QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: mol.105.020636v1 69/5/1725    most recent Submit a response View responses Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kohlstedt, K. Articles by Fleming, I. Search for Related Content PubMed PubMed Citation Articles by Kohlstedt, K. Articles by Fleming, I.

Angiotensin-Converting Enzyme (ACE) Dimerization Is the Initial Step in the ACE Inhibitor-Induced ACE Signaling Cascade in Endothelial Cells

Vascular Signalling Group, Institut für Kardiovaskuläre Physiologie (K.K., C.G., R.B., I.F.) and Institut für Biochemie II (M.F., W.M.-E.), Johann Wolfgang Goethe-Universität, Frankfurt am Main, Germany; and Institut National de la Santé et de la Recherche Médicale U367, Paris, France (F.A.-G.)

Received for publication November 8, 2005.

Accepted for publication February 13, 2006.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
The binding of angiotensin-converting enzyme (ACE) inhibitors to ACE initiates a signaling cascade that involves the phosphorylation of the enzyme on Ser1270 as well as activation of the c-Jun NH₂-terminal kinase (JNK) and leads to alterations in gene expression. To clarify how ACE inhibitors activate this pathway, we determined their effect on the ability of the enzyme to dimerize and the role of ACE dimerization in the initiation of the ACE signaling cascade. In endothelial cells, ACE was detected as a monomer as well as a dimer in native gel electrophoresis and dimerization/oligomerization was confirmed using the split-ubiquitin assay in yeast. ACE inhibitors elicited a rapid, concentration-dependent increase in the dimer/monomer ratio that correlated with that of the ACE inhibitorinduced phosphorylation of ACE. Cell treatment with galactose and glucose to prevent the putative lectin-mediated self-association of ACE or with specific antibodies shielding the N terminus of ACE failed to affect either the basal or the ACE inhibitor-induced dimerization of the enzyme. In ACE-expressing Chinese hamster ovary cells, ACE inhibitors elicited ACE dimerization and phosphorylation as well as the activation of JNK with similar kinetics to those observed in endothelial cells. However, these effects were prevented by the mutation of the essential Zn²⁺-complexing histidines in the C-terminal active site of the enzyme. Mutation of the N-terminal active site of ACE was without effect. Together, our data suggest that ACE inhibitors can initiate the ACE signaling pathway by inducing ACE dimerization, most probably via the C-terminal active site of the enzyme.

All of the ACE isoforms hydrolyze circulating peptides and catalyze the extracellular conversion of the decapeptide angiotensin I to the octapeptide angiotensin II, which is a potent vasopressor. ACE also inactivates the vasodilator peptides bradykinin and kallidin, which are derived from kininogen by the action of kallikreins (for review, see Bernstein et al., 2005). Inhibition of ACE is expected to prevent the formation of angiotensin II and to potentiate the hypotensive response to bradykinin, which would lead to the lowering of blood pressure. Somatic ACE also plays a role in vascular remodeling, effects best highlighted by the fact that the in vivo gene transfer of ACE into the uninjured rat carotid artery results in the development of vascular hypertrophy independent of systemic factors and hemodynamic effects (Morishita et al., 1994). Selective overexpression of ACE in the heart also results in morphological changes in the atria, arrhythmia, and sudden death (Xiao et al., 2004). Antisense oligonucleotides against ACE, in contrast, are reported to prevent neointimal formation after balloon angioplasty (Morishita et al., 2000), and ACE inhibitors decrease vascular hypertrophy in hypertensive animals (Clozel et al., 1991). Furthermore, ACE inhibitors, such as ramiprilat, exert beneficial effects on endothelial function and vascular remodeling (Schartl et al., 1994; Mancini et al., 1996) and protect against the progression of atherosclerosis and the occurrence of cardiovascular events in humans (Heart Outcomes Prevention Evaluation Study Investigators, 2000).

Although the latter effects of ACE inhibitors are generally attributed to a decrease in the ACE-mediated generation of angiotensin II and the accumulation of bradykinin (Wiemer et al., 1991), a number of the effects of this class of compounds cannot be accounted for by inhibition of the enzyme per se. For example, it is not generally appreciated that during long-term ACE inhibition angiotensin II levels are not greatly reduced (Lee et al., 1999), and more than 80% of the angiotensin II formation in the human heart and more than 60% of that in arteries is reported to be chymase-dependent (Petrie et al., 2001). Many other reports published over the past 10 years have also shown that ACE inhibitors can amplify responses to bradykinin in situations in which the accumulation of the peptide cannot be assumed to occur. These effects have been attributed to the reactivation of the deactivated B₂ receptor and/or the cross-talk between ACE and intracellular signaling cascades and may be linked to interference with the sequestration of activated G protein-coupled receptors into caveolae (Benzing et al., 1999). We have since established that ACE functions as a signal transduction molecule and that the binding of an ACE inhibitor to ACE initiates a series of events, including the phosphorylation of ACE on Ser1270 and the activation of ACE-associated JNK, that eventually result in changes in gene expression (Kohlstedt et al., 2004, 2005). This ACE signaling cascade occurs independently of the involvement of angiotensin II and bradykinin or any of their receptors. It is however not clear how the binding of an ACE inhibitor is able to initiate the processes described. Because ACE can exist in a dimeric form in vitro (Kost et al., 2003), the aim of the present investigation was to determine whether or not ACE dimerizes in endothelial cells and whether dimerization is involved in ACE inhibitor-induced ACE signaling.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Materials. The ACE monoclonal antibodies (clone 9B9, 3A5, and 5F1) recognizing different epitopes of the N-terminal domain of the enzyme were from Chemicon International (Temecula, CA). The monoclonal antibody (C78) used for Western blotting also recognized the N-terminal domain and was provided by Dr. Peter Bünning (Sanofi-Aventis, Frankfurt, Germany). The mono- and polyclonal antibodies recognizing the cytoplasmic tail of ACE were generated in mice and rabbits immunized with a synthetic peptide corresponding to amino acids 1250 to 1277 in the ACE C terminus. The phosphospecific p-Ser1270 antibody was generated from the peptide sequence HGPQFGpSEVELR (position 1263 to 1275 in human somatic ACE protein) by Eurogentec (Seraing, Belgium). The antibodies against CD31 and JNK-1 were from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). The anti-green fluorescent protein antibody used for immunoprecipitation of yellow green fluorescent protein-tagged ACE was from Roche Applied Science (Mannheim, Germany). All other substances were obtained from Sigma-Aldrich (St. Louis, MO).

