Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
39 Clinical Science (1984) 66,39-45 Vascular angiotensin-converting enzyme activity in man and other species MIZUO MIYAZAKI, HIDEKI OKUNISHI, KAZUO NISHIMURA AND NOBORU TODA Department of Pharmacology, Shiga University of Medical Science, Seta, Ohtsu, Japan (Received 11April 1983;accepted 18 July 1983) Summary 1. Angiotensin-converting enzyme (ACE) activity in blood vessels of different species was determined. 2. ACE was solubilized by Nonidet P-40, and assayed by reversible phase high performance liquid chromatography. Approximately 98% ACE was recovered in the liquid phase by the use of the detergent. 3. The ACE activity varied with chloride ion (Cl-) concentrations; the maximum activities in dog, human, monkey and rabbit tissues were obtained at the concentrations of 800, 600, 600 and 300mmol/l respectively. The optimal C1concentration was quite similar in different tissues and plasma obtained from the same species. 4. The ACE activity in the cerebral,mesenteric, pulmonary and renal arteries was in a range between 1.01 and 1.60m-units/mg of protein in dogs and between 0.43 and 0.94m-unit/mg of protein in monkeys. The activity in dog aortae was 0.20 f 0.02 m-unit/mg of protein, and the activity in aortic endothelial cells was 2.61 f 0.65 m-units/mg of protein. ACE activities in the dog lung, kidney cortex and cerebral cortex were 28.6k2.6, 15.753.0 and 3.5 *0.6m-units/mg’of protein respectively. SA-446, a captopril-like ACE inhibitor, reduced the ACE activity in arteries in a dose-dependent manner. 5. Vascular ACE appears to be concentrated in the endothelium and may contribute to regulate vascular muscle tone and local blood flow by a conversion of angiotensin I into 11. Correspondence: Dr Mizuo Miyazaki, Department of Pharmacology, Shiga University of Medical Science, Seta, Ohtsu 520-21, Japan. Key words: angiotensin-converting enzyme, blood vessels, chloride, endothelium. Abbreviations: ACE, angiotensin-converting enzyme; HHL, hippuryl-L-histidyl-L-leucine. Introduction An angiotensin-converting enzyme, EC 3.4.15.1 (ACE), was first purified partially from horse plasma by Skeggs et al. [ 11 . The enzyme cleaves the carboxyl-terminal dipeptide, His-Leu, from the decapeptide, angiotensin I (ANG I), to generate a biologically active octapeptide, angiotensin I1 (ANG 11). Erdos [ 2 ] and Soffer [3] have clarified that ACE is functionally identical with kininase 11, which inactivates vasodepressor bradykmin. It is suggested that functional ANG I1 is not produced in plasma but in the vasculature of the lung [4]. Morphological and immunological studies have demonstrated that ACE is predominantly localized in the luminal surface of the endothelial cells which line blood vessels [5-71. A possible conversion of ANG I into ANG I1 in dog mesenteric [8] and intrarenal circulation [9] and in human forearm circulation [ 101 has been indicated. However, whether the Fonversion takes place in plasma or in the vascular wall is not clarified. From studies on isolated vessels exposed to artificial nutrient solutions, Aiken & Vane [ l l ] and Toda et al. [12] have suggested a local conversion of ANG I into ANGII, since ANGI induced vascular contractions are suppressed by ACE inhibitors and ANG I1 antagonists. In the present study, we attempted to apply an efficient biochemical method by using reversedphase high performance liquid chromatography [13] for measurement of the ACE activity in 40 M. Miyazaki et al. different blood vessels isolated from the human, monkey, dog and rabbit. The ACE activity in tissues including brains, kidneys, lungs and plasma was also assayed for comparison. Method Preparation of assay samples from dog, monkey, rabbit and human materials Mongrel dogs of both sexes, weighing 10-1 5 kg, were anaesthetized with intravenous injections of sodium pentobarbital (50 mg/kg) and killed by bleeding from the common carotid arteries. The