Distribution and properties of endo-oligopeptidases A and B in the human neuroendocrine system M. S.
Medeiros, N.
Departments of Clinical
Iazigi,
A. C. M. Camargo and E. B. Oliveira Medicine* and Biochemistryî of the Faculty of Medicine of Ribeiräo Preto,
University of Säo Paulo, Sao Paulo, Brazil tDepartment of Pharmacology, Institute of Biomedicai Sciences, University of Säo Paulo, Säo Paulo, Brazil (M. S. Medeiros to whom requests for offprints should be addressed is now at Department of Biochemistry and Molecular Biology, University of Leeds, Leeds ls2 9jt, U.K.) received
25 March 1992
ABSTRACT
The thiol activated endo-oligopeptidases A and B were studied in the soluble fraction of human hypothalamus and various endocrine glands. For the identification, characterization and purification of the enzymes, Z-Gly-Pro-NH-Np and bradykinin were used as substrates. Endo-oligopeptidase B showed a molecular mass ranging from 55\m=.\5 to 65\m=.\5kDa and isoelectric point from 4\m=.\7to 4\m=.\95.Its activity in tissues was highest in the testis, with intermediate levels in the thyroid, neurohypophysis, adenohypophysis and hypothalamus and the lowest activity in the pineal gland. Endo-oligopeptidase A, 467-fold purified by immunoaffinity chromatography, exhibited a molecular mass of 65\m=.\5 kDa in the adenohypophysis but
58\m=.\5 kDa in other tissues. The isoelectric point ranged from 5\m=.\22 to 5\m=.\50. High endo-oligopeptidase A activity was observed in the adenohypophysis, testis and hypothalamus with lesser activity in the neurohypophysis and thyroid and the lowest in the pineal gland. Endo-oligopeptidase A cleaved the bonds Phe-Ser of bradykinin, Met-Arg of BAM-12P and Arg-Arg of neurotensin as described for rabbit brain and heart and bovine brain enzymes. This work shows that endo-oligopeptidase A also hydrolysed the bonds Tyr\x=req-\ Gly of LH-releasing hormone, Pro-Phe of angiotensin I and Tyr-Ile of angiotensin II. Journal of Endocrinology (1992) 135, 579\p=n-\588
INTRODUCTION
another endopeptidase, now designated endopeptidase-24.15 (EC 3.4.24.15), which is highly active in rat brain and pituitary (Orlowski, Michaud & Chu, 1983) and which has recently been cloned and sequenced (Pierotti, Dong, Glucksman et al 1990). The conclusion that endopeptidase-24.15 is identical to endo-oligopeptidase A and to another activity, 4phenylazobenzyloxycarbonyl-peptidase (Barrett & Tisljar, 1989; Tisljar, Camargo, Costa & Barrett, 1989; Barrett & Brown, 1990) was later contested by Camargo (1991), who found that biochemical dis¬ tinctions can be made between endo-oligopeptidase A and endopeptidase-24.15, including physical separa¬
endo-oligopeptidases A (EC 3.4.22.19) and (prolyl-endopeptidase, EC 3.4.21.26) have been implicated in the processing and metabolism of neuroendocrine peptides. Endo-oligopeptidase A was originally described as a cysteine endopeptidase (kininase) present in a brain cytosol preparation which was capable of hydrolysing the Phe5-Ser6 bond of bradykinin (Camargo, Shapanka & Greene, 1973). Subsequently, endo-oligopeptidase A was shown to Both
hydrolyse the Arg8-Arg9 bond of neurotensin (Camargo, Caldo & Emson, 1983) and to generate [Met]enkephalin from several enkephalin- containing peptides such as the 12-residue opioid peptide from bovine adrenal medulla (BAM-12P) by cleavage at the Met5-Arg6 position (Camargo, Ribeiro & Schwartz, 1985; Camargo, Oliveira, Toffoletto
1987). Endo-oligopeptidase
A resembles in
et al
specificity
tion. The cytosol of nervous tissue also contains another thiol-activated endopeptidase named endooligopeptidase (Oliveira, Martins & Camargo, 1976) further characterized as prolyl-endopeptidase (Greene, Spadaro, Martins et al 1982) which hydro¬ lyses bradykinin at the Pro7-Phe8 bond (Oliveira et al
1976) as well as the carboxyl group of proline in a number of biologically active peptides (Orlowski, Wilk, Pearce & Wilk, 1979; Greene et al 1982). An important feature of both endo-oligopeptidases A and or prolyl-endopeptidase is that they are selec¬ tive for peptides smaller than 20 amino acid residues, the size of the majority of known biologically active peptides. The abundance of these enzymes in brain and their substrate specificity indicates a key role in the processing and metabolism of neuroendocrine
peptides. Furthermore, prolyl-endopeptidase appears to be under the control of gonadal steroids (Jaramillo, Rodrigues & Camargo, 1984). The present work provides the first information on
the distribution of these enzymes in the human neuro¬ endocrine system as well as some of their properties. Some neuropeptide-degrading peptidases, for exam¬ ple angiotensin-converting enzyme (kininase II, EC 3.4.15.1), exist in distinct isoforms of different molecular weights in different organs (Stewart, Weare & Erdös, 1981; Lanzillo, Dasarathy, Stevens et al 1985 ; Lanzillo, Stevens, Dasarathy et al 19856). We have therefore also examined the molecular weights, as well as the isoelectric points, of both endooligopeptidases A and in each of the human tissues examined here.
MATERIALS AND METHODS
Materials Endocrine glands (testis, thyroid, pituitary, pineal) and hypothalami were obtained from ten male indivi¬ duals aged 18 to 37 years, submitted for post-mortem examination at the Instituto Médico Legal de Porto Alegre, Rio Grande do Sul, Brazil after violent death. These organs were collected from 7 to 11 h after death, freed from extraneous tissue by dissection and stored at 20 °C before the enzyme extraction step.
Bradykinin (BK), BK(l-5), BK(6-9) and angio¬ tensins I and II (AI and All) were synthesized by —
Dr A. C. M. Paiva and L. Juliano (Escola Paulista de Medicina, Säo Paulo, Brazil). BAM-12P, [Met]enkephalin, neurotensin (NT), NT(l-9), NT(9-13) and luteinizing hormone-releasing hormone (LHRH) were purchased from Cambridge Research Biochemicals (Northwich, Cheshire, U.K.). Ribonuclease A, chymotrypsinogen, ovalbumin, bovine serum albumin, Sephacryl S-200, DEAESephacel, Sepharose-4B and Sephadex G-25 were obtained from Pharmacia Fine Chemicals (Piscata-
NJ, U.S.A.). Blue-Sepharose, dithiothreitol, ampholyte mixture components, and Tris(hydroxymethyl) aminomethane were purchased from Sigma Chemical way,
Company (St Louis, MO, U.S.A.). Amino acid and peptide analyser reagents were obtained from Pierce Chemical Co. (Rockford, IL, U.S.A.). pBondapak
Cíe columns were obtained from Waters Associates (Milford, MA, U.S.A.).
High-performance liquid chromatography (HPLC) was purchased from Laboratory Data Control (Milford, MA, U.S.A.). W3 ion-exchange resin was purchased from Beckman Instrument Co (Fullerton, CA, U.S.A.). apparatus
Methods Extraction
of enzymes and partial purification of endo-oligopeptidase A The organs were homogenized in distilled water (1:5, w/v) in an Ultra-Turrax homogenizer in an ice bath, at 20 000 r.p.m. for 30 s. Following centrifugation of the homogenate at 25 000 g for 60 min at 4 °C, the pH of the supernatant was adjusted to 5 by dropwise addition of acetic acid (0-5 mol/1). After 1 h at 4 °C, the precipitate was removed by centrifugation (900 g x 15 min) and more than 85% of the total kininase activity was recovered in the supernatant fraction which was neutralized by the addition of Tris-base solution (1 mol/1) and percolated through a blue-Sepharose column, equilibrated in Tris-HCl buffer (0-025 mol/1) pH7-5 to remove albumin (Böhme, Kopperschläger, Schulz & Hofmann, 1972) and particulate matter. The immunoaffinity resin prepared with rabbit polyclonal antibodies raised against bovine endooligopeptidase A has been described previously (Camargo et al. 1987). The immunoaffinity chro¬ matography (Fig 1) was carried out at 4 °C in a column (0-9 10 cm) with capacity to adsorb approxi¬ mately ten units of bovine brain enzyme activity. Following sample application, the unbound material was thoroughly removed by washing the resin with six column volumes of running buffer (Tris-HCl buffer (0-025 mol/1) pH7-5). The adsorbed material was then eluted with running buffer containing Nal (2 mol/1), dithiothreitol (0-5 mmol/1) and 20% glycerol. The active fractions were pooled and equilibrated by gel filtration on a Sephadex G-25 column ( 1 5 x 40 cm) with Tris-acetate buffer (0-025 mol/1) pH 7-0. The endo-oligopeptidase A preparations obtained from all tissues by the immunoaffinity procedure were stored at 20 °C in the presence of 20% glycerol until use. ·
-
Kininase
activity determination
activity was determined by using an enzymatic bioassay with isolated guinea-pig ileum (Camargo, Ramalho Pinto & Greene, 1972). The kininase
5) formed in each reaction was measured using an amino acid analyser, as described previously (Alonzo & Hirs, 1968; Carvalho & Camargo, 1981). One mU endo-oligopeptidase A activity was defined as the amount of enzyme required to release 1 nmol BK(1 5) per min under the conditions described.
