133
Biochimica et Biophysica Acta, I ! 19 (1992) 133-139 © 1992 Elsevier Science Publishers B.V. All rights reserved 0167-4838/92/$05.00
BBAPRO M I ~
A radioassay for aminoacylproline hydrolase (aminopeptidase P) activity James W. Ryan, Alfred Y.K. Chung, Pierre Berryer and Donald H. Sheffy, Jr. Department of Medicine, Unicersityof Miami Schoolof Medicine, Miami, FL (U.S.A.) (Received 13 September 1990) (Revised manuscript received 22 August 1991)
Key words: Aminoacylproline hydrolase; Aminopeptidase P; Bradykinin; Radioassay substrate; First-order enzyme kinetics; (Rat); (Guinea pig)
A mdionssay was devdoped in which aminoacylproline hydrolase acts on Arg-Pro-Pro-[aH]benzylamide to yield ~rglnine plus Pro-Pro-[aH]benzylamide. By Mopping the reaction with base (0.1 M NaOH), the radioactive product is deprotonated to an organophilic form and is separable from the hydrophilic substrate by extraction of the alkaline aqueous solution with an organic solvent. When scintillants are included in the organic solvent, the enzyme:substrate reaction, extraction and quantification of Pro-Pro-[aH]benzylamide can all be conducted using a single liquid scintillation vial. Thus, aminoacylproline hydrolase activity is measured in terms of the rate of release of Pro-Pro-[aH]benzTlamide. The substrate is obtainable at > 20 C i / m m o l , which enables its use under conditions of first-order enzyme kinetics. Conditions of near-zero order kinetics are readily attained by adding unlabeled substrate (K_ 0.7/tM). The substrate is highly reactive (a 1:2000 dilution of guinea pig plasma hydrolyzed > 10% of the substrate during a 10 min incubation at 37°C) and specific in that it is not degraded by leucine aminopeptidase, aminopeptidase A or N, dipeptidyl peptidase IV nor prolyl endopeptidase. The assay was used to measure aminoacTIproline hydrolase specific activities in tissues of rat and guinea pig. Activity was found in virtually all major tissues of both species, and some guinea pig tissues (e.g. kidney and plasma) were found to be notably rich sources of the e n z ~ e .
Introduction During a single passage through the rat pulmonary vascular bed, bradykinin (ArgLPro2-Pro3-Gly4-Phe 5Ser6-ProLPheS-Arg 9) is hydrolyzed at 5 of its 8 peptide bonds [1,2]. One hydrolysis occurs at the Argl-Pro 2 bond and is catal~ed by an aminoacylproline hydrolase [1-4]. An aminoacylproline hydrolase has been isolated from Escherich3a coli, and others have been found in tissues such as pig kidney, rat intestine, rat and bovine lung and human serum [5-10]. Whether these enzymes are structurally related is not known, but all are called aminopeptidase P, the name given the bacterial enzyme (EC 3.4.11.9) [5]. Several methods have been developed to measure aminoacylproline hydrolase activity, but none is at once
Correspondence: J.W. Ryan, 1399 N.W. 17th Avenue, No. 200 Miami, FL 33125, U.S.A.
simple and convenient. Initially, the bacterial enzyme was measured by its ability to release proline from poly-L-proline [5]. The pig kidney enzyme was measured in terms of the release of glycine from Gly-ProHyp [7]. Lasch et al. [10] have described a two step chromogenic assay in which the aminoacylproline hydrolase acts on Gly-Pro-Pro-[4-NO2]anilide to form Gly and Pro-Pro-[4-NOe]anilide in the first step, and the latter product is hydrolyzed by an excess of dipeptidyl aminopeptidase IV (dipeptidyl peptidase IV, EC 3.4.14.5) to release the chromophore 4-NO2-aniline. Yaron and colleagues [8,9] have introduced a series of relatively insoluble internally-quenched fluorogenic substrates, of which Lys[e-DNP)-Pro-Pro-NH-CH2CH2-NH-CO-CoH4(2'-NH2) was the most conveniently used. Yoshimoto et al. [11] have reported that Gly-Pro-7-amido-4-methylcoumarin can be used to measure aminoacyiproline hydrolase activity (in the presence of excess proline iminopeptidase, EC 3.4.11.5) when dipeptidyl aminopeptidase IV is not present or is fully inhibited.
134 To capitalize on the finding that bradykinin is highly reactive with the aminoacylproline hydrolase of rat lung [1-4], we have prepared the bradykinin N-terminal homolog Arg-Pro-Pro-[aH]benzylamide. The substrate is hydrolyzed to yield arginine plus Pro-Pro[3H]benzylamide. At alkaline pH, Pro-Pro-benzylamide is readily extractable into organic solvents. Thus, the hydrophilic substrate can be used to measure aminoacylproline hydrolase activity in terms of the rate of formation of the organophilic product. An abstract describing the new assay procedure has been published I12l. Materials and Methods
Pro-[3 '-iodo]benzylamide A suspension of Boc-Pro-OH (451 mg, 2 mmol) and 540 mg of [3'-iodo]benzylamine hydrochloride in 5 ml of dichloromethane at 0°C was neutralized with 0.272 ml (2 mmol) of N-ethyl morpholine. Dicyclohexylcarbodiimide, 412 mg (2 mmol) in 2 ml of dichloromethane was added. The reaction mixture was stirred at 0°C for 1 h and then at room temperature overnight. Insoluble dicyclohexylurea and N-ethyl morpholine hydrochloride were removed by filtration, and the precipitate was washed with ethyl acetate. The filtrates were combined, washed until neutral and then dried over MgSO4. Solvent was removed in vacuo resulting in a clear oily residue. The Boc group was removed in 5.5 M hydrogen chloride in ethyl acetate, 2 h at 0°C. Solvent was removed in vacuo. Pro-[3'iodo]benzylamide was obtained as white crystals in ethanol and diethyl ether; m.p. 177-177.5°C (decomp), weight 503 mg; for C~z H 16N2IC10: calculated C 39.31, H 4.40, N 7.64, l 34.61, Cl 9.67, O 4.36; found C 39.39, H 4.51, N 7.68, 1 34.53, CI 9.50.
N~-Boc.Arg( Tos).Pro.OH Dicyclohexylcarbodiimide, 2.06 g (10 mmol), in 3 ml of dimethylformamide was added with stirring to a solution of 4.285 g of N~-Boc-Arg(Tos)-OH and 1.35 g of 1-hydroxybenzotriazole in dimethyformamide in an ice-acetone bath. To this mixture was added 2.66 g of proline benzyl ester hydrochloride and 1.1 ml of Nmethyl morpholine in dimethylformamide. The resulting mixture was stirred in the ice-acetone bath for 1 h and then at 4°C overnight. The reaction mixture was filtered, and the precipitate was washed with ethyl acetate. The filtrates were combined, washed until neutral and then dried over MgSO4. Solvent was removed in vacuo. An oily residue was obtained by rotary evaporation. The residue was triturated with isopropyl ether and then filtered to yield a white powder, 4.57 g; m.p. 55-57°C; amino acid analysis: Arg I Pro v The benzyl ester group was removed by hydrogenolysis with 100 mg of 10% Pd on carbon as catalyst in absolute
ethanol at 25 psi for 1 h. The catalyst was removed by filtration, and solvent of the filtrate was removed in vacuo to yield a white foam. On silica gel thin-layer chromatography (methanol/chloroform (1:1 by vol.)), the product N~'-Boc-Arg(Tos)-OH behaved as a single substance, R r 0.54 (o-tolidine/Kl stain).