brain, kidneys and lungs were rapidly removed. Basilar, middle cerebral and posterior cerebral arteries were isolated from the brain, and interlobar branches of the renal artery were isolated from the kidneys. Distal portions of the mesenteric artery, pulmonary artery and pulmonary vein were isolated. Outside diameters of these vessels were in a range from 0.3 to 1.0mm. The tissues and vessels were also obtained from Japanese monkeys (Macaca fitscata) of either sex, weighing 7-1 2 kg, anaesthetized with ketamine (10 mg/kg intramuscularly) and sodium pentobarbital (20 mg/kg intravenously). The thoracic aorta and lung were obtained from albino rabbits under sodium pentobarbital anaesthesia. The cerebral and gastroepiploic arteries were obtained from a 48 year-old female during autopsy 3 h after death, and the gastroepiploic artery was obtained from a totally resected cancer stomach of a 52 year-old male. AU materials, weighing 150-500 mg, were minced into small pieces and immediately placed into an ice-cold glass homogenizer containing 5 vol. of the homogenizing Tris-HC1 buffer solution. The solution consisted of Tris-HC1 buffer (20 mmol/l, pH 8.3), Mg(CH3C00)2 (5 mmol/l), KCl (30 mmol/l), sucrose (250 mmol/l) and 0.5% Nonidet P-40 [13, 141. The suspended solution was well homogenized in flaked ice at approx. 0°C and stored overnight at 4°C. The homogenized sample was centrifuged for 20min at 2 0 0 0 0 g at 4°C next morning. The supernatant was then incubated with substrate. Blood samples were taken (before anaesthesia) into heparinized tubes from the forelimb of monkeys and dogs and the ear artery of rabbits, and centrifuged for 15 min at 3000 rev./min at 4°C. Human plasma samples were obtained from healthy males. Preparation of endothelial cells from dog aortae Endothelial cells were collected from the thoratic aorta by the method described by Ody & Junod [7] with some modification. The aorta segment was divided into two parts. One part was used for the ACE assay of the whole aorta, and the other was used for the isolation of endothelial cells. This second part was rinsed with the Krebs-Ringer bicarbonate buffer, pH 7.4, containing collagenase at a concentration of 1 mg/ml. Both ends of the aorta segment were ligated, and the aorta was incubated for 20 min at 37°C with shaking. After the incubation, the collagenase solution was collected, and the lumen was rinsed three times with the Krebs-Ringer bicarbonate buffer at 4°C. All the solutions were pooled and centrifuged at 600g for 10 min at 4°C. The supernatant was discarded, and the pellet was again rinsed twice with the same buffer. The final pellet was suspended in the Tris-HC1 buffer solution containing 0.5% Nonidet P-40, and sonicated for 5 s. The solution was stored overnight at 4"C, and ACE activity in the supernatant was assayed. Measurement of ACE activity The ACE activity was determined as the production rate of hippuric acid from hippurylchistidylr-leucine (HHL), by the method described by Cushman & Cheung [15]. Assay mixture [0.25 ml:0.2 ml of potassium phosphate buffer, 100 mmol/l, pH 8.3, with optimal concentration of NaCl (300, 600, 600 and 800 mmol/l for rabbit, human, monkey and dog tissues), HHL (5 mmol/l) and 0.05 ml of the sample solution] was incubated for 30 min at 37°C. For the study of the effect of pH on the activity of ACE, KOH powder in different amounts was added to a solution of KH2P04 (100 mmol/l) and NaCl containing the sample plus HHL to make the final pH from 6.0 to 10.0. The C1- concentration was raised from 0 to l000mmol/l. The sample solution for experiments on C1- dependency was dialysed overnight against potassium phosphate buffer (1 mmol/l), pH7.8. The enzyme reaction was terminated by the addition of 3% metaphosphoric acid, and the incubation mixture was kept in iced water for 10 min. The blank sample was prepared by adding the metaphosphoric acid before incubation. It was determined whether or not the