Angiotensinase Activity of Dipeptidyl Aminopeptidase I (Cathepsin C) of Rat Liver

SUMMARY Dipeptidyl aminopeptidase I (cathepsin C) purified from rat liver was shown to have marked angiotensinase activity arising from its ability to catalyze the rapid removal of two dipeptide fragments, in succession, from the NH, terminus of a variety of angiotensin II analogues. The most rapid rates of degradation were observed on Asn’-angiotensin II (Hypertensin, Ciba), angiotensin II (bovine), and Ile5-angiotensin II (human). Kinetic studies with the first (K, = 0.44 mM) and the last (K, = 0.34 mM) showed essentially the same turnover number (5220 minP) for the removal of the tist dipeptide at pH 5.0 and 37”. The K,,, for a-Asp-Arg-P-naphthylamide (a model, fluorogenic substrate) was 0.31 rnM. Unnatural, biologically active analogues such as P-Asp’-angiotensin II and a-D-Asp’-angiotensin II were also degraded, although at lower rates. No action was detected on Asnl,n-Arg2-angiotensin II. were conducted as described dipeptidyl aminopep- tidase I giving molar ratio of 1.3 X 10m5. hydrolysis B-n-Asp’-angiotensin-II. The shown for this are similar to those obtained on a-n-Aspi-angiotensin II. These reactions were conducted 20 -DIP%‘-treated dipeptidyl aminopeptidase I, giving a molar ratio of 3.3 X l@‘.


Dipeptidyl
aminopeptidase I (cathepsin C) purified from rat liver was shown to have marked angiotensinase activity arising from its ability to catalyze the rapid removal of two dipeptide fragments, in succession, from the NH, terminus of a variety of angiotensin II analogues. The most rapid rates of degradation were observed on Asn'-angiotensin II (Hypertensin, Ciba), angiotensin II (bovine), and Ile5angiotensin II (human). Kinetic studies with the first (K, = 0.44 mM) and the last (K, = 0.34 mM) showed essentially the same turnover number (5220 minP) for the removal of the tist dipeptide at pH 5.0 and 37". The K,,, for a-Asp-Arg-P-naphthylamide (a model, fluorogenic substrate) was 0.31 rnM. Unnatural, biologically active analogues such as P-Asp'-angiotensin II and a-D-Asp'-angiotensin II were also degraded, although at lower rates.
No action was detected on Asnl,n-Arg2-angiotensin II. The degradation of Asnlangiotensin II and Iles-angiotensin II by dipeptidyl aminopeptidase I was pronounced over a wide range of pH (3 to 7.5) with a maximum between pH 5 and 6. About 20% of the (pH 7.3) activity of the purified enzyme was manifested when added to rat blood plasma that did not contain added Cl-and -SH activators.
The predominant products and course of angiotensin II degradation at pH 5.5 by isolated rat liver lysosomes were identical with those produced by purified dipeptidyl aminopeptidase I, thereby demonstrating that this enzyme is probably the major contributor to the angiotensinase activity of liver lysosomes.

Numerous investigators
(l-6) have attempted to show that the concentration of circulating angiotensin II, Asp'-A+Va13-Tyr4-Va15-Hi&Pro'-Phe*, the hypertensive component of the renin-angiotensin system, is governed by the relative activities of converting and inactivating enzymes in the blood. More recent studies have shown, however, that the blood level of angiotensin II is probably under the control of peptidases residing in tissue vascular beds. Ng and Vane (7,8) have shown, for example, that angiotensin I is rapidly converted to angiotensin II in the pulmonary circulation, and not by an enzyme in the blood.
The formed angiotensin II subsequently disappears in other vascular beds iu which its rate of disappearance is much too rapid to be caused by augiotensinases iu the blood (8-10). iYumerous studies (11)(12)(13)(14)(15) have shown a central role for the liver in the metabolism of circulating augiotensiu II. Metabolic studies (16)(17)(18) with radioactive analogues have ascribed its disappearance to the dcgradative activity of tissue peptidases.
Earlier reports (19)(20)(21) from this laboratory have described the ability of dipeptidyl amiuopeptidasc I to degrade numerous polypeptide hormones by catalyzing the successive removal of dipeptide moieties from their h'H, termiui.
As was previously noted (19), ~~sll'-angiote~~sil~ II was similarly degraded by dipeptidyl aminopcptidase I. It appears, however, that the angiotensinase activity of this enzyme has remaiued unrecognized by workers seeking to identify the enzyme (or enzymes), in liver and kidney responsible for the iuactivation of circulating angiotensin II. Thus far, the acid angiotensiuase activity of the liver (22, 23) and kidney (24) has beeu attributed primarily to a lysosomal carbosypeptidase.
