We utilized mice with homozygous disruption of angiotensin-converting enzyme (ACE) (−/−), mice with heterozygous deletion of ACE (+/−), and wild-type mice (+/+) to test the hypothesis that genetic variation in ACE modulates tissue and plasma angiotensin (ANG) II concentrations. With the use of ANG I as substrate, kidney, heart, and lung ACE activity was reduced 80% in −/− mice compared with +/+ mice. However, ANG II concentrations and ANG II-to-ANG I ratios in the kidney, heart, and lung did not differ among genotypes. In contrast, plasma ANG II concentrations in −/− mice were <2 fmol/ml, whereas plasma ANG I concentrations were extremely high (765 fmol/ml). Chymase activity was increased 14-fold in the kidney (P < 0.05) and 1.5-fold in the heart (P < 0.05) of −/− versus +/+ mice but did not differ among genotypes in the lung. ANG II formation from enzymes other than ACE and chymase contributed <2% of total ANG II formation in all genotypes. These data suggest that ACE is essential to ANG II formation in the vascular space, whereas chymase may provide an important mechanism in maintaining steady-state ANG II levels in tissue.
- angiotensin I
- angiotensin-converting enzyme
the octapeptide angiotensin ii (ANG II) exerts a wide range of physiological and developmental effects on the cardiovascular, renal, endocrine, and nervous systems (16). These organ systems possess intrinsic ANG II-generating mechanisms (3), which include angiotensin-converting enzyme (ACE) ANG II-forming pathways as well as others that are not dependent on ACE (8). In particular, a serine protease with an extremely high affinity for ANG I, “chymostatin-sensitive angiotensin-generating enzyme,” has been identified in multiple organs of all mammals (22). Furthermore, chymase has catalytic activity for the conversion of ANG I to ANG II that is 20-fold higher than ACE and substantially greater than other ANG II-forming enzymes (7, 24). Whereas ACE has been clearly demonstrated to be the major enzyme responsible for ANG II generation from ANG I in the intravascular space (1, 26), the relative contribution of ACE versus chymase to ANG II generation in tissue remains controversial (2, 25, 26).
Mammalian chymases occur as two distinct isoenzyme groups, α and β (5), which differ in species distribution and substrate specificity. α-Chymase converts ANG I to ANG II by cleaving the Phe8-His9 bond in ANG I (15, 24), whereas β-chymase cleaves the Tyr4-Ile5 and Phe8-His9 bonds in ANGs with near equal efficiencies. As a result, only α-chymase results in net ANG II generation. Humans, baboons, dogs, hamsters, rats, and mice contain a single α-chymase-encoding gene. Humans and baboons do not contain a β-chymase gene, whereas rats contain two genes and mice contain four genes. The predominance of the β-chymase isoform in the mouse may account for the observation that only 15% of in vitro ANG II formation from extracts of the wild-type mouse heart is chymostatin inhibitable (2). However, in vitro tissue assay methodologies may not accurately reflect ANG II formation in vivo, because they utilize pharmacological doses of ANG I substrate, disrupt tissue compartmentalization of enzymes, alter biochemical conditions and pH optima, and incorporate inhibitors of other potentially important ANG II-forming pathways (e.g., tonin, cathepsin G).
The advent of induced mutation technology has provided a new way to study the molecular regulation of ANG II formation in the animal in vivo. Krege and co-workers (13, 21) used gene targeting to insertionally disrupt exon 14 of ACE and thereby markedly decrease ACE expression. Mice would be expected to be particularly sensitive to alterations in tissue ACE, because ACE accounts for >80% of ANG II-generating capacity in heart tissue extracts in this species in vitro (2). In the current study, we utilized mice with homozygous deletion of ACE (−/−), mice with heterozygous deletion of ACE (+/−), and wild-type mice (+/+) to test the hypothesis that genetic variation in ACE modulates tissue and plasma ANG II concentrations.
Generation of mice.
F2 mice (129 × C57BL/6) +/− and +/+ for an insertional disruption of exon 14 of the Ace (mouse) gene (9,11) were provided by Drs. J. H. Krege and O. Smithies (Dept. of Pathology, Univ. of North Carolina, Chapel Hill, NC). Offspring of +/+ or +/− F2 males and +/− F2 females were produced at the Vascular Biology and Hypertension Program, University of Alabama at Birmingham (Birmingham, AL). At an age of 21 days, mice were weaned and genotyped for ACE as previously described in our laboratory (21). Animals were provided chow (Teklad LM-485 sterilizable mouse diet) and water ad libitum and were maintained on a 12:12-h light-dark cycle. Experiments were approved by the Institutional Animal Care and Use Committee at the University of Alabama at Birmingham and are consistent with the National Institutes of Health Guide for the Care and Use of Laboratory Animals(NIH Publication No. 85-23, Revised 1985).
