ACEH/ACE2 is a novel mammalian metallocarboxypeptidase and a homologue of angiotensin-converting enzyme insensitive to ACE inhibitors

ACEH/ACE2 is a novel mammalianmetallocarboxypeptidase and a homologue ofangiotensin-converting enzyme insensitive to ACEinhibitors1

Anthony J. Turner, Sarah R. Tipnis, Jodie L. Guy, Gillian I. Rice, andNigel M. Hooper

Abstract: A human zinc metalloprotease (termed ACEH or ACE2) with considerable homology to angiotensin-converting enzyme (ACE) (EC 3.4.15.1) has been identified and subsequently cloned and functionally expressed. Thetranslated protein contains an N-terminal signal sequence, a single catalytic domain with zinc-binding motif (HEMGH),a transmembrane region, and a small C-terminal cytosolic domain. Unlike somatic ACE, ACEH functions as acarboxypeptidase when acting on angiotensin I and angiotensin II or other peptide substrates. ACEH may function inconjunction with ACE and neprilysin in novel pathways of angiotensin metabolism of physiological significance. Incontrast with ACE, ACEH does not hydrolyse bradykinin and is not inhibited by typical ACE inhibitors. ACEH isunique among mammalian carboxypeptidases in containing an HEXXH zinc motif but, in this respect, resembles a bac-terial enzyme, Thermus aquaticus (Taq) carboxypeptidase (EC 3.4.17.19). Collectrin, a developmentally regulated renalprotein, is homologous with the C-terminal region of ACEH but has no similarity with ACE and no catalytic domain.Thus, the ACEH protein may have evolved as a chimera of a single ACE-like domain and a collectrin domain. Thecollectrin domain may regulate tissue response to injury whereas the catalytic domain is involved in peptide processingevents.

Key words: ACEH, ACE2, metalloprotease, collectrin, carboxypeptidase, angiotensin II.

Résumé : Une métalloprotéase à zinc humaine (nommée ECAH ou ECA2) ayant une forte homologie avec l’enzymede conversion de l’angiotensine (ECA; EC 3.4.15.1) a été identifiée, puis clonée et fonctionnellement exprimée. Laprotéine traduite contient une séquence signal à son extrémité N-terminale, un domaine catalytique comportant un motifde fixation du zinc (HEMGH), une région transmembranaire et un petit domaine cytosolique à son extrémitéC-terminale. Contrairement à l’ECA somatique, l’ECAH fonctionne comme une carboxypeptidase lorsqu’elle agit surl’angiotensine I et l’angiotensine II, ou sur d’autres substrats peptidiques. L’ECAH pourrait agir conjointement avecl’ECA et la néprilysine (NEP) dans de nouvelles voies du métabolisme de l’angiotensine d’importance physiologique.Contrairement à l’ECA, l’ECAH n’hydrolyse pas la bradykinine et n’est pas inhibée par les inhibiteurs typiques del’ECA. L’ECAH est la seule carboxypeptidase de mammifère comportant un motif zinc HEXXH; toutefois, sur cepoint, elle ressemble à l’enzyme bactérienne Thermus aquaticus (Taq) carboxypeptidase (EC 3.4.17.19). La collectrine,une protéine rénale régulée au cours du développement, a une homologie avec l’extrémité C-terminale de l’ECAH,mais n’a aucune similarité avec l’ECA et aucun domaine catalytique. Ainsi, la protéine ECAH pourrait avoir évoluésous la forme d’une chimère d’un domaine de type ECA et d’un domaine de la collectrine. Le domaine de lacollectrine pourrait réguler la réponse tissulaire à une lésion alors que le domaine catalytique intervient dans la matura-tion peptidique.

Mots clés : ECAH, ECA2, métalloprotéase, collectrine, carboxypeptidase, angiotensine II.

[Traduit par la Rédaction] 353

Turner et al.

Can. J. Physiol. Pharmacol. 80: 346–353 (2002) DOI: 10.1139/Y02-021 © 2002 NRC Canada

346

Received 20 August 2001. Published on the NRC Research Press Web site at http://cjpp.nrc.ca on 5 April 2002.

A.J. Turner,2 S.R. Tipnis, J.L. Guy, G.I. Rice, and N.M. Hooper. Proteolysis Research Group, School of Biochemistry andMolecular Biology, University of Leeds, Leeds LS2 9JT, U.K.

1This paper has undergone the Journal’s usual peer review process.2Corresponding author (e-mail: [email protected]).