N ~-Boc-Arg( Tos)-Pro-Pro-[ 3'-iodo ] benzylamide . diacetate Dicyclohexylcarbodiimide 0.206 g (1 mmol) in 2 ml of dichloromethane was added to a stirred solution of N"-Boc-Arg(Tos)-Pro-OH, 1 mmol, and 135 mg of 1-hydroxybenzotriazole in 3 ml of dimethylformamide/dichloromethane (1:2 by vol) at -10°C. To this mixture was added 367 mg of HCI-Pro-[3'iodo]benzylamide and 0.11 mi of N-ethylmorpholine in 3 ml of dimethylformamide. The reaction mixture was stirred at - 10°C for 1 h, at 4°C overnight, and then at room temp for 2 h. The precipitate was removed by filtration and was washed with water and ethyl acetate. The organic phase of the combined filtrates was washed until neutral and then dried over MgSO a. Solvent was removed in vacuo to yield 0.61 g of a foam-like material. Amino acid analysis: Arg t ProL85.
Arg-Pro-Pro-[ 3 '-iodo ]benzylamide The Boc and tosyl protecting groups were removed by treating 280 mg of the protected peptide with 10 ml of HF in the presence of 0.6 ml of anisole at 0°C for 1 h. HF was removed in vacuo and the residue was partitioned in diethyl e t h e r / 2 % (by vol.) acetic acid. The crude material in the lower phase was applied to an LH-20 column (1.2 × 95 cm) equilibrated and developed with 6% 1-butanol. Fractions containing product were pooled and solvent was removed in vacuo to yield 178 mg of residue. The residue was applied to a Sephadex G-25 column (1.2 × 100 era) equilibrated with the upper phase of 1-butanol/acetic acid/water (4: 1 : 5 by vol.). The column was washed with the upper phase and then product, 139 rag, was eluted with the lower phase. Amino acid analysis: Arg I Proi.s7. On silica gel thin-layer chromatography (1-butanol/pyridine/aeetie acid/water (15: 10:3 : 12 by vol.), 1-butanol/ethyl ace t a t e / a c e t i c a c i d / w a t e r (1 : 1 : 1 : 1) and ethyl acetate/pyridine/acetie acid/water (5: 5: 1 : 3)), the product behaved as a single substance; respective R F values: 0.51, 0.42 and 0.67 (ninhydrin, Sakaguchi and o-tolidine/KI staining).
Arg-Pro-Pro-[ JH]benzylamide Arg-Pro-Pro-[3'-iodo]ben~iamide, 11 nag in 2 ml of dimethylformamide/water (1:1 by vol), was dehalogenated in 10 Ci of 3H 2 gas using 10% Pd on carbon as catalyst to yield Arg-Pro-Pro-[3'-3H]benzylamide, 350 mCi at 22.4 Ci/mmol. The tritiated product was dis-
135 solved in 70% ethanol (0.5 mCi/ml), in which it was stable on storage at -20°(2 for more than 3 years.
Other peptides Pro-benzylamide was prepared as described above for Pro-[3'-iodo]benzylamide by substituting benzylamine for [Y-iodo]benzylamine. Pro-Pro-benzylamide was prepared by reacting Z-Pro-OH with Pro-benzylamide via dicyclohexylcarbodiimide coupling in the presence of 1-hydroxybenzotriazole in dichloromethane. The Z-protecting group was removed by hydrogenolysis in methanol using 10% Pd on carbon as catalyst.
Enzyme sources Except where noted, we used as an enzyme source rat "plasma (anticoagulated with heparin), guinea pig plasma or homogenate of rat lung (2 ml of water containing 1% (by vol.) of Nonidet P-40 per g of rat lung; homogenized by three 15-s bursts of a Polytron tissue disrupter at setting 3).
Assay buffer Except for measurements of Kin, Arg-Pro-Pro[3H]benzylamide was used in concentrations within the range of first-order enzyme kinetics (typically at about 20 nM). In Hepes buffers, the pH optimum of aminoacylproline hydrolase was found to be broad (pH 8.0 to at least 9.3). Thus, subsequent assays used 0.05 M Hepes buffer (pH 8.3) containing 0.15 M NaCI and Hibiclens (Stuart Laboratories), 0.01% by vol. (assay buffer). NaCl was used to make the buffer near-isotonic for assay of enzymic activities of intact cells, and Hibiclens (chlorhexidine gluconate) was used as a bacteriostat.
Product formation To examine product formation, 2 ml of the substrate in assay buffer (10.7 #Ci, 0.24/~M) was incubated at 37°C with 2 ml of rat lung homogenate diluted 1:200 in assay buffer; protein concentration of 109 ttg/ml. At timed intervals (0-60 min), 10 /zl samples of the incubation mixture were spotted on a thin-layer silica gel plate. After all samples were collected, 5 ttg samples of authentic benzylamine, Pro-Bz, Pro-Pro-Bz and Arg-Pro-Pro-Bz were applied to the plate, and the plate was developed with I-butanol/acetic acid/water (75 : 13 : 12) or 1-butanol/ethyl a c e t a t e / a c e t i c acid/water (1 : 1 : 1 : 1). In parallel, 10 ttl samples of the reaction mixture were added to a series of 7 m l liquid scintillation vials, each containing 0.1 ml of 0.2 M NaOH. Each vial then received 1.5 ml of toluene/ethyl acetate (2:1 by vol.). The vials were capped, inverted twenty times and then uncapped. A 1 ml sample of each organic phase was dried under N z.
Each residue was dissolved in 50 #! of methanol and I0 tzl of the resulting solution was applied to a silica gel thin-layer chromatography plate. Authentic standards were applied and the plate was developed as described above. Hydrolysis of Arg-Pro-Pro-[~H]benzylamide was essentially completed within 30 min. When, at times > 30 rain, samples of reaction mixture were applied directly to a silica gel plate, 3H was associated with two products: Pro-Pro-benzylamide and benzylamine in a ratio of about 9 • 1. When similar samples were added to the NaOH solution and worked up as described, 3H was associated with the same two products in a similar ratio. These data indicate that Pro-Pro-[aH]benzylamide is the primary radiolabeled product formed by aminoacylproline hydrolase. Pro-Pro-[3H]benzylamide may itself be reactive with the aminoacylproline hydrolase, but under the conditions of our experiment, Pro[3H]benzylamide was not formed. Conceivably, the small amount of [3H]benzylamine formed during the reaction arose artifactually via a Pro-Pro diketopiperazinc rearrangement (Ref. 13 and references stated therein). However, Lasch et al. [10] have reported that Pro-Pro-[4'-NO_,]anilide is highly reactive with dipeptidyl aminopeptidase IV and Pro-Pro-[3H]benzylamide may be reactive as well. Rat lung contains dipeptidyl aminopeptidase IV [14], which may account for the formation of small amounts of [3H]benzylamine.
