Regulation of Local Angiotensin II Formation in the Human Heart in the Presence of Interstitial Fluid
Inhibition of Chymase by Protease Inhibitors of Interstitial Fluid and of Angiotensin-Converting Enzyme by Ang-(1-9) Formed by Heart Carboxypeptidase A–Like Activity
Background Data from in vitro studies suggest that both chymase and ACE contribute to the local generation of angiotensin (Ang) II in the heart. The enzyme kinetics under in vivo conditions are unclear. We thus studied the generation of Ang II by cardiac tissue in the presence of interstitial fluid (IF) that contains a variety of naturally occurring protease inhibitors.
Methods and Results Ang I was incubated with heart homogenate in the presence of IF. IF obtained from human skin contained substantial amounts of protease inhibitors and ACE activity, the concentration of α1-antitrypsin being 35% and the activity of ACE 24% of the corresponding serum values. When heart homogenate was incubated with Ang I, three enzymes were responsible for its metabolism: heart chymase and heart ACE converted Ang I to Ang II, and heart carboxypeptidase A (CPA)–like activity degraded Ang I to Ang-(1-9). Incubation of heart homogenate in the presence of IF led to practically full inhibition of heart chymase–mediated Ang II formation by the natural protease inhibitors present in IF. In contrast, heart CPA–like activity was not blocked, as reflected by the continued generation of Ang-(1-9). In addition, both heart ACE– and IF ACE–mediated Ang II formation were strongly inhibited. This inhibition was shown to be due to the Ang-(1-9) formed.
Conclusions The present experimental study defines two novel inhibitory mechanisms of Ang II formation in the human heart interstitium. Heart chymase–mediated Ang II formation is strongly inhibited by the natural protease inhibitors present in the IF. Similarly, both heart ACE– and IF ACE–mediated Ang II formation appear to be inhibited by the endogenous inhibitor Ang-(1-9) formed by heart CPA–like activity. These inhibitory mechanisms provide additional information about how the Ang II concentration in the heart interstitium may be controlled.
The mechanism by which the activated renin-angiotensin system mediates its effects in the pathophysiology of heart failure is believed to be hemodynamic overload due to vasoconstriction and volume retention. However, several lines of evidence suggest that the active metabolite Ang II may have various important local effects on the heart tissue. These include myocyte hypertrophy1 and interstitial fibrosis,2 3 which lead to pathological structural remodeling and thereby predispose to ventricular dysfunction and symptomatic heart failure.
Moreover, it has been suggested that, in addition to the Ang II taken up from the circulation, Ang II may be formed locally in the heart.4 Furthermore, it has been argued that the major Ang II–forming enzyme (80% to 90% of the Ang II–forming activity) in the heart tissue is not ACE but chymase, a chymotrypsin-like serine protease that is not affected by ACE inhibitors.5 6 7
The cellular sites of synthesis and storage of heart chymase are cardiac mast cells and endothelial cells.8 9 Since most of the chymase activity in the heart has been found to be localized to the extracellular matrix,8 chymase must have been actively secreted into the heart interstitium by these cell types. Besides chymase and ACE, heart tissue contains other enzymes capable of affecting local angiotensin metabolism. Thus, mast cells alone contain two other enzymes besides chymase that are known to act on Ang I. These enzymes are cathepsin G,10 which is known to convert Ang I to Ang II,11 and CPA, a metalloprotease that removes the carboxyterminal Phe or Leu residues from peptides such as Ang I.12 13 Indeed, Ang I has been used as a model substrate in CPA studies and is degraded by this enzyme to Ang-(1-9).12 13 Fig 1⇓ summarizes the known cleavage sites of Ang I by ACE, chymase, cathepsin G, and CPA.
The heart chymase–mediated conversion of Ang II is suggested to take place in the IF.8 However, the IF contains high concentrations of protease inhibitors, which may affect angiotensin metabolism by neutral proteases.14 15 Therefore, Ang II formation in human heart tissue may be subjected to local regulation. To mimic the conditions of angiotensin metabolism in the heart interstitium, we incubated Ang I with heart homogenate in the presence of IF. As a representative of IF, in this study we used tissue fluid obtained from skin by the suction blister method.14 15 16 17 18
The synthetic angiotensin peptides FAPGG and FAP were purchased from Bachem. Furanacrylic acid was obtained from Fluka. Soybean trypsin inhibitor, chymostatin, captopril, aprotinin, EDTA, human α1-antitrypsin, and CPI were purchased from Sigma. Dulbecco’s PBS was obtained from Gibco. Murine anti-chymase monoclonal antibody was purchased from Chemicon.