In this report, a highly purified preparation of rat liver dipeptidyl amiuopeptidase I h-as used to characterize the kinetics aud substrate specificity of this lysosomal enzyme on a variety of augiotensiu II analogues, and to show that the acid angiotensinase activity contained in rat liver lysosomes is primarily attributable to dipeptidyl aminopeptidase I.

MATERIALS ANI) METIIOl)S
Preparation and Assay of Dipeptidyi rlminopeptidase I-The enzyme was prepared from the livers of fasted rats using a method based on that reported by Metrione et al. for the preparation of the beef spleen enzyme (25). The purification results, properties, and substrate specificity for the rat liver enzyme have already been described (19). The euzyme used in this study had a specific activity of at least 20 units per mg of protein. One unit of dipeptidyl aminopeptidase I is defined as the amouut of enzyme hydrolyzing 1 pmole of Gly-Phe-/3-naphthylamide per min at pH 6.0 under the conditions of the assay (19 1). Aliquots equivalent to about 0.75 pmole of angiotensin II were dried under vacuum and recovered in 1 ml of 0.2 M sodium citrate-HCI buffer, pH 2.2. The products were preparatively separated on the lo-cm basic column by the analytical procedure described above using an analyzer equipped for stream division such that one-fifth of the eluent was used for color development, and four-fifths collected in O.&ml fractions.
Peak fractions plates (20 X 20 cm) from Brinkmann Instruments, Inc. (West-were pooled and desalted by ion exchange chromatography and bury, N. Y.). Digest aliquots equivalent to 3 nmoles of angio-extraction procedures as already reported (31). Each dipeptide tensin II were spotted at each origin, and the plates were was identified and quantitated by acid hydrolysis and amino developed with 2-butanol-3 y0 aqueous NH3 (75: 30, v/v) (concen-acid analysis using standard procedures. Known amounts of trated NHaOH was assumed to be 30% NHB). The peptides the dipeptides were then used to establish the elution times and were detected with a polychromic ninhydrin reagent (30) con color values given above. The Asp-Arg dipeptide color value taining 50 ml of 0.2% ninhydrin in absolute ethanol, 10 ml of was determined on the standard obtained commercially. glacial acetic acid, and 2 ml of 2,4,6-collidine.
The plates were Preparation of Rat Liver Lysosomes on Sucrose Density Gradiheated for 2 min at 110'. ent-Sprague-Dawley rats were decapitated and their livers Identijication of Angiotensin Fragments-A preparative amount quickly transferred to cold 0.25 M sucrose. A 5% (w/v) susof each split product of angiotensin II was isolated as a band on pension of minced liver in sucrose was homogenized according thin layers using the chromatographic system described above. The bottom of the tube while the tube was immersed in an ice-water edges of the plate were sprayed to locate the peptide bands.
slurry. The homogenate was centrifuged at 500 x g for 15 min The peptides thus located were recovered for acid hydrolysis by to remove nuclei, erythrocytes, and unbroken cells. The superscraping from the plate the appropriate bands of cellulose. natant was aspirated and centrifuged at 20,000 x g for 15 min The peptide was eluted from the cellulose, contained in a small to sediment the lysosome-rich fraction. This precipitate, which glass filter tube, with about 1.5 ml of 10 rnx HCl.
The eluates contained about 96% of the dipeptidyl aminopeptidase I activity were dried under vacuum and the peptide residues subjected to originally contained in the 500 X g supernatant, was resusacid hydrolysis.
The constituent amino acids were identified pended in cold 0.25 M sucrose to one-fifth its original volume. by comparing their mobilities with standards in two solvent One milliliter of this suspension was layered on the surface of a systems: methanol-chloroform-9r/, aqueous NH3 (2:2:1, v/v), chilled 30.ml sucrose gradient prepared with a Beckman gradient and methanol-water-pyridine (20:5: 1, v/v). The characteristic former with limits of about 30 and 557, (w/w) sucrose. Equicolor reactions obtained with the polychromic ninhydrin spray librium-density centrifugation was performed in the SW 25-1 also served to confirm the identity of the amino acids.