Tissue ACE activity using ANG I as substrate.
ACE activity using ANG I as substrate was performed as previously described in our laboratory (2). ACE was extracted from homogenized tissues with detergent, and the reaction product ANG II was isolated from the reaction mixture by reverse-phase HPLC, thus eliminating interference from the detergent, the substrate ANG I, and unreacted reaction byproducts. To exclude all other potential enzymes that could form ANG II, tissue extract (100 μl) samples were preincubated with enzyme inhibitor solutions that contained 1 mM phenanthroline, 20 μM aprotonin, 300 mM sodium chloride, 0.01% Triton X-100, and 200 μM zinc chloride and chymostatin for 30 min at room temperature. Captopril-inhibitable ANG II formation was considered to represent the ACE activity (nmol ANG II formed · min−1 · g tissue wet wt−1).
Chymase-like activity using ANG I as substrate.
Chymase activity was measured using a procedure that has been performed in our laboratory (1, 2, 7). Generated ANG II was quantitated using a reverse-phase Alltima 5-μm phenyl-HPLC column (Alltech; Deerfield, IL). The peak area corresponding to a synthetic ANG II standard was integrated to calculate ANG II formation with chymostatin and with chymostatin plus captopril. Chymostatin-inhibitable ANG II formation was considered to represent the chymase-like activity, and chymostatin-inhibitable plus captopril-inhibitable ANG II formation was considered to represent all other ANG II-forming pathways (nmol ANG II formed · min−1 · g tissue wet wt−1).
ANG peptide concentrations.
Hearts, lungs, and kidneys from +/+, +/−, and −/− mice were removed after decapitation, rapidly frozen in liquid nitrogen, and stored at −70°C. All peptide assays used three pooled organ samples for ANG peptide assays. ANG I and ANG II concentrations were determined by a method described in our laboratory (7, 14) combining solid-phase extraction, reverse-phase HPLC, and radioimmunoassay.
All data are presented as means ± SE. Differences among groups were assessed by ANOVA with post hoc analysis by Student-Newman-Keuls test using SigmaStat software (Jandel Scientific; San Rafael, CA). Results were considered significant at a level of P < 0.05.
Tissue ACE activity.
ACE activities in the kidney, heart, and lung are shown in Fig.1. ACE activities in the kidney, heart, and lung of +/+ mice were 12.4, 11.8, and 4.8 nmol · min−1 · g−1, respectively. ACE activities were decreased 80% in the kidney, heart, and lung of −/− mice compared with +/+ mice (2.6, 2.4, and 0.97 nmol · min−1 · g−1, respectively).
Tissue chymase activity.
In the kidney, chymase activity was 0.02 ± 0.01 nmol · min−1 · g−1 in +/+ and +/− mice and increased 14-fold in −/− mice (0.28 ± 0.06 nmol · min−1 · g−1,P < 0.001; Fig. 2). Chymase activity did not differ in the heart of +/+ and +/− mice (2.77 ± 0.39 and 2.57 ± 0.40 nmol · min−1 · g−1) but increased 1.5-fold in −/− mice (4.01 ± 0.40 nmol · min−1 · g−1,P < 0.05). In the lung, chymase activity did not differ across +/+, +/−, and −/− mice (1.68 ± 0.42 vs. 1.20 ± 0.28 vs. 1.63 ± 0.28 nmol · min−1 · g−1).
ANG II formation from enzymes other than ACE and chymase.
Total ANG II formation from the kidney, heart, and lung was calculated in the absence of inhibitors and in the presence of the ACE inhibitor captopril and the chymase inhibitor chymostatin. In the presence of captopril and chymostatin, ANG II formation accounted for <2% of total ANG II in the kidney, lung, and heart, and this percent contribution did not differ in +/+, +/−, and −/− mice. Thus we found no evidence to support substantial contribution of other known potential sources of ANG II formation such as trypsin, tonin, cathepsin-G, tissue plasminogen activator, kallikrein, and elastase for ANG II formation in these tissues in our in vitro assays. These results confirm that ACE and chymase are the predominant ANG II-forming mechanisms in these organs in vitro.
Pooled plasma from six −/− mice produced two samples for plasma ANG peptide concentrations. Plasma ANG II was undetectable; however, ANG I concentrations were extremely high at 765 and 717 fmol/ml. In the tissue of the kidney, heart, and lung, ANG I levels ranged between 250 and 420 fmol/g and ANG II levels ranged between 150 and 250 fmol/g in +/+, +/−, and −/− mice (Fig. 3,A and B). Tissue ANG I and ANG II concentrations did not differ in the kidney, heart, and lung of +/+, +/−, and −/− mice. In addition, there were no differences in the ANG II-to-ANG I ratios in tissues of the three Ace genotypes (Fig.3 C).