I:\cjpp\Cjpp-80\80-04\Y02-021.vp28-Mar-2002 1:48:45 PM

Color profile: DisabledComposite Default screen

Renin–angiotensin system and kallikrein–kininsystem

The renin–angiotensin system (RAS) mediates numerouseffects in the cardiovascular system and is believed to oper-ate on two levels, a circulating level and a local tissue level.In humans, the circulating RAS plays a crucial role in regu-lating blood pressure and fluid and electrolyte homeostasis(Campbell 1987). In contrast, the tissue RAS may provide amore chronic, localised influence on regulation of vasculartone or renal, cardiac, neuronal, adrenal, or intestinal func-tion (Johnston 1992). Kidney-derived renin, a proteolytic en-zyme, catalyses the conversion of the large globular proteinangiotensinogen to the decapeptide angiotensin I (Ang I)(Erdös 1975) (Fig. 1). At the circulating level of the RAS,renin represents the rate-limiting step in production of an-giotensin II (Ang II) from Ang I (Peters 1995). However, atthe tissue level of the RAS, it is the activity of another en-zyme, angiotensin-converting enzyme (ACE), that is impor-tant in determining local levels of Ang II. ACE generatesAng II by cleaving the carboxy-terminal dipeptide His-Leufrom Ang I (Ehlers and Riordan 1989).

Ang II is a bioactive peptide with a crucial role in theRAS, mediated by binding to the Ang II type 1 and 2 (AT1and AT2) receptors. The binding of Ang II to the AT1 recep-tor mediates many components of cardiovascular regulationincluding regional blood flow, vascular smooth muscle cellproliferation and migration (Griendling et al. 1993), and regu-lation of local sympathetic activity, pressor, and tachycardicresponses (Dieguez-Lucena et al. 1996). It may also be in-volved in platelet activation and aggregation (Dzau 1994)and maintenance of cardiovascular structure and repair(Fluharty et al. 1995). AT2 is thought to oppose several func-tions mediated by the AT1 receptor (Carey et al. 2000a;Searles and Harrison 1999). The AT2 receptor inhibits cellproliferation, promotes cell differentiation, and may mediateapoptosis, although this remains somewhat controversial(Carey et al. 2000a). Stimulation of the AT2 receptor causesvasodilation and natriuresis by stimulating release of renalbradykinin, which in turn stimulates the endothelial bradykinintype 2 (BK2) receptor resulting in activation of the nitric ox-ide (NO) – cyclic GMP system (Carey et al. 2000a, 2000b;Tsutsumi et al. 1999). NO is continuously released from theendothelium and provides a constant counteracting force tovasoconstrictor substances such as Ang II (Drexler et al.1995). The AT2 receptor also influences the metabolism ofthe vasodilator prostaglandin (Carey et al. 2000a). Thus,there is cross-talk between the AT1 and AT2 receptors in me-diating the physiological effects of Ang II, and the balancebetween vasodilation and vasoconstriction will depend onthe local AT1 and AT2 receptor density.

Angiotensin (1–7) (Ang (1–7)) is the predominant peptidegenerated from Ang I in the vascular endothelium (Brosnihan1998), as opposed to formation of Ang II by ACE. Levels ofAng (1–7) are increased by up to 30% following ACE inhib-itor therapy (Ferrario et al. 1997; Iyer et al. 1998). This isdue to both inhibition of the conversion of Ang I to Ang IIby ACE and subsequent buildup of Ang I and inhibition ofthe conversion of Ang (1–7) to Ang (1–5) by ACE (Allred etal. 2000; Deddish et al. 1998). Ang (1–7) is very short-lived

in the circulation (Chappell et al. 2000; Yamada et al. 1998),and there are low levels of Ang (1–7) in plasma. However,raised levels of Ang (1–7) occur locally, for example, in thekidney and urine (Ferrario et al. 1998). Ang (1–7) stimulatesrelease of prostaglandins and NO (Brosnihan 1998) and po-tentiates the effects of bradykinin (Deddish et al. 1998). Infact, Ang (1–7) may stimulate release of bradykinin andtherefore NO via the BK2 receptor (Brosnihan et al. 1996;Gorelik et al. 1998), but the receptor signalling pathwaysinvolved are not fully understood (Deddish et al. 1998;Fernandes et al. 2001). Evidence suggests that there may bea distinct Ang1–7 receptor (Tallant et al. 1997), but this hasnot been isolated and characterised. Ang (1–7) opposes theactions of Ang II by causing vasodilation (Iyer et al. 1998;Ueda et al. 2000), antiproliferation (Strawn et al. 1999; Tallantet al. 1999), and apoptosis (Chappell et al. 2000). The majorenzyme involved in the formation of Ang (1–7) from Ang Iin vivo is neutral endopeptidase 24.11 (neprilysin) (NEP)(Brosnihan 1998; Deddish et al. 1998). NEP is located in theheart, vascular endothelium, and kidney and is therefore lo-cated in organs where Ang (1–7) is known to be produced(Campbell et al. 1998; Chappell et al. 2000). NEP was im-plicated in cardiovascular regulation when it was shown tohydrolyse atrial natriuretic peptide. Atrial natriuretic peptidehas beneficial diuretic, natriuretic, and vasorelaxant effectsand inhibits renin and aldosterone secretion; therefore, NEPinhibition is beneficial in the treatment of heart disease (Huand Ertl 1999). NEP inhibition increases Ang II and Ang Ilevels in plasma (Campbell et al. 1998) and is also involvedin bradykinin metabolism (Campbell et al. 1998).