Assay protocol Except where noted, assays of aminoacylproline hydrolase activities were performed as follows: To a 7 ml liquid scintillation vial (test vial) was added 100 #1 of buffered enzyme source (e.g., rat lung homogenate diluted 1:200 in assay buffer). For a blank reaction, 100 p.1 of buffer was added to a 7 ml vial (blank vial). Blank and test vials were transferred to a 37°C shaking water bath and reactions were started by adding 100/~! of buffered substrate (1.0 tzCi of Arg-Pro-Pro-[3H]ben zylamide per ml of assay buffer; about 22.3 nM in the final reaction mixture) to each vial. Reactions were stopped (typically after 10 min of incubation) by adding 200/~1 of 0.2 M NaOH. Each vial then received 3 ml of Ventrex LSC #1 (a proprietary toluene/ethyl acetate solution containing scintillants; Ventrex Laboratories, Inc.). The vials were capped and inverted 20-times to partition unhydrolyzed subs~rate and 3H-labeled product. The initial substrate concentration So was measured by add,~ng 100 ttl of buffered substrate to a 7 ml vial containing 5 ml of an aqueous compatible liquid scintillation cocktail such as Scint A (Packard Instruments). All vials were submitted for liquid scintillation counting. The substrate concentration (22.3 nM) is well within the range of first-order enzyme kinetics (Km 7" 10 - 7
136 M; see below), and assay results were computed using the integrated first-order rate equation I/ma,~lKm = {I/r)lntSo/S)
(I)
where S is the concentration of substrate remaining at time t. The extraction of product into the upper ('counting') phase is not quantitative. The fractional extraction of product (fp) is 0.8. Similarly, a small fraction of unhydrolyzed substrate (fs = 0.07) enters the counting phase. To correct for fp and f~ and to take into account that radioactivities and not concentrations are obtained, Eqn. 1 is modified to
Vma~/rm=cl/r)ln(So/[So-[tT-B)/(fD-f,)]])
(2)
where S o is initial substrate in dpm (disintegrations per min); T is the dpm of the test vial and B is the dpm of the blank vial [15]. Eqn. 2 can be simplified to Vm~,/rm = (l/r)t.[((/~So)- B)/((f~So)- r)]
(3)
In experiments in which inhibitors or alternative substrates were added or believed to be present, the leftmost term of Eqn. 3 was understood to be apparent (VmJKm).
TABLE !
Effects of Xaa-Pro-Pro tripeptides and related compounds on aminoacylproline hydrolase activity A given test peptide (or buffer alone for the control reaction) was added to Arg-Pro-Prd3H~en~larMde, and the reaction was started by adding rat lung homogenate diluted in buffer. Each peptide was tested in replicate at 12 different concentrations to enable elaboration of a plot of percent inhibition vs. log peptide concentration. Results of replicate assays differed by less than 5%. ICs0 values were obtained by interpolation. Xaa denotes the N-terminal amino acid residue. XaaArgAspGluAlaMetlieLeuTyrPhehPhe- ~ N-methyI-PheTrpCys-
ICso (/,tM) 103 1082 2770 354 97 367 434 323 588 114 619 980 478
Other compounds
Results and Discussion
l-butanoyl-Pro-Pro hydrocin namoyI-Pro-Pro benzoyI-Pro-Pro Pro-Pro Pro-Pro diketopiperazine h
Apparent K m
a hPhe: homophenylalanine (2-amino-4-phenylbutanoic acid). b No inhibition at 6000 p.M.
To measure apparent Kin, 12 serial 1:2 dilutions of unlabeled Arg-Pro-Pro-benzylamide were added to a single concentration of radiolabeled substrate to yield a concentration range of 22 nM-100 #M. With any amount of rat lung or plasma or guinea pig plasma, the apparent K m was 7" 10 -7 M (for rat plasma enzyme, estimated via Lineweaver-Burk, Hanes and EadieHofstee plots, K m 0.712, 0.729 and 0.709/.tM, respectively). The close agreement of results obtained using three crude sources of enzyme supports the view that the assay substrate is highly selective for aminoacylproline hydrolase. Further, the apparent K m values reported here are in good agreement with those measured using apparently pure guinea pig serum aminoacylproline hydrolase [22].
234 1330 3 820 624 >> 6000
K i values for competitive, noncompetitive and mixed competitive-noncompetitive inhibitors [16]. Of the Xaa-Pro-Pro tripeptides tested, those in which Xaa-were acidic residues (e.g., Glu-) were least preferred. The heterocyclic side chain of Trp- also provided weak binding. The relatively high binding affinities of Arg.-, Met-, hPhe- and butanoyl-Pro-Pro suggest that relatively long aliphatie side chains assist binding, even when a free a-amino function is not present. The comparable binding affinities of Pro-Pro and the Phe- and N-methyl-Phe-Pro-Pro peptides suggest that a secondary a-amino group binds to aminoacylproline hydrolase as well as a primary a-amino function.
Effects of Xaa-Pro-Pro tripeptides and related compounds
Effects of inhibitors
Using a similar approach, we examined for effects of peptides. Results are expressed as the concentration of peptide at which enzyme activity was inhibited by 50% (ICs0) and are shown in Table I. Because the enzyme:substrate reactions were conducted under first-Grder enzyme kinetics, the ICs0 values should approximate K m values for alternative substrates and
The aminoacylproline hydrolase activity of rat lung homogenate was not inhibited by any of amastatin, bestatin, phosphoramidon or RAC-X-65 in concentrations that strongly inhibit leucine aminopeptidase (EC 3.4.11.1), aminopeptidases A and N (EC 3.4.11.7 and EC 3.4.11.2), endopeptidase EC 3.4.24.11 and angiotensin converting enzyme (EC 3.4.15.1) [17,18] (see
137 Table II). The activity was readily inhibited by ophenanthroline and simple mercapto compounds including captopril. Under conditions of our experiments, the reaction was notably sensitive to effects of 2-mercaptoethanol (IC.~n 5 / t M ) . Although our results were consistent with the proposal that aminoacylproline hydrolase is a metallo enzyme [5-7], EDTA was a poor inhibitor unless preincubated with the enzyme for 30-60 rain (IC5o of 500 ~tM after a 30 min preincubation). Preineubation of the other inhibitors with the enzyme source did not yield IC50 values that differed by more than 8% from those shown in Table II. Others have reported distinctly weaker inhibitory activities for 2-mercaptoethanol, dithiothreitol and o-phenanthroline [3,4]. These differences may be owing to the use of different assay conditions. Under conditions of first-order reaction kinetics, virtually no protection of the enzyme catalytic site is afforded by substrate binding. Under conditions of near-zero enzyme kinetics, TABLE I!
Effects o f inhibitors on aminoacylproline hydrolase acth'ity Arg-Pro-Pro-[3H]benzylamide was reacted with rat lung homogenate in the presence or absence of an inhibitor. Enzyme activity in the presence of an inhibitor is expressed as percent of control. Results of replicate assays differed by less than 7%. Addition None Amastatin Bestatin Phosphoramidon Captopril RAC-X-65 b EDTA
Concentration (/.tM) -
Relative activity (%) 100
50 a
96
73 ~ 184 a
101 100
29 7
47 70
! 13 a
100
10000 1000
66 89
o-Phenanthroline
250 70
2 41
2-Mereaptoethanol
116 32 6 1
0 14 45 81
Dithiothreitol
164 41 11
4 25 58
2-MGP ¢
100 26 7
62 83 92
Highest concentration used. b RAC-X-65: N-[l(S)earboxy-3-carboxanilidopropyi]-Ala-Pro, a tight binding inhibitor of angiotension converting enzyme [18]. c 2-MGP: 2-mercaptomethyl-3-guanidinoethylthiopropanoic acid, an inhibitor of basic carboxypeptidase enzymes [23].
TABLE III
Aminoaeylproline hydrolase acticities of rat tissues Tissues of five Sprague-Dawley female rats were collected and homogenized as described in the text for rat lung. Five or more serial ! :2 dilutions of each tissue were reacted in re,~licate with Arg-ProPro-[3H]benzylamide (22 nM in the final reaction mixture) and results were expressed as V,,,,~/K., in m i n - i. Results of replicate assays differed by less than 4%. Homogenate protein was measured by the BCA method (Pierce Chemical) and expressed in g of protein/liter of homogenate. Specific activities are in units of liters g - t rain- t. Tissue
Specific activity ( 1 g - i min - I )
Kidney Lungs Jejunum Ileum Plasma Liver Thymus Aorta Brain Duodenum Heart Diaphragm Adrenal gland Colon Spleen Skin Esophagus Ovary Ve na cava Pancreas Fat Stomach Uterus Skeletal muscle Thyroid
0.894 0.622 0.341 0.206 0.117 0.080 0.058 0.051 0.033 0.032 0.027 0.026 0.025 0.024 0.023 0.019 0.019 0.018 0.014 0.012 0.011 0.010 0.009 0.006 0.004
the near saturation of enzyme with substrate may greatly reduce the rate at which the putative metal cofactor can be removed. The results shown in Table 11 indicate that neither dipeptidyl aminopeptidase IV nor prolyl endopeptidase (EC 3.4.21.26) contributes to the hydrolysis of Arg-ProPro-[3H]benzylamide. Both of the latter enzymes are fully active in the presence of much higher concentrations of 2-mercaptoethanol and o-phenanthroline than were used in this study [4,19,20].