Cell Culture. Human umbilical vein endothelial cells were isolated and cultured as described previously (Busse and Lamontagne, 1991). Because ACE expression decreases with time in culture, all of the experiments in human endothelial cells were performed using primary cultures. Porcine aortic endothelial cells stably transfected with ACE or the nonphosphorylatable S1270A ACE mutant were generated and cultured as described previously (Kohlstedt et al., 2002). Although the porcine endothelial cells no longer endogenously expressed ACE or functional AT₁ and B₂ receptors, they expressed a number of characteristic endothelial cell proteins (von Willebrand factor, CD31, the endothelial nitric-oxide synthase, and VE-cadherin).

Histidinyl and glutamyl residues in the active sites of ACE are essential for their enzymatic activity. Inactive ACE mutants were generated by either the mutation of the histidine doublet in the HEMGH sequence to lysine (H361,365K in the N domain and H959,963K in the C domain) or by mutation of the active site glutamate to aspartate (E362D in the N domain or E960D in the C domain). In addition, an ACE mutant in which both glutamate residues in the N and C domains were mutated (E362,960D) was generated and stably overexpressed in CHO cells (Wei et al., 1991). For the sake of clarity, the mutants are referred to as follows: N_His (H361,365K), C_His (H959,963K), N_Glu (E362D), C_Glu (E960D), and N+C_Glu (E362,960D).

Immunoblotting and Immunoprecipitation. Cells were lysed in Nonidet lysis buffer containing 20 mM Tris-HCl, pH 8.0, 137 mM NaCl, 25 mM -glycerophosphate, 10% (v/v) glycerol, 1 mM Na₄P₂O₇, 10 nM okadaic acid, 2 mM Na₃VO₄, 2 µg/ml leupeptin, 2 µg/ml pepstatin A, 10 µg/ml trypsin inhibitor, 44 µg/ml phenylmethylsulfonyl fluoride, and 1% (v/v) Nonidet P-40 and left on ice for 10 min and centrifuged at 10,000g for 10 min. After preclearing with protein A/G-Sepharose, proteins were immunoprecipitated from the cell supernatant or from whole cell lysates with their respective primary antibodies as detailed under Results. Proteins in the cell supernatant or immunoprecipitates were heated with SDS-PAGE sample buffer and separated by SDS-PAGE as described previously (Kohlstedt et al., 2002). Proteins were detected using their respective antibodies and were visualized by enhanced chemiluminescence using a commercially available kit (GE Healthcare, Little Chalfont, Buckinghamshire, UK). To reprobe Western blots with alternative antibodies, the nitrocellulose membranes were incubated at 50°C for 30 min in a buffer containing 67.5 mM Tris-HCl, pH 6.8, 100 mM -mercaptoethanol, and 2% SDS.

For native gels, cells were lysed by the addition of H₂O supplemented with 2 µg/ml leupeptin, 2 µg/ml pepstatin A, 10 µg/ml trypsin inhibitor, and 44 µg/ml phenylmethylsulfonyl fluoride and three cycles of freeze-thawing. The cells were then harvested by scraping, and appropriate volumes of concentrated phosphate-buffered saline were added to give a final concentration of 140 mM NaCl, 10 mM Na₂HPO₄, 2.68 mM KCl, and 1.47 mM K₂HPO₄ at pH 7.0. Lysates were then centrifuged (10 min; 4°C; 10,000g), and supernatants were used for analysis. Equal amounts of protein (20 µg) were loaded on a 6% acrylamide gel without SDS at 4°C 1) in SDS- and DTT-free sample buffer, 2) after treatment with 10 U/ml trypsin at room temperature for 1 h before addition of SDS- and DDT-free sample buffer, and 3) after boiling for 10 min in SDS-PAGE buffer. After electrophoresis (4°C; 20 mA), gels were incubated in SDS-transfer buffer for 10 min, and proteins were transferred to nitrocellulose membranes as described above and blotted with the C78 anti-ACE antibody. The molecular masses were estimated by comparison with those of the reference proteins thyroglobulin (669 kDa), ferritin (440 kDa), and catalase (232 kDa).

Chemical Cross-Linking. Cross-linking was performed using the cross-linking agents disuccinimidyl suberate (DSS) or bis(sulfosuccinimidyl) suberate (BS₃; Pierce Chemical, Rockford, IL) according to the manufacturer's instructions. The cross-linking of ACE on the surface of confluent cells was performed at room temperature for 30 min using 1 to 2 mM DSS or BS₃. After quenching the reaction with 20 mM Tris-HCl, pH 7.4, for 15 min at room temperature, cells were lysed and immunoblotting performed as described above.

Split-Ubiquitin Assay. The split-ubiquitin (Ub) assay was carried out as described previously (Eckert and Johnsson, 2003) using the Saccharomyces cerevisiae strain JD53 (Mat ura 3-52, leu2-3, -112his3 200, lys2-801, trp63). For the expression of a fusion protein of ACE and C-terminal portion of ubiquitin (C_ub), the human ACE-cDNA was cloned into the vector pMet-Ste14-C_ub-RUra3, replacing Ste14 (Wittke et al., 1999). Fusion constructs of ACE and the N-terminal portion of ubiquitin (N_ub) were generated by replacing the Ubc6 sequence by the human full-length ACE-cDNA within the pCUP314-Ubc6 plasmid. To enhance the discriminating power of the assay system, we used two N_ub mutants containing alanine (N_ua) or glycine (N_ug) in position 13, which have a lower affinity for C_ub than wild-type N_ub possessing isoleucine in this position (Johnsson and Varshavsky, 1994). The coding sequence for N_ub, N_ua, and N_ug followed by an ADH terminator sequence was cloned in-frame to the 3' end of the ACE cDNA. Transformed JD53 cells were grown at 30°C in selective medium containing 50 mg/ml uracil (ura) to an optical density₆₀₀ of 1.0, and 4 µl of 10-fold dilutions was spotted on agar plates lacking uracil or containing 1 mg/ml 5'-fluororotic acid (5-FOA) and 50 mg/ml uracil. Plates were incubated at 30°C for 2 to 4 days unless otherwise stated. The plasmids were generous gifts from Dr. Nils Johnsson (Westfälische Wilhelms-Universität Münster, Münster, Germany).