rotor of a Spinco model L ultracentrifuge at 90,000 x g (bottom Measurement of Hydrolysis Rates on Angiotensin II-Aliquots of tube) for 3 hours. The gradient was then fractionated into (0.1 ml) were taken at specific periods from reaction mixtures to l-ml fractions using a density gradient fractionator, model 640 establish the rate of removal of the NH?-terminal dipeptide (Instrumentation Specialties Company, Lincohi, Neb.). The (AsnArg or Asp-Arg) as well as the penultimate dipeptide sucrose concentration in each fraction was determined by frac-(Val-Tyr) from the angiotensin II substrates. Aliquots were tionating a companion gradient (in which 0.25 M sucrose replaced inactivated with 0.4 ml of 0.2 M sodium citrate-HCl buffer, pII the sample) and measuring the index of refraction (ni') of each 2.2, and analyzed on a Beckman/Spinco model 12OB amino acid fraction. The distribution of protein in the gradient was apanalyzer equipped with a (570 mm) high sensitivit,y cuvette and prosimated by means of a direct scan at 280 nm. expanded range recorder.
This method was an adaptation of dssay oj Prolylcarboxypeptidase-The distribution of this one previously reported (31). The dipeptides Asn-ilrg and Val-enzyme in sucrose gradients was established by combining 1 part Tyr were conveniently analyzed on the IO-cm basic column filled of each gradient fraction with 3 parts of 14 rnhr Z-Pro-Phe in 0.2 to 5.5 cm with Beckman/Spinco type AA-27 spherical resin M sodium acetate buffer, pI1 5.5. The mixture was incubated maintained at 57". To separate Asn-drg and Val-Tyr from at 37", and O.l-ml aliquots were taken at appropriate time contaminants the following development was necessary: Step 1, intervals for the calorimetric determination of free phenylalanine to follow the sample onto the column with 3.0 ml of 0.2 M sodium (in the range of 0.05 to 0.5 kmole) by reaction with trinitrocitrate-HCl buffer, pH 3.25; Step 2, to elute with 8.65 ml of this benzenesulfonic acid using an adaptation (33) of the method same buffer for about 5 min; and then Step 3, to complete the described by Okugama and Satake (34). development with 0.35 M sodium citrate-HCl buffer, pH 5.25, at Source of Substrates and Peptides-L-Ilej-angiotensin II (hua flow rate of 68 ml per hour. Val-Tyr eluted at 24 min with a man) was purchased from Schwarz-Mann.
Its identity and color value of 5.55 A min per pmole (compared with 60.16 for a purity were established by amino acid analysis and by thin leucine reference), and hsnhrg eluted at 65 min with a color layer chromatography as described in Fig. 1. L-Asn*-angiovalue of 16.01 d min per pmole.
To analyze for Asp-Arg, tensin II (Ciba, hypertensin) was a gift from Dr. Albert J. simply omit Step 2 above.
Asp-rirg eluted at 20 min with a Plummer of Ciba Pharmaceutical Co. (Summit, i\'. J.). This color value of 66.15 ii min per pmole and was adequately sepa-preparation was said to contain 17% ammonium acetate. The rated from NH3 (64 min) and other contaminants. concentration of the peptidc was established by amino acid these peptides. However, they appeared to be pure by thin layer chromatography (Fig. I), and the enzymatically derived peptide fragments showed the espected amino acid composition. N*-Benzyloxycarbonyl-Pro-Phe was purchased from Cycle Chemical Co. (Los Angeles, Ca.); a--4sp-Arg-fi-naphthylamide (free base) and cy-Asp-Arg dipeptide from Bachem, Inc. (Ifarina Del Rey, Ca.) ; and Gly-Arg-fl-naphthylamide diacetate from Fox Chemical Company (Los Angeles, Ca.). Their purity and identity were confirmed by thin layer chromatography and amino acid analysis. The amount of free /%naphthylamine in the acid hydrolysates was determined fluorometrically.
Unless otherwise indicated, all amino acid residues should be considered to have the L configuration, and aspartyl residues the LY linkage. a, hvdrolysis of Asni-angiotensin II (Ciba hypertensin).
A 100-J &antity of 3 mM Asni-angiotensin II contained in 5 mM dithiothreitol-10 mM NaCl-100 mM NazHPOd-50 mM citric acid, pH 5.0, was brought to 37". The reaction was initiated by adding 1 ~1 (0.4 pg) of DIPF-treated dipeptidyl aminopeptidase I. Based on a molecular weight of 200,000 for the rat liver enzyme (19), the molar ratio of enzyme to substrate was estimated to be 6.3 X 1OV. b, hydrolysis of angiotensin II. The results shown for the bovine analogue are similar to those found for the human (IleE) analogue. These reactions were conducted as described above, except that 0.8 pg of DIPF-treated dipeptidyl aminopeptidase I was used, giving a molar ratio of 1.3 X 10m5. c, hydrolysis of B-n-Asp'-angiotensin-II.