The major finding of this study is that intravascular ANG II concentration is markedly depressed, whereas kidney, heart, and lung ANG II concentrations are preserved in mice with genetic ACE deficiency. Our finding of depressed plasma ANG II levels in −/− ACE mice is consistent with previous observations that ACE mediates intravascular conversion of ANG I to ANG II (1, 26). This is consistent with the predominant location of ACE bound to the cell membranes of endothelial cells with its catalytic site exposed to the luminal surface (12). Intravascular ANG I is thus more accessible to ACE than to chymase, which is rapidly inactivated in plasma by serine protease inhibitors (23). Conversely, our finding of preserved tissue ANG II levels provides important new evidence that non-ACE mechanisms play an important role maintaining normal steady-state ANG II levels in tissue.
We did find that ACE activity in the kidney, heart, and lung of −/− mice was reduced to 20% of +/+ mice. This is consistent with our previous finding that −/− mice had low levels of ACE mRNA that were reduced in size and included a readthrough of the neomycin-resistance gene that was inserted in the 3′-end of exon 14 of the ACE gene (21). This construct could produce small amounts of a functional, nonmembrane-bound ACE molecule within the interstitium of organs because the insertion resulted in exclusion of the membrane-anchoring portion of ACE while preserving the catalytic domains. Although this could contribute to tissue ANG II formation, the failure to detect increasing levels of ANG I with decreasing ACE activity in the tissues of ACE-deficient mice strongly supports the activity of an ANG I convertase other than ACE. Chymase is the most likely candidate because ANG II formation from enzymes other than ACE and chymase contributed <2% of total tissue ANG II formation in vitro. In situ hybridization and electrophoretic mobility immunocytochemical studies in the human heart demonstrate that chymase is located in the extracellular matrix and is stored and synthesized in mast cells and endothelial cells, with secretion directed to the interstitium (22). Taken together, the presence of chymase within the interstitium coupled with its high activity profile and increased expression in the heart and kidney in the setting of severe reductions of ACE in vivo support its role in maintaining steady-state ANG II levels in −/− mice.
Additional support for the concept that chymase may be an important regulator of tissue ANG II generation comes from studies of the response of ANG II and chymase to hemodynamic stress. Increased tissue chymase activity in concert with increased tissue ANG II has been demonstrated in left ventricular (LV) volume overload in the dog (7) and mouse (14) and LV pressure overload (20) and cardiomyopathy (18, 19) in the hamster. LV ACE is also upregulated in these models. However, we have previously reported that +/− and +/+ ACE transgenic mice develop similar elevations in LV ANG II and chymase after 4 wk of volume overload from aortocaval fistula despite threefold higher LV ACE activity in +/+ mice (14). Thus LV ANG II levels and cardiac hypertrophy appear to correlate better with LV chymase than with LV ACE in response to hemodynamic stress; however, blood pressure control is clearly under the influence of ACE in these mice.
We (21) have previously reported that +/+, +/−, and −/− ACE mice differ in their blood pressure and heart rate response to exogenous ANG I and bradykinin, indicating that the quantitative variation in ACE function does affect the in vivo metabolism of ACE substrates and acute blood pressure responses. In addition, Wyss and Carlson (4) found an ACE copy-dependent relationship between both basal arterial pressure and arterial pressure responses to a high NaCl diet, further supporting the important role of the ACE gene in chronic arterial blood pressure control and salt and water balance. Bernstein and colleagues (10) also showed that mice lacking tissue-bound ACE (but expressing circulating ACE) exhibit a significant reduction in blood pressure comparable to the ACE knockout mice. From this differential expression of tissue versus plasma ACE, it was presumed that blood pressure reduction was dependent on tissue ANG II and thus would appear to contradict our results. However, these studies did not measure tissue ANG II and, therefore, could not rule out vascular compartmentalization of ANG II generation that is ACE dependent versus tissue ANG II generation that is ACE and non-ACE dependent.
Indeed, previous studies in the isolated rabbit thoracic aorta support a vascular compartmentalization of ANG II generation that is ACE dependent (11). In these studies, ANG I was applied only to the intraluminal vascular border, and ANG II generation and its identification in the media of the vessel wall occurred in the presence of an intact endothelium but not after removal of the endothelium nor in the presence of an ACE inhibitor (11). Other studies in the isolated rabbit and rat aorta also demonstrated a functional importance of ANG II generation from the adventitial side of the blood vessel wall (9, 17). The results of the current investigation further support the hypothesis that ANG II formation in the vascular space is dependent on endothelium-bound ACE, whereas ANG II formation within tissue relies upon both ACE and chymase.
Address for reprint requests and other correspondence: L. J. Dell'Italia, Univ. of Alabama at Birmingham, Dept. of Medicine, Division of Cardiology, 834 MCLM, 1918 University Blvd, Birmingham, AL 35295-0007 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2002 the American Physiological Society