ACE is also a pivotal enzyme in the kallikrein–kinin path-way (Fig. 2). The kallikrein–kinin system is an autocrine andparacrine mediator of local metabolic needs for cellular func-tions (Pellacani et al. 1994). Bradykinin causes vasodilationand hypotension by stimulating the production of arachidonicacid metabolites, NO, and endothelium-derived hyperpolarisingfactor in vascular endothelium (Brown and Vaughan 1998),actions that are mediated via BK2 receptors (Pellacani et al.1994). ACE and NEP both cleave bradykinin to form the in-active peptide BK (1–7). The C-terminal arginine of bradykininis removed by carboxypeptidase N (Kaplan and Silverberg1987), otherwise known as kininase I, to formdes-Arg9-bradykinin (BK (1–8)). BK (1–8) is also active,mediating its effects via bradykinin type 1 (BK1) receptors.BK1 receptors are expressed mainly in pathological condi-tions such as tissue injury and are thought to mediate the in-flammatory and pain-producing effects of kinins (Dell’Italiaand Oparil 1999). ACE or NEP subsequently degrades BK(1–8) to inactive peptides (Kaplan and Silverberg 1987).

Thus, it can be seen that the RAS and kallikrein–kininsystems are very complex enzyme cascades with balances andcounterbalances at every step. The balance of the systembetween coronary risk factors (vasoconstriction, cellular pro-liferation, and hypertrophy) and coronary protection(vasodilation, antiproliferation, and apoptosis) will dependon the levels of individual peptides, enzymes, and their in-hibitors in each local tissue system. There may therefore bedifferent balances in different systems depending on thepresence and levels of different components of the RAS.Levels of ACE, NEP, and any other enzymes involved in the

© 2002 NRC Canada

Turner et al. 347



RAS may influence the balance, as will AT1, AT2, and Ang1–7receptor expression.

Discovery of a homologue of ACE

The most recently reported member of the ACE family,which may prove to be a novel player in the RAS, is theACE homologue, designated ACEH or ACE2. The identifi-cation of human ACEH was first reported in 2000 by two in-dependent groups: ourselves (Tipnis et al. 2000) andDonoghue et al. (2000). We originally identified the cDNAencoding ACEH as a novel zinc metalloprotease, in a propri-etary EST database, as it contains a HEXXH zinc-bindingmotif and has high homology to ACE around this region.Subsequently, a partial ACEH cDNA was isolated from ahuman lymphoma cDNA library. This was then used to ob-tain the full-length cDNA sequence from a human kidney li-brary. Donoghue et al. (2000) isolated their ACE2 cDNA ina comparable fashion from a human heart failure ventricularlibrary. Northern blotting studies have shown the highest lev-els of ACEH mRNA expression to be in kidney, heart, andtestis (Fig. 3A). It is of interest to note that the cDNA librar-ies in which ACEH has been identified to date have been de-rived from these tissues. Immunoblotting analysis, using anantipeptide antibody raised against amino acids 19–32 of the

published sequence, shows that ACEH protein is expressedin human kidney microvillar membranes (Fig. 3B, lane 2), arich source of membrane peptidases. It should be noted thatthis antibody does not cross-react with purified human kid-ney ACE (Fig. 3B, lane 1). Immunohistochemical studieshave shown that ACEH is localised to the endothelium ofmost of the intramyocardial vessels of the heart and it is alsopresent in the endothelium and proximal tubule epithelialcells of the kidney (Donoghue et al. 2000).

Characterisation of ACEH/ACE2

Although a physiological function remains to be estab-lished for ACEH, initial biochemical characterisation hassuggested that this enzyme may have an important functionin the cardiovascular system. The predicted protein sequenceof ACEH shows that it consists of 805 amino acids, with aputative C-terminal transmembrane anchor, a 17 amino acidN-terminal signal sequence, and the obligatory HEXXH mo-tif (amino acids 374–378) (Fig. 4A). Like ACE, ACEH is aglycoprotein with seven putative N-linked glycosylation sitesthat are all located on the extracellular domain of the protein.These may play a crucial role in the stability of the enzyme.Full-length ACEH expressed in Chinese hamster ovary cellscan be solubilised from the membrane by treatment with Triton

© 2002 NRC Canada

348 Can. J. Physiol. Pharmacol. Vol. 80, 2002

Fig. 1. Renin–angiotensin cascade highlighting Ang (1–7) formation and metabolism by ACE and ACEH.