activity of the enzyme obtained was inhibited by SA-446, an ACE inhibitor [16]. After centrifugation of the reaction mixture for 10min at 12000g, 2 0 d of the supernatant was applied to a reversed-phase column (LS 410-K, TOYOSODA: 30 cm x 0.40 cm i.d.; 10 pm particle size) and eluted at 38°C with KH2 PO4 (1 0 mmol/l)/methanol (1 : 1, pH 3 .O) at a rate of 0.7 ml/min. Hippuric acid was detected by U.V. absorbance at 228 nm. The activity of ACE was expressed in m-units per mg of protein or per ml of plasma. One unit of ACE activity was Vascular angiotensin-converting enzyme 41 defined as the amount of enzyme that cleaved 1 p o l of hippuric acid in 1 min at 37°C. The protein concentration was measured by the method of Lowry et al. [171. liquid chromatography (h.p.1.c.). Peaks for hippuric acid and HHL were 5 min and 7.25 min respectively. Precipitates of protein, metaphosphoric acid and Nonidet P-40 did not affect the elution pattern by h.p .1.c. The Nonidet P-40 was used to obtain high Materials recovery of ACE in the supernatant from homoHippuryl-histidyl-leucine (HHL), hippuric acid genized tissues. The percentage extraction of ACE was defined as the ratio of ACE activity in the and histidyl-leucine were purchased from Protein supernatant to the sum of ACE activities in the Research Foundation, Osaka, Japan. (2R, 4R)-2(2-hydroxyphenyl)-3-(3-mercaptopropionyl)-4-thi- supernatant and the pellet. ACE in dog pulmonary and renal artery homogenates was solubilized in the azolidine carboxylic acid (SA-446) was provided supernatant fraction to the greatest extent (98%) by Santen Pharmaceutical Co. Ltd, Osaka, Japan. Polyoxyethylenephenylether (Nonidet P-40) was with 0.5% Nonidet P-40. In the absence of the detergent, the percentage extraction was as low as supplied from Shell Chemicals, Manchester, U.K. Collagenase (CLS-III) was purchased from Millipore 30%. The production of hippuric acid from HHL was increased linearly for at least 30min, up to Corp. NJ, U.S.A. 150 m-units/ml. The ACE activity in pulmonary and renal artery Results homogenates increased with increasing pH from 6 to 8. The additional increase in the pH inhibited Fig. 1 shows a typical elution pattern of the the activity (Fig.2). C1- was one of important incubation mixture of HHL substrate and sample determinants for the ACE activity. The ACE preparation by reversed-phase high-performance activity in tissues and plasma was related directly to the ion concentrations, ranging from 0 to 800 mmol/l in dog tissue homogenates, from 0 to 600 mmol/l in monkey and human tissue homogenates and from 0 to 300mmol/l in rabbit tissue homogenates. The maximum activities in dog, monkey, human and rabbit materials were obtained at C1concentrations of 800, 600,600 and 300 mmol/l 1 l! 0 5 10 Time (min) FIG. 1. Ultraviolet absorbance profile of an h.p.1.c. separation of the reaction products in vascular homogenate-HHL (hippuryl-L-histidyl-L-leucine) incubation. After a 30 min incubation period, 20pl of the supernatant was applied to a reversedphase column (LS 410-K, TOYOSODA: 30 cm x 0.40 cm i.d.; lOpm particle size). The mobile phase consisted of KH2P04(10 mmol/l)/methanol (1 ;1> adjusted to pH 3.0 with phosphoric acid. The sample was eluted at 38 C at a rate of 0.7 ml/min and hippuric acid was detected by U.V. absorbance at 228 nm. Peak 1, hippuric acid; peak 2, HHL. OL 1 1 1 1 I 6 7 8 9 10 PH FIG. 2. Effect of incubation buffer pH on angiotensin-converting enzyme activity in dog pulmonary ( 0 ) and renal artery ( 0 ) homogenates. Final concentrations in the incubation mixture: HHL, 5 mmol/l; sodium chloride, 800 mmol/l; potassium phosphate buffer, 100 mmol/l. M, Miyazaki et al. 42 o Pulmon. 0 2 4 o Plasma a. x Cerebral a. Mesent. a. I 8 10x10' 6 Gast.-epip. 0 2 [Cl-] (mmol/l) I 6 4 . . . 8 a. . 10xlOz [Cl-] (mmol/l) x + 3 100 loo Rabbit w u 4 o Plasma a Lung Aorta 0 2 4 6 8 [Cl-] (rnrnol/l) 10x10' 1 , 0 , . 2 Lung . 4 , . , o Aorta . . 