Search

A series of experiments were conducted to investigate the regulation of the primary secretory protein of the canine prostate, arginine esterase, by androgens and/or new antiandrogen under development. In the first experiment, castration decreased (P less than 0.05) prostatic arginine esterase levels relative to intact controls (0.26 +/- 0.1 and 17.0 +/- 0.1 mumole/min/mg protein, respectively). Treatment of castrate dogs with either 5, 10, or 20 silastic capsules (8 cm length) containing the androgen 5 alpha-androstane-3 alpha, 17 beta-diol (3 alpha-diol) plus 1 capsule containing estradiol-17 beta (E2) or the i.m. injection of 25 mg 3 alpha-diol and 0.25 mg E2 for 12 weeks resulted in a dose-dependent increase (P less than 0.05) in prostatic arginine esterase activity (6.8 +/- 1.7, 19.0 +/- 3.6, 21.3 +/- 0.9, and 14.2 +/- 0.7 mumole/min/mg protein, respectively). In the second experiment, steroid treatment (10 3 alpha-diol plus 1 E2 silastic capsules) of castrate dogs for 12 weeks resulted in prostatic arginine esterase activity of 17.8 +/- 2.3 mu mole/min/mg. Co-administration of the steroidal androgen receptor antagonist. Win 49,596 (WIN) at doses of 0.625, 2.5, 10, or 40 mg/kg/day p.o., dose-dependently inhibited (P less than 0.05) prostatic arginine esterase activity (14.9 +/- 1.1, 14.3 +/- 1.3, 3.4 +/- 1.9, and 0.21 +/- 0.1 mumole/min/mg, respectively) to levels similar to that observed in castrate controls (0.14 +/- 0.03 mumole/min/mg). Administration of the nonsteroidal androgen receptor antagonist flutamide at 10 mg/kg/day p.o. to steroid-induced dogs also inhibited (P less than 0.05) arginine esterase activity (0.07 +/- 0.02 mumole/min/mg). In the last experiment, treatment of intact dogs with WIN at 0.625, 2.5, 10, and 40 mg/kg/day for 16 weeks dose-dependently reduced (P less than 0.05) arginine esterase levels (17.0 +/- 1.0, 16.3 +/- 1.5, 10.2 +/- 1.2, and 3.9 +/- 2.5 mumole/min/mg, respectively) compared to intact controls (14.4 +/- 1.2 mumole/min/mg). Histomorphologic and ultrastructural evaluation of prostates from dogs indicated that antiandrogen treatment resulted in glandular epithelial atrophy as well as a reduction in the number of secretory granules. The results of these experiments support that canine prostatic arginine esterase activity is under androgenic control, can be inhibited by antiandrogen treatment and may serve as a functional marker of the androgenic state of the prostate. Whether the effects of androgen and antiandrogens on prostatic arginine esterase is direct or indirect due to a general inhibitory effect on secretory epithelial cell function requires additional study. Furthermore, subject to further evaluation, the steroidal androgen receptor antagonist.(ABSTRACT TRUNCATED AT 400 WORDS)

Previously purified arginine esterase from dog seminal plasma was characterized enzymatically. The enzyme was found to have a rather narrow specificity for arginine esters, much less for lysine esters and was practically devoid of activity towards tyrosine esters, casein, albumin and azocoll. It had a broad optimum pH between 8 and 9. It presented no kallikrein-like activities either in the blood pressure test in dog or in the rat uterus contraction test. It was inhibited by bovine pancreas trypsin inhibitor, aprotinin, phenylalanylprolyl arginine chloromethyl ketone, diisopropylfluorophosphate, phenylmethylsulfonyl fluoride, sodium dodecyl sulfate and leupeptin, but not by soybean trypsin inhibitor, tosyllysine chloromethyl ketone, tosylamide-2-phenylethyl chloromethyl ketone, iodoacetamide, Triton X-100 and EDTA. Experiments involving incubation of prostatic cytosol with purified arginine esterase showed that actin was the only important prostatic protein that was extensively hydrolyzed by this enzyme. It is not known presently whether the hydrolysis of actin is related to a true physiological function of the enzyme and whether actin and arginine esterase ever come into contact with each other in vivo. These properties indicate that arginine esterase from dog seminal plasma is different from other known proteinases including classical kallikreins, although it presents many similarities with this class of enzyme.

We have isolated secretory granules from dog prostate homogenates and have determined whether a major portion of arginine esterase was localized in this fraction and if it was enzymatically active. Secretory granules were purified by density gradient centrifugation on sucrose, metrizamide, or Percoll. A major proportion of whole prostate homogenate arginine esterase was found in the granule fractions. Furthermore, the specific enzymatic activity in the granules was similar to the one observed in seminal plasma. No evidence could be found for the existence of significant amount of a zymogen inactive form of arginine esterase. These results suggest that arginine esterase could be active within the secretory granules in vivo and that it could hydrolyze protein substrates contained in this organelle.