JNK Activity Assay. JNK was immunoprecipitated, and its in vitro kinase activity was measured using 2 µg of glutathione S-transferase-c-Jun as substrate as described previously (Kohlstedt et al., 2004). Reactions were stopped, and the products were resolved by 12% SDS-PAGE. The incorporation of ³²P was visualized by autoradiography and quantified by scanning densitometry.

Statistical Analysis. Data are expressed as the mean ± S.E.M., and statistical evaluation was performed using Student's t test for unpaired data or one-way analysis of variance followed by a Bonferroni t test where appropriate. Values of p < 0.05 were considered statistically significant.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Evaluation of ACE Dimerization by Native Gel Electrophoresis and the Split-Ubiquitin Assay. To determine whether ACE exists as a homodimer in endothelial cells, porcine aortic endothelial cells stably overexpressing human somatic ACE were subjected to native gel electrophoresis. Two ACE forms were apparent: one dominant form of approximately 270 kDa, the putative monomer; and a minor form of approximately 520 kDa, the putative dimer (Fig. 1a). Because the quality of the signals obtained with ACE, which is a highly glycosylated transmembrane protein, was relatively poor, we confirmed that the ACE dimer (520 kDa) could only be detected in SDS- and DTT-free conditions whereas under denaturating conditions (+SDS; +95°C), the lower molecular mass band (monomeric ACE) was exclusively detectable. Moreover, the treatment of the endothelial cells with trypsin to cleave ACE at the C-terminal stalk region (Woodman et al., 2000) resulted in the loss of both the 520- and the 270-kDa signals. It is noteworthy that the stimulation of endothelial cells with ramiprilat (100 µM; 7 min) enhanced the relative amount of 520-kDa ACE by 137 ± 11% (n = 5; p < 0.05) compared with the situation in unstimulated cells (Fig. 1a).

View larger version (35K): [in this window] [in a new window]   Fig. 1. Dimerization of ACE under basal conditions and in the presence of ramiprilat. a, representative Western blots showing somatic ACE monomers (M) and dimers (D) in lysates of ACE-overexpressing porcine aortic endothelial cells treated with either solvent (S) or ramiprilat (R; 100 nM; 7 min) (left side) or ACE overexpressing (+ACE) and ACE-deficient (-ACE) porcine aortic endothelial cells (right side). Experiments were performed under nonreducing conditions (-SDS-DTT). As a control, lysates were treated with trypsin (10 U/ml; 1 h) before adding nonreducing sample buffer or were heated for 10 min in Laemmli SDS sample buffer (+SDS +95°C) before application. b, serial dilutions of JD53 cells coexpressing ACE-Nub, ACE-Nug, or ACE-Nua and ACE-Cub were spotted on plates lacking Trp/His (CTL), or uracil and 5-FOA (ura- 5-FAO-), or on plates containing uracil and 5-FOA (ura+ 5-FOA+). Interaction between ACE monomers is indicated by cell growth on ura+ 5-FOA+ plates and growth inhibition on ura- 5-FAO- plates and vice versa. Similar results were obtained in at least three additional experiments.

To demonstrate homodimer formation of ACE in living cells, we used the split-ubiquitin system. In brief, we used a ura auxotrophic yeast strain (JD53) lacking orotidylate decarboxylase (Ura3p) (i.e., the enzyme catalyzing the final step of UMP biosynthesis and also converting 5-FOA to the toxic metabolite 5-fluorouracil) (Wittke et al., 1999). The interaction assay is based on the reassembly of the N- and C-terminal halves, N_ub and C_ub, respectively, of Ub that are fused to the cytosolic portions of full-length human ACE (i.e., ACE-N_ub and ACE-C_ub). Uracil prototrophy is reconstituted through the fusion of a modified version of Ura3p, containing an extra N-terminal residue of arginine (RUra3p), to the C-terminal half of ACE-C_ub, thereby generating ACE-C_ub- RUra3p. Dimerization-induced apposition of N_ub and C_ub halves in cotransformants creates a quasi-native Ub that is recognized by ubiquitin-specific proteases splitting off the C-terminally attached reporter RUra3p. Because of its extra arginine residue at the N terminus, released RUra3p is subject to rapid degradation by proteases of the N-end rule. In this way, the split-Ub system provides a direct readout of the N_ub-C_ub association state; i.e., the nonassociated Ub halves retain RUra3p activity, which allows cells to grow on uracildeficient (ura^-) medium and inhibits growth on ura⁺ plates containing 5-FOA, because of the accumulation of toxic 5-fluorouracil, whereas the associated Ub halves produced an inverse growth pattern.

Transformed JD53 cells expressed both ACE-N_ub and ACE-C_ub proteins, as determined by Western blotting (data not shown). Cells coexpressing ACE-N_ub and ACE-C_ub were unable to grow on ura^- plates but expanded on 5-FOA plates, indicating that ACE forms stable dimers (and/or higher oligomers) in intact cells (Fig. 1b). By contrast, cells expressing ACE-C_ub and an unrelated construct in which the ubiquitinconjugating enzyme Ubc6 was fused to the N-terminal portion of N_ub grew on ura^- plates but not on ura⁺, 5-FOA⁺ plates, indicating that ACE and Ubc6 do not interact under these conditions. To enhance the discriminating power of the assay system, we also used two N_ub mutants containing alanine (N_ua) or glycine (N_ug) in position 13, which have a lower affinity for C_ub than wild-type N_ub. Using these modified probes, we observed a robust interaction between ACE-N_ua or ACE-N_ug and ACE-C_ub, confirming that ACE dimerizes/oligomerizes in living cells (Fig. 1b).