X-100 (S.R. Tipnis et al. unpublished observations), lendingcredence to the hypothesis that ACEH is a membrane pro-tein. In addition, Donoghue et al. (2000) reported that whenthe full-length protein is expressed in Chinese hamster ovarycells, a secreted form can be detected in the conditioned me-dium. This indicates that there may be a secretase that canpost-translationally cleave and release the protein from thecell membrane, resulting in a soluble, circulating form (Hooperet al. 1997). Whether ACE secretase, the enzyme responsi-ble for the shedding of ACE (Oppong and Hooper 1993;Parvathy et al. 1997), also cleaves ACEH is debatable, asACEH has a juxtamembrane stalk region different from thatof ACE (Woodman et al. 2000).

Although ACEH shares significant sequence homology withACE (approximately 40% identity), there are differences thatmay be affecting substrate binding or catalysis, since it doesnot appear to share the same substrate specificity, nor is itinhibited by the typical ACE inhibitors such as enalaprilat,lisinopril, and captopril. Despite the ability of ACEH to hy-drolyse Ang I and Ang II, in contrast with ACE, ACEH doesnot cleave bradykinin or the typical ACE substrateHip-His-Leu. When cleaving Ang I and Ang II, ACEH actsas a carboxypeptidase, removing a single amino acid residuefrom the C-terminus of the substrate. Other substrates thatare effectively cleaved in this manner include kinetensin,des-Arg9 bradykinin, and neurotensin (Donoghue et al.2000; S.R. Tipnis et al., unpublished observations). Thesesubstrates all indicate that ACEH has a preference for aC-terminal hydrophobic residue (Fig. 5). In the substratestested to date, proline occupies either position P1 or P3 and

may therefore form part of the substrate motif recognised byACEH. Although ACEH appears to act as a carboxy-peptidase, its specificity is distinct from that of carboxy-peptidase A, as it does not effectively hydrolyse the typicalcarboxypeptidase A substrate Hip-Phe. In addition, the hy-drolysis of Ang I by ACEH is not inhibited by classicalcarboxypeptidase inhibitors such as benzyl succinate or thecarboxypeptidase inhibitor from potato tuber (Tipnis et al.2000; J.L. Guy et al., unpublished observations). Thecarboxypeptidase A zinc-binding motif (Vendrell et al.2000) is also distinct from that of ACEH.

The high level of ACEH mRNA expression in the testisimplies that there may be a role for ACEH in fertility. A spe-cific, single domain isoform of ACE is also expressed in tes-tis (Ehlers et al. 1989); however, as ACE knockouts in malemice result in infertility (Krege et al. 1995), there may notbe an overlap of function between ACE and ACEH withinthis organ. ACEH knockouts would certainly go some wayto clarifying this issue.

ACEH is a unique mammaliancarboxypeptidase but resembles Taqcarboxypeptidase

ACEH is distinct from other mammalian carboxypepti-dases, as it contains a HEXXH zinc-binding motif (Tipnis etal. 2000). However, there is some similarity between ACEHand a thermostable, bacterial metallocarboxypeptidase fromThermus aquaticus, which has previously been characterised(Lee et al. 1992). Carboxypeptidase Taq (EC 3.4.17.19) with

© 2002 NRC Canada

Turner et al. 349

Fig. 2. Kallikrein–kinin system highlighting bradykinin metabolism by ACE and ACEH.



© 2002 NRC Canada


Fig. 3. Northern and Western blotting analysis of ACEH expression. (A) Multiple tissue Northern blots (CLONTECH) containing 2 µgof poly(A)+ RNA per lane were probed with 32P-labeled cDNA fragments of either ACEH or β-actin. Lanes: 1, heart; 2, brain; 3, pla-centa; 4, lung; 5, liver; 6, skeletal muscle; 7, kidney; 8, pancreas; 9, spleen; 10, thymus; 11, prostate; 12, testis; 13, ovary; 14, smallintestine; 15, colon. The highest levels of ACEH expression are in heart (lane 1), kidney (lane 7), and testis (lane 12). (B) Proteinswere subject to Western blotting, carried out using an antipeptide antibody raised against amino acids 19–32 of ACEH. Lanes: 1, 1 µgof purified human kidney ACE; 2, 25 µg of human kidney microvillar membranes.

A

Testis ACE C

732

Somatic ACE C1306

CACEH

805

Collectrin CN

222

N

HEMGH

1

HEMGH HEMGH

N

N

N

1

N

HEMGH

1

N

1

B

Fig. 4. Comparative diagram and sequence alignment of ACEH and homologous proteins. (A) Somatic ACE, testis ACE, ACEH, andcollectrin. The signal peptide is indicated by a dotted box and the transmembrane domain is represented by a black box. Regions ofhomology between ACE and ACEH are denoted by white boxes and between ACEH and collectrin by diagonal lines. Note theN-terminal testis ACE-specific sequence (grey box). (B) Amino acid sequence alignment of the C-terminal part of ACEH withcollectrin. Identical residues between the two sequences are in bold. Gaps (represented by dots) have been introduced to maximise thealignment. The transmembrane domain is underlined. The numbers on the left refer to the amino acid sequences of the two proteins.Collectrin shows no homology with ACE.