Preparation of Human Heart Homogenates
Human heart tissue was obtained from the excised hearts of patients (n=10) undergoing cardiac transplantation at the University Central Hospital, Helsinki. The patients had end-stage congestive heart failure due to coronary heart disease (n=5), idiopathic dilated cardiomyopathy (n=3), or congenital heart disease (n=2). The cause of the congenital heart disease in both cases was ventricular septal defect. All patients were men 17 to 60 years old. Before transplantation, the patients with coronary heart disease and dilated cardiomyopathy but not those with congenital heart disease had received ACE inhibitor therapy. The use of these tissues was approved by the Internal Review Committee of the University Central Hospital, Helsinki. After excision, the hearts were thoroughly flushed with ice-cold cardioplegia solution, and cylindrical pieces of tissue weighing ≈200 mg were immediately cut from the left ventricles with a biopsy punch (diameter, 6 mm). The heart tissue was stored at −50°C. Heart homogenates were prepared by homogenization of the tissue in PBS at 4°C (100 mg tissue/mL PBS) with an Ultra-Turrax T25 homogenizer (IKA-Labortechnik, Staufen, Germany) at 13 500 rpm for 1 minute. The concentration of each heart homogenate is expressed in terms of its protein concentration. The protein concentrations of the different homogenates varied between 8 and 11 mg/mL.
Preparation of Human IF
Human IF was obtained on seven separate occasions from the skin of three healthy donors by the suction blister method, as described by Kiistala and Mustakallio.16 Blisters were generated between the epidermis and the dermis by mild suction with a Dermovac suction blister device (Instrumentarium). Locally warmed abdominal skin was subjected to suction pressures of 100 to 200 mm Hg for 2 to 3 hours. The standard procedure produced 20 to 30 blisters with diameters of 5 mm, each containing 20 to 30 μL fluid. IF was collected by aspiration, extensively dialyzed against PBS, filtered through a 0.22-μm filter, and stored at −50°C. IF and the corresponding serum were analyzed for their concentrations of total protein and α1-antitrypsin and for their Ang II–forming capacities.
Determination of Ang I Conversion
The standard assay was conducted at 37°C in 50 μL PBS (in mmol/L: NaCl 137, KCl 2.7, Na2HPO4 8.1, CaCl2 0.9, KH2PO4 1.1, MgCl2 0.5; pH 7.3) containing heart homogenate (25 μg protein), IF (10 to 25 μL), or a mixture of the two; 5 nmol Ang I; and the indicated concentrations of inhibitors. After incubation for the indicated times, the reactions were stopped by addition of 300 μL ice-cold ethanol and incubated at 4°C for 30 minutes, and the precipitated proteins were centrifuged at 15 000g for 10 minutes at 4°C. The supernatants were then collected for peptide analysis by RP-HPLC.
Determination of ACE Activity
ACE activity in both heart homogenate and IF was also measured, with FAPGG as substrate. FAPGG has been widely used to measure ACE activity by spectrophotometry.19 However, this method proved to be insensitive to ACE activities in the heart homogenate. Therefore, the degradation of FAPGG to FAP by the ACE was monitored by RP-HPLC. The standard assay was conducted at 37°C in 50 μL PBS containing heart homogenate (25 μg protein), IF (25 μL), or a mixture of the two; 5 to 10 nmol FAPGG; and the indicated concentrations of Ang-(1-9). After incubation for the indicated times, the reactions were stopped with ice-cold ethanol, and the samples were prepared for RP-HPLC analysis as described for angiotensin peptides.
For RP-HPLC analysis, the supernatants containing angiotensin peptides or FAPGG/FAP were evaporated to dryness and finally dissolved in 100 μL 0.1% trifluoroacetic acid. Eighty-five microliters of each sample was analyzed on a reverse-phase column (Spherisorb S5X C18, 5 μm/30 nm, 3×150 mm). The chromatographic apparatus consisted of two pumps (Applied Biosystems solvent delivery system 400) controlled by an Applied Biosystems 738 detector/gradient controller. The column was eluted at a flow rate of 0.5 mL/min with an increasing linear gradient of acetonitrile (0% to 32% in 40 minutes) containing 0.075% trifluoroacetic acid, and the eluate was monitored at 214 nm. Ang I–derived peptides were identified by comparison of their retention times with those of synthetic standards and by N-terminal sequence analysis. Formation of Ang II and Ang-(1-9) was quantified by measurement of peak area relative to synthetic standards. The results are expressed as nmol Ang II or Ang-(1-9) formed per minute per milligram homogenate protein or per milliliter IF. Formation of FAP was quantified by measurement of peak height relative to a known standard. Under these conditions, FAPGG eluted at 34 minutes and FAP at 39 minutes. The results are expressed as nmol FAP formed per minute per milligram homogenate protein or per milliliter IF.
