 3. LUKAC, J., AND KOREN, E. (1979) J. Reprod. Fe&L 56,501-506. 4. GEIGER, R., AND CLAUSNITZER, B. (1981) HoppeSeyler ‘s 2. PhysioL Chem. 362,1279-1283. 5. FINK, E., AND SCHILL, W.-B. (1983) Adv. Exp. Med. Biol. 156B, 1175-1180. 6. FINK, E., SCHILL, W.-B., FIEDLER, F., KRASSNIGG, F., GEIGER, R., AND SHIMAMOTO, K. (1985) BioL Chem. Hoppe-Seyler 366,917-924. 7. LAURELL, C.-B., WEIBER, H., OHLSSON, K., AND RANNEVIK, G. (1982) Clin. Chim Acta 126,161170. 8. ERD&, E. G., SCHULZ, W. W., GAFFORD, J. T., AND DEFENDINI, R. (1985) Lab. Invest. 52,437-447. 9. DEPIERRE, D., BARGETZI, J.-P., AND ROTH, M. (1978) Biochim Biophys. Acta 523,469-476. 10. SCHILL, W.-B., AND HABERLAND, G. L. (1974) Hop pe-Seyler !s 2. PhysioL Chem 355,229-231. 11. SCHILL, W.-B. (1975) Andrologia 7,229-236. 12. PALM, S., SCHILL, W.-B., WALLNER, O., PRINZEN, R., AND FRITZ, H. (1976) in Kinins: Pharmacodynamics and Biological Roles (Sicuteri, F., Back, N., and Haberland, G. L., Eds.), pp. 271279, Plenum, New York. 13. FRAIOLI, F., FABBRI, A., GNESSI, L., SILVESTRONI, L., MORETTI, C., REDI, F., AND ISIDORI, A. (1984) Ann. N. Y Acad. Sci. 438,365-370. 14. RAMA SASTRY, B. V., JANSON, V. E., OWENS, L. K., AND TAYEB, 0. S. (1982) Biochem. Pharmacol. 31,3519-3522. 15. SINGER, R., BRUCHIS, S., BARNET, M., SAGIV, M., KAUFMAN, H., AND SERVADIO, C. (1985) Experentiu 41,64-65.

IN HUMAN

SEMINAL

PLASMA

667

16. SKIDGEL, R. A., DAVIS, R. M., AND ERD&, E. G. (1985) Fed. Proc. 44,1242. 17. LEVIN, Y., SKIDGEL, R. A., AND ERD~S, E. G. (1982) Proc. NatL Acad Sci. USA 79,4618-4622. 18. SKIDGEL, R. A., AND ERDGS, E. G. (1984) in Methods of Enzymatic Analysis (Bergmeyer, H. U., Ed.), Vol. V, pp. 60-72, Verlag Chemie, Weinheim. 19. SKIDGEL, R. A., JOHNSON, A. R., AND ERDBS, E. G. (1984) Biochem. PharmacoL 33,3471-3478. 20. LAEMMLI, U. K. (1970) Nature (London) 227,680685. 21. MORRISSEY, J. H. (1981) Anal. B&hem. 117,307310. 22. LAURELL, C. B. (1965) Anal. Biochem. 10,358-361. 23. BRADFORD, M. M. (1976) Anal. Biochem. 72, 24% 254. 24. PLUMMER, T. H., AND RYAN, T. J. (1981) Biochem. Biophys. Res. Cornmun. 98,448-454. 25. OSHIMA, G., KATO, J., AND ERDBS, E. G. (1975) Arch. Biochem. Biophys. 170,132-138. 26. SKIDGEL, R. A., DAVIS, R. M., AND ERD&, E. G. (1984) AnaL Biochem. 140,520-531. 27. HOFMANN, W., AND GEIGER, R. (1983) Hoppe-Seyler’s 2. Physiol. Chem. 364,413-423. 28. TAM, L.-T., ENGELBRECHT, S., TALENT, J. M., GRACY, R. W., AND ERD&, E. G. (1985) Biothem. Biophys. Res. Commun 133,1187-1192. 29. POLAKOSKI, K. L., SYNER, F. N., AND ZANEVELD, L. J. D. (1976) in Human Semen and Fertility Regulation in Men (Hafez, E. S. E., Ed.) pp. 133-143, Mosby, St. Louis. 30. MARINKOVIC, D. V., MARINKOVIC, J. N., ERDBS, E. G., AND ROBINSON, C. J. G. (1977) Biochem. J. 163,253-260. 31. FRICKER, L. D. (1985) Z’rena!s Neurosci. 8,210-214. 32. SKIDGEL, R. A., AND DAVIS, R. M. (1986) Fed. Proc. 45,790. 33. STEINER, D. F., DOCHERTY, K., AND CARROLL, R. (1984) J. Cell. Biochem. 24,121-130. 34. VAALASTI, A., LINNOILA, I., AND HERVONEN, A. (1980) Histochemistry 66.89-98. 35. ERDBS, E. G., AND SLOANE, E. M. (1962) B&hem. Pharmacol. 11,585-592. 36. LILJA, H., LALJRELL, C.-B., AND JEPPSSON, J.-O. (1984) Stand. J. Clin. Lab. Invest. 44,439-446. 37. LILJA, H., AND LAURELL, C.-B. (1984) Stand J. Clin. Lab. Invest. 44,447-452. 38. MANN, T., AND LUTWAK-MANN, C. (1981) Male Reproductive Function and Semen, pp. 269-336, Springer-Verlag, New York. 39. KELLER, D. W., AND POLAKOSKI, K. L. (1975) Biol. Reprod. 13,154-157. 40. SCHACHTER, A., FRIEDMAN, S., GOLDMAN, J. A., AND ECKERLING, B. (1973) Znt. J. GynecoL Obstet. 11,206-209. 41. KOREN, E., AND LUKAC, J. (1979) J. Reprod. Fertil 56,493-499.