8 6 x10' [Cl-] (mmol/l) FIG. 3. Effects of chloride on angiotensin-converting enzyme activity in dog, monkey, human and rabbit tissues. Final concentrations in the incubation mixture: HHL, 5 mmol/l; potassium phosphate buffer (pH 8.3), 100 mmol/l. Sample homogenates were dialysed overnight against potassium phosphate buffer (1 mmol/l), pH 7.8, before incubation respectively. C1--dependent alterations in the ACE activity were quite similar in different tissues and plasma obtained from the same species (Fig. 3). ACE activities of dog and monkey tissues measured under optimal conditions are summarized in Tables 1 and 2. In cerebral, mesenteric, pulmonary and renal arteries, ACE activities were in the range between 1.01 and 1.60m-units/mg of protein in dogs and between 0.43 and 0.94 in monkeys. The pulmonary vein in these species contained significantly higher activities. The ACE activity of dog aorta was significantly less than that of arteries and vein studied. However, endothelial cells isolated from the aorta contained the higher activity of ACE, the mean value being 13.1 times that of the values in the whole aorta (Table 3). Gastroepiploic arteries isolated from two patients contained ACE activities of 0.59 and 0.58 m-unit/mg of protein. TABLE 1. Anpotensin-converting enzyme ( A C E ) activity in dog tissues n , Number of animals used, Data presented are mean values Tissues ACE activity (rn-units/mg of protein) n Lung Renal cortex 14 14 Cerebral cortex 10 Choroid plexus 10 Cerebral artery Mesenteric artery Pulmonary artery 11 Pulmonary vein Renal artery Aorta 11 15 Plasma* * SEM. 15 15 9 6 * m-units/ml. 28.6 t 2.6 15.7 t 3.0 3.47 f 0.62 17.2 t 2.6 1.60 t 0.46 1.54 t 0.19 1.13 t 0.22 3.01 f 0.49 1.01 * 0.09 0.20 ? 0.02 8.83 t 1.06 Vascular angiotensin-converting enzyme 43 TABLE2. Angiotensin-converting enzyme (ACE) activity in monkey tissues n , Number of animals used. Data presented are mean values f SEM . Tissues n ACE activity (m-unitslmg of protein) Lung Renal cortex Cerebral cortex Choroid plexus Cerebral artery MesenteIic artery Pulmonary artery Pulmonary vein Renal artery Aorta Plasma* 14 12 12 11 8 I 12 8 9 14 14 113.8 i 11.0 64.2 i 8.2 1.39 i 0.26 0.69 f 0.07 0.72 f 0.18 0.43 + 0.10 0.93 f 0.20 4.63 f 1.31 0.94 f 0.22 0.10 f 0.01 57.5 f 3.4 * m-unitslml. TABLE3. Angiotensin-converting enzyme (ACE) activity in the aorta and aortic endothelium of dogs ( n = 9 ) ACE activity Ratio (m-units/mg of protein) ~ Aorta Endothelium 0.20 f 0.02 2.51 f 0.65 1 13.1 As compared with the ACE activity in arteries, vein and aortae, the activities in the lung, renal cortex and cerebral cortex from dogs and monkeys were appreciably higher (Tables 1 and 2). The lung contained the highest activity in the tissues studied. The activity in the choroid plexus was very high in dogs but not in monkeys. Dog plasma contained ACE activity of O.lOm-unit/mg of protein or 8.83 f 1.06 m-units/ml (n = 6 ) , which was significantly lower than that of 57.5 f 5.4 m-units/ml (n = 14) in monkey plasma. ACE activities of plasma obtained from two healthy males were 23.8 and 16.3 m-units/ml. The ACE activity of dog mesenteric, pulmonary and renal arteries was inhibited by treatment with a specific ACE inhibitor, SA-446 (10-9-10-7mol/l), in a concentration-dependent manner; at mol/l, the activity was abolished almost completely (Fig. 4). Concentrations of SA-446 sufficient to inhibit the ACE activity by 50%in three mesenteric, three pulmonary and three renal arteries averaged 4.2 x mol/l, 7.9 x mol/l and 5.6 x lod9 mol/l respectively. Discussion The ACE activity in various tissues including arteries and veins of dogs and monkeys was Concn. of SA 446 (mol/l) FIG. 4. Inhibition by SA-446, an angiotensin-converting enzyme inhibitor, of vascular angiotensinconverting enzyme activity. Mean values of SA-446 concentrations inducing half-maximum inhibition (IDSO)in three dogmesenteric (*),three pulmona ( 0 ) and three renal (a) arteries were 4.2 X 10mol/l, 7.9 x lO-*mol/l and 5 . 