Endo-oligopeptidase A cleavage of regulatory peptides
Elution volume
(ml)
Immunoaffinity chromatography was carried in a column (0-1x10 cm) with capacity to adsorb approximately 10 units of bovine brain enzyme activity. Following sample application, the unbound material was thoroughly removed by washing the resin with six column volumes of running buffer (Tris-HCl buffer (0025 mol/1), pH 7-5). The adsorbed material was then eluted with running buffer containing Naf (2 mol/1), dithiothreitol (5 mmol/1) and 20% glycerol. figure
1.
out at 4 °C
Endo-oligopeptidase A assay determination endo-oligopeptidase A activity in tissue homo¬ genates was determined by measuring the amount of the fragment BK(l-5) formed when bradykinin was used as the substrate. Bradykinin (100 pmol/l) was The
incubated at 37 °C for 30 min in Tris-HCl buffer (005 mol/1) pH 7-5 containing NaCl (0-1 mol/1) and dithiothreitol (0-5 mmol/1) with an appropriate aliquot of the 25 000 g x 60 min supernatant fraction obtained from the hypothalamus, testis, thyroid, pineal, anterior and posterior pituitaries, to result in 50-80% hydrolysis of the peptide. The reactions were terminated by acidification and the fragment BK(1-
The immunoaffinity purified endo-oligopeptidase A from human testis (1075 mU/mg) was used for verify¬ ing the cleavage site of the enzyme on the following peptides: BK, BAM-12P, NT, LHRH, AI and AIL The peptides (10-30 pmol/l) were incubated with a given amount of the enzyme (0-25-6 mU) for 3045 min at 37 °C in Tris-HCl buffer (0-025 mol/1) pH7-5, containing dithiothreitol (0-5 mrhol/1). The reaction was terminated by addition of 5µ1 50% (w/v) H3PO4. The products formed in each reaction were separated by reverse-phase HPLC on a C-18 pBondapak column (0-39x30 cm) developed by an initial 5-min isocratic elution in 0-1% H3PO4, pH 2-7 containing 5% (v/v) acetonitrile followed by a 20-min linear gradient of acetonitrile up to 35% (v/v) in 0-1% H3PO4. The flow rate was kept constant at 2 ml/min and the emerging peptides were monitored by their absorbance at 220 nm. The fragments generated by endo-oligopeptidase A digestion of the regulatory peptides were identified either by comparing their retention times on the HPLC with those of synthetic standard peptides or by subjecting the corresponding peptides obtained by HPLC to acid hydrolysis (Simpson, Neuberger & Liu, 1976) followed by amino acid anaylsis (Alonzo & Hirs, 1968). The cleavage sites on the parent peptides could be readily assigned by determining the amino acid composition of the derived fragments.
Endo-oligopeptidase
assay
The endo-oligopeptidase activity in tissue homogen¬ ates was determined spectrophotometrically by meas¬ uring the amount of />-nitroaniline released from the chromogenic substrate Z-Gly-Pro-NH-Np. Z-GlyPro-NH-Np (200 pmol/l) was incubated at 37 °C in Tris-HCl buffer (0-025 mol/1) pH7-5, containing dithiothreitol (0-5 mmol/1) with an appropriate aliquot of 25 000 g 60 min supernatant fraction obtained from the hypothalamus, testis, thyroid, pineal, anterior and posterior pituitaries to hydrolyse 5-10% of the substrate in 10-15 min. After terminat¬ ing the reaction by acidification with 0-1 ml 50% acetic acid, the product formed was determined by its absorbance at 410 nm (Erlanger, Kokowsky &
Cohen, 1961). Endo-oligopeptidase
defined
as
the amount of enzyme
(1 mU)
required
was
to release
1 nmol
^-nitroaniline/min
under the conditions
described.
as
Physicochemical characterization Molecular
A in bradykinin, was measured described in the Materials and Methods. The high¬ est enzymatic activity per g wet tissue was found in the anterior pituitary followed by the testis and hypo¬ thalamus (Table 1). The lowest activity was found in the pineal gland corresponding to 12% ofthat found in the anterior pituitary. When the activity was expressed relative to the protein content, the tissue displaying the highest value was the hypothalamus and the lowest was the thyroid, corresponding to 6% of the activity found in the hypothalamus.
endo-oligopeptidase
mass
determination. The MT values of both
endo-oligopeptidases A and were estimated by gel filtration (Andrews, 1970) on a Sephacryl S-200 column (1-2 x 110 cm) equilibrated and developed at 4 °C with Tris-HCl buffer (0-05 mol/1) pH 7-5, con¬ taining NaCl (0-1 mol/1) at a flow rate of 6-0 ml/h.
The Mr markers used were: bovine serum albumin (67 kDa), ovalbumin (45 kDa), chymotrypsinogen (25 kDa) and ribonuclease A (13-7 kDa). The endooligopeptidase A sample for each tissue was obtained by immunoaffinity chromatography, as described, and its elution volume measured by following its kininase activity. The source of endo-oligopeptidase from each tissue was the non-adsorbed protein fraction from the immunoaffinity column and the correspond¬ ing elution volume was determined by following the activity towards the specific substrate Z-Gly-Pro-NH-
Np.
Isoelectric point determination. The pi values for both endo-oligopeptidases A and were determined by isoelectric focusing (Prestidge & Hearn, 1979) on granulated gel using an ampholyte mixture for gener¬ ating a stable pH gradient in the range of 3 to 10. The sources of endo-oligopeptidase A and were the pH 5 supernatant fraction and the non-adsorbed protein fraction from the immunoaffinity column correspond¬ ing to each tissue respectively. The affinity purified fraction of endo-oligopeptidase A was inactivated on isoelectric focusing. The fractions containing the focused endo-oligopeptidase A activity were deter¬ mined by a two-step procedure, in which the kininase activity of each fraction was measured by the bioassay with isolated guinea-pig ileum first, followed by moni¬ toring the ability of the active fractions to release stoichiometric amounts of the bradykinin fragments BK(1 5) and BK(6 9) on HPLC. The fractions con¬ taining the focused endo-oligopeptidase activity were determined by their activities towards the sub¬
Z-Gly-Pro-NH-Np. Determination of protein concentration
1. Distribution of endo-oligopeptidases A and activities in different human neuroendocrine tissues
table
Endo-oligopeptidase Endo-oligopeptidase A activity activity mU/mg mU/g mU/mg mu/g protein tissue protein tissue Source Testis
Thyroid Hypothalamus Anterior pituitary Posterior pituitary Pineal
1-8 0-2 3-4 1-3 0-7 0-7
65-5 25-9 63-1 82-4 38-0 9-8
82-3 4-5 141 6-2 7-6 6-7
2927-8 577-8 258-9 385-4 416-4 800
endo-oligopeptidase A activity in tissue homogenates was determined amino acid analyser by measuring the amount of the fragment BK(l-5) formed from the hydrolysis of bradykinin (BK). 1 mU endooligopeptidase A is defined as the amount of enzyme required to release 1 nmol BK(1 5)/min under the conditions described in Materials and Methods. Z-Gly-Pro-NH-Np was used to determine the activity of endooligopeptidase B. 1 mU endo-oligopeptidase is defined as the amount of enzyme required to release 1 nmol /7-nitroaniline/min under the conditions The
on an
described in Materials and Methods.
Another enzyme that contributes to bradykinin inactivation by tissue peptidases is prolyl-endopeptidase or endo-oligopeptidase B. This enzymatic activity can be specifically measured in tissue homogenates by using the chromogenic substrate Z-Gly-Pro-NH-Np. The highest activity per g wet tissue and per mg protein was found in the testis, being 5 to 25 times higher than the activity found in other tissues
(Table 1).
strate
Protein concentration was determined by the proce¬ dure of Lowry, Rosebrough, Farr & Randall (1951) using bovine serum albumin as the standard. RESULTS
Distribution of endo-oligopeptidases A and neuroendocrine system
in human
The hydrolysis of the Phe5-Ser6 bond of bradykinin, which typically represents the only site of cleavage by
Purification of endo-oligopeptidase A by immunoaffinity chromatography from human endocrine glands and hypothalami The cross-reactivity of the antibody raised against endo-oligopeptidase A from bovine brain against the human enzyme permitted purification of the human endo-oligopeptidase A by immunoaffinity chromatography.
The 25 000 g x 60 min supernatant fraction of the tissue homogenate was subjected to immunoaffinity chromatography resulting in approximately 500-fold
table
2. Purification of endo-oligopeptidase A from human testis Total
protein (mg)
Fractions 25 000 g supernatant pH 5 supernatant
81-0 45-1 30-8 0-2
Blue-Sepharose Immunoaffinity chromatography
Protein recovery
(%) 1000 55-7 38-0 0-25
(mU)
activity
Specific activity (mU/mg protein)
184-5 155-1 1400 215-0
1000 84-1 75-9 116-5
2-3 3-4 4-5 10750
Total kininase
activity
Kininase
Purification
(fold) 10 1-5 2-0 467-4
The above results are provided from a 10 g testis homogenate. The total volume of the homogenate was 55 ml and the total volume of 25 000 g 60 min supernatant fraction was 50 ml. From the latter was taken an aliquot of 15 ml for endo-oligopeptidase A purification. The enzymatic reaction was followed by the kininase assay described in Materials and Methods. Bradykinin (5 µ / ) was incubated with each fraction at 37 °C in Tris-HCl buffer (0-05 mol/1), pH 7· 5, containing NaCl (0· 1 mol/1) and dithiothreitol (0- 5 mmol/1). The concentration of enzyme was selected to result in 50 80% of bradykinin inactivation in 10 mins.
purification with
total recovery of enzyme activity. Table 2 illustrates the purification of endo-oligo¬ peptidase A from testis. As the amount of enzyme applied to the column (220 mU) was less than the immunoadsorbent capacity (10 units) the nonadsorbed protein was devoid of activity able to hydrolyse the Phe5-Ser6 bond of bradykinin. Endooligopeptidase A was completely recovered from the column in an active form after the addition of the chaotropic agent Nal (2 mol/1).