N-Terminal Sequence Analysis
The angiotensin peptide fractions obtained from RP-HPLC analysis were subjected to an automatic sequence analysis with an Applied Biosystems Procise 494 protein sequencing system and a model 610 data analysis system.
Protein concentrations were determined by the procedure of Lowry et al,20 with BSA as standard. α1-Antitrypsin concentrations were determined by nephelometry as described by the manufacturer (Orion Diagnostica).
Degradation of Ang I by Human Heart Tissue
We first studied the degradation of Ang I by heart tissue in the absence of IF. Fig 2⇓ shows a typical RP-HPLC analysis of Ang I–derived peptides after incubation of Ang I with heart homogenate for 3 hours at 37°C. At this time, 90% of the Ang I (elution time, 40 minutes) was degraded. The elution profile disclosed two major peptide peaks, one eluting at 32 and the other at 36 minutes. The peptides were then identified by N-terminal sequence analysis. The peptide eluting at 36 minutes was found to be Ang II, suggesting an effect of heart ACE, chymase, or cathepsin G on Ang I (see Fig 1⇑). The peptide eluting at 32 minutes was Ang-(1-9), suggesting an effect of heart CPA–like activity on Ang I. The peptide eluting at 11 minutes was His-Leu, a dipeptide formed in parallel with Ang II.
The results presented in Fig 2⇑ show that degradation of Ang I by human heart tissue leads to formation of two major metabolites, Ang II and Ang-(1-9). We next assessed the formation of these two peptides as a function of time. As shown in Fig 3A⇓, Ang I was rapidly degraded by the heart tissue. The rate of Ang I degradation was closely followed by formation of the two degradation products, Ang II and Ang-(1-9) (Fig 3B⇓). At the end of incubation, 200 nmol Ang I had been degraded per milligram heart tissue. Of the Ang I degraded (170 nmol/mg heart tissue), 85% had been converted into Ang II and Ang-(1-9). The residual 15% was represented by the various peptides eluting between 19 and 29 minutes (see Fig 2⇑). The formation of these peptides was not inhibited by inhibitors of chymase, ACE, or CPA, and their formation was not studied further. The rates of formation of Ang II and Ang-(1-9) were linear for at least 30 minutes. Accordingly, in the subsequent experiments the incubation time was 30 minutes.
The abilities of the 10 failing hearts obtained at cardiac transplantation to form Ang II and Ang-(1-9) were tested. The patients had had end-stage congestive heart failure due to coronary heart disease (n=5), idiopathic dilated cardiomyopathy (n=3), or congenital heart disease (n=2). Their abilities to degrade Ang I were similar irrespective of the cause of the heart failure. Thus, the amounts of Ang II and Ang-(1-9) formed per sample were 1.37±0.23 and 1.60±0.40 nmol·min−1·mg−1 (±SD; n=10), respectively.
Inhibition of Ang II and Ang-(1-9) Formation by Enzyme Inhibitors
The findings shown in Figs 2⇑ and 3⇑ suggested that degradation of Ang I by human heart tissue is mediated by several enzymes. To study the contribution of the enzymes probably involved, the degradation of Ang I was studied in the presence of various enzyme inhibitors.
In a control experiment, we showed that lowering of pH from 7.3 to 4.0 completely inhibited formation of Ang II and Ang-(1-9), indicating that lysosomal enzymes were not involved in the formation of these peptides by heart homogenate (data not shown). In addition, we showed that >90% of the Ang II– and Ang-(1-9)–forming activities could be sedimented at 40 000g (data not shown). This finding accords with that of Urata et al,5 who showed that chymase activity in heart homogenate resides in the 40 000g membrane preparation.
As shown in the Table⇓, formation of Ang II was effectively inhibited by soybean trypsin inhibitor, which inhibits serine proteases, and by chymostatin, a specific inhibitor of chymotrypsin-like enzymes.6 Moreover, formation of Ang II was inhibited by murine anti-chymase monoclonal antibody (data not shown). Neither of these inhibitors affected the formation of Ang-(1-9). Captopril, a specific ACE inhibitor, also slightly reduced Ang II formation. Similarly, if the experiment was repeated with another ACE inhibitor, lisinopril (1 mmol/L), Ang II formation was reduced by only 10% (data not shown). In addition, aprotinin (0.4 mg/mL), an inhibitor of cathepsin G but not of chymase,10 had no effect on the formation of Ang II and Ang-(1-9), showing that cathepsin G was not involved in Ang I metabolism. These findings indicated that most (≈90%) of the Ang II formed was contributed by heart chymase and very little by heart ACE.