© 2002 NRC Canada

Turner et al. 351

an approximate molecular mass 56 kDa possesses a singlezinc-dependent catalytic domain bearing the motif HEMGH(Fig. 6) (Lee et al. 1994). Mutagenesis studies confirmedthat the HEXXH sequence of carboxypeptidase Taq has es-sentially the same role in enzyme function as the active-sitemotif of thermolysin-like zinc peptidases (Lee et al. 1996).

A consensus sequence, EXIXD, for the glutamate thirdzinc ligand has been identified in the thermolysin family ofmetallopeptidases (Jiang and Bond 1992), which is locatedon the C-terminal side of the active-site motif (Fig. 6). Theconserved aspartate is thought to have an indirect effect oncatalysis with a possible role in the accessibility or precisepositioning of the zinc ion (Colman et al. 1972). With theabsence of such a motif in carboxypeptidase Taq, two resi-dues, Glu-298 and Glu-332, were explored as potential zincligands through mutation; although a reduction in the zinccontent was observed in both cases, a negligible decrease inactivity resulted (Lee et al. 1996). The amino acid sequenceof ACEH indicates the presence of a consensus sequence forthe third zinc ligand that is similar to that in ACE. This isseparated from the zinc-binding motif by a 24-residue spacer;however, the aspartate is replaced by a second glutamate(Fig. 6). To ascertain the functional importance of this motifwithin the enzyme would further our understanding of theproteolytic action of ACEH.

Carboxypeptidase Taq exhibits a broad substrate specific-ity; with the exception of proline, it cleaves C-terminal neu-tral, basic, and acidic amino acids. However, amino acidswith large side chains were most readily hydrolysed (Lee etal. 1992). In contrast, ACEH appears to specifically hydro-lyse peptide bonds of C-terminal hydrophobic residues alone(Donoghue et al. 2000; Tipnis et al. 2000). The penultimateamino acid residue affected the rate of peptide cleavage by

carboxypeptidase Taq and this is therefore a factor that needsconsidering in relation to ACEH specificity.

ACEH: a possible role in Ang (1–7)production and des-Arg9-bradykininmetabolism?

ACE has two active sites with slightly different specificitiesand activation characteristics. In contrast, ACEH has onlyone active site, which has 40% identity with the first activesite of ACE (Tipnis et al. 2000). We have shown that ACEHcleaves the C-terminal amino acid from Ang I to form Ang(1–9) and from Ang II to form Ang (1–7) (Fig. 1) (Tipnis etal. 2000). Thus, it can be seen that ACEH may have a role inthe formation of Ang (1–7). ACEH may form Ang (1–9)from Ang I, which may then be converted to Ang (1–7) byACE (Donoghue et al. 2000). In this way, ACEH may alsoact with NEP to alter the balance of the RAS from vasocon-striction, mediated by Ang II and AT1, towards vasodilation,mediated by Ang (1–7) and the Ang1–7 and AT2 receptors.ACEH does not cleave bradykinin (Tipnis et al. 2000) but isable to cleave the C-terminal amino acid from BK (1–8) toform an inactive peptide, BK (1–7) (Figs. 2 and 5) (Donoghueet al. 2000). Therefore, ACEH may also have a role inbradykinin peptide metabolism via degradation of the activepeptide BK (1–8).

Collectrin: a developmentally regulatedACEH homologue?

Recently, collectrin, a novel homologue of ACEH, wasidentified (Zhang et al. 2001). It comprises 222 amino acidswith an apparent signal peptide and transmembrane domain(Fig. 4A). The sequence is conserved in mouse, rat, and hu-man and shares 81.9% identity. Human collectrin has 47.8%identity with part of the extracellular, transmembrane, andcytosolic domains of ACEH but has no similarity with ACE.A partial sequence alignment is shown in Fig. 4B. Interest-ingly, unlike ACE and ACEH, collectrin lacks a catalyticallyactive domain, suggesting that it has other physiological roles

Collectrin mRNA transcripts are expressed exclusively inthe kidney, and this developmentally regulated renal proteinis reported to be highly localised to collecting ducts. Isola-tion of the cDNA encoding ACEH from a human heart failureventricle library (Donoghue et al. 2000) and the upregulatedexpression of collectrin after renal ablation suggest that thecollectrin domain of ACEH may regulate tissue response toinjury whereas the catalytic domain is involved in peptideprocessing events.