Effect of Ramiprilat on the Dimerization of ACE in Endothelial Cells. To determine the effects of the ACE inhibitor ramiprilat on the dimerization of ACE, cells were stimulated with the ACE inhibitor and then treated with the chemical cross-linkers DSS or BS₃ before ACE dimerization was assessed by SDS-PAGE and Western blotting.

Ramiprilat elicited a rapid (within 2 min) 2-fold increase in the dimerization of ACE, which was maintained for up to 60 min (Fig. 2a). In parallel experiments, the phosphorylation of ACE on Ser1270 was monitored, and, as reported previously (Kohlstedt et al., 2004), the ramiprilat-induced phosphorylation of ACE was maximal 2 min after the addition of the inhibitor but declined back to baseline over the next 10 to 15 min (Fig. 2b). The phosphorylation of ACE increases again after 24 to 48 h of ACE-inhibitor treatment (Kohlstedt et al., 2004), and this was paralleled by an increased dimerization of ACE (data not shown). A similar ramiprilat-induced dimerization of ACE was observed using primary cultures of human umbilical vein endothelial cells and was also initiated by other ACE inhibitors (i.e., enalaprilat, quinaprilat, and captopril, albeit with varying efficacies) (Fig. 2c). Primary cultures of human umbilical vein endothelial cells endogenously express ACE, and the fact that the ACE inhibitor-induced dimerization of ACE could be detected in these cells indicates that the effects observed cannot be attributed to an artifact related to the overexpression system.

View larger version (41K): [in this window] [in a new window]   Fig. 2. Effect of inhibitors on ACE dimerization and phosphorylation in endothelial cells. Time course (2-60 min) of the ramiprilat (100 nM)-induced dimerization (assessed after chemical cross-linking) (a) and phosphorylation of ACE on Ser1270 (p-Ser1270) (b) in porcine aortic endothelial cells stably overexpressing human somatic ACE. Results were quantified by calculating the D/M ratio and comparing changes relative to the ratio detected in unstimulated endothelial cells (CTL) or by calculating the p-Ser1270ACE/ACE signal ratio by densitometry and comparing changes relative to the signal obtained in CTL cells. The bar graphs summarize the results of four to six independent experiments; *, p < 0.05; **, p < 0.01 versus CTL. c, effect of ramiprilat (Rami) and enalaprilat (Enala; each 100 nM; 2 and 7 min) on ACE dimer formation in primary cultures of human umbilical vein endothelial cells, after cross-linking with BS3. The blots were reprobed with CD31 to demonstrate equal loading of each lane. Identical results were obtained in four to six additional experiments. d, effect of Rami (100 nM; 2 and 7 min) on ACE dimerization in porcine aortic endothelial cells stably overexpressing the nonphosphorylatable ACE mutant S1270A. Porcine aortic endothelial cells stably overexpressing human ACE were used as a positive control (+ve CTL).

View larger version (63K): [in this window] [in a new window]   Fig. 3. Concentration-dependent effect of different ACE inhibitors on ACE dimerization. Representative Western blots showing the concentration-dependent effects of ramiprilat, quinaprilat, captopril, and enalaprilat (0.1 nM-10 µM; 2 min) on ACE dimerization, assessed after chemical cross-linking. Similar results were obtained in three additional experiments.

Effect of Carbohydrates and Antibodies on ACE Dimerization and Shedding. We next assessed the role of the putative carbohydrate recognition domain (Kost et al., 2003), a lectin-like structure residing within the N-terminal portion of somatic ACE, in enzyme dimerization. Because ACE dimer formation in vitro in inverse micelles has been reported to be inhibited by galactose and other carbohydrates (Kost et al., 2000), we incubated ACE-overexpressing porcine aortic endothelial cells with 10 µM galactose, 10 µM glucose, or 10 µM mannitol (as an osmotic control) for 30 min, before the addition of 100 nM ramiprilat for 2 min. The basal or ramiprilat-induced dimerization of ACE in endothelial cells was not influenced by either galactose, glucose, or mannitol (Fig. 4). Furthermore, the pretreatment of cells with the monosaccharides did not affect either the basal or the ramiprilat-induced phosphorylation of ACE on Ser1270 (data not shown). The carbohydrate recognition domain that was suggested to be important for ACE dimerization is recognized and thus shielded by the monoclonal antibody 9B9 (Kost et al., 2003). We therefore determined the effect of different monoclonal ACE antibodies (9B9, 3A5, 5F1, and C78) directed to different epitopes within the N domain (Danilov et al., 1994) on ACE dimer formation. The antibodies tested had no acute effect on ACE dimerization per se (Fig. 5a). However, the 9B9 and 3A5 antibodies (but not 5F1) significantly enhanced the amount of ACE recovered from the endothelial cell supernatant (Fig. 5b), reflecting the enhanced cleavage/secretion of the enzyme. None of the antibodies tested were able to prevent the ramiprilat-induced dimerization of ACE in ACE-overexpressing porcine aortic endothelial cells (Fig. 5c), excluding the previously speculated correlation between ACE dimerization and shedding (Kost et al., 2003).

View larger version (61K): [in this window] [in a new window]   Fig. 4. Effect of carbohydrates on the dimerization of ACE in ACE-overexpressing endothelial cells. a, representative Western blots showing the effect of pretreatment with galactose (Gal), glucose (Glc), or mannitol (Man) (each 10 µM; 30 min) on ramiprilat (R; 100 nM; 2 min)-induced ACE dimerization. The results were quantified by calculating the crosslinked D/M ratio and comparing changes relative to the ratio detected in unstimulated endothelial cells (S, solvent). The bar graphs summarize results obtained in four to six different experiments; *, p < 0.05; **, p < 0.01 versus CTL.