6 10-9mol/l ~ respectively. Values in mesenteric, pulmonary and renal arteries are not significantly different at each concentration of SA-446. 7 measured quantitatively by a modified biochemical method; the recovery of ACE activity in supernatant of the arterial homogenate solubilized by Nonidet P-40 increased to 98%. Without such a treatment, the recovery was only 30%. In our samples obtained from homogenates of arterial tissues, the optimum pH t o release hippuric acid from HHL was approx. 8 , which is in good agreement with the values reported with the rabbit lung (8.3) [15], endothelial cells from the dog aorta and lung (8.0) [7], human serum (8.3) [18] and the rat lung (8.3) [ 191. Skeggs et ul. [20] showed that C1- was a determining factor in activating ACE in horse plasma. The dependency of ACE activity on C1- was determined in different animal tissues under various conditions, the optimal C1- concentration being 300 mmol/l in rabbit lung [ 151,600 mmol/l in rat lung [ 191, 800 mmol/l in human lung [ 181, 700-1000mmol/l in human kidney [21] and 100-150 mmol/l in cerebral microvessels of rabbit (221 at pH 8.3 and 1000 mmol/l in the human prostate at pH 7.8 1231, with Hip-His-Leu used as a substrate. On the other hand, when angiotensin I 44 M. Miyazaki et al. was used as a substrate, the optimal ion concentration in rabbit lung was 30 mmol/l at pH 7.5 [24]. In the present study, C1- dependency and the optimal C1- concentration were clarified in a variety of tissues (PH 8.3 and HHL as a substrate). The ACE activities were greatest with C1- concentrations of 300 mmol/l and 800 mmol/l in rabbit and dog tissues respectively. The value in rabbit tissues was identical with that reported previously [24]. The optimal concentration for ACE of the human plasma and arteries was in agreement with that for ACE of the monkey plasma, arteries and lung (600 mmol/l). Different tissues obtained from the same species showed quite similar dependency of ACE activity on Cl- concentrations. The activity of ACE in samples obtained from dog arteries was inhibited by an ACE inhibitor, SA-446, in a dose-dependent manner. IDso values of SA-446 did not differ appreciably in mesenteric (4.2 x 10-7mol/l),pulmonary(7.9 x 10-7mol/l)and mol/l) arteries. The IDs0 values in renal (5.6 x rabbit lung ACE is reportedly 6 x lo-’ mol/l [ 161. ACE activities of the dog lung and kidney were 28.6 f 2.6 and 15.7 f 3.0 m-units/mg of protein respectively. These values are appreciably higher than the values obtained in dog lungs and kidneys by Cushman & Cheung (14 and 2.6 m-units/mg of protein respectively) [24]. The discrepancy is not only due to the concentrations of chloride used (600 mmol/l in our experiments vs 300 mmol/l in theirs) but possibly to efficiency of enzyme extraction from these tissues, since the ACE activities obtained at 300 mmol/l C1- concentration in the present study (Fig.3) were still greater than their values (18.1 vs 14 m-units/mg of protein in lungs and 9.1 vs 2.6m-units/mg of protein in kidneys). The present study revealed that ACE activities in plasma in different species measured under optimal conditions differed; the activity was in the order monkey (57.5 m-units/ml or 0.64 m-units/mg of protein)> human (23.8 and 16.3 m-units/ml) > dogs (8.8 m-units/ml or 0.10 m-units/mg of protein). The activity in rat serum (1.03 [25] and 1.6 m-units/mg of protein [24]) is appreciably greater than the activities in monkey, human and dog. The value in human plasma here was similar to the values of 32.2 [26], 27.4 [27] and 19.2 m-units/ml [281 previously reported. Whereas ACE activities in monkey plasma, lungs and kidneys were markedly greater than those in dog plasma and tissues, the enzyme activities in the cerebral, pulmonary and renal arteries and the pulmonary vein obtained from dogs and monkeys were similar, and the activities were greater in dog mesenteric arteries and aortae