Conclusions

The discovery of ACEH as the first human homologue ofACE provides an intriguing perspective to the regulation ofthe RAS system. The observation that ACEH displays a dis-tinct substrate specificity from ACE provides additional di-versity in the processing of both angiotensin and bradykininpeptides and perhaps a new therapeutic option in cardiovas-cular disease. However, the possibility cannot be dismissedthat ACEH has other distinct physiological roles as a novelcarboxypeptidase. Much remains to be learned of the molec-

Fig. 5. Peptides hydrolysed by ACEH. ACEH appears to func-tion as a carboxypeptidase with a preference for a C-terminalhydrophobic residue (shown in white); the arrow indicates thecleavage site.



© 2002 NRC Canada


ular and cell biology of ACEH; however, the tools are nowavailable to make rapid advances in our understanding of itsphysiology. If it even begins to approach ACE in biomedicalimportance, it will provide a flourishing field of research foryears to come.

Acknowledgements

We thank the British Heart Foundation and the U.K. Med-ical Research Council for financial support.

References

Allred, A.J., Diz, D.I., Ferrario, C.M., and Chappell, M.C. 2000.Pathways for angiotensin-(1–7) metabolism in pulmonary andrenal tissue. Am. J. Physiol. 279: F841–F850.

Brosnihan, K.B. 1998. Effect of the angiotensin-(1–7) peptide onnitric oxide release. Am. J. Cardiol. 82: 17S–19S.

Brosnihan, K.B., Li, P., and Ferrario, C.M. 1996. Angiotensin-(1–7)dilates canine coronary arteries through kinins and nitric oxide.Hypertension, 27: 523–528.

Brown, N.J., and Vaughan, D.E. 1998. Angiotensin-converting en-zyme inhibitors. Circulation, 97: 1411–1420.

Campbell, D.J. 1987. Tissue renin–angiotensin system: sites of an-giotensin formation. J. Cardiovasc. Pharmacol. 10: S1–S8.

Campbell, D.J., Anastasopoulos, F., Duncan, A.-M., James, G.M.,Kladis, A., and Briscoe, T.A. 1998. Effects of neutral endopeptidaseinhibition and combined angiotensin converting enzyme andneutral endopeptidase inhibition on angiotensin and bradykininpeptides in rats. J. Pharmacol. Exp. Ther. 287: 567–577.

Carey, R.M., Wang, Z.-Q., and Siragy, H.M. 2000a. Role of the an-giotensin type 2 receptor in the regulation of blood pressure andrenal function. Hypertension, 35: 155–163.

Carey, R.M., Jin, X.-H., Wang, Z.-Q., and Siragy, H.M. 2000b. Ni-tric oxide: a physiological mediator of the type 2 (AT2) angio-tensin receptor. Acta Physiol. Scand. 168: 65–71.

Chappell, M.C., Tallant, E.A., Diz, D.I., and Ferrario, C.M. 2000.The renin–angiotensin system and cardiovascular homeostasis.In Drugs, enzymes and receptors of the renin–angiotensin sys-tem: celebrating a century of discovery. Edited by A. Husain andR. Graham. Harwood Academic Publishers, Amsterdam, Neth-erlands. pp. 3–22.

Colman, P.M., Jansonius, J.N., and Matthews, B.W. 1972. Thestructure of thermolysin: an electron density map at 2–3 Å reso-lution. J. Biol. Chem. 70: 701–724.

Deddish, P.A., Marcic, B., Jackman, H.L., Wang, H.-Z., Skidgel,R.A., and Erdös, E.G. 1998. N-domain-specific substrate andC-domain inhibitors of angiotensin-converting enzyme. Angioten-sin-(1–7) and keto-ACE. Hypertension, 31: 912–917.

Dell’Italia, L.J., and Oparil, S. 1999. Bradykinin in the heart. Friendor foe? Circulation, 100: 2305–2307.

Dieguez-Lucena, J.L., Aranda-Lara, P., Ruiz-Galdón, M.,García-Villanova, J., Morell-Ocaña, M., and Reyes-Engel, A.1996. Angiotensin I-converting enzyme genotypes and angioten-sin II receptors. Response to therapy. Hypertension, 28: 98–103.

Donoghue, M., Hsieh, F., Baronas, E., Godbout, K., Gosselin, M.,Stagliano, N., Donovan, M., Woolf, B., Robinson, K., Jeyaseelan,R., Breitbart, R.E., and Acton, S. 2000. A novel angioten-sin-converting enzyme-related carboxypeptidase (ACE2) con-verts angiotensin I to angiotensin 1–9. Circ. Res. 87: e1–e9.

Drexler, H., Kurz, S., Jeserich, M., Münzel, T., and Hornig, B.1995. Effect of chronic angiotensin-converting enzyme inhibi-tion on endothelial function in patients with chronic heart fail-ure. Am. J. Cardiol. 76: 13E–18E.

Dzau, V.J. 1994. Cell biology and genetics of angiotensin in car-diovascular disease. J. Hypertens. 12: S3-S10.