The experiment shown in the Table⇑ was performed with heart homogenate from a patient who had received ACE inhibitors before heart transplantation. Since prolonged therapy with ACE inhibitors may affect the observed ACE activity in heart homogenates, we also carried out inhibitor experiments with homogenates derived from two patients who had not received ACE inhibitors before transplantation. The results, however, were identical to those described above. Thus, >90% of the Ang II–forming activity in the heart tissue was inhibited by soybean trypsin inhibitor and by chymostatin and very little by captopril (data not shown).
Formation of Ang-(1-9) was effectively inhibited by EDTA, an inhibitor of metalloproteases, such as carboxypeptidases. A similar inhibitory effect was found with CPI, a specific inhibitor of carboxypeptidases. Since the carboxypeptidase activity found in the heart homogenate cleaved the carboxyterminal leucine from Ang I, this enzymatic activity in human heart tissue is thereafter referred to as CPA-like activity (Table⇑). Furthermore, formation of Ang-(1-9) was inhibited by about 30% by the ACE inhibitor captopril (1 mmol/L) owing to the ability of captopril in high concentrations to inhibit CPA-like activity.21
The results shown in the Table⇑ are typical, obtained with one homogenate. In addition, we tested heart homogenates from seven other donors with similar results (data not shown). Taken together, the above results show that Ang I metabolism in human heart tissue is mediated by three enzymes: heart chymase and heart ACE converting Ang I into Ang II, and heart CPA–like activity degrading Ang I to Ang-(1-9).
Inhibition of Chymase-Mediated Ang II Formation by IF
To mimic the conditions of angiotensin metabolism in the heart interstitium, we incubated Ang I with heart homogenate in the presence of IF. Human IF was obtained from the skin of three healthy donors as described in “Methods.” The total protein concentration of the IF was 27±2.3 mg/mL (±SD; n=7; obtained from the three donors on seven occasions), ie, 31% of the protein concentration of the corresponding sera. In IF, the concentration of the protease inhibitor α1-antitrypsin was 0.56±0.06 mg/mL (±SD), representing 35% of the corresponding serum concentration. IF also contained significant Ang II–forming activity: 0.62±0.07 nmol·min−1·mL−1 (±SD), ie, 24% of the capacity of the corresponding sera. Ang II formation was completely inhibited by 1 mmol/L captopril but not by 100 μmol/L chymostatin, indicating the presence of soluble ACE in IF.
We decided to study first the effect of IF and more specifically the effect of protease inhibitors on the heart chymase–mediated formation of Ang II. For this purpose, Ang I and heart homogenate were incubated with various concentrations of IF (from 0.5% to 100% vol/vol) in the presence of captopril. Captopril inhibits the ACE activity present in the heart tissue and in the IF but leaves chymase activity intact. As shown in Fig 4⇓, heart chymase–mediated Ang II formation was effectively inhibited in the presence of IF. Of the Ang II–forming activity of heart chymase, 50% was inhibited by as little as 0.5% (vol/vol) IF, and >95% was inhibited at concentrations of IF >50%. In contrast, only 40% of Ang-(1-9) formation was inhibited by the presence of 100% IF, showing that the CPA-like activity in heart homogenate was largely preserved (Fig 4⇓).
Fig 4⇑ shows the results obtained with one IF preparation and one heart homogenate. In all, we performed nine measurements in which we tested six different IF preparations derived from three donors with heart homogenates from four donors. It could be shown that the IF preparations were equally effective in their inhibitory capacity. Thus, in the presence of 50% IF, Ang II formation was inhibited by 90±5.6% and Ang-(1-9) formation by 30±13% (±SD; n=9).
The results indicate that IF is able to effectively inhibit heart chymase–mediated formation of Ang II. The major physiological inhibitor of human chymase has been shown to be α1-antitrypsin.22 The hypothesis that α1-antitrypsin is also the major inhibitor of chymase-mediated Ang II formation in the presence of IF was tested by incubating Ang I and heart homogenate with purified human α1-antitrypsin at the same concentration (0.56 mg/mL) of this inhibitor as was present in IF. In this experiment, 82% of the chymase-mediated Ang II formation was inhibited, strongly suggesting that the major inhibitor of chymase-mediated Ang II formation in the presence of IF was α1-antitrypsin (data not shown).
Inhibition of ACE-Mediated Ang II Formation by Ang-(1-9)
In the next series of experiments, we studied ACE-mediated Ang II formation in the presence of IF. When Ang I was incubated with heart homogenate alone, 0.56 nmol Ang II per assay was formed (Fig 5A⇓, left column). As shown in the Table⇑, 90% of this Ang II formation was mediated by heart chymase. When Ang I was incubated with 50% IF alone, 0.47 nmol Ang II per assay was formed by the ACE present in IF (middle column). Thus, potentially, the capacity of heart homogenate and IF to form Ang II should be 1.03 nmol per assay. However, when heart homogenate and 50% IF were incubated together, only 0.07 nmol Ang II per assay was formed (right column). Thus, 93% of the potential capacity of heart homogenate and IF to form Ang II was inhibited when the two were incubated together. From the results in Fig 4⇑, one would expect to obtain >95% inhibition of heart chymase–mediated Ang II formation in the presence of 50% IF. However, considerable amounts of Ang II should be formed by the ACE present in IF in the presence of heart homogenate. Thus, the results suggest the presence of an ACE inhibitor in the mixture, either the heart homogenate itself containing an ACE inhibitor or an ACE inhibitor being formed from Ang I during incubation.