View larger version (31K): [in this window] [in a new window]   Fig. 5. Effect of antibodies on ACE dimerization and cleavage. a, effect of solvent (S), ramiprilat (R; 100 nM; 7 min), monoclonal antibodies to the ACE N domain C78, 9B9, 3A5, and 5F1 (each 10 µg/ml; 7 min) and an unrelated monoclonal antibody (mAb; anti-caveolin-1), on ACE dimerization in overexpressing porcine aortic endothelial cells. b, effect of the 9B9, 3A5, and 5F1 antibodies (each 10 µg/ml; 4 h) on the amount of soluble ACE (sACE) recovered by immunoprecipitation from the supernatant of ACE-overexpressing porcine aortic endothelial cells. c, effect of pretreatment with solvent and antibodies 9B9, 3A5, and 5F1 (each 10 µg/ml; 4 h) on the R (100 nM; 2 min)-induced dimerization of ACE, assessed after chemical cross-linking. The bar graphs summarize the results of three to six independent experiments; *, p < 0.05; **, p < 0.01 versus the respective buffer alone.

Mutation of the glutamyl residues within the C and/or N domains (N_Glu, C_Glu, and N+C_Glu) was without effect on the ramiprilat-induced dimerization of ACE or on its phosphorylation on Ser1270 (data not shown). Replacement of the two histidyl residues within the C domain of ACE (C_His) attenuated the basal formation of ACE dimers as well as Ser1270 phosphorylation of the ACE mutant, whereas the N_His mutant showed unchanged dimerization status and increased basal Ser1270 phosphorylation (Fig. 6, a and b). Moreover, although the ramiprilat-induced dimerization (Fig. 6a) and phosphorylation of ACE (Fig. 6b) were apparent in cells expressing the wild-type ACE or the N_His mutant, neither was observed in CHO cells expressing the C_His ACE mutant. Consistent with these effects, ramiprilat was also unable to elicit the activation of JNK in C_His-expressing cells (Fig. 7). The effects of ramiprilat treatment on cells expressing wildtype ACE were similar to those observed in cells expressing N_His, or any of the Glu mutants (N_Glu,C_Glu, and N+C_Glu; Fig. 7), indicating that ramiprilat was able to elicit signaling as long as the two His residues mediating Zn²⁺ binding of the active site in the C domain were present.

View larger version (43K): [in this window] [in a new window]   Fig. 6. Effect of ramiprilat on dimerization and phosphorylation of NHis and CHis mutants. Representative Western blots showing the effect of solvent (S) and ramiprilat (R; 100 nM; 2 min) on ACE dimerization (a) and phosphorylation (b) at Ser1270 (p-Ser1270) in CHO cells overexpressing human somatic ACE (wtACE) or NHis and CHis mutants. and D forms of ACE are M indicated. The bar graphs summarize results obtained in four to six independent experiments; **, p < 0.01 versus basal wild-type ACE dimerization or phosphorylation.

View larger version (48K): [in this window] [in a new window]   Fig. 7. Effect of ramiprilat on signaling by ACE mutants. Autoradiography of 32P-labeled c-Jun and statistical analysis showing the effect of solvent (S) and ramiprilat (R; 100 nM; 7 min) on the activity of JNK immunoprecipitated from CHO cells overexpressing human somatic ACE (wtACE) or one each of the NHis, CHis, NGlu, CGlu, or N + CGlu mutants. To ensure equal immunoprecipitation of JNK, blots were reprobed with an antibody against JNK. The bar graph summarize the results obtained in four independent experiments; *, p < 0.05 versus respective solvent; , p < 0.05; , p <0.01 versus basal JNK activity in cells expressing wild-type ACE.

	Discussion

Top Abstract Materials and Methods Results Discussion References

Mature somatic ACE exists at the endothelial cell surface and has a calculated molecular mass of 146.6 kDa; however, the enzyme is highly glycosylated, and the molecular mass of the enzyme is between 170 and 180 kDa when assessed by denaturating polyacrylamide gel electrophoresis (Das and Soffer, 1975). Other forms of human somatic ACE exist, such as an immature form found in intracellular compartments (170 kDa) and a soluble form (175 kDa) found in the plasma representing a C-terminally truncated form of the enzyme that is generated by the proteolytic cleavage of its stalk region and the subsequent release from the plasma membrane (Alhenc-Gelas et al., 1989). Aggregates of ACE that might reflect enzyme dimers and/or higher oligomers have been found during purification of ACE from human lung (Nishimura et al., 1977) as well as in in vitro experiments by using inverse micelles and ACE-expressing COS cells (Kost et al., 2000, 2003).

Using native gel electrophoresis, we found that the ACE endogenously expressed in endothelial cells migrates not only as monomers but also as dimers. The treatment of cell lysates with SDS, to denature ACE dimers, resulted in failure to detect the larger protein, whereas the use of trypsin, which cleaves ACE at the C-terminal stalk region (Woodman et al., 2000), resulted in failure to detect the enzyme at all. The ability of ACE to homodimerize was also demonstrated using the split-ubiquitin system in yeast as well as by chemical cross-linking in primary cultures of human endothelial cells that endogenously express ACE. Moreover, the data obtained with the split ubiquitin assay indicate that ACE dimerization is greater in living cells and a more pronounced phenomenon than the in vitro experiments would suggest. The relatively low dimer levels detected in the cross-linking experiments can probably be attributed to methodological problems related to the detection of the dimeric form of ACE by the antibodies used. Similar problems have been described when assessing the dimer (D)/monomer (M) ratio of the endothelial nitric-oxide synthase by low-temperature SDS-PAGE (Zou et al., 2002).