Enzyme activities of rat and guinea pig tissues Table Ill shows specific activities of aminoacylproline hydrolase in major organs and tissues of female rats. Kidney, lung, small intestine and plasma are relatively rich sources of the enzyme. A comparison of specific activities of the enzyme in guinea pig tissues (Table IV) indicates some significant differences. Firstly, the guinea pig enzyme specific activities in spleen, kidney, plasma and most other tissues are much higher than those of rats. Secondly, lung is not among
138 TABLE IV
Amittoacylproline hydrolase actirities of guinea pig tissues Tissues of two female guinea pigs of the Hartley-Ft. Detrick strain were collected and homogenized as described for rat lung. Assays were perfi,rmed as described in the text and legend of Table Ill. Tissue
Specific activity (1 g I rain- ;)
Spleen Kidney Liver Plasma Ileum Jejunum Colon Duodenum Lungs Ve na cava Fat Thymus Uterus Aorta Esophagus Pancreas Stomach Ovary Heart Diaphragm Adrenal Thyroid Skeletal muscle Brain Skin
7.617 4. i 95 1.259 0.869 0.698 0.604 0.513 0.499 0.415 0.4 ! 4 11.266 11.256 0.256 11.1811 11.15I O. 149 0.143 0.132 0.1211 0.118 0.112 0.109 0.107 0.1164 11.059
the richest sources of the guinea pig enzyme. Although the meaning of these interspeeific differences is not known, it is possible that aminoacylproline hydrolase plays a more prominent role in the physiology of guinea pigs than in that of rats.
Performance of the assay To examine the reproducibility of the assay, we measured the aminoacylproline hydrolase activities of plasmas of 20 guinea pigs: m e a n Vmax/K m in m i n - t 38.685 + 2.264 (S.E.), S.D. 10.127. For a single plasma sample assayed in replicate 10 times over 10 days, the mean Vm~JK,, was 50.4 min-i, S.D. 1.76. Favorable lower limits of detection were suggested by the fact that a 1 : 2000 dilution of guinea pig plasma of l/max//Km of 50.4 min- ~ hydrolyzed 22.2% of the substrate over a 10 min incubation. Using the value of kcat/K m (1.78" 108 M-t min-~) provided in the accompanying report [22], the latter guinea pig plasma contained aminoacylproline hydrolase at a concentration of 0.283/~M.
Comment The assay procedure described here is an endpoint assay and may be disadvantageous for certain kinds of kinetic studies (e.g., examination of effects of slow tight-binding inhibitors [21]). Further, the procedure
requires access to a liquid scintillation counter and provision for disposal of organic solvents and nonvolatile 3H. However, the assay is simple to perform, provides favorable lower limits of detection, and, for the enzyme sources used thus far, is apparently specific. The assay is, for many purposes, best performed under conditions of first-order enzyme kinetics. By so doing, the first-order rate constant, Vmax/Km, is measured directly. By definition, possible inhibitory effects of substrate or products are eliminated or greatly reduced. Under first-order enzyme kinetics (in the absence of inhibitors or alternative substrates), fractional substrate utilization per unit time is a direct function of the second-order rate constant, kc~t/Km [16]. When an inhibitor is present, the IC5o of the inhibitor can be taken as its K~ value or a very close approximation. As noted previously, the ICs0 of an alternative substrate can be taken as its Kin. Given the high reactivity of the substrate, the simplicity and apparent specificity of the assay, and the ease with which certain kinetic parameters and constants (e.g., l/max/Kin, K m and K i) c a n be measured under conditions of first-order enzyme kinetics, the radioassay described here should materially assist future efforts to characterize aminoacylproline hydrolase in great detail. As described in the accompanying report [22], the radioassay has been used to monitor the purification of guinea pig serum aminoacylproline hydrolase.
Acknowledgements We thank Ms. Magda Guzman for preparing the typescript. This work was supported in part by the U.S. Public Health Service (grants HL 39684 and EY 07135).
References 1 Ryan, J.W., Roblero, J. and Stewart, J.M. (1968) Biochem. J. 110, 795 -797. 2 Ryan, J.W., Roblero, J. and Stewart, J.M. 119701 Adv. Exp. Med. Biol. 8, 240-262. 30rawski, A.T., Susz, J.P. and Simmons, W.H. 119871 Mol. Cell. Biochem. 75, 123-132. 40rawski, A.T., Susz, J.P. and Simmons, W.H. 119891 Adv. Exp. Med. Biol. 247, 355-364. 5 Yaron, A. and Mlynar. D. 119681 Biochem. Biophys. Res. Commun. 32, 658-663. 6 Nordwig, A. and Dehm, P. (19681 Biochim. Biophys. Acta 160, 293-295. 7 Dehm, P. and Nordwig, A. 119701 Eur. J. Biochem. 17, 364-371. 8 Fleminger, G., Carmel, A., Goldenberg, D. and Yaron, A. (1982) Eur. J. Biochem. 125, 609-615. 9 Holtzman, E.J., Pillay, G., Rosenthal, T. and Yaron, A. 119871 Anal. Biochem. 162, 476-484. 10 Lasch, J., Koelsch, R., Steinmetzer, T., Neumann, U. and Demuth, H.-U. 119881 FEBS Left. 227, 171-174.
139 11 Yoshimo!o, T., Murayama, N. and Tsuru, D. (1988) Agric. Biol. Chem. 52, 1957-1963. 12 Ryan0 J.W., Chung, A.Y.K., Berryer, P. and Sheffy, D. (1989) FASEB J. 3, A!026. 13 Steinberg, S.M. and Bada, J.L. (1983} J. Org. Chem. 48, 22952298. 14 Krepela, E., Vicar, J., Zizk~wa, L., Dazafirek, E., Kolar, Z. and Lichncwsky, V. (1985) Lung 163, 33-54. 15 Ryan, J.W. (1988} in lmmunochemical Techniques. Part M, Chemotaxis and Inflammation, Methods in Enz3,mology (Di Sabato, G., ed.), Vol. 163, pp. 194-210, Academic Press, San Diego. 16 Cornish-Bowden, A. (1979) Fundamentals of Enzyme Kinetics, p. 22, p. 79, p. 84, Bulterworths, London.
17 Kenny, A.J., Stephenson, S.L. and Turner, A.J. (1987) Mammalian Ectoenzymes: Res. Monogr. Cell Tissue Physiol. 14, 169210. 18 Ryan, J.W., Chung, A.Y.K., Berryer, P., Murray, M.A. and Ryan, J.P.A. (1986) Adv. Exp. Med. Biol. 198, 419-425. 19 Walter, R., Simmons, W.H. and Yoshimoto, T. (1980) Mol. Ceil. Biochem. 30. 111-127. 20 Yoshimoto. T., Orlt~wski, R.C. and Walter, R. (1977) Biochemistry 16, 2942-2948. 21 Shapiro, R. and Riordan, J.F. (1984) Biochemistry 23, 5234-5240. 22 Ryan, J.W., Valido, F., Berryer, P., Chung, A.Y.K. and Ripka, J. (1992) Biochim. Biophys. Acta, 1119, 140-147. 23 Plummer. T.H. and Ryan, T.J. (1981) Biochem. Biophys. Res. Commun. 98, 448-454.