table
3. Molecular mass determination of endoand from human hypothalamus and
oligopeptidase A endocrine glands Source
A
Endo-oligopeptidase (Mr)
Endo-oligopeptidase
Testis
58 58 58 65 58
500 ± 1500 500 ± 1500 500 ± 1500 500 ± 1500 500 ± 1500
55 62 62 58 65
Thyroid Hypothalamus Anterior pituitary Posterior pituitary
(Mr)
500 ± 1500 000 ± 2000 000 ± 2000 500 ± 1500 500 ± 1500
endo-oligopeptidase A activity obtained by immunoaffinity chromatography and endo-oligopeptidase present in the non-adsorbed protein fraction from the immunoaffinity column, of each tissue, were applied separately to a Sephacryl S-200 column (1-2x110 cm) equilibrated and developed at 4 °C in Tris-HCl buffer (005 mol/1), pH 7-5, containing NaCl (0-1 mol/1) at a flow rate of 6 ml/h. Fractions of 1-5 ml were collected and assayed for kininase activity. The hydrolysis of Z-Gly-ProNH-Np was also assayed in all fractions. The M, of the enzymes were estimated according to Andrews (1970). Haifa fraction volume was taken as the margin of error in molecular mass determination. The
Determination of \f and pi of endo-oligopeptidases from human endocrine glands and
A and
hypothalami Endo-oligopeptidase A purified by immunoaffinity chromatography and endo-oligopeptidase present in the non-adsorbed protein fraction from the immuno¬ affinity column of each tissue were subjected separ¬ ately to gel filtration on Sephacryl S-200. The gel
filtration results indicated
an
apparent Mr for endo-
oligopeptidase A of 58 500 for the testis, thyroid, hypothalamus and posterior pituitary and 65 500 for the anterior pituitary (Table 3). The values for endooligopeptidase varied from 55 500 for the testis to 65 500 for the posterior pituitary (Table 3). Endooligopeptidase A from the testis, thyroid and hypo¬ thalamus was isoelectric in the pH range 5-22-5-50 whereas endo-oligopeptidase of the same tissues was isoelectric at a pH significantly lower: 4-70-4-95 (Fig. 2). Sites of cleavages
on neuropeptides by endooligopeptidase A Immunoaffinity purified endo-oligopeptidase A obtained from human endocrine glands and hypothal¬
ami was incubated with BK, NT, BAM-12P, LHRH, AI and AIL The products of peptide hydrolysis were separated by HPLC (Fig. 3) and collected separately. The peptides corresponding to each peak were
subjected to acid hydrolysis followed by amino acid analysis. The results indicate that BK, NT, BAM-12P and AI and All were hydrolysed at a single peptide bond. LHRH appeared to be cleaved at more than one peptide bond. Table 4 illustrates the cleavage sites of neuropeptides by glandular and hypothalamic endo-oligopeptidase A. DISCUSSION
Purification methods: affinity chromatography In this work a simplified purification procedure was followed for separating endo-oligopeptidase A from endo-oligopeptidase producing the former in a highly purified form. The procedure involved suc¬ cessive affinity chromatography steps on blueSepharose and an immunoadsorbent column. The endo-oligopeptidase A preparation obtained was devoid of other peptidase activities as indicated by the stoichiometric recovery of the fragments derived from
Testis
Testis
Fraction
Fraction
2. The pf values of endo-oligopeptidase A and B were determined by isoelectric focusing granulated gel, using an ampholyte mixture to generate a stable pH gradient in the range of 3-10. The activity of endo-oligopeptidase A was determined by a two-step procedure: the bioassay of isolated guinea-pig ileum followed by monitoring the ability of the active fractions to release stoichiometric amounts of bradykinin fragments BK(l-5) and BK(6-9) on HPLC. The activity of endo-oligopeptidase was verified with Z-Gly-Pro-NH-Np as substrate. figure on
BK, BAM-12P and NT. When BK was completely hydrolysed only the complementary fragments BK(l-5) and BK(6 9) were recovered, indicating the absence of aminopeptidase and carboxypeptidase activities as well as post-proline endopeptidase or and angiotensin-converting endo-oligopeptidase enzyme. The cross-reactivity of the antibody prepared against bovine brain endo-oligopeptidase A with the human enzyme has now allowed purification of the human endo-oligopeptidase A. Highly purified endo-oligopeptidase A (approximately 500-fold purification) was obtained after immunoaffinity chromatography, leading to virtually complete recov¬ ery of activity in all the tissues used.
Characterization of human endo-oligopeptidases A and Endo-oligopeptidases A and differed in isoelectric point (about 5-4 and 4-8 respectively), whether from
or testis, as previously found for the rabbit brain enzymes. The observed differences between the pi of these enzymes make the isoelectric focusing a useful method for separating these two enzymatic activities. As Table 3 shows, the Mr values of the two enzymes from human hypothalamus and endocrine glands are similar to each other and do not differ much with tissue. Nevertheless anterior pituitary tissue does
thyroid, hypothalamus
possess
a
significantly larger endo-oligopeptidase
A
004
0-04-
004
5
10
15
Time
(min)
20
neurotensin (NT), BAM-12P, LH-releasing hormone (LHRH), angiotensin incubated with 0-75-6-0 mU endo-oligopeptidase A purified by immunoaffinity chromatography (specific activity towards BK:1075 mU/mg) at 37 °C in Tris-HCl buffer (0025 mol/1), pH 7-5, containing dithiothreitol (005 mmol/1) for 30-45 min. The hydrolysis products of the substrates were analysed by reverse-phase HPLC as described in Materials and Methods. Typical chromatograms show the production of: (a) BK(l-5), BK(6-9) from BK; (b) NT(l-8) and NT(9-13) from NT; (c) BAM-12P(6-12) and [Met]enkephalin (ME) from BAM-12P; (d) LHRH(l-5) and LHRH(6 10) from LHRH; (e) AI(l-7) and Al(8-10) from AI; (/) AII(l-4) from AH. figure
3.
(AI) and
Bradykinin (BK), (10-30 nmol)
were
and testis a significantly smaller endo-oligopeptidase B, than the other tissues. These small differences in Mr from enzymes of various sources may be due to the presence of isoenzymatic forms of each enzyme in different tissues. There are precedents for the occurrence of tissue-specific forms of peptidases of differing molecular size. For example, angiotensinconverting enzyme exists as a polypeptide of Mr 180 000 in kidney and lung, 180 000 and 170 000 in brain, and as a truncated form of Mr 110 000 in testis (Stewart et al 1981; Lanzillo et al. 1985 ) which arises by alternative splicing (Kumar, Kusari,
Roy et al 1989). Alternatively, spliced forms of endopeptidase-24.11 of differing molecular weights may
also exist (Llorens-Cortes, Giros & Schwartz, 1990). It is noteworthy that molecular exclusion chro¬ matography cannot be used to separate these enzymes, because their Mr values are so close.
Hydrolysis of neuroendocrine peptides by endo-oligopeptidase A from human testis The use of HPLC and amino acid analysis allowed us to show that testicular endo-oligopeptidase A,
table
4.
Cleavage
sites of peptides
hydrolysed by endo-oligopeptidase A from human testis
P4 P, P2 P, Pi' P2' P3' P/
I Arg-Pro-Pro-Gly-Phe-Ser-Pro-Phe-Arg I Tyr-Gly-Gly-Phe-Met-Arg-Arg-Val-Gly-Arg-Pro-Glu I pGlu-Leu-Tyr-Glu-Asn-Lys-Pro-Arg-Arg-Pro-Tyr-Ile-Leu I pGlu-His-Trp-Ser-Tyr-Gly-Leu-Arg-Pro-GlyNH2 1 Asp-Arg-Val-Tyr-Ile-His-Pro-Phe-His-Leu ï Asp-Arg-Val-Tyr-Ile-His-Pro-Phe
bradykinin BAM-12P
neurotensin LHRH
angiotensin
I
angiotensin
II
The table shows the amino acid sequences of peptides hydrolysed by endo-oligopeptidase A. The cleavage site was determined by combination of HPLC and amino acid analysis of the hydrolysis products. The residue positions (Pi_4 and P'1-4) on either side of the scissile bond are named using the nomenclature of Schechter & Berger (1967). The observed cleavage site is indicated by an arrow between the two amino acids in bold print in positions Pi and Pi'. BAM-12P is the 12-residue opioid peptide from bovine adrenal medulla. LHRH is LH-releasing hormone.
purified by immunoaffinity chromatography, hydro¬ lyses the Phe-Ser bond of bradykinin, the Met-Arg bond of BAM-12P and the Arg-Arg bond of neuro¬ tensin. LHRH, AI and All were also hydrolysed at the Tyr-Gly, Pro-Phe and Tyr-Ile bonds respectively. The presence of bradykinin in the incubation mixture inhibited the hydrolysis of all regulatory peptides used in this work, supporting the occurrence of a single activity hydrolysing all the peptides. There is no simple predictor of the cleavage sites within peptides by endo-oligopeptidase A. From analysis of the sites of peptide hydrolysis (Table 4) it can be suggested that there is a chymotryptic-like specificity for hydrolysis of BK, LHRH and All but, apparently, a tryptic-like specificity for NT. The hydrolysis by endo-oligopeptidase A of the Pro-Phe bond of AI is not due to contamination by prolyl endopeptidase or endo-oligopeptidase since this enzyme was not detectable after 24 h by using as sub¬ strate Z-Gly-Pro-NH-Np or even after 100% hydro¬ lysis of BK. Concerning the secondary specificity of endooligopeptidase A, the alignment of amino acid resi¬
dues does not allow verification of whether the enzyme has any structural requirements conforming to the model of Schechter & Berger (1967). Other features of the substrate such as size, charge and con¬ formation may be more important than the simple alignment of the peptide chain. The broad specificity of endo-oligopeptidase A suggests a general role in the metabolism of regulatory peptides with different target peptide substrates in the various endocrine organs.
Relationship of endo-oligopeptidase A to other endopeptidases Clearly there are considerable similarities in substrate specificity between endo-oligopeptidase A and endo-
peptidase-24.15. The latter enzyme, however, is class¬ ified as a metallopeptidase and contains the amino acid sequence typical of a zinc-dependent metallopeptidase (Pierotti et al 1990). However, the presence of a cysteine residue close to the catalytic centre of endopeptidase-24.15 might explain the activation of the enzyme by 2-mercaptoethanol and its inhibition by pmercuribenzoate and TV-ethylmaleimide. Although it has been claimed that endo-oligopeptidase A and endopeptidase-24.15 may be identical (Barrett & Brown, 1990), an antibody to endo-oligopeptidase A does not precipitate endopeptidase-24.15 (Toffoletto, Metters, Oliveira et al 1988) and the two activities are separable by ion exchange chromatography (Camargo, 1991 ). The relationship between these two
activities remains unresolved, therefore, and must await the cloning and sequencing of endo-oligopep¬ tidase A. The involvement of cysteine in the catalytic mechanism also needs to be explored in detail.