The results shown for this analogue are similar to those obtained on a-n-Aspi-angiotensin II. These reactions were conducted with 20 pg of -DIP%'-treated dipeptidyl aminopeptidase I, giving a molar ratio of 3.3 X l@'.
analysis. Angiotensin II (bovine), /3-L-Asp'-angiotensin II, cu-n-Asp'-angiotensin II, and L-Asnl , n-Arg2-angiotensin II were kindly provided by Drs. W. R. Rittel and B. Riniker of Ciba-Geigy (Basel, Switzerland). In view of their limited supply, no direct attempt was made to confirm the identity and purity of RESULTS Products and Relafive Rates of Hydrolysis of Angiotensin II Analogues by Dipeptidyl Aminopeptidase I-As illustrated in Fig. la, dipeptidyl aminopeptidase I catalyzed the consecutive removal of the first two dipeptides, Asn-Arg and Val-Tyr, from the NH2 terminus of Asnl-angiotensin II (Ciba, hypertensin). The initial appearance of free Asn-hrg was accompanied by the concomitant accumulation of a transient hexapeptide intermediate. After 30 min of reaction, this intermediate was further degraded, giving rise to the simultaneous appearance of split products identified as Val-Tyr and Val-His-Pro-Phe. The latter was resistant to further hydrolysis, which was taken as evidence that dipeptidyl aminopeptidase I was unable to attack the imino bond of proline. The histidyl bond should otherwise have been susceptible to attack (21).
As seen in Fig. lb, dipeptidyl aminopeptidase I exhibited a similar course of hydrolysis on Asp-terminal analogues such as Asp', Valh-angiotensin II (bovine) and ilsp' , Iles-angiotensin II (human). However, on these substrates, the transient hexapeptide intermediates did not accumulate to the same extent; this observation indicated that at pH 5.0, the Asp' analogues, compared with the Asnl analogues, compete less effectively with the formed hexapeptide intermediate. The cleavage of the Tyr4-He5 bond in human angiotensin II occurred as readily as the Tyr4-Valj in the bovine analogue. The tetrapeptide product, Ile-His-Pro-Phe, was similarly resistant to further breakdown by dipeptidyl aminopeptidase I.
The hydrolysis of two unnatural, but biologically-active analogues of angiotensin II are shown in Fig. lc. Both the /~-L-ASP' and ol-n-Asp1 analogues of angiotensin II were also hydrolyzed, and at comparable rates, by dipeptidyl aminopeptidase I. Although at least 25 times as much enzyme was required to achieve these rates, this level of activity may still be physiologically significant, since the molar ratio of enzyme to substrate (3.3 x lo-') is still relatively low, and the liver is especially rich in dipeptidyl aminopeptidase I. By comparison, absolutely no activity was detected when the same amount of enzyme was incubated with As&n-.4rg*-angiotensin II. This analogue resembles that used in Fig. la, except that the penultimate arginyl residue has the D configuration. Enzyme and substrate controls for the reactions described in Fig. 1 were invariably negative.
Eflect of pH--Dipeptidyl aminopeptidase I exhibited angiotensinase activity over a wide range of pH in a citrate-phosphate buffer system containing sulfhydryl and chloride activators as employed in Fig. 1. A substantial rate of removal of the first dipeptide occurred between pH 3 and 7 on both human (Asp'-Ile5) and Ciba (Asnl-Val5) angiotensin II. Although the pH curves were broad and lacked a well defined maximum, this was not attributed to a possible inhibitory effect of the competitive (hexapeptide) intermediate.
A similarly broad pH curve was previously observed (19) for the removal of Gly-Arg from Gly-Arg-fi-naphthylamide.
The latter determination was based on initial rates of hydrolysis established by direct fluorometric analysis, and the rates were not complicated by the production of an inhibitory product.