Biochimica et Biophysica Acta, I ! 19 (1992) 133-139 © 1992 Elsevier Science Publishers B.V. All rights reserved 0167-4838/92/$05.00
BBAPRO M I ~
A radioassay for aminoacylproline hydrolase (aminopeptidase P) activity James W. Ryan, Alfred Y.K. Chung, Pierre Berryer and Donald H. Sheffy, Jr. Department of Medicine, Unicersityof Miami Schoolof Medicine, Miami, FL (U.S.A.) (Received 13 September 1990) (Revised manuscript received 22 August 1991)
Key words: Aminoacylproline hydrolase; Aminopeptidase P; Bradykinin; Radioassay substrate; First-order enzyme kinetics; (Rat); (Guinea pig)
A mdionssay was devdoped in which aminoacylproline hydrolase acts on Arg-Pro-Pro-[aH]benzylamide to yield ~rglnine plus Pro-Pro-[aH]benzylamide. By Mopping the reaction with base (0.1 M NaOH), the radioactive product is deprotonated to an organophilic form and is separable from the hydrophilic substrate by extraction of the alkaline aqueous solution with an organic solvent. When scintillants are included in the organic solvent, the enzyme:substrate reaction, extraction and quantification of Pro-Pro-[aH]benzylamide can all be conducted using a single liquid scintillation vial. Thus, aminoacylproline hydrolase activity is measured in terms of the rate of release of Pro-Pro-[aH]benzTlamide. The substrate is obtainable at > 20 C i / m m o l , which enables its use under conditions of first-order enzyme kinetics. Conditions of near-zero order kinetics are readily attained by adding unlabeled substrate (K_ 0.7/tM). The substrate is highly reactive (a 1:2000 dilution of guinea pig plasma hydrolyzed > 10% of the substrate during a 10 min incubation at 37°C) and specific in that it is not degraded by leucine aminopeptidase, aminopeptidase A or N, dipeptidyl peptidase IV nor prolyl endopeptidase. The assay was used to measure aminoacTIproline hydrolase specific activities in tissues of rat and guinea pig. Activity was found in virtually all major tissues of both species, and some guinea pig tissues (e.g. kidney and plasma) were found to be notably rich sources of the e n z ~ e .
Introduction During a single passage through the rat pulmonary vascular bed, bradykinin (ArgLPro2-Pro3-Gly4-Phe 5Ser6-ProLPheS-Arg 9) is hydrolyzed at 5 of its 8 peptide bonds [1,2]. One hydrolysis occurs at the Argl-Pro 2 bond and is catal~ed by an aminoacylproline hydrolase [1-4]. An aminoacylproline hydrolase has been isolated from Escherich3a coli, and others have been found in tissues such as pig kidney, rat intestine, rat and bovine lung and human serum [5-10]. Whether these enzymes are structurally related is not known, but all are called aminopeptidase P, the name given the bacterial enzyme (EC 3.4.11.9) [5]. Several methods have been developed to measure aminoacylproline hydrolase activity, but none is at once
Correspondence: J.W. Ryan, 1399 N.W. 17th Avenue, No. 200 Miami, FL 33125, U.S.A.
simple and convenient. Initially, the bacterial enzyme was measured by its ability to release proline from poly-L-proline [5]. The pig kidney enzyme was measured in terms of the release of glycine from Gly-ProHyp [7]. Lasch et al. [10] have described a two step chromogenic assay in which the aminoacylproline hydrolase acts on Gly-Pro-Pro-[4-NO2]anilide to form Gly and Pro-Pro-[4-NOe]anilide in the first step, and the latter product is hydrolyzed by an excess of dipeptidyl aminopeptidase IV (dipeptidyl peptidase IV, EC 3.4.14.5) to release the chromophore 4-NO2-aniline. Yaron and colleagues [8,9] have introduced a series of relatively insoluble internally-quenched fluorogenic substrates, of which Lys[e-DNP)-Pro-Pro-NH-CH2CH2-NH-CO-CoH4(2'-NH2) was the most conveniently used. Yoshimoto et al. [11] have reported that Gly-Pro-7-amido-4-methylcoumarin can be used to measure aminoacyiproline hydrolase activity (in the presence of excess proline iminopeptidase, EC 3.4.11.5) when dipeptidyl aminopeptidase IV is not present or is fully inhibited.
134 To capitalize on the finding that bradykinin is highly reactive with the aminoacylproline hydrolase of rat lung [1-4], we have prepared the bradykinin N-terminal homolog Arg-Pro-Pro-[aH]benzylamide. The substrate is hydrolyzed to yield arginine plus Pro-Pro[3H]benzylamide. At alkaline pH, Pro-Pro-benzylamide is readily extractable into organic solvents. Thus, the hydrophilic substrate can be used to measure aminoacylproline hydrolase activity in terms of the rate of formation of the organophilic product. An abstract describing the new assay procedure has been published I12l. Materials and Methods
Pro-[3 '-iodo]benzylamide A suspension of Boc-Pro-OH (451 mg, 2 mmol) and 540 mg of [3'-iodo]benzylamine hydrochloride in 5 ml of dichloromethane at 0°C was neutralized with 0.272 ml (2 mmol) of N-ethyl morpholine. Dicyclohexylcarbodiimide, 412 mg (2 mmol) in 2 ml of dichloromethane was added. The reaction mixture was stirred at 0°C for 1 h and then at room temperature overnight. Insoluble dicyclohexylurea and N-ethyl morpholine hydrochloride were removed by filtration, and the precipitate was washed with ethyl acetate. The filtrates were combined, washed until neutral and then dried over MgSO4. Solvent was removed in vacuo resulting in a clear oily residue. The Boc group was removed in 5.5 M hydrogen chloride in ethyl acetate, 2 h at 0°C. Solvent was removed in vacuo. Pro-[3'iodo]benzylamide was obtained as white crystals in ethanol and diethyl ether; m.p. 177-177.5°C (decomp), weight 503 mg; for C~z H 16N2IC10: calculated C 39.31, H 4.40, N 7.64, l 34.61, Cl 9.67, O 4.36; found C 39.39, H 4.51, N 7.68, 1 34.53, CI 9.50.
N~-Boc.Arg( Tos).Pro.OH Dicyclohexylcarbodiimide, 2.06 g (10 mmol), in 3 ml of dimethylformamide was added with stirring to a solution of 4.285 g of N~-Boc-Arg(Tos)-OH and 1.35 g of 1-hydroxybenzotriazole in dimethyformamide in an ice-acetone bath. To this mixture was added 2.66 g of proline benzyl ester hydrochloride and 1.1 ml of Nmethyl morpholine in dimethylformamide. The resulting mixture was stirred in the ice-acetone bath for 1 h and then at 4°C overnight. The reaction mixture was filtered, and the precipitate was washed with ethyl acetate. The filtrates were combined, washed until neutral and then dried over MgSO4. Solvent was removed in vacuo. An oily residue was obtained by rotary evaporation. The residue was triturated with isopropyl ether and then filtered to yield a white powder, 4.57 g; m.p. 55-57°C; amino acid analysis: Arg I Pro v The benzyl ester group was removed by hydrogenolysis with 100 mg of 10% Pd on carbon as catalyst in absolute
ethanol at 25 psi for 1 h. The catalyst was removed by filtration, and solvent of the filtrate was removed in vacuo to yield a white foam. On silica gel thin-layer chromatography (methanol/chloroform (1:1 by vol.)), the product N~'-Boc-Arg(Tos)-OH behaved as a single substance, R r 0.54 (o-tolidine/Kl stain).