Given the existence of ACE dimers, we next determined whether the binding of an ACE inhibitor to ACE altered the dimer/monomer ratio in endothelial cells. Our results showed that the treatment of endothelial cells with ramiprilat increased the dimer/monomer ratio and that the formation of ACE dimers correlated temporally with the onset of the inhibitor-induced phosphorylation of the enzyme at Ser1270. The effects described do not represent a transient cellular response, because the ACE inhibitor-induced phosphorylation of ACE is biphasic and consists of a transient peak after 2 to 7 min followed by maintained elevation in phosphorylation after 6 to 48 h (Kohlstedt et al., 2004). The activation of this signaling cascade has previously been linked to changes in endothelial cell gene expression and the enhanced production of the vasoprotective autacoid prostacyclin (Kohlstedt et al., 2005). The ability to induce ACE dimer formation was not restricted to ramiprilat and was also observed after cell stimulation with enalaprilat, quinaprilat, and captopril, indicating that this is an effect of this class of compounds. Dimer formation was observed in an endothelial cell line that overexpressed human somatic ACE as well as in primary cultures of human umbilical vein endothelial cells that endogenously express ACE, indicating that the effects described cannot be simply attributed to an artifact associated with the overexpression system.

Somatic ACE contains 17 potential sites for N-glycosylation, mainly of the complex type (Das and Soffer, 1975). Because ACE dimerization has been attributed to interactions between its carbohydrate side chains and a carbohydrate recognition domain at its N domain that are sensitive to treatment with galactose (Kost et al., 2000), we tested the effects of galactose, glucose, or mannitol on either the basal or the ramiprilat-induced dimerization of ACE, but we were unable to find any significant interference with ACE dimerization. It has also been suggested that monoclonal antibody 9B9 directed to the putative carbohydrate recognition domain region can interfere with ACE self-association (Kost et al., 2003). We therefore assessed the effect of various antibodies directed to the N domain of ACE, on dimer formation. The monoclonal antibodies tested had no effect on the basal or ramiprilat-induced dimerization of ACE but exerted differential effects on the cleavage of the enzyme. The latter effect has been reported previously (Kost et al., 2003) and linked to the antibody-mediated inhibition of dimerization. However, although we clearly observed antibody-induced ACE shedding, we were unable to detect any link between shedding and dimerization. To determine whether the dimerization of ACE was linked to the recently described ACE signaling pathway, which involves the ACE inhibitor-induced phosphorylation of the enzyme on Ser1270 as well as the activation of the JNK pathway (Kohlstedt et al., 2002, 2004), we compared the effects of ramiprilat on the dimerization of the wild-type ACE as well as of the S1270A ACE mutant. Ramiprilat elicited dimerization of both the wildtype and the S1270A ACE, indicating that the phosphorylation status of the cytoplasmic tail of the enzyme does not influence dimer formation.

Other type I membrane glycoproteins, such as the -secretase (-site amyloid precursor protein cleaving enzyme) are also able to dimerize and thus regulate their intracellular and/or extracellular functions (Schmechel et al., 2004). Like ACE, -site amyloid precursor protein cleaving enzyme can be phosphorylated on a Ser residue near its C terminus and can also be cleaved from the cell surface (Capell et al., 2000). We therefore determined whether interfering with the dimerization of ACE affected the ACE inhibitor-induced phosphorylation of the enzyme and/or the subsequent activation of JNK. Analysis of different C- or N-domain inactive ACE mutants overexpressed in CHO cells revealed that the C domain rather than the N domain of ACE seems to be implicated in its dimerization, inasmuch as inactivation of the C domain active site via mutation of the two essential Zn²⁺-complexing histidyl residues of the HEMGH consensus sequence rendered the enzyme unable to dimerize either under basal conditions or in response to ramiprilat. Mutation within the C domain also resulted in the loss of inhibitor-induced ACE signaling; i.e., we failed to observe a ramiprilat-induced increase in Ser1270 phosphorylation or activation of the downstream JNK pathway. Both aspects of the ACE signaling pathway remained intact after mutation of the N domain. At present, we do not know how ACE inhibitors can effect ACE dimerization via the C domain; however, it is possible that their binding induces conformational changes that eventually result in the exposure of a dimerization domain, as has been reported for the ligand-induced dimerization of the epidermal growth factor receptor (Ferguson et al., 2003).

Although the N and C domains of ACE arose as a consequence of gene duplication (Lattion et al., 1989), there is at least circumstantial evidence suggesting that there are marked differences in the function of the two sites. For example, although both the N and C domains contribute to the degradation of bradykinin, angiotensin I conversion takes place preferentially within the C domain, and selective C domain inhibition is sufficient to prevent angiotensin I-induced vasoconstriction (van Esch et al., 2005). There is also evidence suggesting that the N domain of ACE may be functionally less important, because the RXP407 peptide that specifically inhibits the ACE N domain active site has no effect on blood pressure (Junot et al., 2001).

Together, the results of the present investigation suggest that the ACE inhibitor-induced dimerization of ACE, via the C domain of the enzyme, represents the initial step in the ACE signaling pathway that involves the activation of the JNK/c-Jun pathway and leads to changes in endothelial cell gene expression.

	Acknowledgements

 
We are indebted to Marie von Reutern and Isabel Winter for expert technical assistance.

	Footnotes

 
The experimental work described in this report was supported by the Deutsche Forschungsgemeinschaft (FL 364/1-2 and SFB 642, TP5), the Heinrich and Fritz Riese-Stiftung, a research grant from Sanofi-Aventis, and by the European Vascular Genomics Network, a Network of Excellence supported by the European Community's Sixth Framework Program (contract LSHM-CT-2003-503254).

ABBREVIATIONS: ACE, angiotensin-converting enzyme; JNK, c-Jun NH₂-terminal kinase; CHO, Chinese hamster ovary; PAGE, polyacrylamide gel electrophoresis; DTT, dithiothreitol; DSS, disuccinimidyl suberate; BS₃, bis(sulfosuccinimidyl)suberate; ura, uracil; 5-FOA, fluoroorotic acid; Ub, ubiquitin; CTL, control; D, dimer; M, monomer.