The hexapeptide intermediates of both the Ile5 and Asnl analogues accumulated to a detectable extent only on the acid side of the pH curve, indicating that the more acidic conditions favored the removal of the first dipeptide (Asp-Arg or Asn-Arg) as compared with the second dipeptide (Val-Tyr) Kinetic Studies-Initial velocities, v, were measured over a critical range of substrate concentration to determine the rate of removal of the NHp-terminal dipeptide moieties from As& angiotensin II, Ile5-angiotensin II, and Asp-Arg-fi-naphthylamide by dipeptidyl aminopeptidase I in the same reaction medium as described in Fig. 1. The V,,, values obtained for the two angiotensin analogues were similar; based on these values, the turnover number was estimated to be about 87 pmoles of angiotensin II hydrolyzed (NHz-terminal dipeptide released) per s per pmole of dipeptidyl aminopeptidase I, using a molecular weight of 200,000 for the rat liver enzyme (19). The K, values were estimated to be 0.44 mM for As+angiotensin II and 0.34 mM for Ile5-angiotensin II at pH 5.0 and 37". A somewhat greater rate response to substrate concentration was obtained on Asp-Arg-/I-naphthylamide.
However, as a result of the substrate inhibition that occurred at concentrations in excess of 0.5 mM, its rate of hydrolysis at 1 mM was essentially the same as the rates observed on the angiotensin analogues at 1 mM. The K, value for Asp-Arg-P-naphthylamide was estimated to be 0.31 mM. Since the values for K, and V,,, resembled those obtained for the analogues of angiotensin II, it became apparent that Asp-Arg-fl-naphthylamide might serve as a useful fluorogenic model substrate for estimating the angiotensinase activity of dipeptidyl aminopeptidase I in relatively crude preparations. E$ect of Plasma on Activity of Purijied Dipeptidyl Aminopeptidase I-When it was found that rat liver dipeptidyl aminopeptidase I possessed significant angiotensinase activity at pH 7.0 (>85% of the pH 5.5 rate), an attempt was made to determine whether the purified enzyme retained activity when combined with blood, in which competitive substrates or inhibitory components might obscure its activity.
For this purpose, extremely small amounts of purified dipeptidyl aminopeptidase I were added to aliquots of rat plasma, and the proportion of detectable activity was assessed fluoromet.rically.
To achieve maximum sensitivity, Gly-Arg-/3-naphthylamide was used as the assay substrate of choice since, as previously reported, it possesses the lowest K, (0.1 mM) and the highest turnover number (1,300 s-1) of all the fluorogenic substrates thus far reported for dipeptidyl aminopeptidase I (19). Consequently, it was possible to subject aliquots of the plasma-containing digests to large dilutions before making fluorescence measurements to quantitate the amount of P-naphthylamine arising from the hydrolysis of Gly-Arg-fi-naphthylamide.
It was thereby possible to dilute out the fluorescence and quenching contributions made by the normal components of plasma. These dilutions also rendered undetectable the contributions of endogenous plasma aminopeptidases also capable of degrading Gly-Arg-fi-naphthylamide, but at a relatively low rate compared with the added dipeptidyl aminopeptidase I. As seen in Fig. 2 2. Rate of hydrolysis of Gly-Arg-p-naphthylamide by purified rat liver dipeptidvl aminonentidase I (DAP I) added to ;at plasma. React,ions w&e initiated by adding 2 ~1 '(1.9 rg) of fully activated dipeptidyl aminopeptidase I t,o l-ml reaction mixtures containing 0.8 ml of rat blood plasma and 0.2 ml of 10 mM Gly-Arg-o-naphthylamide in 0.5 M phosphate buffer, pH 7.3 and 37". Reactions were also conducted with mercaptoethylamine hydrochloride substituted for plasma at 10 mM, as well as with plasma in the absence of added dipeptidyl aminopeptidase I. At 1-min intervals, the concentration of liberated P-naphthylamine was estimat,ed fluorometrically by diluting 10.~1 aliquoti of the reaction mixture with 4 ml of 50 mM uhosnhate buffer. nH 6.5. and reading the fluorescence relative to-a standard curie-in the range of 10e6 to 10-T M fi-naphthylamine. rate of Gly-Arg-P-naphthylamide hydrolysis was exhibited by the system containing purified rat liver dipeptidyl aminopeptidase I at about 2 pg per ml in phosphate-buffered saline, pH 7.3, containing the necessary sulfhydryl and chloride activators. This rate was taken to represent 100% of the added activity. When undiluted rat plasma was substituted for 0.9% NaCl solution, comprising 80% of the reaction volume, about 20% of the added activity was still manifest.
This amount was considered to represent a significant proportion of the added activity, since only a very small amount of enzyme had been added to preclude the masking of plasma inhibitors, esogenous sulfhydryl and chloride activators had not been added to the plasma, and the plasma had not been protected from the effects of acration. Under the conditions of the assay, buffered plasma alone (wibhout added enzyme) exhibited an undetectable level of activity on Gly-Arg-fi-naphthylamide.