Ehlers, M.R.W., and Riordan, J.F. 1989. Angiotensin-convertingenzyme: new concepts concerning its biological role. Biochem-istry, 28: 5311–5318.

Ehlers, M.R.W., Fox, E.A., Strydom, D.J., and Riordan, J.F. 1989.Molecular cloning of human testicular angiotensin-convertingenzyme: the testis isozyme is identical to the C-terminal half ofendothelial angiotensin-converting enzyme. Proc. Natl. Acad.Sci. U.S.A. 86: 7741–7745.

Erdös, E.G. 1975. Angiotensin I converting enzyme. Circ. Res. 36:247–255.

Fernandes, L., Fortes, Z.B., Nigro, D., Tostes, R.C.A., Santos, R.A.S.,and de Carvalho, M.H.C. 2001. Potentiation of bradykinin byangiotensin-(1–7) on arterioles of spontaneously hypertensiverats studied in vivo. Hypertension, 37: 703–709.

Ferrario, C.M., Chappell, M.C., Tallant, E.A., Brosnihan, K.B., andDiz, D.I. 1997. Counterregulatory actions of angiotensin-(1–7).Hypertension, 30: 535–541.

Ferrario, C.M., Martell, N., Yunis, C., Flack, J.M., Chappell, M.C.,Brosnihan, K.B., Dean, R.H., Fernandez, A., Novikov, S.V., Pinillas,C., and Luque, M. 1998. Characterization of angiotensin-(1–7)in the urine of normal and essential hypertensive subjects. Am.J. Hypertens. 11: 137–146.

Fluharty, S.J., Reagan, L.P., and Yee, D.K. 1995. The angiotensintype 1 and type 2 receptor families. Siblings or cousins? In Tis-sue renin–angiotensin systems. Edited by A.K. Mukhopadhyayand M.K. Raizada. Plenum Press, New York. pp. 193–211.

Gorelik, G., Carbini, L.A., and Scicli, A.G. 1998. Angiotensin 1–7induces bradykinin-mediated relaxation in porcine coronary ar-tery. J. Pharmacol. Exp. Ther. 286: 403–410.

Griendling, K.K., Lassègue, B., and Alexander, R.W. 1993. Thevascular angiotensin (AT1) receptor. Thromb. Haemost. 70:188–192.

Fig. 6. Sequence alignment of the active site domain of ACEH and related proteins. The sequence around the zinc-binding site of hu-man and rat ACEH are compared with those in the N- and C-terminal domains of ACE and Taq carboxypeptidase (CPase). Numbersrefer to the amino acid position in each protein sequence. Conserved active-site motifs are shown in bold, and italicised residues repre-sent the three zinc-coordinating ligands.



© 2002 NRC Canada

Turner et al. 353

Hooper, N.M., Karran, E.H., and Turner, A.J. 1997. Membraneprotein secretases. Biochem. J. 321: 265–279.

Hu, K., and Ertl, G. 1999. Potential role of mixed ACE and neutralendopeptidase inhibitor in the treatment of heart failure.Cardiovasc. Res. 41: 503–505.

Iyer, S.N., Chappell, M.C., Averill, D.B., Diz, D.I., and Ferrario,C.M. 1998. Vasodepressor actions of angiotensin-(1–7) unmaskedduring combined treatment with lisinopril and losartan. Hyper-tension, 31: 699–705.

Jiang, W., and Bond, J.S. 1992. Families of metalloendopeptidasesand their relationships. FEBS Lett. 312: 110–114.

Johnston, C.I. 1992. Renin–angiotensin system: a dual tissue andhormonal system for cardiovascular control. Franz Volhard Lec-ture. J. Hypertens. 10: S13–S26.

Kaplan, A.P., and Silverberg, M. 1987. The coagulation–kinin path-way of human plasma. Blood, 70: 1–15.

Krege, J.H., John, S.W.M., Langenbach, L.L., Hodgin, J.B., Hagaman,J.R., Bachman, E.S., Jennette, J.C., O’Brien, D.A., and Smithies,O. 1995. Male–female differences in fertility and blood pressurein ACE-deficient mice. Nature (Lond.), 375: 146–148.

Lee, S.-H., Minagawa, E., Taguchi, H., Matsuzawa, H., Ohta, T.,Kaminogawa, S., and Yamauchi, K. 1992. Purification and char-acterization of a thermostable carboxypeptidase (carboxypeptidaseTaq) from Thermus aquaticus YT-1. Biosci. Biotechnol. Biochem.56: 1839–1844.

Lee, S.-H., Taguchi, H., Yoshimura, E., Minagawa, E., Kaminogawa,S., Ohta, T., and Matsuzawa, H. 1994. Carboxypeptidase Taq, athermostable zinc enzyme, from Thermus aquaticus YT-1; molec-ular cloning, sequencing, and expression of the encoding gene inEscherichia coli. Biosci. Biotechnol. Biochem. 58: 1490–1495.