To test this hypothesis, we used a well-characterized specific ACE substrate, FAPGG, to measure ACE activity both in the heart tissue and in IF.19 ACE was found to hydrolyze the Phe-Gly bond, with formation of FAP, as detected by RP-HPLC analysis (see “Methods”). Of the FAP formed, 95% was inhibited both in heart tissue and in IF by 1 mmol/L captopril but not by 100 μmol/L chymostatin or by 1 mg/mL soybean trypsin inhibitor, indicating that hydrolysis of the substrate was due to ACE activity, chymase being inactive against this substrate. Moreover, no further degradation of FAP to furanacrylic acid by heart CPA–like activity was observed (data not shown).
Fig 5B⇑ shows an experiment similar to that described in 5A, except that FAPGG was used as substrate instead of Ang I. When FAPGG was incubated with heart homogenate alone, 0.10 nmol FAP per assay was formed by the heart ACE (Fig 5B⇑). When FAPGG was incubated with 50% IF alone, 1.63 nmol FAP per assay was formed by the ACE present in IF. Thus, the potential capacity of the heart homogenate and IF to form FAP would be 1.73 nmol per assay. When the heart homogenate and 50% IF were incubated together, 1.14 nmol FAP per assay was formed, representing 66% of the potential ACE activity present in the heart homogenate and IF. In contrast, if 2 nmol Ang-(1-9) was added to the mixture of heart tissue and IF, 97% of the potential ACE activity was inhibited (right column). Thus, the results strongly suggest that the observed inhibition of the ACE-mediated Ang II formation in the mixture of heart and IF was due to generation of Ang-(1-9) in the assay (Fig 5A⇑). The inset in Fig 5A⇑ shows the formation of Ang-(1-9) by the heart homogenate, by IF, and by a mixture of the two when incubated with Ang I. Most (61%) of the Ang-(1-9) formed by the heart homogenate was preserved in the presence of IF, whereas no detectable amount of Ang-(1-9) was formed by IF alone.
It has been shown previously that Ang-(1-9) is a competitive inhibitor of ACE.23 In the next series of experiments, we studied the effect of Ang-(1-9) as an inhibitor of the ACE activity present in the heart tissue and IF in greater detail. In Fig 6A⇓, the effect of Ang-(1-9) on IF ACE was studied. Ten microliters of IF was incubated with 5 nmol (100 μmol/L) Ang I in the presence of increasing concentrations (from 1 to 40 μmol/L) of Ang-(1-9). It was demonstrated that IF ACE–mediated Ang II formation was effectively inhibited by Ang-(1-9), 50% of the Ang II–forming capacity being inhibited with 2 μmol/L Ang-(1-9). This concentration of Ang-(1-9) was only 1/50 of the concentration of Ang I in the incubation mixture. Ang II formation was fully inhibited when the concentration of Ang-(1-9) was 40 μmol/L.
The inset in Fig 6A⇑ shows a control experiment in which the effect of Ang-(1-9) on heart chymase–mediated Ang II formation was studied. Of the Ang-(1-9) formed by the CPA-like activity present in the heart homogenate, 95% was inhibited by CPI. About 0.1 nmol Ang-(1-9) per assay was formed even in the presence of CPI. However, addition of Ang-(1-9) to a concentration of 40 μmol/L had no effect on heart chymase–mediated Ang II formation.
To study the effect of Ang-(1-9) on heart ACE, we used FAPGG as a substrate instead of Ang I. This was prompted by the observation that on incubation of heart homogenate with Ang I, even in the presence of a CPI, sufficient amounts of Ang-(1-9) were formed to inhibit the heart ACE. Therefore, heart homogenate was incubated with 5 nmol (100 μmol/L) of FAPGG in the presence of increasing amounts of Ang-(1-9). Like the ACE activity in IF, the ACE activity present in the heart homogenate was effectively inhibited by Ang-(1-9) (Fig 6B⇑).
Taken together, the experiments illustrated in Figs 4 through 6⇑⇑⇑ show that the presence of IF effectively inhibits local Ang II formation in the heart homogenate; the natural protease inhibitors of IF effectively block heart chymase–mediated Ang II formation, and the Ang-(1-9) formed by heart CPA–like activity effectively inhibits heart ACE– and IF ACE–mediated Ang II formation at concentrations that represent only a fraction of that of Ang I.