Hydrolysis of A&-Angiotensin II by Extract of Rat Liver Lysosomes-Equilibrium density centrifugation was used to obtain a lysosome-rich fraction from rat liver.
The details of the procedure are described under "Materials and hlethods." The distribution of activities in the sucrose gradient are shown in Fig. 3. Dipeptidyl aminopeptidase I, an established lysosomal enzyme (21), was located using Asp-Arg-P-naphthylamide as a model of the NH2 terminus of angiotensin II.
The lysosomal carboxypeptidase of rat liver that inactivates angiotensin II by removing the COOH-terminal phenylalanine was assayed on Z-Pro-Phe at pH 5.5. This activity has been observed in liver and kidney preparations by various workers, and has been reported under such names as acid angiotensinase (22), angiotensinase C (24), prolylcarboxypeptidase (35), and catheptic carboxypeptidase C (28). Since these activities are, in all probability, attributable to a common enzyme, the term "prolylcarboxypeptidase," as introduced by Yang and Erdijs by guest on August 16, 2020 http://www.jbc.org/ Downloaded from (35), is used herein to refer to the DIPF-sensitive, lysosomal carboxypeptidase activity measured on Z-Pro-l'he at pH 5.5. As illustrated in Fig. 3, dipeptidyl aminopeptidase I and prolylcarboxypeptidase had a coincident location in the sucrose gradient, with maximum activities occurring in tube 16 (d = 1.21). Assays conducted with Gly-l'he-@-naphthylamide, a more conventional substrate for dipeptidyl aminopeptidase I, also showed maximal activity in tube 16 following equilibrium density centrifugation of a rat liver lysosome fraction on a 30-ml sucrose gradient having limits of 30 and 55% (w/w) sucrose. Dipeptidyl aminopeptidase I was assayed on Asp-Arg-fl-naphthylamide at pH G.0 and prolylcarboxypeptidase on Z-Pro-Phc at pH 5.5. For both activities, a unit of enzyme hydrolyzes 1 pmole of substrate per min at 37". Procedures are described under "Materials and Methods." PCMS, p-chloromercuriphenyl sulfonate.
activity of dipeptidyl amiuopeptidase I in this tube was strongly inhibited by p-chloromercuriphenyl sulfonate, and the activity of prolylcarboxypeptidase by DII'F. As shown in Fig. 3, the activity of dipeptidyl nminopeptidase I contained in tube 16, as measured on ,~sp-Arg-~-~~apl~th~la~~~itle, was about 26 times greater thau the activity of prolylcarbosypeptidase measured on Z-Pro-Phe.
An aliquot, of the rat liver lysosome fraction contained in tube 16 was incubated wit'11 Asn*-augiotcnsin II in an attempt to elucidate the primary route of angiotensin II degradation at. pH 5.5 by the coucerted action of all of the lgsosomal peptidases. As illust'rated in Fig. 4, the time course chromatographic analysis of the products of augiotensiu II degradatiou revealed products, the RF values aud order of appearance of which could only be attributed to the action of dipeptidyl amiuopeptidase I. Typical of the time course results obtained with purified dipeptidyl aminopeptidase I (Fig. l), the substrate disappeared with the concomitant appearance of products with RF values correspouding to split products such as Asn-Arg and the hexapeptide intermediate.
The Asu-hrg spots near the origin were somewhat distorted by the relatively high level of solute contained in this crude lysosomal digest.
The fragments typically produced by the action of purified dipeptidyl aminopeptidase I, together with some free amino acids, were included as standards on the chromatogram.
It can be seen that the product corresponding to the hexapeptide intermediate subsequently disappeared with the concomitant appearance of products corresponding to Val-Tyr and Val-His-Pro-l'he.
The amount of the tetrapeptide was seen to become more faint with time and in longer time courses (not shown) actually disappeared.
This was attributed to the terminal action of prolylcarboxypeptidase, inasmuch as free I c FIG. 4. Time course chromatography of the products of Asniangiotensin II hydrolysis at pH 5.5 by an extract of rat liver lysosomes. The incubation mixture was prepared by combining 40 ~1 of 100 mM NatHPOd-citric acid buffer, pH 5.5, containing NaCl at 20 mM and dithiothreitol at 10 mM, with 20 ~1 of 30 mM Asnl-angiotensin II in water. The mixture was brought to 37", and the reaction was initiated by adding 50 ~1 of a lysosomal extract.