Address correspondence to: Dr. Ingrid Fleming, Vascular Signalling Group, Institut für Kardiovaskuläre Physiologie, Johann Wolfgang Goethe-Universität, Theodor-Stern-Kai 7, D-60590 Frankfurt am Main, Germany. E-mail: fleming{at}em.uni-frankfurt.de

	References

Top Abstract Materials and Methods Results Discussion References

 
Alhenc-Gelas F, Soubrier F, Hubert C, Allegrini J, and Corvol P (1989) The peculiar characteristics of the amino acid sequence of angiotensin I-converting enzyme, as determined by cDNA cloning of the human endothelial enzyme. J Cardiovasc Pharmacol 14: S6-S9.

Benzing T, Fleming I, Blaukat A, Muller-Esterl W, and Busse R (1999) Angiotensin-converting enzyme inhibitor ramiprilat interferes with the sequestration of the B₂ kinin receptor within the plasma membrane of native endothelial cells. Circulation 99: 2034-2040.[Abstract/Free Full Text]

Bernstein KE, Xiao HD, Frenzel K, Li P, Shen XZ, Adams JW, and Fuchs S (2005) Six truisms concerning ACE and the renin-angiotensin system educed from the genetic analysis of mice. Circ Res 96: 1135-1144.[Free Full Text]

Busse R and Lamontagne D (1991) Endothelium-derived bradykinin is responsible for the increase in calcium produced by angiotensin-converting enzyme inhibitors in human endothelial cells. Naunyn-Schmiedeberg's Arch Pharmacol 344: 126-129.[Medline]

Capell A, Steiner H, Willem M, Kaiser H, Meyer C, Walter J, Lammich S, Multhaup G, and Haass C (2000) Maturation and pro-peptide cleavage of beta-secretase. J Biol Chem 275: 30849-30854.[Abstract/Free Full Text]

Clozel J-P, Hess P, Michael C, Schietinger K, and Baumgartner HR (1991) Inhibition of converting enzyme and neointima formation after vascular injury in rabbits and guinea pigs. Hypertension 18: II55-II59.[Medline]

Danilov S, Jaspard E, Churakova T, Towbin H, Savoie F, Wei L, and Alhenc-Gelas F (1994) Structure-function analysis of angiotensin I-converting enzyme using monoclonal antibodies. Selective inhibition of the amino-terminal active site. J Biol Chem 269: 26806-26814.[Abstract/Free Full Text]

Das M and Soffer RL (1975) Pulmonary angiotensin-converting enzyme. Structural and catalytic properties. J Biol Chem 250: 6762-6768.[Abstract/Free Full Text]

Eckert JH and Johnsson N (2003) Pex10p links the ubiquitin conjugating enzyme Pex4p to the protein import machinery of the peroxisome. J Cell Sci 116: 3623-3634.[Abstract/Free Full Text]

Ferguson KM, Berger MB, Mendrola JM, Cho HS, Leahy DJ, and Lemmon MA (2003) EGF activates its receptor by removing interactions that autoinhibit ectodo-main dimerization. Mol Cell 11: 507-517.[CrossRef][Medline]

Heart Outcomes Prevention Evaluation Study Investigators (2000) Effects of ramipril on cardiovascular and microvascular outcomes in people with diabetes mellitus: results of the HOPE study and MICRO-HOPE substudy. Heart Outcomes Prevention Evaluation Study Investigators. Lancet 355: 253-259.[CrossRef][Medline]

Hubert C, Houot AM, Corvol P, and Soubrier F (1991) Structure of the angiotensin I-converting enzyme gene. Two alternate promoters correspond to evolutionary steps of a duplicated gene. J Biol Chem 266: 15377-15383.[Abstract/Free Full Text]

Johnsson N and Varshavsky A (1994) Split ubiquitin as a sensor of protein interactions in vivo. Proc Natl Acad Sci USA 91: 10340-10344.[Abstract/Free Full Text]

Junot C, Gonzales MF, Ezan E, Cotton J, Vazeux G, Michaud A, Azizi M, Vassiliou S, Yiotakis A, Corvol P, et al. (2001) RXP 407, a selective inhibitor of the N-domain of angiotensin I-converting enzyme, blocks in vivo the degradation of hemoregulatory peptide acetyl-Ser-Asp-Lys-Pro with no effect on angiotensin I hydrolysis. J Pharmacol Exp Ther 297: 606-611.[Abstract/Free Full Text]

Kohlstedt K, Brandes RP, Muller-Esterl W, Busse R, and Fleming I (2004) Angiotensin-converting enzyme is involved in outside-in signaling in endothelial cells. Circ Res 94: 60-67.[Abstract/Free Full Text]

Kohlstedt K, Busse R, and Fleming I (2005) Signaling via the angiotensin-converting enzyme enhances the expression of cyclooxygenase-2 in endothelial cells. Hypertension 45: 126-132.[Abstract/Free Full Text]

Kohlstedt K, Shoghi F, Muller-Esterl W, Busse R, and Fleming I (2002) CK2 phosphorylates the angiotensin-converting enzyme and regulates its retention in the endothelial cell plasma membrane. Circ Res 91: 749-756.[Abstract/Free Full Text]

Kost OA, Balyasnikova IV, Chemodanova EE, Nikolskaya II, Albrecht RF, and Danilov SM (2003) Epitope-dependent blocking of the angiotensin-converting enzyme dimerization by monoclonal antibodies to the N-terminal domain of ACE: possible link of ACE dimerization and shedding from the cell surface. Biochemistry 42: 6965-6976.[CrossRef][Medline]

Kost OA, Bovin NV, Chemodanova EE, Nasonov VV, and Orth TA (2000) New feature of angiotensin-converting enzyme: carbohydrate-recognizing domain. J Mol Recognit 13: 360-369.[CrossRef][Medline]

Lattion AL, Soubrier F, Allegrini J, Hubert C, Corvol P, and Alhenc-Gelas F (1989) The testicular transcript of the angiotensin I-converting enzyme encodes for the ancestral, non-duplicated form of the enzyme. FEBS Lett 252: 99-104.[CrossRef][Medline]