Poly(A)+ rich RNA was isolated from prostate of adult dogs and translated in the rabbit reticulocyte lysate cell-free protein-synthesizing system. Two-dimensional gel electrophoresis of the translation products showed that a protein with a molecular weight of 31 000 was predominantly synthesized. This protein was immunoprecipitated with antibodies directed against purified arginine esterase from dog seminal plasma. mRNA isolated from the prostate of animals castrated for 1 or 2 weeks was unable to direct the synthesis of arginine esterase. However, the synthesis of the enzyme could be stimulated by androgens in castrated animals, presumably by increasing prostatic concentrations of arginine esterase mRNA. The single chain translation product could be further processed in vitro by the addition of dog pancreas microsomes and purified arginine esterase. This procedure yielded split chains of arginine esterase which had identical electrophoretic mobilities as seminal plasma enzyme by two-dimensional gel electrophoresis. When prostatic tissue slices were incubated with tunicamycin, the unglycosylated arginine esterase obtained had a lower molecular weight than the in vitro translation product, suggesting that a signal peptide had been removed in the living cells. These results indicate that arginine esterase processing may include the following steps: removal of a signal peptide, glycosylation, and splitting of the polypeptide chain by active arginine esterase in the secretory granules or outside the cell.

This work was undertaken to determine the identity of the major androgen-dependent 15,000 molecular weight protein previously observed on SDS polyacrylamide gel electrophoresis of both dog prostate cytosol and dog seminal plasma. The protein was identified as one of the two chains of arginine esterase on the basis of its ability to bind 3H-diisopropylphosphofluoridate (DFP), an active site titrant of serine proteases. Furthermore, since the other polypeptide chain was heterogeneous, at least five distinct peaks of arginine esterase activity could be separated by chromatofocusing under nonreducing conditions. The molecular weight of the seminal plasma protein was estimated at 29,500 by Sephadex G-100 gel filtration, and at 25,000 by SDS polyacrylamide gel electrophoresis in the absence of mercaptoethanol. In the presence of mercaptoethanol, two major peaks were observed with molecular weights of 15,000 and 14,000. These results show that arginine esterase of dog seminal plasma is a serine protease composed of two different chains linked by disulfide bridges. One of the chains contains the reactive serine group. The other one is probably glycosylated since it presents several isoelectric points.

We have developed a radioimmunoassay to determine serum levels of arginine esterase, the major secretory product of the canine prostate. Although the antibodies recognized other related proteins in immunoblotting conditions, particularly in the pancreas, the radioimmunoassay appeared specific for arginine esterase of prostatic origin and reacted only slightly or not at all with cytosolic proteins from salivary glands, liver, skeletal muscles, and kidney. In serum samples, the sensitivity limit of the assay was about 2.5 ng per ml. The interassay and intraassay variations for serum samples that contained concentrations of arginine esterase ranging from 6 to 21 ng per ml were, respectively, 14% and 24%. Sera from 30 adult males and 14 adult females contained, respectively, 71 +/- 9 and 5.3 +/- 0.4 ng per ml. In a group of 12 dogs treated daily for 36-39 days with 50 micrograms of D-Trp6-luteinizing hormone-releasing hormone ethylamide, the concentrations of arginine esterase in the serum decreased progressively and became identical with those observed in the females. These results constitute the first evidence for the presence of a serum marker of the normal prostate in an experimental model. It should be particularly useful to test the effects of various agents on the prostate.

At equimolar ratio of enzyme/substrate, actin, tropomyosin, fibronectin and myosin were extensively hydrolyzed during an incubation of one hour at 37 degrees C. Dog serum albumin, ovalbumin, bovine gamma-globulin and human prostatic acid phosphatase were not hydrolyzed. The activity of arginine esterase towards actin at pHs 6.5, 7.1 and 7.6 was respectively 60, 74 and 84% of the one found at optimum pH 8.2. The cleavage products of actin by arginine esterase and trypsin were similar although trypsin activity was 5000-fold higher. Kallikrein produced a major fragment of actin not observed with arginine esterase and trypsin. It is concluded that arginine esterase has a low trypsin-like activity towards structural proteins and that this activity may have a physiological significance.

Canine prostate fluids and seminal plasma contain a major androgen-dependent protein which was identified as a proteolytic enzyme exhibiting an Arg-esterase activity. This protease, as characterized, is shown to be present as a two-chain structure held together by at least one disulfide bridge and composed of approximately 220 amino acids. Amino acid sequence determination of both chains has revealed a clear homology to other known amino acid sequences of serine proteases. Furthermore, the comparison of the presented 58 amino acids of the Arg-esterase with the other sequences revealed a very strong homology (larger than 50%) to members of the kallikrein family. The two chain structure could thus result from autolysis of a single chain enzyme in the 'kallikrein autolysis loop'. Amino acid composition of the canine prostatic enzyme suggests that it is related, but not identical, to pancreatic canine kallikrein.