The extract was prepared from lysosomes derived from a sucrose gradient (tube 16 of Fig. 3). The lysosomes were ruptured with Triton X-100 at O.l%, and then dialyzed overnight at 4" against 0.5% NaCl. The 50 ~1 of extract used to initiate the reaction contained 1 mg of protein measured by the Lowry method. The final reaction volume was 110 ~1. At the indicated time intervals, 1-J aliquots of the reaction were spotted on a thin layer chromatogram. Enzyme and substrate controls were spotted at zero time and after 4 hours of incubation.
The typical products resulting from the breakdown of Asnl-angiotensin-II by purified dipeptidyi aminooentidase I CDAP 1) are also shown. The standards include phenylalanine was seen to appear in conjunction with a spot with an RF value corresponding to a,product previously identi-alanine was the only product detected; the pH optimum was not fied as Val-His-Pro (28 A lysosomal enzyme with similar properties was also found in swine kidney cortex by It is evident from this study that purified dipeptidyl amino- Yang et al. (24). Initially, the enzyme was named "angiotenpeptidase I from rat liver has potent angiotensinase activity on sinase C," but was subsequently characterized (41) and reboth the bovine (Asp1,Va15) and human (Asp', Ile5) forms of named (36) "prolylcarboxypeptidase," a term that was adopted angiotensin II, as well as on Ciba hypertensin (Asnl ,Vals). in the present report.
Considering the fact that liver and kid-Such results were anticipated in the light of earlier findings (19) ney are rich sources of dipeptidyl aminopeptidase I (21,42), it is that showed rat liver dipeptidyl aminopeptidase I to have its surprising that its angiotensinase activity went unrecognized. greatest rates of hydrolysis on dipeptide derivatives with penulti-As in other studies, it appears that the essential activators were mate basic residues, for example Gly-Arg-/3-naphthylamide and lacking. Gly-Lys-OMe.
Unlike the traditional aminoacyl aminopepti-Notably, the bonds cleaved in angiotensin II by dipeptidyl dases, dipeptidyl aminopeptidase I also hydrolyzed, although at aminopeptidase I correspond to the sites of attack reported (43) lower rates, unnatural analogues of angiotensin II such as the for trypsin (Arg2-Va13) and chymotrypsin (Tyr4-Vals). The ,&~-Asp* and ~-Asp' analogues.
Such a lack of specificity possibility therefore exists that the split products arising from could have physiological significance since it has been reported the breakdown of angiotensin II by the action of dipeptidyl (10) that the vasopressor activity of fl-L-Asp'-angiotensin II, aminopeptidase I contained in tissue extracts and impure prepawhich is resistant to the angiotensinase activity of plasma (36) rations could be mistakenly attributed to the action of endoand purified aminopept'idase A (37), disappears from the circula-peptidases.
Such an occurrence could account for the finding tion almost as readily as that of its LY analogue.
of Matsunaga et al. that both liver and kidney lysosomes, in the Based on the known minimal structural requirements for the presence of added dithiothreitol, exhibited "chymotrypsin-like" biological activity of angiotensin II (38), a complete loss of and "trypsirl-like" activity on Asn'-angiotensin II. pressor activity would be expected to result from the removal of Our studies suggest that t.he acid angiotensinase activity of the first dipeptide by dipeptidyl aminopeptidase I. This effect liver lysosomes is primarily attributable to dipeptidyl aminocan be compared with a resultant 50% loss of activity for the peptidase I, as indicated by the early appearance in lysosomal removal of the NHZ-terminal residue only. digests of angiotensinasc II split products typical of those result-Although the L configuration was not rcquircd at the NHz ing from an attack by purified dipeptidyl aminopeptidase I. terminus of angiotensin II to permit hydrolysis by dipeptidyl Using Asp-Arg-fl-naphthylamide and Z-Pro-Phc as model subaminopeptidase I, it was essential that the penultimate residue strates for the NH2 and COOH termini of angiotensin II, it was have the natural configuration.
Such a requirement was evi-estimated that the rate of removal of Asp-Arg at pH 5.5 by an dent from the inability of the enzyme to hydrolyze Asn*, extract of liver lysosomes exceeded the rate of phenylalanine u-Arg2-angiotensin II, a result that agrees with a similar2 lim-cleavage by about 26.fold. itation found for n-Tyr2, Lys 17~18-P1-1%orticotropinamide. Dipeptidyl aminopeptidase I has been shown to have a lyso-Dipeptidyl aminopeptidase I was found to have angioten-somal localization in rat liver, as demonstrated by biochemical sinase activity over a wide range of pII.
The activity at yH procedures (44) and by electron microscopy (21) using both his-7.5, measured in terms of the rate of Asp-Arg removal, was tochemical and immunohistochemical techniques.