Distribution of endo-oligopeptidase A and proline endopeptidase in human hypothalamus and endocrine glands: physiological role The distribution of endo-oligopeptidases A and differed among the organs studied. The endooligopeptidase A predominated in anterior pituitary, testis and hypothalamus and endo-oligopeptidase in testis, thyroid and posterior pituitary. Recently, a number of regulatory peptides such as an LHRH-like peptide and ß-endorphin have been demonstrated in the testes of various animals (Sharp, Pekari, Meyer & Hershman, 1980; Sharp & Pekari, 1981; Sharpe, Fraser, Cooper & Rommerts, 1982).
[Met]enkephalin,
[Leu]enkephalin,
ß-endorphin,
substance and calcitonin were also found in human sperm and semen, suggesting that they may play an important regulatory function in the male reproduc¬ tive system, including sperm motility (Rama Sastry,
Janson, Owens & Tayeb, 1982; Fraioli, Fabbri,
Gnessi et al 1984). Studies in rats on the steroid regulation of hypo¬ thalamic LHRH degradation (Advis, Krause & McKelvy, 1982, 1983) have shown that LHRH degra¬ dation products, in all physiological situations, revealed cleavage of the Tyr-Gly bond. The same hydrolysis site was observed in this work for endooligopeptidase A on this peptide. Jaramillo et al (1984) have suggested the participation of adeno¬ hypophysis prolyl endopeptidase in the control of gonadotrophin secretion. Therefore, it is possible that both enzymes could be implicated in the feedback mechanisms of the hypothalamic-hypophysialgonadal axis. Finally, the action of both enzymes on oligopeptides, together with their distribution in the human neuroendocrine system, suggest that they have specific functions related to regulatory peptide metabolism. These observations indicate that studies in vivo and further cellular and subcellular studies are needed to define their physiological role in the metabolism of
peptides.
ACKNOWLEDGEMENTS
M.S.M. was in receipt of a Fellowship from CAPESBrazil. Mrs E. S. Gray is thanked for efficient sec¬ retarial assistance. REFERENCES
Advis, J. P., Krause, J. E. & McKelvy, J. F. (1982). Luteinizing
hormone-releasing hormone peptidase activities in discrete hypothalamic regions and anterior pituitary of the rat : appar¬ ent regulation during the prepubertal period and first estrous cycle at puberty. Endocrinology 110, 1238-1245. Advis, J. P., Krause, J. E. & McKelvy, J. F. (1983). Evidence that endopeptidase-catalysed luteinizing hormone-releasing hormone cleavage contributes to the regulation of median eminence LHRH levels during positive steroid feedback. Endocrinology 112, 1147-1150. Alonzo, . & Hirs, C. H. W. (1968). Automation of sample appli¬ cation in amino acid analyzers. Analytical Biochemistry 23, 272-288.
Andrews, P. (1970). Estimation of molecular size and molecular
weights of biological compounds by gel filtration. In Methods of Biochemical Analysis, vol. 18, pp. 1-53. Ed. D. Glick. New York : Wiley-Interscience. Barrett, A. J. & Brown, M. A. (1990). Chicken liver Pz-peptidase, a thiol-dependent metallo-endopeptidase. Biochemical Journal
271,701-706. Barrett, A. J. & Tisljar, U. (1989). The activities of 'Pz-peptidase' and 'endopeptidase-24.15' are due to a single enzyme. Biochemical Journal 261, 1047-1050. Böhme, . J., Kopperschläger, G., Schulz, J. & Hofmann, E. (1972). Affinity chromatography of phospho-fructokinase using Cibacron blue F3G-A. Journal of Chromatography 69, 209-214.
Camargo, A. C. M. (1991). Distinction between endooligopeptidase A (EC 3.4.22.19) and soluble metalloendopeptidase (EC 3.4.24.15). Biochemical Journal 277, 294-295. Camargo, A. C. M., Caldo, H. & Emson, P. C. (1983). Degrada¬ tion of neurotensin by rabbit brain endo-oligopeptidase A and endo-oligopeptidase (proline-endopeptidase). Biochemical and Biophysical Research Communications 116, 1151-1159. Camargo, A. C. M., Oliveira, E. B., Toffoletto, O., Metters, . M. & Rossier, J. (1987). Brain endo-oligopeptidase A, a putative enkephalin-converting enzyme. Journal of Neurochemistry 48, 1258-1263.
Camargo, A. C. M., Ramalho Pinto, F. J. & Greene, L. J. (1972). Brain peptidases : conversion and inactivation of kinin hor¬ mones. Journal of Neurochemistry 19, 37-49. Camargo, A. C. M., Ribeiro, M. J. V. F. & Schwartz, W. N. (1985). Conversion and inactivation of opioid peptides by rab¬ bit brain endo-oligopeptidase A. Biochemical and Biophysical Research Communications 130, 932-938. Camargo, A. C. M., Shapanka, R. & Greene, L. J. (1973).
Preparation, assay and partial characterization of a neutral endopeptidase from rabbit brain. Biochemistry 12, 1838-1844. Carvalho, K. M. & Camargo, A. C. M. (1981). Purification of rabbit brain endo-oligopeptidases and preparation of antienzyme antibodies. Biochemistry 20, 7082-7088. Erlanger, . F., Kokowsky, . & Cohen, W. (1961). The prep¬ aration and properties of two new chromogenic substrates of trypsin. Archives of Biochemistry and Biophysics 95, 271-278. Fraioli, F., Fabbri, ., Gnessi, L., Silvestroni, L., Moretti, C, Redi, F. & Isidori, . (1984). Beta-endorphin, Met-enkephalin,
and calcitonin in human semen : evidence for a possible role in human sperm motility. Annals of the New York Academy of Sciences 438, 365-370. Greene, L. J., Spadaro, A. C. C, Martins, A. R., Jesus, W. D. P. & Camargo, A. C. M. (1982). Brain endo-oligopeptidase B: a post-proline cleaving enzyme that inactivates angiotensin I and II. Hypertension 4, 178-184. Jaramillo, P. L., Rodrigues, J. A. & Camargo, A. C. M. (1984). Effect of gonadal steroids on hypothalamus and anterior pitu¬ itary endo-oligopeptidase (proline-endopeptidase) activity in castrated female rats. Peptides 5, 1017-1019. Kumar, R. S., Kusari, J., Roy, S. N., Soffer, R. L. & Sen, G. C. (1989). Structure of testicular angiotensin-converting enzyme. A segmental mosaic isozyme. Journal of Biological Chemistry 264, 16754-16758. Lanzillo, J. J., Dasarathy, Y., Stevens, J., Bardin, C. W. & Fanburg, B. L, (1985a). Human testicular angiotensinconverting enzyme is a mixture of two molecular weight forms. Only one is similar to the seminal plasma enzyme. Biochemical and Biophysical Research Communications 128, 457 463. Lanzillo, J. J., Stevens, J., Dasarathy, Y., Yotsumoto, H. & Fanburg, B. L. (1985¿>). Angiotensin-converting enzyme from human tissues'. Physiochemical, catalytic and immunological properties. Journal of Biological Chemistry 260, 14938-14944. Llorens-Cortes, C, Giros, . & Schwartz, J. C. (1990). A novel potential metallopeptidase derived from the enkephalinase gene by alternative splicing. Journal of Neurochemistry 55, 2146-2148. Lowry, O. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951). Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry 93, 265-275. Oliveira, E. B., Martins, A. R. & Camargo, A. C. M. (1976). Isolation of brain endopeptidases : influence of size and sequence of substrates structurally related to bradykinin.
Biochemistry 15, 1967-1974. Orlowski, M., Michaud, C. & Chu,
metalloendopeptidase from
T. G.
(1983).
A soluble
rat brain. Purification of the
enzyme and determination of specificity with synthetic and natural peptides. European Journal of Biochemistry 135, 81-88.
Orlowski, M., Wilk, E., Pearce, S. & Wilk, S. (1979). Purification
and properties of a prolyl endopeptidase from rabbit brain. Journal of Neurochemistry 33, 461-469. Pierotti, ., Dong, K.-W., Glucksman, M. J., Orlowski, M. & Roberts, J. L. (1990). Molecular cloning and primary struc¬ ture of rat metalloendopeptidase EC 3.4.24.15. Biochemistry 29, 10323-10329. Prestidge, R. L. & Hearn, M. T. W. (1979). Preparative flatbed electrofocusing in granulated gels with natural pH gradients generated from simple buffers. Analytical Biochemistry 97, 95-102. Rama Sastry, B. V., Janson, V. E., Owens, L. K. & Tayeb, O. S. (1982). Enkephalin and substance P-like immunoreactivities of mammalian sperm and accessory sex glands. Biochemical Pharmacology 31, 3519-3522. Schechter, I. & Berger, A. (1967). On the size of the active site in proteases. I. Papain. Biochemical and Biophysical Research Communications 21, 157-162. Sharp, B. & Pekari, A. E. (1981). ß-endorphin 6i 9, and other ß-endorphin immuno-reactive peptides in human semen. Journal of Clinical Endocrinology and Metabolism 52, 586-588.
Sharp, B., Pekari, A. E., Meyer, . V. & Hershman, J. M. (1980). ß-endorphin in male rat reproductive organs. Biochemical and Biophysical Research Communications 95, 618-623. Sharpe, R. M., Fraser, H. M., Cooper, I. & Rommerts, F. F. G. (1982). The secretion, measurement and function of a testicu¬ lar LHRH-like factor. Annals of the New York Academy of Sciences 383, 272-292. R. J., Neuberger, M. R. & Liu, T. Y. (1976). Complete amino acid analysis of proteins from a single hydrolysate. Journal of Biological Chemistry 251, 1936-1940. Stewart, . ., Weare, J. A. & Erdös, E. G. (1981). Purification and characterization of human converting enzyme (kininase II). Peptides 2, 145-152. Tisljar, U., de Camargo, A. C, da Costa, C. A. & Barrett, A. J. (1989). Activity of Pz-peptidase and endo-oligopeptidase are due to the same enzyme. Biochemical and Biophysical Research Communications 162, 1460-1464. Toffoletto, O., Metters, K.M., Oliveira, E.B., Camargo, A.C. & Rossier, J. (1988). Enkephalin is liberated from metorphamide and dynorphin A, 8 by endo-oligopeptidase A, but not by metalloendopeptidase EC 3.4.24.15. Biochemical Journal 252, 35-38.