N ~-Boc-Arg( Tos)-Pro-Pro-[ 3'-iodo ] benzylamide . diacetate Dicyclohexylcarbodiimide 0.206 g (1 mmol) in 2 ml of dichloromethane was added to a stirred solution of N"-Boc-Arg(Tos)-Pro-OH, 1 mmol, and 135 mg of 1-hydroxybenzotriazole in 3 ml of dimethylformamide/dichloromethane (1:2 by vol) at -10°C. To this mixture was added 367 mg of HCI-Pro-[3'iodo]benzylamide and 0.11 mi of N-ethylmorpholine in 3 ml of dimethylformamide. The reaction mixture was stirred at - 10°C for 1 h, at 4°C overnight, and then at room temp for 2 h. The precipitate was removed by filtration and was washed with water and ethyl acetate. The organic phase of the combined filtrates was washed until neutral and then dried over MgSO a. Solvent was removed in vacuo to yield 0.61 g of a foam-like material. Amino acid analysis: Arg t ProL85.
Arg-Pro-Pro-[ 3 '-iodo ]benzylamide The Boc and tosyl protecting groups were removed by treating 280 mg of the protected peptide with 10 ml of HF in the presence of 0.6 ml of anisole at 0°C for 1 h. HF was removed in vacuo and the residue was partitioned in diethyl e t h e r / 2 % (by vol.) acetic acid. The crude material in the lower phase was applied to an LH-20 column (1.2 × 95 cm) equilibrated and developed with 6% 1-butanol. Fractions containing product were pooled and solvent was removed in vacuo to yield 178 mg of residue. The residue was applied to a Sephadex G-25 column (1.2 × 100 era) equilibrated with the upper phase of 1-butanol/acetic acid/water (4: 1 : 5 by vol.). The column was washed with the upper phase and then product, 139 rag, was eluted with the lower phase. Amino acid analysis: Arg I Proi.s7. On silica gel thin-layer chromatography (1-butanol/pyridine/aeetie acid/water (15: 10:3 : 12 by vol.), 1-butanol/ethyl ace t a t e / a c e t i c a c i d / w a t e r (1 : 1 : 1 : 1) and ethyl acetate/pyridine/acetie acid/water (5: 5: 1 : 3)), the product behaved as a single substance; respective R F values: 0.51, 0.42 and 0.67 (ninhydrin, Sakaguchi and o-tolidine/KI staining).
Arg-Pro-Pro-[ JH]benzylamide Arg-Pro-Pro-[3'-iodo]ben~iamide, 11 nag in 2 ml of dimethylformamide/water (1:1 by vol), was dehalogenated in 10 Ci of 3H 2 gas using 10% Pd on carbon as catalyst to yield Arg-Pro-Pro-[3'-3H]benzylamide, 350 mCi at 22.4 Ci/mmol. The tritiated product was dis-
135 solved in 70% ethanol (0.5 mCi/ml), in which it was stable on storage at -20°(2 for more than 3 years.
Other peptides Pro-benzylamide was prepared as described above for Pro-[3'-iodo]benzylamide by substituting benzylamine for [Y-iodo]benzylamine. Pro-Pro-benzylamide was prepared by reacting Z-Pro-OH with Pro-benzylamide via dicyclohexylcarbodiimide coupling in the presence of 1-hydroxybenzotriazole in dichloromethane. The Z-protecting group was removed by hydrogenolysis in methanol using 10% Pd on carbon as catalyst.
Enzyme sources Except where noted, we used as an enzyme source rat "plasma (anticoagulated with heparin), guinea pig plasma or homogenate of rat lung (2 ml of water containing 1% (by vol.) of Nonidet P-40 per g of rat lung; homogenized by three 15-s bursts of a Polytron tissue disrupter at setting 3).
Assay buffer Except for measurements of Kin, Arg-Pro-Pro[3H]benzylamide was used in concentrations within the range of first-order enzyme kinetics (typically at about 20 nM). In Hepes buffers, the pH optimum of aminoacylproline hydrolase was found to be broad (pH 8.0 to at least 9.3). Thus, subsequent assays used 0.05 M Hepes buffer (pH 8.3) containing 0.15 M NaCI and Hibiclens (Stuart Laboratories), 0.01% by vol. (assay buffer). NaCl was used to make the buffer near-isotonic for assay of enzymic activities of intact cells, and Hibiclens (chlorhexidine gluconate) was used as a bacteriostat.
Product formation To examine product formation, 2 ml of the substrate in assay buffer (10.7 #Ci, 0.24/~M) was incubated at 37°C with 2 ml of rat lung homogenate diluted 1:200 in assay buffer; protein concentration of 109 ttg/ml. At timed intervals (0-60 min), 10 /zl samples of the incubation mixture were spotted on a thin-layer silica gel plate. After all samples were collected, 5 ttg samples of authentic benzylamine, Pro-Bz, Pro-Pro-Bz and Arg-Pro-Pro-Bz were applied to the plate, and the plate was developed with I-butanol/acetic acid/water (75 : 13 : 12) or 1-butanol/ethyl a c e t a t e / a c e t i c acid/water (1 : 1 : 1 : 1). In parallel, 10 ttl samples of the reaction mixture were added to a series of 7 m l liquid scintillation vials, each containing 0.1 ml of 0.2 M NaOH. Each vial then received 1.5 ml of toluene/ethyl acetate (2:1 by vol.). The vials were capped, inverted twenty times and then uncapped. A 1 ml sample of each organic phase was dried under N z.
Each residue was dissolved in 50 #! of methanol and I0 tzl of the resulting solution was applied to a silica gel thin-layer chromatography plate. Authentic standards were applied and the plate was developed as described above. Hydrolysis of Arg-Pro-Pro-[~H]benzylamide was essentially completed within 30 min. When, at times > 30 rain, samples of reaction mixture were applied directly to a silica gel plate, 3H was associated with two products: Pro-Pro-benzylamide and benzylamine in a ratio of about 9 • 1. When similar samples were added to the NaOH solution and worked up as described, 3H was associated with the same two products in a similar ratio. These data indicate that Pro-Pro-[aH]benzylamide is the primary radiolabeled product formed by aminoacylproline hydrolase. Pro-Pro-[3H]benzylamide may itself be reactive with the aminoacylproline hydrolase, but under the conditions of our experiment, Pro[3H]benzylamide was not formed. Conceivably, the small amount of [3H]benzylamine formed during the reaction arose artifactually via a Pro-Pro diketopiperazinc rearrangement (Ref. 13 and references stated therein). However, Lasch et al. [10] have reported that Pro-Pro-[4'-NO_,]anilide is highly reactive with dipeptidyl aminopeptidase IV and Pro-Pro-[3H]benzylamide may be reactive as well. Rat lung contains dipeptidyl aminopeptidase IV [14], which may account for the formation of small amounts of [3H]benzylamine.