Lee, S.-H., Taguchi, H., Yoshimura, E., Minagawa, E., Kaminogawa,S., Ohta, T., and Matsuzawa, H. 1996. The active site ofcarboxypeptidase Taq possesses the active-site motifHis-Glu-X-X-His of zinc-dependent endopeptidases and amino-peptidases. Protein Eng. 9: 467–469.

Oppong, S.Y., and Hooper, N.M. 1993. Characterization of a secretaseactivity which releases angiotensin-converting enzyme from themembrane. Biochem. J. 292: 597–603.

Parvathy, S., Oppong, S.Y., Karran, E.H., Buckle, D.R., Turner,A.J., and Hooper, N.M. 1997. Angiotensin-converting enzymesecretase is inhibited by zinc metalloprotease inhibitors and re-quires its substrate to be inserted into a lipid bilayer. Biochem.J. 327: 37–43.

Pellacani, A., Brunner, H.R., and Nussberger, J. 1994. Plasma kininsincrease after angiotensin-converting enzyme inhibition in hu-man subjects. Clin. Sci. 87: 567–574.

Peters, J. 1995. Molecular basis of human hypertension: the role ofangiotensin. Baillière’s Clin. Endocrin. Metab. 9: 657–678.

Searles, C.D., and Harrison, D.G. 1999. The interaction of nitricoxide, bradykinin, and the angiotensin II type 2 receptor: les-sons learned from transgenic mice. J. Clin. Invest. 104:1013–1014.

Strawn, W.B., Ferrario, C.M., and Tallant, E.A. 1999. Angioten-sin-(1–7) reduces smooth muscle growth after vascular injury.Hypertension, 33: 207–211.

Tallant, E.A., Lu, X., Weiss, R.B., Chappell, M.C., and Ferrario,C.M. 1997. Bovine aortic endothelial cells contain an angioten-sin-(1–7) receptor. Hypertension, 29: 388–393.

Tallant, E.A., Diz, D.I., and Ferrario, C.M. 1999. Antiproliferativeactions of angiotensin-(1–7) in vascular smooth muscle. Hyper-tension, 34: 950–957.

Tipnis, S.R., Hooper, N.M., Hyde, R., Karran, E., Christie, G., andTurner, A. J. 2000. A human homolog of angiotensin convertingenzyme. Cloning and functional expression as a captopril-insensitivecarboxypeptidase. J. Biol. Chem. 275: 33 238 – 33 243.

Tsutsumi, Y., Matsubara, H., Masaki, H., Kurihara, H., Murasawa,S., Takai, S., Miyazaki, M., Nozawa, Y., Ozono, R., Nakagawa,K., Miwa, T., Kawada, N., Mori, Y., Shibasaki, Y., Tanaka, Y.,Fujiyama, S., Koyama, Y., Fujiyama, A., Takahshi, H., and Iwasaka,T. 1999. Angiotensin II type 2 receptor overexpression activatesthe vascular kinin system and causes vasodilation. J. Clin. In-vest. 104: 925–935.

Ueda, S., Masumori-Maemoto, S., Ashino, K., Nagahara, T., Gotoh,E., Umemura, S., and Ishii, M. 2000. Angiotensin-(1–7) attenuatesvasoconstriction evoked by angiotensin II but not by noradrenalinein man. Hypertension, 35: 998–1001.

Vendrell, J., Querol, E., and Aviles, F.X. 2000. Metallocarboxy-peptidases and their protein inhibitors. Structure, function andbiomedical properties. Biochim. Biophys. Acta, 1477: 284–298.

Woodman, Z.L., Oppong, S.Y., Cook, S., Hooper, N.M., Schwager,S.L.U., Brandt, W.F., Ehlers, M.R.W., and Sturrock, E.D. 2000.Shedding of somatic angiotensin-converting enzyme (ACE) isinefficient compared to testis ACE despite cleavage at identicalstalk sites. Biochem. J. 347: 711–718.

Yamada, K., Iyer, S.N., Chappell, M.C., Ganten, D., and Ferrario,C.M. 1998. Converting enzyme determines plasma clearance ofangiotensin-(1–7). Hypertension, 32: 496–502.

Zhang, H., Wada, J., Hida, K., Tsuchiyama, Y., Hiragushi, K.,Shikata, K., Wang, H.Y., Lin, S., Kanwar, Y.S., and Makino, H.2001. Collectrin, a collecting duct-specific transmembraneglycoprotein, is a novel homologue of ACE2 and is develop-mentally regulated in embryonic kidneys. J. Biol. Chem. 276:17 132 – 17 139.



Documents

ACEH/ACE2 is a novel mammalian metallocarboxypeptidase and a homologue of angiotensin-converting enzyme insensitive to ACE inhibitors