Simpson,
Medeiros, N.
Departments of Clinical
Iazigi,
A. C. M. Camargo and E. B. Oliveira Medicine* and Biochemistryî of the Faculty of Medicine of Ribeiräo Preto,
University of Säo Paulo, Sao Paulo, Brazil tDepartment of Pharmacology, Institute of Biomedicai Sciences, University of Säo Paulo, Säo Paulo, Brazil (M. S. Medeiros to whom requests for offprints should be addressed is now at Department of Biochemistry and Molecular Biology, University of Leeds, Leeds ls2 9jt, U.K.) received
25 March 1992
ABSTRACT
The thiol activated endo-oligopeptidases A and B were studied in the soluble fraction of human hypothalamus and various endocrine glands. For the identification, characterization and purification of the enzymes, Z-Gly-Pro-NH-Np and bradykinin were used as substrates. Endo-oligopeptidase B showed a molecular mass ranging from 55\m=.\5 to 65\m=.\5kDa and isoelectric point from 4\m=.\7to 4\m=.\95.Its activity in tissues was highest in the testis, with intermediate levels in the thyroid, neurohypophysis, adenohypophysis and hypothalamus and the lowest activity in the pineal gland. Endo-oligopeptidase A, 467-fold purified by immunoaffinity chromatography, exhibited a molecular mass of 65\m=.\5 kDa in the adenohypophysis but
58\m=.\5 kDa in other tissues. The isoelectric point ranged from 5\m=.\22 to 5\m=.\50. High endo-oligopeptidase A activity was observed in the adenohypophysis, testis and hypothalamus with lesser activity in the neurohypophysis and thyroid and the lowest in the pineal gland. Endo-oligopeptidase A cleaved the bonds Phe-Ser of bradykinin, Met-Arg of BAM-12P and Arg-Arg of neurotensin as described for rabbit brain and heart and bovine brain enzymes. This work shows that endo-oligopeptidase A also hydrolysed the bonds Tyr\x=req-\ Gly of LH-releasing hormone, Pro-Phe of angiotensin I and Tyr-Ile of angiotensin II. Journal of Endocrinology (1992) 135, 579\p=n-\588
INTRODUCTION
another endopeptidase, now designated endopeptidase-24.15 (EC 3.4.24.15), which is highly active in rat brain and pituitary (Orlowski, Michaud & Chu, 1983) and which has recently been cloned and sequenced (Pierotti, Dong, Glucksman et al 1990). The conclusion that endopeptidase-24.15 is identical to endo-oligopeptidase A and to another activity, 4phenylazobenzyloxycarbonyl-peptidase (Barrett & Tisljar, 1989; Tisljar, Camargo, Costa & Barrett, 1989; Barrett & Brown, 1990) was later contested by Camargo (1991), who found that biochemical dis¬ tinctions can be made between endo-oligopeptidase A and endopeptidase-24.15, including physical separa¬
endo-oligopeptidases A (EC 3.4.22.19) and (prolyl-endopeptidase, EC 3.4.21.26) have been implicated in the processing and metabolism of neuroendocrine peptides. Endo-oligopeptidase A was originally described as a cysteine endopeptidase (kininase) present in a brain cytosol preparation which was capable of hydrolysing the Phe5-Ser6 bond of bradykinin (Camargo, Shapanka & Greene, 1973). Subsequently, endo-oligopeptidase A was shown to Both
hydrolyse the Arg8-Arg9 bond of neurotensin (Camargo, Caldo & Emson, 1983) and to generate [Met]enkephalin from several enkephalin- containing peptides such as the 12-residue opioid peptide from bovine adrenal medulla (BAM-12P) by cleavage at the Met5-Arg6 position (Camargo, Ribeiro & Schwartz, 1985; Camargo, Oliveira, Toffoletto
1987). Endo-oligopeptidase
A resembles in
et al
specificity
tion. The cytosol of nervous tissue also contains another thiol-activated endopeptidase named endooligopeptidase (Oliveira, Martins & Camargo, 1976) further characterized as prolyl-endopeptidase (Greene, Spadaro, Martins et al 1982) which hydro¬ lyses bradykinin at the Pro7-Phe8 bond (Oliveira et al
1976) as well as the carboxyl group of proline in a number of biologically active peptides (Orlowski, Wilk, Pearce & Wilk, 1979; Greene et al 1982). An important feature of both endo-oligopeptidases A and or prolyl-endopeptidase is that they are selec¬ tive for peptides smaller than 20 amino acid residues, the size of the majority of known biologically active peptides. The abundance of these enzymes in brain and their substrate specificity indicates a key role in the processing and metabolism of neuroendocrine
peptides. Furthermore, prolyl-endopeptidase appears to be under the control of gonadal steroids (Jaramillo, Rodrigues & Camargo, 1984). The present work provides the first information on
the distribution of these enzymes in the human neuro¬ endocrine system as well as some of their properties. Some neuropeptide-degrading peptidases, for exam¬ ple angiotensin-converting enzyme (kininase II, EC 3.4.15.1), exist in distinct isoforms of different molecular weights in different organs (Stewart, Weare & Erdös, 1981; Lanzillo, Dasarathy, Stevens et al 1985 ; Lanzillo, Stevens, Dasarathy et al 19856). We have therefore also examined the molecular weights, as well as the isoelectric points, of both endooligopeptidases A and in each of the human tissues examined here.
MATERIALS AND METHODS
Materials Endocrine glands (testis, thyroid, pituitary, pineal) and hypothalami were obtained from ten male indivi¬ duals aged 18 to 37 years, submitted for post-mortem examination at the Instituto Médico Legal de Porto Alegre, Rio Grande do Sul, Brazil after violent death. These organs were collected from 7 to 11 h after death, freed from extraneous tissue by dissection and stored at 20 °C before the enzyme extraction step.
Bradykinin (BK), BK(l-5), BK(6-9) and angio¬ tensins I and II (AI and All) were synthesized by —
Dr A. C. M. Paiva and L. Juliano (Escola Paulista de Medicina, Säo Paulo, Brazil). BAM-12P, [Met]enkephalin, neurotensin (NT), NT(l-9), NT(9-13) and luteinizing hormone-releasing hormone (LHRH) were purchased from Cambridge Research Biochemicals (Northwich, Cheshire, U.K.). Ribonuclease A, chymotrypsinogen, ovalbumin, bovine serum albumin, Sephacryl S-200, DEAESephacel, Sepharose-4B and Sephadex G-25 were obtained from Pharmacia Fine Chemicals (Piscata-
NJ, U.S.A.). Blue-Sepharose, dithiothreitol, ampholyte mixture components, and Tris(hydroxymethyl) aminomethane were purchased from Sigma Chemical way,
Company (St Louis, MO, U.S.A.). Amino acid and peptide analyser reagents were obtained from Pierce Chemical Co. (Rockford, IL, U.S.A.). pBondapak
Cíe columns were obtained from Waters Associates (Milford, MA, U.S.A.).
High-performance liquid chromatography (HPLC) was purchased from Laboratory Data Control (Milford, MA, U.S.A.). W3 ion-exchange resin was purchased from Beckman Instrument Co (Fullerton, CA, U.S.A.). apparatus
Methods Extraction
of enzymes and partial purification of endo-oligopeptidase A The organs were homogenized in distilled water (1:5, w/v) in an Ultra-Turrax homogenizer in an ice bath, at 20 000 r.p.m. for 30 s. Following centrifugation of the homogenate at 25 000 g for 60 min at 4 °C, the pH of the supernatant was adjusted to 5 by dropwise addition of acetic acid (0-5 mol/1). After 1 h at 4 °C, the precipitate was removed by centrifugation (900 g x 15 min) and more than 85% of the total kininase activity was recovered in the supernatant fraction which was neutralized by the addition of Tris-base solution (1 mol/1) and percolated through a blue-Sepharose column, equilibrated in Tris-HCl buffer (0-025 mol/1) pH7-5 to remove albumin (Böhme, Kopperschläger, Schulz & Hofmann, 1972) and particulate matter. The immunoaffinity resin prepared with rabbit polyclonal antibodies raised against bovine endooligopeptidase A has been described previously (Camargo et al. 1987). The immunoaffinity chro¬ matography (Fig 1) was carried out at 4 °C in a column (0-9 10 cm) with capacity to adsorb approxi¬ mately ten units of bovine brain enzyme activity. Following sample application, the unbound material was thoroughly removed by washing the resin with six column volumes of running buffer (Tris-HCl buffer (0-025 mol/1) pH7-5). The adsorbed material was then eluted with running buffer containing Nal (2 mol/1), dithiothreitol (0-5 mmol/1) and 20% glycerol. The active fractions were pooled and equilibrated by gel filtration on a Sephadex G-25 column ( 1 5 x 40 cm) with Tris-acetate buffer (0-025 mol/1) pH 7-0. The endo-oligopeptidase A preparations obtained from all tissues by the immunoaffinity procedure were stored at 20 °C in the presence of 20% glycerol until use. ·
-
Kininase
activity determination
activity was determined by using an enzymatic bioassay with isolated guinea-pig ileum (Camargo, Ramalho Pinto & Greene, 1972). The kininase
5) formed in each reaction was measured using an amino acid analyser, as described previously (Alonzo & Hirs, 1968; Carvalho & Camargo, 1981). One mU endo-oligopeptidase A activity was defined as the amount of enzyme required to release 1 nmol BK(1 5) per min under the conditions described.