Inhibition of Heart Chymase–Mediated Ang II Formation in the Presence of IF
In the absence of IF, the major Ang II–forming enzyme in the heart tissue was chymase, as also reported by Urata et al.5 In a recent study, however, Zisman et al24 claimed that in the absence of IF, the major Ang II–forming enzyme in the heart tissue was not chymase but ACE. The reason for the lack of chymase activity in that study appears to be that the procedure used to prepare the solubilized heart membrane preparation resulted in a significant loss of chymase activity. In our laboratory, at least 90% of the chymase activity was lost when the above-described procedure was used (data not shown).
In the presence of IF, practically no heart chymase–mediated Ang II formation was observed. To represent cardiac IF, we used tissue fluid obtained from human skin by the suction blister method.16 Several lines of evidence indicate that tissue fluid obtained from skin by the suction blister method can be regarded as IF.14 15 17 18 Skin IF–to-serum ratios of various proteins reveal that skin IF is an ultrafiltrate of plasma. The total protein concentration of skin IF is about 20% to 40% of the corresponding serum values, and its composition resembles that of the peripheral lymph.14 15 The concentration in skin IF of the major physiological inhibitor of human chymase, α1-antitrypsin,22 has been reported to be 29% of the corresponding serum value.15 In IF used in this study, the concentrations of total protein and of α1-antitrypsin were in accord with the above findings, being 31% and 35% of the corresponding serum values. Whether the IF of the human heart differs from that of human skin in protein composition is not known, because no method for obtaining cardiac IF has yet been devised. Interestingly, skin IF has recently been used to represent IF of human arterial intima.18
Rat chymase complexed to heparin is more resistant to natural protease inhibitors than purified enzyme alone.25 We have shown that after homogenization of human heart tissue, chymase and CPA-like activity remain complexed to heparin proteoglycan (J.S. and P.T.K., unpublished observations). However, as shown in the present study, human heart chymase was effectively inhibited in the presence of IF, suggesting that rat and human chymases may differ in their sensitivities to protease inhibitors. Therefore, we performed additional experiments with purified rat and human chymases complexed and not complexed to commercial heparin. It could be shown that although rat chymase complexed to heparin was more resistant to natural protease inhibitors of plasma than the purified enzyme alone, no such difference was found with human chymase (L. Lindstedt and P.T. Kovanen, unpublished observations). Thus, it seems evident that, in contrast to rat chymase, complexing of human chymase with heparin does not protect it against natural protease inhibitors.
Although the concentrations of protease inhibitors were lower in IF than in the corresponding serum, they were nevertheless so high that heart chymase–mediated Ang II formation was practically fully inhibited. In the present study, we showed that chymase-mediated Ang II formation could be inhibited with purified α1-antitrypsin by 82%, with the same concentration of α1-antitrypsin (0.56 mg/mL) as is present in IF. Thus, we conclude that α1-antitrypsin is the major inhibitor of chymase-mediated Ang II formation.
Inhibition of ACE-Mediated Ang II Formation by Ang-(1-9)
Naturally occurring ACE inhibitors have previously been reported by Ikemoto et al26 and Snyder and Wintroub.23 Interestingly, Snyder and Wintroub showed that Ang-(1-9) is able to competitively inhibit ACE. The present data demonstrated that both heart ACE and IF ACE could be fully inhibited by Ang-(1-9). In competitive inhibition, the degree of inhibition depends on the relative concentrations of the inhibitor and the substrate. In this study, the IF ACE was inhibited by 50% and 100% by Ang-(1-9) at concentrations of 1/50 and 2/5, respectively, of the concentration of Ang I. The actual concentrations of Ang-(1-9) in such tissues as human heart are currently not known. In humans and rats, plasma concentrations of Ang-(1-9) are very low.27 28 In rat kidney, however, the concentration of Ang-(1-9) is half of that of Ang I,28 indicating the presence of a CPA-like activity in that organ. Accordingly, at least within the kidney interstitium, the ACE activity should be fully inhibited.
Role of CPA-Like Enzyme Activity in Angiotensin Metabolism
The present study demonstrated that in the presence of IF, the major Ang I–metabolizing enzyme in the heart homogenate was CPA-like activity and the major metabolite of Ang I was not Ang II but Ang-(1-9). This was because the two Ang II–forming enzymes chymase and ACE were strongly suppressed, but CPA-like activity was resistant to the protease inhibitors present in IF.