than in the monkey arteries and aortae. Human gastroepiploic arteries possessed activities of ACE (0.59 to 0.58 m-unit/ mg of protein) similar to those of monkey mesenteric arteries. The ACE activity in the choroid plexus of the dog was significantly higher than that of the monkey choroid plexus. Igic et al. [29] reported that ACE in the choroid plexus of the rabbit and human was less active than in the plexus of the dog. Arteries and veins isolated from the monkey and dog contained significant amounts of ACE, which, however, were less than those seen in the lung, kidney and brain. The activity in the endothelium of the aorta was 13 times higher than that in the whole aorta. Localization of ACE in the vascular endothelium has been demonstrated immunohistochemically [5,6]. The positive activity of ACE was chemically detected in endothelial cells of the pig pulmonary artery and aorta [7]. These authors demonstrated that the ACE activities in endothelial cells of pulmonary arteries and aortae did not differ. These findings may explain the least ACE activity in aortae and the highest activity in veins, since the endothelial-vascular wall thickness ratio is least in aortae and greatest in veins. ACE activities in the endothelium are comparable with those of lungs, kidneys and brain; therefore, physiological roles of vascular ACE in the local production of ANG I1 would have to be stressed. The addition of ANG I to isolated dog arteries exposed t o artificial solutions produces a rapidly developing contraction, which is suppressed by treatment with ACE inhibitors and ANG I1 antagonists [ l l , 121. ACE present in the arterial wall appears to be sufficient t o convert ANG I rapidly into ANG 11, which acts on angiotensin receptors in smooth muscles responsible for arterial contractions [30] and those in adrenergic nerve terminals responsible for the accelerated release of transmitter noradrenaline [31]. Acknowledgments This work was supported in part by a Research Grant for Cardiovascular Disease (56A-7) from the Ministry of Health and Welfare of Japan, a Grantin-Aid for Cooperative Research 5637-0009 from the Ministry of Education, Science and Culture and the Takeda Medical Research Foundation. References 1. Skeggs, L.T., Kahan, J.R. & Shumway, N.P. (1956) The preparation and function of the hypertensin converting enzyme. Journal of Experimental Medicine, 103,295-299. 2. Erdds, E.V. (1975) Angiotensin I converting enzyme. Circulation Research, 36,247 -25 5 . Vascular angiotensin-converting enzyme 3. Soffer, R.L. (1976) Angiotensin-converting enzyme and the regulation of vasoactive peptides. Annual Review of Biochemistry, 45,73-94. 4.Bakhle, Y.S., Reynard, A.M. & Vane, J.R. (1968) Metabolism of the angiotensin in isolated perfused tissues. Nature (London), 222,956-959. 5. Ryan, J.W., Ryan, U.S., Schults, D.R., Whitaker, C. & Chung, A. (1975) Subcellular localization of pulmonary angiotensin-converting enzyme (kininase 11). Biochemical Journal, 146,497-499. 6. Caldwell, P.R.B., Seegal, B.C., Hsu, K.C., Das, M. & Soffer, R.L. (1976) Angotensin-converting enzyme: vascular endothelial localization. Science, 191, 10501051. 7.Ody, C. & Junod, A.F. (1977) Converting enzyme activity in endothelial cells isolated from pig pulmonary artery and aorta. American Journal of Physiology, 232, C95-C98. 8. DiSalvo, J. & Montefusco, C.B. (1971) Conversion of angiotensin I to angiotensin I1 in the canine mesenteric circulation. American Journal of Physiology, 221, 1576-1 5 79. 9. Franklin, W.G., Peach, M.J. & Gilmore, J.P. (1970) Evidence for the renal conversion of angiotensin I in the dog. Circulation Research, 21, 321-324. 10. Collier, J.G. & Robinson, B.F. (1974) Comparison of effects of locally infused angiotensin I and I1 on hand veins and forearm arteries in man: evidence for converting enzyme activity in limb vessels. CIinicalScience and Molecular Medicine, 41, 189-192. 