Endo-oligopeptidase A cleavage of regulatory peptides
Elution volume
(ml)
Immunoaffinity chromatography was carried in a column (0-1x10 cm) with capacity to adsorb approximately 10 units of bovine brain enzyme activity. Following sample application, the unbound material was thoroughly removed by washing the resin with six column volumes of running buffer (Tris-HCl buffer (0025 mol/1), pH 7-5). The adsorbed material was then eluted with running buffer containing Naf (2 mol/1), dithiothreitol (5 mmol/1) and 20% glycerol. figure
1.
out at 4 °C
Endo-oligopeptidase A assay determination endo-oligopeptidase A activity in tissue homo¬ genates was determined by measuring the amount of the fragment BK(l-5) formed when bradykinin was used as the substrate. Bradykinin (100 pmol/l) was The
incubated at 37 °C for 30 min in Tris-HCl buffer (005 mol/1) pH 7-5 containing NaCl (0-1 mol/1) and dithiothreitol (0-5 mmol/1) with an appropriate aliquot of the 25 000 g x 60 min supernatant fraction obtained from the hypothalamus, testis, thyroid, pineal, anterior and posterior pituitaries, to result in 50-80% hydrolysis of the peptide. The reactions were terminated by acidification and the fragment BK(1-
The immunoaffinity purified endo-oligopeptidase A from human testis (1075 mU/mg) was used for verify¬ ing the cleavage site of the enzyme on the following peptides: BK, BAM-12P, NT, LHRH, AI and AIL The peptides (10-30 pmol/l) were incubated with a given amount of the enzyme (0-25-6 mU) for 3045 min at 37 °C in Tris-HCl buffer (0-025 mol/1) pH7-5, containing dithiothreitol (0-5 mrhol/1). The reaction was terminated by addition of 5µ1 50% (w/v) H3PO4. The products formed in each reaction were separated by reverse-phase HPLC on a C-18 pBondapak column (0-39x30 cm) developed by an initial 5-min isocratic elution in 0-1% H3PO4, pH 2-7 containing 5% (v/v) acetonitrile followed by a 20-min linear gradient of acetonitrile up to 35% (v/v) in 0-1% H3PO4. The flow rate was kept constant at 2 ml/min and the emerging peptides were monitored by their absorbance at 220 nm. The fragments generated by endo-oligopeptidase A digestion of the regulatory peptides were identified either by comparing their retention times on the HPLC with those of synthetic standard peptides or by subjecting the corresponding peptides obtained by HPLC to acid hydrolysis (Simpson, Neuberger & Liu, 1976) followed by amino acid anaylsis (Alonzo & Hirs, 1968). The cleavage sites on the parent peptides could be readily assigned by determining the amino acid composition of the derived fragments.
Endo-oligopeptidase
assay
The endo-oligopeptidase activity in tissue homogen¬ ates was determined spectrophotometrically by meas¬ uring the amount of />-nitroaniline released from the chromogenic substrate Z-Gly-Pro-NH-Np. Z-GlyPro-NH-Np (200 pmol/l) was incubated at 37 °C in Tris-HCl buffer (0-025 mol/1) pH7-5, containing dithiothreitol (0-5 mmol/1) with an appropriate aliquot of 25 000 g 60 min supernatant fraction obtained from the hypothalamus, testis, thyroid, pineal, anterior and posterior pituitaries to hydrolyse 5-10% of the substrate in 10-15 min. After terminat¬ ing the reaction by acidification with 0-1 ml 50% acetic acid, the product formed was determined by its absorbance at 410 nm (Erlanger, Kokowsky &
Cohen, 1961). Endo-oligopeptidase
defined
as
the amount of enzyme
(1 mU)
required
was
to release
1 nmol
^-nitroaniline/min
under the conditions
described.
as
Physicochemical characterization Molecular
A in bradykinin, was measured described in the Materials and Methods. The high¬ est enzymatic activity per g wet tissue was found in the anterior pituitary followed by the testis and hypo¬ thalamus (Table 1). The lowest activity was found in the pineal gland corresponding to 12% ofthat found in the anterior pituitary. When the activity was expressed relative to the protein content, the tissue displaying the highest value was the hypothalamus and the lowest was the thyroid, corresponding to 6% of the activity found in the hypothalamus.
endo-oligopeptidase
mass
determination. The MT values of both
endo-oligopeptidases A and were estimated by gel filtration (Andrews, 1970) on a Sephacryl S-200 column (1-2 x 110 cm) equilibrated and developed at 4 °C with Tris-HCl buffer (0-05 mol/1) pH 7-5, con¬ taining NaCl (0-1 mol/1) at a flow rate of 6-0 ml/h.
The Mr markers used were: bovine serum albumin (67 kDa), ovalbumin (45 kDa), chymotrypsinogen (25 kDa) and ribonuclease A (13-7 kDa). The endooligopeptidase A sample for each tissue was obtained by immunoaffinity chromatography, as described, and its elution volume measured by following its kininase activity. The source of endo-oligopeptidase from each tissue was the non-adsorbed protein fraction from the immunoaffinity column and the correspond¬ ing elution volume was determined by following the activity towards the specific substrate Z-Gly-Pro-NH-
Np.
Isoelectric point determination. The pi values for both endo-oligopeptidases A and were determined by isoelectric focusing (Prestidge & Hearn, 1979) on granulated gel using an ampholyte mixture for gener¬ ating a stable pH gradient in the range of 3 to 10. The sources of endo-oligopeptidase A and were the pH 5 supernatant fraction and the non-adsorbed protein fraction from the immunoaffinity column correspond¬ ing to each tissue respectively. The affinity purified fraction of endo-oligopeptidase A was inactivated on isoelectric focusing. The fractions containing the focused endo-oligopeptidase A activity were deter¬ mined by a two-step procedure, in which the kininase activity of each fraction was measured by the bioassay with isolated guinea-pig ileum first, followed by moni¬ toring the ability of the active fractions to release stoichiometric amounts of the bradykinin fragments BK(1 5) and BK(6 9) on HPLC. The fractions con¬ taining the focused endo-oligopeptidase activity were determined by their activities towards the sub¬
Z-Gly-Pro-NH-Np. Determination of protein concentration
1. Distribution of endo-oligopeptidases A and activities in different human neuroendocrine tissues
table
Endo-oligopeptidase Endo-oligopeptidase A activity activity mU/mg mU/g mU/mg mu/g protein tissue protein tissue Source Testis
Thyroid Hypothalamus Anterior pituitary Posterior pituitary Pineal
1-8 0-2 3-4 1-3 0-7 0-7
65-5 25-9 63-1 82-4 38-0 9-8
82-3 4-5 141 6-2 7-6 6-7
2927-8 577-8 258-9 385-4 416-4 800
endo-oligopeptidase A activity in tissue homogenates was determined amino acid analyser by measuring the amount of the fragment BK(l-5) formed from the hydrolysis of bradykinin (BK). 1 mU endooligopeptidase A is defined as the amount of enzyme required to release 1 nmol BK(1 5)/min under the conditions described in Materials and Methods. Z-Gly-Pro-NH-Np was used to determine the activity of endooligopeptidase B. 1 mU endo-oligopeptidase is defined as the amount of enzyme required to release 1 nmol /7-nitroaniline/min under the conditions The
on an
described in Materials and Methods.
Another enzyme that contributes to bradykinin inactivation by tissue peptidases is prolyl-endopeptidase or endo-oligopeptidase B. This enzymatic activity can be specifically measured in tissue homogenates by using the chromogenic substrate Z-Gly-Pro-NH-Np. The highest activity per g wet tissue and per mg protein was found in the testis, being 5 to 25 times higher than the activity found in other tissues
(Table 1).
strate
Protein concentration was determined by the proce¬ dure of Lowry, Rosebrough, Farr & Randall (1951) using bovine serum albumin as the standard. RESULTS
Distribution of endo-oligopeptidases A and neuroendocrine system
in human
The hydrolysis of the Phe5-Ser6 bond of bradykinin, which typically represents the only site of cleavage by
Purification of endo-oligopeptidase A by immunoaffinity chromatography from human endocrine glands and hypothalami The cross-reactivity of the antibody raised against endo-oligopeptidase A from bovine brain against the human enzyme permitted purification of the human endo-oligopeptidase A by immunoaffinity chromatography.
The 25 000 g x 60 min supernatant fraction of the tissue homogenate was subjected to immunoaffinity chromatography resulting in approximately 500-fold
table
2. Purification of endo-oligopeptidase A from human testis Total
protein (mg)
Fractions 25 000 g supernatant pH 5 supernatant
81-0 45-1 30-8 0-2
Blue-Sepharose Immunoaffinity chromatography
Protein recovery
(%) 1000 55-7 38-0 0-25
(mU)
activity
Specific activity (mU/mg protein)
184-5 155-1 1400 215-0
1000 84-1 75-9 116-5
2-3 3-4 4-5 10750
Total kininase
activity
Kininase
Purification
(fold) 10 1-5 2-0 467-4
The above results are provided from a 10 g testis homogenate. The total volume of the homogenate was 55 ml and the total volume of 25 000 g 60 min supernatant fraction was 50 ml. From the latter was taken an aliquot of 15 ml for endo-oligopeptidase A purification. The enzymatic reaction was followed by the kininase assay described in Materials and Methods. Bradykinin (5 µ / ) was incubated with each fraction at 37 °C in Tris-HCl buffer (0-05 mol/1), pH 7· 5, containing NaCl (0· 1 mol/1) and dithiothreitol (0- 5 mmol/1). The concentration of enzyme was selected to result in 50 80% of bradykinin inactivation in 10 mins.
purification with
total recovery of enzyme activity. Table 2 illustrates the purification of endo-oligo¬ peptidase A from testis. As the amount of enzyme applied to the column (220 mU) was less than the immunoadsorbent capacity (10 units) the nonadsorbed protein was devoid of activity able to hydrolyse the Phe5-Ser6 bond of bradykinin. Endooligopeptidase A was completely recovered from the column in an active form after the addition of the chaotropic agent Nal (2 mol/1).