Previous reports on angiotensin metabolism of the human heart5 24 make no mention of CPA-like activity. The reason for the lack of CPA-like activity in the study by Zisman et al24 appears to be that in the solubilized heart membrane preparations, such activity, like the activity of chymase (see above), was greatly reduced. Indeed, as with chymase, in our hands ≥90% of the CPA-like activity was lost when we prepared heart membranes by the above-mentioned procedure (data not shown). The factors contributing to the discrepancy between our findings and those of Urata et al,5 ie, the presence and absence of CPA-like activity, respectively, remain obscure. One factor may be the methodological differences between the two studies: we used heart homogenate instead of a heart membrane preparation, and we used higher concentrations of Ang I than Urata et al.5 However, when we used a heart membrane preparation identical to that described by Urata et al, the two major Ang I–derived metabolites were Ang II and Ang-(1-9), reflecting the presence of active chymase and CPA-like activity in the preparation. This was so even when the concentration of Ang I was lowered to 10 nmol/L; at this concentration of the substrate, the formation of Ang II was 228±14.8 and that of Ang-(1-9) 200±54.5 fmol·min−1·mg−1 (±SD; n=3). In this series of experiments, we labeled Ang I with 14C by reductive methylation.29
In the tissues studied so far, the only source of CPA-like activity is the TC mast cells,30 ie, cells that contain both tryptase and chymase. Immunohistochemical studies have shown that 90% of all heart mast cells are TC mast cells.31 32 Thus, it seems evident that the CPA-like activity now observed in the heart tissue is also the activity of mast cell–derived CPA. The amount of CPA in human mast cells is very high. It has been estimated that TC mast cells contain 16 pg CPA/cell, compared with 5 pg chymase/cell.13 In mast cells, CPA is located within the secretory granules together with chymase and tryptase, all three enzymes being bound to heparin proteoglycans.33 After mast cell stimulation and degranulation, chymase and CPA, unlike tryptase, remain bound to proteoglycans, thus forming extracellular chymase/CPA/proteoglycan complexes.33
In the heart homogenate used in this study, the mast cells were artificially disrupted, and as a result, CPA-like activity and chymase had free access to the substrate, Ang I. Normally, these enzymes are not secreted constitutively but are located intracellularly within the secretory granules of mast cells. Accordingly, for CPA-like activity and chymase to exert their action in the heart interstitium, the mast cells must have been stimulated to degranulate. However, the fact that most of the chymase activity in the heart has been found to be localized to the extracellular matrix8 reveals that mast cells have degranulated and actively secreted granules, ie, complexes containing both chymase and CPA, into the heart interstitium.
Although Ang I has been used as a model substrate in studies of mast cell CPA, in which Ang I is readily hydrolyzed to Ang-(1-9) (Km=60 μmol/L, kcat=37 s−1, kcat/Km=0.62 [μmol/L]−1·s−1),12 13 34 the physiological substrate of mast cell CPA has remained unknown. Our results suggest that CPA may play an important role in local angiotensin metabolism in the heart interstitium.
The present in vitro study describes two novel mechanisms that effectively inhibited Ang II formation in conditions mimicking those existing in the human heart interstitium. One inhibitory mechanism, protease inhibitor–mediated suppression of chymase, should be effective in vivo, because the heart interstitium is constantly bathed by IF containing protease inhibitors in concentrations sufficiently high to ensure efficient inhibition of this enzyme. Consequently, we speculate that the Ang II concentration in the heart interstitium is regulated by ACE rather than by chymase. This notion is supported by the recent in vivo findings of Zisman et al24 that most of the Ang II formation in the human heart was blocked by an ACE inhibitor (enalaprilat). However, our results also suggest that the ACE activity located within heart interstitium may be suppressed. This suppression of ACE by Ang-(1-9), an endogenous ACE inhibitor, depends on stimulation of myocardial mast cells, with ensuing secretion of CPA-like activity into the interstitium. Under such conditions, the residual heart ACE activity responsible for Ang II formation and inhibitable by ACE inhibitors (eg, enalaprilat) is likely to be the ACE activity located on the luminal surface of the heart capillary endothelium.
Selected Abbreviations and Acronyms
|CPI||=||carboxypeptidase inhibitor from potato tuber|
|RP-HPLC||=||reverse-phase high-performance liquid chromatography|
This study was supported by grants from the Paavo Nurmi Foundation, Helsinki (Dr Kokkonen). We are grateful to Professor Severi Mattila and the heart transplantation team, University Central Hospital of Helsinki, for supplying human heart tissue; to Professor Timo Reunala, Department of Dermatology, University Central Hospital of Helsinki, for supplying skin blister fluid; and to Dr Nisse Kalkkinen, Institute of Biotechnology, University of Helsinki, for performing the N-terminal sequence analysis.
- Received June 17, 1996.
- Revision received September 23, 1996.
- Accepted November 19, 1996.