Its localizaabout half that at the pH 5.5 optimum.
Although clipeptidyl tion in these organelles should not necessarily detract from the aminopeptidase I is known to have polymerase activity between possibility that this enzyme contributes to the noted capacity of pH 7.5 and 8.0 on dipeptide amides (39) and P-napht'hylamidcs the liver for inactivating angiotensin II. As indicated by de (19), its action on the various analogues of angiotensin II re-Duve and Wattiaus (45), it is the normal function of lysosomes, vealcd only split products. III addition, the enzyme was shown through the "vacuolar apparatus," to extrude enzymes from to manifest considerable activity in a plasma environment not cells into surrounding tissues while enzymes of the cytosol are containing esogenous activators. These results suggest that retained. Alteinatively, the vacuolar apparatus is known to enzyme released into the blood as a consequence of injury or provide a means whereby lysosomal enzymes may gain access disease would be free to escrt its angiotensinase activity in the to substrates that are drawn into the cell by means of an endogeneral circulation.
cytic process such as pinocytosis, thereby facilitating an attack Since other studies (8-10) have shown that the rapid inactiva-by lysosomal enzymes to the exclusion of enzymes in the cytosol. tion of circulating angiotensin II cannot be attributed to plasma In this regard, the findings of Robertson and Khairallah (46) pcptidases in nonpathological states, a number of investigators seem particularly significant. They showed that the injection of have therefore attempted to identify angiotensinases residing in 5 to 10 ng of angiotensin II into the heart of a rat produces a the tissues. Since the rapid inactivation of unnatural analogues marked increase in the number of pinocytotic vesicles in endosuch as 0-L.Asp'-angiotensin II precluded a significant contribu-thelial cells of both aorta and coronary arteries, including a tion by (aminoacyl) aminopeptidases, several workers have widening of intercellular gaps. It is therefore tempting to emphasized the importance of a carboxypeptidase attack. For speculate that angiotensin II may similarly perturb the plasma example, Johnson and Ryan (40) reported that rabbit liver membranes of other organs, thereby providing a mechanism for extracts showed a carboxypeptidase attack at pH 7.4. Phenyl-the selective uptake of angiotensin II within pinocytotic vesicles destined to fuse with lysosomes. Such a trapping mechanism 2 J. K. McDonald and S. Ellis, unpublished data.
could account for the rapid disappearance from the blood of by guest on August 16, 2020 http://www.jbc.org/ Downloaded from both natural and unnatural biologically active analogues of angiotensin II. By means of exocytosis, the digested contents of these heterolysosomes could subsequently be extruded. Such a process would be compatible with the results of studies con ducted with radioisotopically-labelled angiotensin II showing that the initial uptake of angiotensin II is soon followed by the appearance of inactive fragments in the circulation.
A possible contribution for the angiotensinase activity of dipeptidyl aminopeptidase I seems particularly probable in various pathological states. de Duve and Beaufay (47) have shown that the rapid release of lysosomal enzymes is one of the earliest detectable changes in the ischemic liver.
Since tissue anoxia and acidosis are central components in the pathophysiology of shock (48), and since these conditions, as reported by Janoff et al. (49), favor the labilization and even rupture of tissue lysosomcs, it is not surprising that lysosomal hydrolases have been reported in the blood of shocked animals (49, 50). Since our studies show that dipeptidyl aminopeptidase I can exert activity in plasma, it seems probable that it could contribute to the elevated plasma angiotensinase activity and lowered blood pressure seen in dogs rendered hypotensive by hemorrhage or injected endotoxin (51), and in patients suffering various forms of liver disease (52).
The foregoing observations do not prove that dipeptidyl aminopeptidase I makes an important contribution to the inactivation of circulating angiotensin II. However, such a role is favored in the light of observations reported here. 011 the other hand, the assumed contributions of certain other plasma and tissue peptidases now seem less likely in the light of more recent studies concerned with the biological fate of administered natural and unnatural analogues of angiotensin II.
In summary, dipeptidyl aminopeptidase I exhibits a strong affinity and high rate of hydrolysis on angiotensin II; it catalyzes the hydrolysis and complete inactivation of various natural and unnatural analogues over a wide range of pH; it is, significantly, most abundant in the liver, but is also present in large amounts in other tissues noted for their angiotensinase activity; and it appears to be responsible for the major part of the angiotensinase activity of liver lysosomes.
It therefore seems most probable that the angiotensinase activity of dipeptidyl aminopeptidase I plays an important role in the metabolism of angiotensin II under both normal and certain pathological conditions.