Search

Selective inhibition of the angiotensin-converting enzyme C-domain (cACE) and neprilysin (NEP), leaving the ACE N-domain (nACE) free to degrade bradykinin and other peptides, has the potential to provide the potent antihypertensive and cardioprotective benefits observed for nonselective dual ACE/NEP inhibitors, such as omapatrilat, without the increased risk of adverse effects. We have synthesized three 1-carboxy-3-phenylpropyl dipeptide inhibitors with nanomolar potency based on the previously reported C-domain selective ACE inhibitor lisinopril-tryptophan (LisW) to probe the structural requirements for potent dual cACE/NEP inhibition. Here we report the synthesis, enzyme kinetic data, and high-resolution crystal structures of these inhibitors bound to nACE and cACE, providing valuable insight into the factors driving potency and selectivity. Overall, these results highlight the importance of the interplay between the S1′ and S2′ subsites for ACE domain selectivity, providing guidance for future chemistry efforts toward the development of dual cACE/NEP inhibitors.

Angiotensin I-converting enzyme (ACE) is central to the regulation of the renin-angiotensin system and is a key therapeutic target for combating hypertension and related cardiovascular diseases. Currently available drugs bind both active sites of its two homologous domains, although it is now understood that these domains function differently in vivo. The recently solved crystal structures of both domains (N and C) open the door to new domain-specific inhibitor design, taking advantage of the differences between these two large active sites. Here we present the first crystal structure at a resolution of 2.25 Å of testis ACE (identical to the C domain of somatic ACE) with the highly C-domain-specific phosphinic inhibitor, RXPA380. Testis ACE retains the same conformation as seen in previously determined inhibitor complexes, but the RXPA380 central backbone conformation is more similar to that seen for the inhibitor captopril than enalaprilat. The RXPA380 molecule occupies more subsites of the testis ACE active site than the previously determined inhibitors and possesses bulky moieties that extend into the S2′ and S2 subsites. Thus the high affinity of RXPA380 for the testis ACE/somatic ACE C domain is explained by the interaction of these bulky moieties with residues unique to these domains, specifically Phe 391, Val 379, and Val 380, that are not found in the N domain. The characterization of the extended active site and the binding of a potent C-domain-selective inhibitor provide the first structural data for the design of truly domain-specific pharmacophores.

Human angiotensin-converting enzyme is an important drug target for which little structural information has been available until recent years. The slow progress in obtaining a crystal structure was due to the problem of surface glycosylation, a difficulty that has thus far been overcome by the use of a glucosidase-1 inhibitor in the tissue culture medium. However, the prohibitive cost of these inhibitors and incomplete glucosidase inhibition makes alternative routes to minimizing the N-glycan heterogeneity desirable. Here, glycosylation in the testis isoform (tACE) has been reduced by Asn-Gln point mutations at N-glycosylation sites, and the crystal structures of mutants having two and four intact sites have been solved to 2.0 A and 2.8 A, respectively. Both mutants show close structural identity with the wild-type. A hinge mechanism is proposed for substrate entry into the active cleft, based on homology to human ACE2 at the levels of sequence and flexibility. This is supported by normal-mode analysis that reveals intrinsic flexibility about the active site of tACE. Subdomain II, containing bound chloride and zinc ions, is found to have greater stability than subdomain I in the structures of three ACE homologues. Crystallizable glycosylation mutants open up new possibilities for cocrystallization studies to aid the design of novel ACE inhibitors.

Eosinophil-derived neurotoxin (EDN) is a catalytically proficient member of the pancreatic ribonuclease superfamily secreted along with other eosinophil granule proteins during innate host defense responses and various eosinophil-related inflammatory and allergic diseases. The ribonucleolytic activity of EDN is central to its antiviral and neurotoxic activities and possibly to other facets of its biological activity. To probe the importance of this enzymatic activity further, specific inhibitors will be of great aid. Derivatives of 5'-ADP are among the most potent inhibitors currently known. Here, we use X-ray crystallography to investigate the binding of four natural nucleotides containing this moiety. 5'-ATP binds in two alternative orientations, one occupying the B 2 subsite in a conventional manner and one being a retro orientation with no ordered adenosine moiety. Diadenosine triphosphate (Ap 3 A) and diadenosine tetraphosphate (Ap4A) bind with one adenine positioned at the B 2 subsite, the polyphosphate chain extending across the P 1 subsite in an ill-defined conformation, and a disordered second adenosine moiety. Diadenosine pentaphosphate (Ap 5 A), the most avid inhibitor of this series, binds in a completely ordered fashion with one adenine positioned conventionally at the B 2 subsite, the polyphosphate chain occupying the P 1 and putative P -1 subsites, and the other adenine bound in a retro-like manner at the edge of the B 1 subsite. The binding mode of each of these inhibitors has features seen in previously determined structures of adenosine diphosphates. We examine the structure-affinity relationships of these inhibitors and discuss the implications for the design of improved inhibitors.

ACE chimeric proteins and N domain monoclonal antibodies (mAbs) were used to determine the influence of the N domain, and particular regions thereof, on the rate of ACE ectodomain shedding. Somatic ACE (having both N and C domains) was shed at a rate of 20%/24 h. Deletion of the C domain of somatic ACE generated an N domain construct (ACEDeltaC) which demonstrated the lowest rate of shedding (12%). However, deletion of the N domain of somatic ACE (ACEDeltaN) dramatically increased shedding (212%). Testicular ACE (tACE) having 36 amino acid residues (heavily O-glycosylated) at the N-terminus of the C domain shows a 4-fold decrease in the rate of shedding (49%) compared to that of ACEDeltaN. When the N-terminal region of the C domain was replaced with the corresponding homologous 141 amino acids of the N domain (N-delACE) the rate of shedding of the ACEDeltaN was only slightly decreased (174%), but shedding was still 3.5-fold more efficient than wild-type testicular ACE. Monoclonal antibodies specific for distinct, but overlapping, N-domain epitopes altered the rate of ACE shedding. The mAb 3G8 decreased the rate of shedding by 30%, whereas mAbs 9B9 and 3A5 stimulated ACE shedding 2- to 4-fold. Epitope mapping of these mAbs in conjunction with a homology model of ACE N domain structure, localized a region in the N-domain that may play a role in determining the relatively low rate of shedding of somatic ACE from the cell surface.

Angiotensin converting enzyme (ACE) plays a critical role in the circulating or endocrine renin-angiotensin system (RAS) as well as the local regulation that exists in tissues such as the myocardium and skeletal muscle. Here we report the high-resolution crystal structures of testis ACE (tACE) in complex with the first successfully designed ACE inhibitor captopril and enalaprilat, the Phe-Ala-Pro analogue. We have compared these structures with the recently reported structure of a tACE-lisinopril complex [Natesh et al. (2003) Nature 421, 551-554]. The analyses reveal that all three inhibitors make direct interactions with the catalytic Zn(2+) ion at the active site of the enzyme: the thiol group of captopril and the carboxylate group of enalaprilat and lisinopril. Subtle differences are also observed at other regions of the binding pocket. These are compared with N-domain models and discussed with reference to published biochemical data. The chloride coordination geometries of the three structures are discussed and compared with other ACE analogues. It is anticipated that the molecular details provided by these structures will be used to improve the binding and/or the design of new, more potent domain-specific inhibitors of ACE that could serve as new generation antihypertensive drugs.