- Copyright © 1997 by American Heart Association
Baker KM, Aceto JF. Angiotensin II stimulation of protein synthesis and cell growth in chick heart cells. Am J Physiol. 1990;259:H610-H618.
Brilla CG, Pick R, Tan LB, Janicki JS, Weber KT. Remodeling of the rat right and left ventricles in experimental hypertension. Circ Res. 1990;67:1355-1364.
Lindpaintner K, Jin MW, Niedermaier N, Wilhelm MJ, Ganten D. Cardiac angiotensinogen and its local activation in isolated perfused beating heart. Circ Res. 1990;67:564-573.
Urata H, Healy B, Stewart RW, Bumbus FM, Husain A. Angiotensin II-forming pathways in normal and failing hearts. Circ Res. 1990;66:883-890.
Urata H, Kinoshita A, Misono KS, Bumbus FM, Husain A. Identification of a highly specific chymase as the major angiotensin II-forming enzyme in the human heart. J Biol Chem. 1990;265:22348-22357.
Hoit BD, Shao Y, Kinoshita A, Gabel M, Husain A, Walsh RA. Effects of angiotensin II generated by an angiotensin-converting enzyme-independent pathway on the left ventricular performance in the conscious baboon. J Clin Invest. 1995;95:1519-1527.
Urata H, Boehm KD, Philip A, Kinoshita A, Gabrovsek J, Bumbus FM, Husain A. Cellular localization and regional distribution of an angiotensin II-forming chymase in the heart. J Clin Invest. 1993;91:1269-1281.
Schecter NM, Irani A-MA, Sprows JL, Abernethy J, Wintroub BU, Schwartz LB. Identification of a cathepsin G-like proteinase in the MCTC type of human mast cell. J Immunol. 1990;145:2652-2661.
Tonnesen MG, Klempner MS, Austen KF, Wintroub BU. Identification of a human neutrophil angiotensin II generating protease as cathepsin G. J Clin Invest. 1982;69:25-30.
Goldstein SM, Kaempfer CE, Proud D, Schwartz LB, Irani A-MA, Wintroub BU. Detection and partial characterization of a human mast cell carboxypeptidase. J Immunol. 1987;139:2724-2729.
Goldstein SM, Kaempfer CE, Kealey JT, Wintroub BU. Human mast cell carboxypeptidase: purification and characterization. J Clin Invest. 1989;83:1630-1636.
Dabbagh AJ, Frei B. Human suction blister interstitial fluid prevents metal ion-dependent oxidation of low density lipoproteins by macrophages and in cell-free systems. J Clin Invest. 1995;96:1958-1966.
Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem. 1951;193:265-275.
Schecter NM, Sprows JL, Schoenberger OL, Lazarus GS, Cooperman BS, Rubin H. Inhibition of human skin chymotrypsin-like proteinase chymase with plasma proteinase inhibitors. J Biol Chem. 1989;264:21308-21315.
Zisman LS, Abraham WT, Meixell GE, Vamvakias BN, Quaife RA, Lowes BD, Roden RL, Peacock SJ, Groves BM, Raynolds MV, Bristow MR, Perryman MB. Angiotensin formation in the intact human heart: predominance of the angiotensin-converting enzyme pathway. J Clin Invest. 1995;96:1490-1498.
Campbell DJ, Lawrence AC, Towrie A, Kladis A, Valentijn AJ. Differential regulation of angiotensin peptide levels in plasma and kidney of the rat. Hypertension. 1991;18:763-773.
Irani A-MA, Goldstein SM, Wintroub BE, Bradford T, Schwartz LB. Human mast cell carboxypeptidase: selective localization to MCTC cells. J Immunol. 1991;147:247-253.
Sperr WR, Bankl HC, Mundigler G, Klappacher G, Großschmidt K, Agis H, Simon P, Laufer P, Imhof M, Radaszkiewicz T, Glogar D, Lechner K, Valent P. The human cardiac mast cell: localization, isolation, phenotype, and functional characterization. Blood. 1994;84:3876-3884.
Patella V, Marinò I, Lampärter B, Arbustini E, Adt M, Marone G. Human heart mast cells: isolation, purification, ultrastructure, and immunologic characterization. J Immunol. 1995;154:2855-2865.
Goldstein SM, Leong J, Schwartz LB, Cooke D. Protease composition of exocytosed human skin mast cell protease-proteoglycan complexes: tryptase resides in a complex distinct from chymase and carboxypeptidase. J Immunol. 1992;148:2475-2482.
Goldstein SM. Mast cell carboxypeptidase: structure and regulation of gene expression. In: Caughey GH, ed. Mast Cell Proteases in Immunology and Biology. New York, NY: Marcel Dekker Inc; 1995:109-126.