table
3. Molecular mass determination of endoand from human hypothalamus and
oligopeptidase A endocrine glands Source
A
Endo-oligopeptidase (Mr)
Endo-oligopeptidase
Testis
58 58 58 65 58
500 ± 1500 500 ± 1500 500 ± 1500 500 ± 1500 500 ± 1500
55 62 62 58 65
Thyroid Hypothalamus Anterior pituitary Posterior pituitary
(Mr)
500 ± 1500 000 ± 2000 000 ± 2000 500 ± 1500 500 ± 1500
endo-oligopeptidase A activity obtained by immunoaffinity chromatography and endo-oligopeptidase present in the non-adsorbed protein fraction from the immunoaffinity column, of each tissue, were applied separately to a Sephacryl S-200 column (1-2x110 cm) equilibrated and developed at 4 °C in Tris-HCl buffer (005 mol/1), pH 7-5, containing NaCl (0-1 mol/1) at a flow rate of 6 ml/h. Fractions of 1-5 ml were collected and assayed for kininase activity. The hydrolysis of Z-Gly-ProNH-Np was also assayed in all fractions. The M, of the enzymes were estimated according to Andrews (1970). Haifa fraction volume was taken as the margin of error in molecular mass determination. The
Determination of \f and pi of endo-oligopeptidases from human endocrine glands and
A and
hypothalami Endo-oligopeptidase A purified by immunoaffinity chromatography and endo-oligopeptidase present in the non-adsorbed protein fraction from the immuno¬ affinity column of each tissue were subjected separ¬ ately to gel filtration on Sephacryl S-200. The gel
filtration results indicated
an
apparent Mr for endo-
oligopeptidase A of 58 500 for the testis, thyroid, hypothalamus and posterior pituitary and 65 500 for the anterior pituitary (Table 3). The values for endooligopeptidase varied from 55 500 for the testis to 65 500 for the posterior pituitary (Table 3). Endooligopeptidase A from the testis, thyroid and hypo¬ thalamus was isoelectric in the pH range 5-22-5-50 whereas endo-oligopeptidase of the same tissues was isoelectric at a pH significantly lower: 4-70-4-95 (Fig. 2). Sites of cleavages
on neuropeptides by endooligopeptidase A Immunoaffinity purified endo-oligopeptidase A obtained from human endocrine glands and hypothal¬
ami was incubated with BK, NT, BAM-12P, LHRH, AI and AIL The products of peptide hydrolysis were separated by HPLC (Fig. 3) and collected separately. The peptides corresponding to each peak were
subjected to acid hydrolysis followed by amino acid analysis. The results indicate that BK, NT, BAM-12P and AI and All were hydrolysed at a single peptide bond. LHRH appeared to be cleaved at more than one peptide bond. Table 4 illustrates the cleavage sites of neuropeptides by glandular and hypothalamic endo-oligopeptidase A. DISCUSSION
Purification methods: affinity chromatography In this work a simplified purification procedure was followed for separating endo-oligopeptidase A from endo-oligopeptidase producing the former in a highly purified form. The procedure involved suc¬ cessive affinity chromatography steps on blueSepharose and an immunoadsorbent column. The endo-oligopeptidase A preparation obtained was devoid of other peptidase activities as indicated by the stoichiometric recovery of the fragments derived from
Testis
Testis
Fraction
Fraction
2. The pf values of endo-oligopeptidase A and B were determined by isoelectric focusing granulated gel, using an ampholyte mixture to generate a stable pH gradient in the range of 3-10. The activity of endo-oligopeptidase A was determined by a two-step procedure: the bioassay of isolated guinea-pig ileum followed by monitoring the ability of the active fractions to release stoichiometric amounts of bradykinin fragments BK(l-5) and BK(6-9) on HPLC. The activity of endo-oligopeptidase was verified with Z-Gly-Pro-NH-Np as substrate. figure on
BK, BAM-12P and NT. When BK was completely hydrolysed only the complementary fragments BK(l-5) and BK(6 9) were recovered, indicating the absence of aminopeptidase and carboxypeptidase activities as well as post-proline endopeptidase or and angiotensin-converting endo-oligopeptidase enzyme. The cross-reactivity of the antibody prepared against bovine brain endo-oligopeptidase A with the human enzyme has now allowed purification of the human endo-oligopeptidase A. Highly purified endo-oligopeptidase A (approximately 500-fold purification) was obtained after immunoaffinity chromatography, leading to virtually complete recov¬ ery of activity in all the tissues used.
Characterization of human endo-oligopeptidases A and Endo-oligopeptidases A and differed in isoelectric point (about 5-4 and 4-8 respectively), whether from
or testis, as previously found for the rabbit brain enzymes. The observed differences between the pi of these enzymes make the isoelectric focusing a useful method for separating these two enzymatic activities. As Table 3 shows, the Mr values of the two enzymes from human hypothalamus and endocrine glands are similar to each other and do not differ much with tissue. Nevertheless anterior pituitary tissue does
thyroid, hypothalamus
possess
a
significantly larger endo-oligopeptidase
A
004
0-04-
004
5
10
15
Time
(min)
20
neurotensin (NT), BAM-12P, LH-releasing hormone (LHRH), angiotensin incubated with 0-75-6-0 mU endo-oligopeptidase A purified by immunoaffinity chromatography (specific activity towards BK:1075 mU/mg) at 37 °C in Tris-HCl buffer (0025 mol/1), pH 7-5, containing dithiothreitol (005 mmol/1) for 30-45 min. The hydrolysis products of the substrates were analysed by reverse-phase HPLC as described in Materials and Methods. Typical chromatograms show the production of: (a) BK(l-5), BK(6-9) from BK; (b) NT(l-8) and NT(9-13) from NT; (c) BAM-12P(6-12) and [Met]enkephalin (ME) from BAM-12P; (d) LHRH(l-5) and LHRH(6 10) from LHRH; (e) AI(l-7) and Al(8-10) from AI; (/) AII(l-4) from AH. figure
3.
(AI) and
Bradykinin (BK), (10-30 nmol)
were
and testis a significantly smaller endo-oligopeptidase B, than the other tissues. These small differences in Mr from enzymes of various sources may be due to the presence of isoenzymatic forms of each enzyme in different tissues. There are precedents for the occurrence of tissue-specific forms of peptidases of differing molecular size. For example, angiotensinconverting enzyme exists as a polypeptide of Mr 180 000 in kidney and lung, 180 000 and 170 000 in brain, and as a truncated form of Mr 110 000 in testis (Stewart et al 1981; Lanzillo et al. 1985 ) which arises by alternative splicing (Kumar, Kusari,
Roy et al 1989). Alternatively, spliced forms of endopeptidase-24.11 of differing molecular weights may
also exist (Llorens-Cortes, Giros & Schwartz, 1990). It is noteworthy that molecular exclusion chro¬ matography cannot be used to separate these enzymes, because their Mr values are so close.
Hydrolysis of neuroendocrine peptides by endo-oligopeptidase A from human testis The use of HPLC and amino acid analysis allowed us to show that testicular endo-oligopeptidase A,
table
4.
Cleavage
sites of peptides
hydrolysed by endo-oligopeptidase A from human testis
P4 P, P2 P, Pi' P2' P3' P/
I Arg-Pro-Pro-Gly-Phe-Ser-Pro-Phe-Arg I Tyr-Gly-Gly-Phe-Met-Arg-Arg-Val-Gly-Arg-Pro-Glu I pGlu-Leu-Tyr-Glu-Asn-Lys-Pro-Arg-Arg-Pro-Tyr-Ile-Leu I pGlu-His-Trp-Ser-Tyr-Gly-Leu-Arg-Pro-GlyNH2 1 Asp-Arg-Val-Tyr-Ile-His-Pro-Phe-His-Leu ï Asp-Arg-Val-Tyr-Ile-His-Pro-Phe
bradykinin BAM-12P
neurotensin LHRH
angiotensin
I
angiotensin
II
The table shows the amino acid sequences of peptides hydrolysed by endo-oligopeptidase A. The cleavage site was determined by combination of HPLC and amino acid analysis of the hydrolysis products. The residue positions (Pi_4 and P'1-4) on either side of the scissile bond are named using the nomenclature of Schechter & Berger (1967). The observed cleavage site is indicated by an arrow between the two amino acids in bold print in positions Pi and Pi'. BAM-12P is the 12-residue opioid peptide from bovine adrenal medulla. LHRH is LH-releasing hormone.
purified by immunoaffinity chromatography, hydro¬ lyses the Phe-Ser bond of bradykinin, the Met-Arg bond of BAM-12P and the Arg-Arg bond of neuro¬ tensin. LHRH, AI and All were also hydrolysed at the Tyr-Gly, Pro-Phe and Tyr-Ile bonds respectively. The presence of bradykinin in the incubation mixture inhibited the hydrolysis of all regulatory peptides used in this work, supporting the occurrence of a single activity hydrolysing all the peptides. There is no simple predictor of the cleavage sites within peptides by endo-oligopeptidase A. From analysis of the sites of peptide hydrolysis (Table 4) it can be suggested that there is a chymotryptic-like specificity for hydrolysis of BK, LHRH and All but, apparently, a tryptic-like specificity for NT. The hydrolysis by endo-oligopeptidase A of the Pro-Phe bond of AI is not due to contamination by prolyl endopeptidase or endo-oligopeptidase since this enzyme was not detectable after 24 h by using as sub¬ strate Z-Gly-Pro-NH-Np or even after 100% hydro¬ lysis of BK. Concerning the secondary specificity of endooligopeptidase A, the alignment of amino acid resi¬
dues does not allow verification of whether the enzyme has any structural requirements conforming to the model of Schechter & Berger (1967). Other features of the substrate such as size, charge and con¬ formation may be more important than the simple alignment of the peptide chain. The broad specificity of endo-oligopeptidase A suggests a general role in the metabolism of regulatory peptides with different target peptide substrates in the various endocrine organs.
Relationship of endo-oligopeptidase A to other endopeptidases Clearly there are considerable similarities in substrate specificity between endo-oligopeptidase A and endo-
peptidase-24.15. The latter enzyme, however, is class¬ ified as a metallopeptidase and contains the amino acid sequence typical of a zinc-dependent metallopeptidase (Pierotti et al 1990). However, the presence of a cysteine residue close to the catalytic centre of endopeptidase-24.15 might explain the activation of the enzyme by 2-mercaptoethanol and its inhibition by pmercuribenzoate and TV-ethylmaleimide. Although it has been claimed that endo-oligopeptidase A and endopeptidase-24.15 may be identical (Barrett & Brown, 1990), an antibody to endo-oligopeptidase A does not precipitate endopeptidase-24.15 (Toffoletto, Metters, Oliveira et al 1988) and the two activities are separable by ion exchange chromatography (Camargo, 1991 ). The relationship between these two
activities remains unresolved, therefore, and must await the cloning and sequencing of endo-oligopep¬ tidase A. The involvement of cysteine in the catalytic mechanism also needs to be explored in detail.