Assay protocol Except where noted, assays of aminoacylproline hydrolase activities were performed as follows: To a 7 ml liquid scintillation vial (test vial) was added 100 #1 of buffered enzyme source (e.g., rat lung homogenate diluted 1:200 in assay buffer). For a blank reaction, 100 p.1 of buffer was added to a 7 ml vial (blank vial). Blank and test vials were transferred to a 37°C shaking water bath and reactions were started by adding 100/~! of buffered substrate (1.0 tzCi of Arg-Pro-Pro-[3H]ben zylamide per ml of assay buffer; about 22.3 nM in the final reaction mixture) to each vial. Reactions were stopped (typically after 10 min of incubation) by adding 200/~1 of 0.2 M NaOH. Each vial then received 3 ml of Ventrex LSC #1 (a proprietary toluene/ethyl acetate solution containing scintillants; Ventrex Laboratories, Inc.). The vials were capped and inverted 20-times to partition unhydrolyzed subs~rate and 3H-labeled product. The initial substrate concentration So was measured by add,~ng 100 ttl of buffered substrate to a 7 ml vial containing 5 ml of an aqueous compatible liquid scintillation cocktail such as Scint A (Packard Instruments). All vials were submitted for liquid scintillation counting. The substrate concentration (22.3 nM) is well within the range of first-order enzyme kinetics (Km 7" 10 - 7
136 M; see below), and assay results were computed using the integrated first-order rate equation I/ma,~lKm = {I/r)lntSo/S)
(I)
where S is the concentration of substrate remaining at time t. The extraction of product into the upper ('counting') phase is not quantitative. The fractional extraction of product (fp) is 0.8. Similarly, a small fraction of unhydrolyzed substrate (fs = 0.07) enters the counting phase. To correct for fp and f~ and to take into account that radioactivities and not concentrations are obtained, Eqn. 1 is modified to
Vma~/rm=cl/r)ln(So/[So-[tT-B)/(fD-f,)]])
(2)
where S o is initial substrate in dpm (disintegrations per min); T is the dpm of the test vial and B is the dpm of the blank vial [15]. Eqn. 2 can be simplified to Vm~,/rm = (l/r)t.[((/~So)- B)/((f~So)- r)]
(3)
In experiments in which inhibitors or alternative substrates were added or believed to be present, the leftmost term of Eqn. 3 was understood to be apparent (VmJKm).
TABLE !
Effects of Xaa-Pro-Pro tripeptides and related compounds on aminoacylproline hydrolase activity A given test peptide (or buffer alone for the control reaction) was added to Arg-Pro-Prd3H~en~larMde, and the reaction was started by adding rat lung homogenate diluted in buffer. Each peptide was tested in replicate at 12 different concentrations to enable elaboration of a plot of percent inhibition vs. log peptide concentration. Results of replicate assays differed by less than 5%. ICs0 values were obtained by interpolation. Xaa denotes the N-terminal amino acid residue. XaaArgAspGluAlaMetlieLeuTyrPhehPhe- ~ N-methyI-PheTrpCys-
ICso (/,tM) 103 1082 2770 354 97 367 434 323 588 114 619 980 478
Other compounds
Results and Discussion
l-butanoyl-Pro-Pro hydrocin namoyI-Pro-Pro benzoyI-Pro-Pro Pro-Pro Pro-Pro diketopiperazine h
Apparent K m
a hPhe: homophenylalanine (2-amino-4-phenylbutanoic acid). b No inhibition at 6000 p.M.
To measure apparent Kin, 12 serial 1:2 dilutions of unlabeled Arg-Pro-Pro-benzylamide were added to a single concentration of radiolabeled substrate to yield a concentration range of 22 nM-100 #M. With any amount of rat lung or plasma or guinea pig plasma, the apparent K m was 7" 10 -7 M (for rat plasma enzyme, estimated via Lineweaver-Burk, Hanes and EadieHofstee plots, K m 0.712, 0.729 and 0.709/.tM, respectively). The close agreement of results obtained using three crude sources of enzyme supports the view that the assay substrate is highly selective for aminoacylproline hydrolase. Further, the apparent K m values reported here are in good agreement with those measured using apparently pure guinea pig serum aminoacylproline hydrolase [22].
234 1330 3 820 624 >> 6000
K i values for competitive, noncompetitive and mixed competitive-noncompetitive inhibitors [16]. Of the Xaa-Pro-Pro tripeptides tested, those in which Xaa-were acidic residues (e.g., Glu-) were least preferred. The heterocyclic side chain of Trp- also provided weak binding. The relatively high binding affinities of Arg.-, Met-, hPhe- and butanoyl-Pro-Pro suggest that relatively long aliphatie side chains assist binding, even when a free a-amino function is not present. The comparable binding affinities of Pro-Pro and the Phe- and N-methyl-Phe-Pro-Pro peptides suggest that a secondary a-amino group binds to aminoacylproline hydrolase as well as a primary a-amino function.
Effects of Xaa-Pro-Pro tripeptides and related compounds
Effects of inhibitors
Using a similar approach, we examined for effects of peptides. Results are expressed as the concentration of peptide at which enzyme activity was inhibited by 50% (ICs0) and are shown in Table I. Because the enzyme:substrate reactions were conducted under first-Grder enzyme kinetics, the ICs0 values should approximate K m values for alternative substrates and
The aminoacylproline hydrolase activity of rat lung homogenate was not inhibited by any of amastatin, bestatin, phosphoramidon or RAC-X-65 in concentrations that strongly inhibit leucine aminopeptidase (EC 3.4.11.1), aminopeptidases A and N (EC 3.4.11.7 and EC 3.4.11.2), endopeptidase EC 3.4.24.11 and angiotensin converting enzyme (EC 3.4.15.1) [17,18] (see
137 Table II). The activity was readily inhibited by ophenanthroline and simple mercapto compounds including captopril. Under conditions of our experiments, the reaction was notably sensitive to effects of 2-mercaptoethanol (IC.~n 5 / t M ) . Although our results were consistent with the proposal that aminoacylproline hydrolase is a metallo enzyme [5-7], EDTA was a poor inhibitor unless preincubated with the enzyme for 30-60 rain (IC5o of 500 ~tM after a 30 min preincubation). Preineubation of the other inhibitors with the enzyme source did not yield IC50 values that differed by more than 8% from those shown in Table II. Others have reported distinctly weaker inhibitory activities for 2-mercaptoethanol, dithiothreitol and o-phenanthroline [3,4]. These differences may be owing to the use of different assay conditions. Under conditions of first-order reaction kinetics, virtually no protection of the enzyme catalytic site is afforded by substrate binding. Under conditions of near-zero enzyme kinetics, TABLE I!
Effects o f inhibitors on aminoacylproline hydrolase acth'ity Arg-Pro-Pro-[3H]benzylamide was reacted with rat lung homogenate in the presence or absence of an inhibitor. Enzyme activity in the presence of an inhibitor is expressed as percent of control. Results of replicate assays differed by less than 7%. Addition None Amastatin Bestatin Phosphoramidon Captopril RAC-X-65 b EDTA
Concentration (/.tM) -
Relative activity (%) 100
50 a
96
73 ~ 184 a
101 100
29 7
47 70
! 13 a
100
10000 1000
66 89
o-Phenanthroline
250 70
2 41
2-Mereaptoethanol
116 32 6 1
0 14 45 81
Dithiothreitol
164 41 11
4 25 58
2-MGP ¢
100 26 7
62 83 92
Highest concentration used. b RAC-X-65: N-[l(S)earboxy-3-carboxanilidopropyi]-Ala-Pro, a tight binding inhibitor of angiotension converting enzyme [18]. c 2-MGP: 2-mercaptomethyl-3-guanidinoethylthiopropanoic acid, an inhibitor of basic carboxypeptidase enzymes [23].
TABLE III
Aminoaeylproline hydrolase acticities of rat tissues Tissues of five Sprague-Dawley female rats were collected and homogenized as described in the text for rat lung. Five or more serial ! :2 dilutions of each tissue were reacted in re,~licate with Arg-ProPro-[3H]benzylamide (22 nM in the final reaction mixture) and results were expressed as V,,,,~/K., in m i n - i. Results of replicate assays differed by less than 4%. Homogenate protein was measured by the BCA method (Pierce Chemical) and expressed in g of protein/liter of homogenate. Specific activities are in units of liters g - t rain- t. Tissue
Specific activity ( 1 g - i min - I )
Kidney Lungs Jejunum Ileum Plasma Liver Thymus Aorta Brain Duodenum Heart Diaphragm Adrenal gland Colon Spleen Skin Esophagus Ovary Ve na cava Pancreas Fat Stomach Uterus Skeletal muscle Thyroid
0.894 0.622 0.341 0.206 0.117 0.080 0.058 0.051 0.033 0.032 0.027 0.026 0.025 0.024 0.023 0.019 0.019 0.018 0.014 0.012 0.011 0.010 0.009 0.006 0.004
the near saturation of enzyme with substrate may greatly reduce the rate at which the putative metal cofactor can be removed. The results shown in Table 11 indicate that neither dipeptidyl aminopeptidase IV nor prolyl endopeptidase (EC 3.4.21.26) contributes to the hydrolysis of Arg-ProPro-[3H]benzylamide. Both of the latter enzymes are fully active in the presence of much higher concentrations of 2-mercaptoethanol and o-phenanthroline than were used in this study [4,19,20].