Lee AF, MacFadyen RJ, and Struthers AD (1999) Neurohormonal reactivation in heart failure patients on chronic ACE inhibitor therapy: a longitudinal study. Eur J Heart Fail 1: 401-406.[CrossRef][Medline]

Mancini GB, Henry GC, Macaya C, O'Neill BJ, Pucillo AL, Carere RG, Wargovich TJ, Mudra H, Luscher TF, Klibaner MI, et al. (1996) Angiotensin-converting enzyme inhibition with quinapril improves endothelial vasomotor dysfunction in patients with coronary artery disease. The TREND (Trial on Reversing ENdothelial Dysfunction) Study. Circulation 94: 258-265.[Abstract/Free Full Text]

Morishita R, Gibbons GH, Ellison KE, Lee W, Zhang L, Yu H, Kaneda Y, Ogihara T, and Dzau VJ (1994) Evidence for direct local effect of angiotensin in vascular hypertrophy. In vivo gene transfer of angiotensin converting enzyme. J Clin Investig 94: 978-984.[Medline]

Morishita R, Gibbons GH, Tomita N, Zhang L, Kaneda Y, Ogihara T, and Dzau VJ (2000) Antisense oligodeoxynucleotide inhibition of vascular angiotensin-converting enzyme expression attenuates neointimal formation: evidence for tissue angiotensin-converting enzyme function. Arterioscler Thromb Vasc Biol 20: 915-922.[Abstract/Free Full Text]

Nishimura K, Yoshida N, Hiwada K, Ueda E, and Kokubu T (1977) Purification of angiotensin I-converting enzyme from human lung. Biochim Biophys Acta 483: 398-408.[Medline]

Petrie MC, Padmanabhan N, McDonald JE, Hillier C, Connell JM, and McMurray JJ (2001) Angiotensin converting enzyme (ACE) and non-ACE dependent angiotensin II generation in resistance arteries from patients with heart failure and coronary heart disease. J Am Coll Cardiol 37: 1056-1061.[Abstract/Free Full Text]

Schartl M, Bocksch WG, Dreysse S, Beckmann S, Franke O, and Hunten U (1994) Remodeling of myocardium and arteries by chronic angiotensin converting enzyme inhibition in hypertensive patients. J Hypertens Suppl 12: S37-S42.[Medline]

Schmechel A, Strauss M, Schlicksupp A, Pipkorn R, Haass C, Bayer TA, and Multhaup G (2004) Human BACE forms dimers and colocalizes with APP. J Biol Chem 279: 39710-39717.[Abstract/Free Full Text]

Soubrier F, Alhenc-Gelas F, Hubert C, Allegrini J, John M, Tregear G, and Corvol P (1988) Two putative active centers in human angiotensin I-converting enzyme revealed by molecular cloning. Proc Natl Acad Sci USA 85: 9386-9390.[Abstract/Free Full Text]

van Esch JHM, Tom B, Dive V, Batenburg WW, Georgiadis D, Yiotakis A, van Gool JMG, de Bruijn RJA, de Vries R, and Danser AHJ (2005) Selective angiotensin-converting enzyme C-domain inhibition is sufficient to prevent angiotensin I-induced vasoconstriction. Hypertension 45: 120-125.[Abstract/Free Full Text]

Wei L, Alhenc-Gelas F, Corvol P, and Clauser E (1991) The two homologous domains of human angiotensin I-converting enzyme are both catalytically active. J Biol Chem 266: 9002-9008.[Abstract/Free Full Text]

Wiemer G, Scholkens BA, Becker RH, and Busse R (1991) Ramiprilat enhances endothelial autacoid formation by inhibiting breakdown of endothelium-derived bradykinin. Hypertension 18: 558-563.[Abstract/Free Full Text]

Wittke S, Lewke N, Muller S, and Johnsson N (1999) Probing the molecular environment of membrane proteins in vivo. Mol Biol Cell 10: 2519-2530.[Abstract/Free Full Text]

Woodman ZL, Oppong SY, Cook S, Hooper NM, Schwager SL, Brandt WF, Ehlers MR, and Sturrock ED (2000) Shedding of somatic angiotensin-converting enzyme (ACE) is inefficient compared with testis ACE despite cleavage at identical stalk sites. Biochem J 347: 711-718.

Xiao HD, Fuchs S, Campbell DJ, Lewis W, Dudley SC Jr, Kasi VS, Hoit BD, Keshelava G, Zhao H, Capecchi MR, et al. (2004) Mice with cardiac-restricted angiotensin-converting enzyme (ACE) have atrial enlargement, cardiac arrhythmia and sudden death. Am J Pathol 165: 1019-1032.[Abstract/Free Full Text]

Zou MH, Shi C, and Cohen RA (2002) Oxidation of the zinc-thiolate complex and uncoupling of endothelial nitric oxide synthase by peroxynitrite. J Clin Investig 109: 817-826.[CrossRef][Medline]

This article has been cited by other articles:

				K. Kohlstedt, C. Gershome, C. Trouvain, W.-K. Hofmann, S. Fichtlscherer, and I. Fleming Angiotensin-Converting Enzyme (ACE) Inhibitors Modulate Cellular Retinol-Binding Protein 1 and Adiponectin Expression in Adipocytes via the ACE-Dependent Signaling Cascade Mol. Pharmacol., March 1, 2009; 75(3): 685 - 692. [Abstract] [Full Text] [PDF]

				P.-a. B. Shih and D. T. O'Connor Hereditary Determinants of Human Hypertension: Strategies in the Setting of Genetic Complexity Hypertension, June 1, 2008; 51(6): 1456 - 1464. [Full Text] [PDF]

eLetters:

This Article Abstract Full Text (PDF) All Versions of this Article: mol.105.020636v1 69/5/1725    most recent Submit a response View responses Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kohlstedt, K. Articles by Fleming, I. Search for Related Content PubMed PubMed Citation Articles by Kohlstedt, K. Articles by Fleming, I.