Distribution of endo-oligopeptidase A and proline endopeptidase in human hypothalamus and endocrine glands: physiological role The distribution of endo-oligopeptidases A and differed among the organs studied. The endooligopeptidase A predominated in anterior pituitary, testis and hypothalamus and endo-oligopeptidase in testis, thyroid and posterior pituitary. Recently, a number of regulatory peptides such as an LHRH-like peptide and ß-endorphin have been demonstrated in the testes of various animals (Sharp, Pekari, Meyer & Hershman, 1980; Sharp & Pekari, 1981; Sharpe, Fraser, Cooper & Rommerts, 1982).
[Met]enkephalin,
[Leu]enkephalin,
ß-endorphin,
substance and calcitonin were also found in human sperm and semen, suggesting that they may play an important regulatory function in the male reproduc¬ tive system, including sperm motility (Rama Sastry,
Janson, Owens & Tayeb, 1982; Fraioli, Fabbri,
Gnessi et al 1984). Studies in rats on the steroid regulation of hypo¬ thalamic LHRH degradation (Advis, Krause & McKelvy, 1982, 1983) have shown that LHRH degra¬ dation products, in all physiological situations, revealed cleavage of the Tyr-Gly bond. The same hydrolysis site was observed in this work for endooligopeptidase A on this peptide. Jaramillo et al (1984) have suggested the participation of adeno¬ hypophysis prolyl endopeptidase in the control of gonadotrophin secretion. Therefore, it is possible that both enzymes could be implicated in the feedback mechanisms of the hypothalamic-hypophysialgonadal axis. Finally, the action of both enzymes on oligopeptides, together with their distribution in the human neuroendocrine system, suggest that they have specific functions related to regulatory peptide metabolism. These observations indicate that studies in vivo and further cellular and subcellular studies are needed to define their physiological role in the metabolism of
peptides.
ACKNOWLEDGEMENTS
M.S.M. was in receipt of a Fellowship from CAPESBrazil. Mrs E. S. Gray is thanked for efficient sec¬ retarial assistance. REFERENCES
Advis, J. P., Krause, J. E. & McKelvy, J. F. (1982). Luteinizing
hormone-releasing hormone peptidase activities in discrete hypothalamic regions and anterior pituitary of the rat : appar¬ ent regulation during the prepubertal period and first estrous cycle at puberty. Endocrinology 110, 1238-1245. Advis, J. P., Krause, J. E. & McKelvy, J. F. (1983). Evidence that endopeptidase-catalysed luteinizing hormone-releasing hormone cleavage contributes to the regulation of median eminence LHRH levels during positive steroid feedback. Endocrinology 112, 1147-1150. Alonzo, . & Hirs, C. H. W. (1968). Automation of sample appli¬ cation in amino acid analyzers. Analytical Biochemistry 23, 272-288.
Andrews, P. (1970). Estimation of molecular size and molecular
weights of biological compounds by gel filtration. In Methods of Biochemical Analysis, vol. 18, pp. 1-53. Ed. D. Glick. New York : Wiley-Interscience. Barrett, A. J. & Brown, M. A. (1990). Chicken liver Pz-peptidase, a thiol-dependent metallo-endopeptidase. Biochemical Journal
271,701-706. Barrett, A. J. & Tisljar, U. (1989). The activities of 'Pz-peptidase' and 'endopeptidase-24.15' are due to a single enzyme. Biochemical Journal 261, 1047-1050. Böhme, . J., Kopperschläger, G., Schulz, J. & Hofmann, E. (1972). Affinity chromatography of phospho-fructokinase using Cibacron blue F3G-A. Journal of Chromatography 69, 209-214.
Camargo, A. C. M. (1991). Distinction between endooligopeptidase A (EC 3.4.22.19) and soluble metalloendopeptidase (EC 3.4.24.15). Biochemical Journal 277, 294-295. Camargo, A. C. M., Caldo, H. & Emson, P. C. (1983). Degrada¬ tion of neurotensin by rabbit brain endo-oligopeptidase A and endo-oligopeptidase (proline-endopeptidase). Biochemical and Biophysical Research Communications 116, 1151-1159. Camargo, A. C. M., Oliveira, E. B., Toffoletto, O., Metters, . M. & Rossier, J. (1987). Brain endo-oligopeptidase A, a putative enkephalin-converting enzyme. Journal of Neurochemistry 48, 1258-1263.
Camargo, A. C. M., Ramalho Pinto, F. J. & Greene, L. J. (1972). Brain peptidases : conversion and inactivation of kinin hor¬ mones. Journal of Neurochemistry 19, 37-49. Camargo, A. C. M., Ribeiro, M. J. V. F. & Schwartz, W. N. (1985). Conversion and inactivation of opioid peptides by rab¬ bit brain endo-oligopeptidase A. Biochemical and Biophysical Research Communications 130, 932-938. Camargo, A. C. M., Shapanka, R. & Greene, L. J. (1973).
Preparation, assay and partial characterization of a neutral endopeptidase from rabbit brain. Biochemistry 12, 1838-1844. Carvalho, K. M. & Camargo, A. C. M. (1981). Purification of rabbit brain endo-oligopeptidases and preparation of antienzyme antibodies. Biochemistry 20, 7082-7088. Erlanger, . F., Kokowsky, . & Cohen, W. (1961). The prep¬ aration and properties of two new chromogenic substrates of trypsin. Archives of Biochemistry and Biophysics 95, 271-278. Fraioli, F., Fabbri, ., Gnessi, L., Silvestroni, L., Moretti, C, Redi, F. & Isidori, . (1984). Beta-endorphin, Met-enkephalin,
and calcitonin in human semen : evidence for a possible role in human sperm motility. Annals of the New York Academy of Sciences 438, 365-370. Greene, L. J., Spadaro, A. C. C, Martins, A. R., Jesus, W. D. P. & Camargo, A. C. M. (1982). Brain endo-oligopeptidase B: a post-proline cleaving enzyme that inactivates angiotensin I and II. Hypertension 4, 178-184. Jaramillo, P. L., Rodrigues, J. A. & Camargo, A. C. M. (1984). Effect of gonadal steroids on hypothalamus and anterior pitu¬ itary endo-oligopeptidase (proline-endopeptidase) activity in castrated female rats. Peptides 5, 1017-1019. Kumar, R. S., Kusari, J., Roy, S. N., Soffer, R. L. & Sen, G. C. (1989). Structure of testicular angiotensin-converting enzyme. A segmental mosaic isozyme. Journal of Biological Chemistry 264, 16754-16758. Lanzillo, J. J., Dasarathy, Y., Stevens, J., Bardin, C. W. & Fanburg, B. L, (1985a). Human testicular angiotensinconverting enzyme is a mixture of two molecular weight forms. Only one is similar to the seminal plasma enzyme. Biochemical and Biophysical Research Communications 128, 457 463. Lanzillo, J. J., Stevens, J., Dasarathy, Y., Yotsumoto, H. & Fanburg, B. L. (1985¿>). Angiotensin-converting enzyme from human tissues'. Physiochemical, catalytic and immunological properties. Journal of Biological Chemistry 260, 14938-14944. Llorens-Cortes, C, Giros, . & Schwartz, J. C. (1990). A novel potential metallopeptidase derived from the enkephalinase gene by alternative splicing. Journal of Neurochemistry 55, 2146-2148. Lowry, O. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951). Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry 93, 265-275. Oliveira, E. B., Martins, A. R. & Camargo, A. C. M. (1976). Isolation of brain endopeptidases : influence of size and sequence of substrates structurally related to bradykinin.
Biochemistry 15, 1967-1974. Orlowski, M., Michaud, C. & Chu,
metalloendopeptidase from
T. G.
(1983).
A soluble
rat brain. Purification of the
enzyme and determination of specificity with synthetic and natural peptides. European Journal of Biochemistry 135, 81-88.
Orlowski, M., Wilk, E., Pearce, S. & Wilk, S. (1979). Purification
and properties of a prolyl endopeptidase from rabbit brain. Journal of Neurochemistry 33, 461-469. Pierotti, ., Dong, K.-W., Glucksman, M. J., Orlowski, M. & Roberts, J. L. (1990). Molecular cloning and primary struc¬ ture of rat metalloendopeptidase EC 3.4.24.15. Biochemistry 29, 10323-10329. Prestidge, R. L. & Hearn, M. T. W. (1979). Preparative flatbed electrofocusing in granulated gels with natural pH gradients generated from simple buffers. Analytical Biochemistry 97, 95-102. Rama Sastry, B. V., Janson, V. E., Owens, L. K. & Tayeb, O. S. (1982). Enkephalin and substance P-like immunoreactivities of mammalian sperm and accessory sex glands. Biochemical Pharmacology 31, 3519-3522. Schechter, I. & Berger, A. (1967). On the size of the active site in proteases. I. Papain. Biochemical and Biophysical Research Communications 21, 157-162. Sharp, B. & Pekari, A. E. (1981). ß-endorphin 6i 9, and other ß-endorphin immuno-reactive peptides in human semen. Journal of Clinical Endocrinology and Metabolism 52, 586-588.
Sharp, B., Pekari, A. E., Meyer, . V. & Hershman, J. M. (1980). ß-endorphin in male rat reproductive organs. Biochemical and Biophysical Research Communications 95, 618-623. Sharpe, R. M., Fraser, H. M., Cooper, I. & Rommerts, F. F. G. (1982). The secretion, measurement and function of a testicu¬ lar LHRH-like factor. Annals of the New York Academy of Sciences 383, 272-292. R. J., Neuberger, M. R. & Liu, T. Y. (1976). Complete amino acid analysis of proteins from a single hydrolysate. Journal of Biological Chemistry 251, 1936-1940. Stewart, . ., Weare, J. A. & Erdös, E. G. (1981). Purification and characterization of human converting enzyme (kininase II). Peptides 2, 145-152. Tisljar, U., de Camargo, A. C, da Costa, C. A. & Barrett, A. J. (1989). Activity of Pz-peptidase and endo-oligopeptidase are due to the same enzyme. Biochemical and Biophysical Research Communications 162, 1460-1464. Toffoletto, O., Metters, K.M., Oliveira, E.B., Camargo, A.C. & Rossier, J. (1988). Enkephalin is liberated from metorphamide and dynorphin A, 8 by endo-oligopeptidase A, but not by metalloendopeptidase EC 3.4.24.15. Biochemical Journal 252, 35-38.
Simpson,