Enzyme activities of rat and guinea pig tissues Table Ill shows specific activities of aminoacylproline hydrolase in major organs and tissues of female rats. Kidney, lung, small intestine and plasma are relatively rich sources of the enzyme. A comparison of specific activities of the enzyme in guinea pig tissues (Table IV) indicates some significant differences. Firstly, the guinea pig enzyme specific activities in spleen, kidney, plasma and most other tissues are much higher than those of rats. Secondly, lung is not among
138 TABLE IV
Amittoacylproline hydrolase actirities of guinea pig tissues Tissues of two female guinea pigs of the Hartley-Ft. Detrick strain were collected and homogenized as described for rat lung. Assays were perfi,rmed as described in the text and legend of Table Ill. Tissue
Specific activity (1 g I rain- ;)
Spleen Kidney Liver Plasma Ileum Jejunum Colon Duodenum Lungs Ve na cava Fat Thymus Uterus Aorta Esophagus Pancreas Stomach Ovary Heart Diaphragm Adrenal Thyroid Skeletal muscle Brain Skin
7.617 4. i 95 1.259 0.869 0.698 0.604 0.513 0.499 0.415 0.4 ! 4 11.266 11.256 0.256 11.1811 11.15I O. 149 0.143 0.132 0.1211 0.118 0.112 0.109 0.107 0.1164 11.059
the richest sources of the guinea pig enzyme. Although the meaning of these interspeeific differences is not known, it is possible that aminoacylproline hydrolase plays a more prominent role in the physiology of guinea pigs than in that of rats.
Performance of the assay To examine the reproducibility of the assay, we measured the aminoacylproline hydrolase activities of plasmas of 20 guinea pigs: m e a n Vmax/K m in m i n - t 38.685 + 2.264 (S.E.), S.D. 10.127. For a single plasma sample assayed in replicate 10 times over 10 days, the mean Vm~JK,, was 50.4 min-i, S.D. 1.76. Favorable lower limits of detection were suggested by the fact that a 1 : 2000 dilution of guinea pig plasma of l/max//Km of 50.4 min- ~ hydrolyzed 22.2% of the substrate over a 10 min incubation. Using the value of kcat/K m (1.78" 108 M-t min-~) provided in the accompanying report [22], the latter guinea pig plasma contained aminoacylproline hydrolase at a concentration of 0.283/~M.
Comment The assay procedure described here is an endpoint assay and may be disadvantageous for certain kinds of kinetic studies (e.g., examination of effects of slow tight-binding inhibitors [21]). Further, the procedure
requires access to a liquid scintillation counter and provision for disposal of organic solvents and nonvolatile 3H. However, the assay is simple to perform, provides favorable lower limits of detection, and, for the enzyme sources used thus far, is apparently specific. The assay is, for many purposes, best performed under conditions of first-order enzyme kinetics. By so doing, the first-order rate constant, Vmax/Km, is measured directly. By definition, possible inhibitory effects of substrate or products are eliminated or greatly reduced. Under first-order enzyme kinetics (in the absence of inhibitors or alternative substrates), fractional substrate utilization per unit time is a direct function of the second-order rate constant, kc~t/Km [16]. When an inhibitor is present, the IC5o of the inhibitor can be taken as its K~ value or a very close approximation. As noted previously, the ICs0 of an alternative substrate can be taken as its Kin. Given the high reactivity of the substrate, the simplicity and apparent specificity of the assay, and the ease with which certain kinetic parameters and constants (e.g., l/max/Kin, K m and K i) c a n be measured under conditions of first-order enzyme kinetics, the radioassay described here should materially assist future efforts to characterize aminoacylproline hydrolase in great detail. As described in the accompanying report [22], the radioassay has been used to monitor the purification of guinea pig serum aminoacylproline hydrolase.
Acknowledgements We thank Ms. Magda Guzman for preparing the typescript. This work was supported in part by the U.S. Public Health Service (grants HL 39684 and EY 07135).
References 1 Ryan, J.W., Roblero, J. and Stewart, J.M. (1968) Biochem. J. 110, 795 -797. 2 Ryan, J.W., Roblero, J. and Stewart, J.M. 119701 Adv. Exp. Med. Biol. 8, 240-262. 30rawski, A.T., Susz, J.P. and Simmons, W.H. 119871 Mol. Cell. Biochem. 75, 123-132. 40rawski, A.T., Susz, J.P. and Simmons, W.H. 119891 Adv. Exp. Med. Biol. 247, 355-364. 5 Yaron, A. and Mlynar. D. 119681 Biochem. Biophys. Res. Commun. 32, 658-663. 6 Nordwig, A. and Dehm, P. (19681 Biochim. Biophys. Acta 160, 293-295. 7 Dehm, P. and Nordwig, A. 119701 Eur. J. Biochem. 17, 364-371. 8 Fleminger, G., Carmel, A., Goldenberg, D. and Yaron, A. (1982) Eur. J. Biochem. 125, 609-615. 9 Holtzman, E.J., Pillay, G., Rosenthal, T. and Yaron, A. 119871 Anal. Biochem. 162, 476-484. 10 Lasch, J., Koelsch, R., Steinmetzer, T., Neumann, U. and Demuth, H.-U. 119881 FEBS Left. 227, 171-174.
139 11 Yoshimo!o, T., Murayama, N. and Tsuru, D. (1988) Agric. Biol. Chem. 52, 1957-1963. 12 Ryan0 J.W., Chung, A.Y.K., Berryer, P. and Sheffy, D. (1989) FASEB J. 3, A!026. 13 Steinberg, S.M. and Bada, J.L. (1983} J. Org. Chem. 48, 22952298. 14 Krepela, E., Vicar, J., Zizk~wa, L., Dazafirek, E., Kolar, Z. and Lichncwsky, V. (1985) Lung 163, 33-54. 15 Ryan, J.W. (1988} in lmmunochemical Techniques. Part M, Chemotaxis and Inflammation, Methods in Enz3,mology (Di Sabato, G., ed.), Vol. 163, pp. 194-210, Academic Press, San Diego. 16 Cornish-Bowden, A. (1979) Fundamentals of Enzyme Kinetics, p. 22, p. 79, p. 84, Bulterworths, London.
17 Kenny, A.J., Stephenson, S.L. and Turner, A.J. (1987) Mammalian Ectoenzymes: Res. Monogr. Cell Tissue Physiol. 14, 169210. 18 Ryan, J.W., Chung, A.Y.K., Berryer, P., Murray, M.A. and Ryan, J.P.A. (1986) Adv. Exp. Med. Biol. 198, 419-425. 19 Walter, R., Simmons, W.H. and Yoshimoto, T. (1980) Mol. Ceil. Biochem. 30. 111-127. 20 Yoshimoto. T., Orlt~wski, R.C. and Walter, R. (1977) Biochemistry 16, 2942-2948. 21 Shapiro, R. and Riordan, J.F. (1984) Biochemistry 23, 5234-5240. 22 Ryan, J.W., Valido, F., Berryer, P., Chung, A.Y.K. and Ripka, J. (1992) Biochim. Biophys. Acta, 1119, 140-147. 23 Plummer. T.H. and Ryan, T.J. (1981) Biochem. Biophys. Res. Commun. 98, 448-454.