11. Aiken, J.W. & Vane, J.R. (1970) The renin-angiotensin system: inhibition of converting enzyme in isolated tissues. Nature (London), 228,30-34. 12. Toda, N., Hayashi, S. & Miyazaki, M. (1978) Contractile responses of isolated dog mesenteric arteries to angiotensin I, I1 and 111. Japanese Journal of PharmaC O I Om~ ,~5 ,27-5 34. 13. Horiuchi, M., Fujimura, K., Terashima, T. & Iso, T. (1982) Method for determination of angiotensinconverting enzyme activity in blood and tissue by high-performance liquid chromatography. Journal of Chromatography, 233,123-130. 14. Soffer, R.L., Reza, R. & Caldwell, P.R.B. (1974) Angiotensin-convertingenzyme from rabbit pulmonary particles. Proceedings of the National Academy of Sciences U.S.A., I, 1720-1724. 15. Cushman, D.W. & Cheung, H.S. (1970) Spectrophotometric assay and properties of the angiotensin-converting enzyme of rabbit lung. Biochemical Pharmacology, 20,1637-1648. 16. Iso, T., Yamauchi, H., Suda,H., Nakata, K., Nishimura, K. & Iwao, J. (1981) A new potent inhibitor of converting enzyme: (2R, 4R)-2-(2-hydroxyphenyl) -3(3mercaptopropionyl)4-thiazolidine-carboxylic acid (SA-446). Japanese Journal of Pharmacology, 31, 875-882. 17. Lowry, O.H., Roseborough, N. J., Farr, A.I. & Randall, R.J. (1951) Protein measurement with the Folinphenol reagent. Journal of Biological Chemistry, 193, 265-275. 45 18. Hurst, P.L. & Lovell-Smith, C.J. (1981) Optimized assay for serum angiotensin convertingenzyme activity. Clinical Chemistry, 21,2048-2051. 19. Hayakari, M., Kondo, Y. & Izumi, H. (1978) A rapid and simple spectrophotometric assay of angiotensinconverting enzyme. Analytical Biochemistry, 84, 36 1-369. 20. Skeggs, L.T., Marsh, W.H., Kahn, J.R. & Shumway, N.P. (1954) The existence of two forms of hypertensin. Journal of Experimental Medicine, 99, 275282. 21.Takada, Y., Unno, M., Hiwada, K. & Kokubu, T. (1982) Biochemical and immunological studies of angiotensin converting enzymes from human, bovine, dog, hog, rabbit, rat and sheep kidneys. Comparative Biochemistry and Physiology, 73B, 189-194. 22. Brecher, P., Tercyak, A. & Chobanian, A.V. (1981) Properties of angiotensin-converting enzyme in intact cerebral microvessels. Hypertension, 3,198-204. 23. Yokoyama, M., Hiwada, K., Kokubu, T., Takaha, M. & Takeuchi, M. (1980) Angiotensin-converting enzyme in human prostate. Clinica Chimica Acta, 100,253-258. 24.Cushman, D.W. & Cheung, H.S. (1972) Studies in vitro of angiotensin-converting enzyme of lung and other tissues. In: Hypertension '72, pp.532-541. Ed. Genest, J. & Koiw, E. Springer-Verlag, Berlin, Heidelberg, New York. 25. Roth, M., Weizmann, A.F. & Piquilloud, Y. (1969) Converting enzyme content of different tissues of the rat. Experientia, 25, 1427. 26.Friedland, J. & Silverstein, E. (1976) A sensitive fluorimetric assay for serum angiotensin-converting enzyme. American Journal of Clinical Pathology, 66, 416-424. 27. Boomsma, F., de Bruyn, J.H.B.,'Derkx, F.H.M. & Schalekamp, M.A.D.H. (1981) Opposite effects of captopril on angiotensin-converting enzyme activity and concentration; relation between enzyme inhibition and long-term blood pressure response. Clinical Science, 60,491-498. 28. Ryder, K.W., Jay, S.T., Jackson, S.A. & Hoke, S.R. (1981) Characterization of a spectrophotometric assay for angiotensin converting enzyme. Clinical Chemistry, 21,530-534. 29. Igic, R.P., Robinson, C.J.G. & Erdas, E.G. (1977) Angiotensin I converting enzyme activity in the choroid plexus and in the retina. In: Central Action of Angiotensin and Related Hormones, pp. 23-27. Ed. Buckly, J.P. & Ferrario, C.M. Pergamon Press, New York, Oxford, Toronto, Sydney, Paris, Frankfurt. 30. Bohr, D.F. (1974) Angiotensin on vascular smooth muscle. In: Angiotensin, pp. 424-440. Ed. Page, I.H. & Bumpus, F.M. Springer-Verlag, Berlin, Heidelberg, New York. 31. Hughes, J. & Roth, R.H. (1971) Evidence that angiotensin enhances transmitter release during sympathetic nerve stimulation. British Journal of Pharmacology, 41.239-255.