Investigating domain-selective angiotensin converting enzyme inhibition and oxidative inactivation

Doctoral Thesis

2018

Permanent link to this Item
Authors
Journal Title
Link to Journal
Journal ISSN
Volume Title
Publisher
Publisher

University of Cape Town

License
Series
Abstract
Angiotensin converting enzyme (ACE) is a zinc metalloprotease comprised of two highly homologous, catalytically active domains (90% active site identity and 60% sequence similarity). The C-domain is responsible for blood pressure regulation via angiotensin I cleavage while the N-domain inactivates an antifibrotic peptide Acetyl-Ser-Asp-Lys-Pro (AcSDKP). Since selective N-domain inhibition will result in AcSDKP accumulation, it shows promise for the treatment of fibrosis without affecting blood pressure. Low bioavailability, however, precludes the use of currently available N-selective ACE inhibitors in a clinical setting. Inhibition of ACE by a phosphinic, peptidomimetic compound, 33RE, was characterized using a continuous assay with quenched fluorogenic substrate. The N-domain displayed nanomolar (Ki = 11.21±0.74nM) and the C-domain micromolar (Ki = 11 278±410nM) inhibition, thus 1000-fold selectivity. Residues predicted to contribute to selectivity based on the N-domain-33RE co-crystal structure were subsequently mutated to their C-domain counterparts. S2 subsite mutation with resulting loss of a hydrogen bond drastically decreased 33RE affinity (Ki = 2794±156nM), yet did not entirely account for the selectivity. Additional substitution of all unique S2’ residues, however, completely abolished N-selectivity (Ki = 10 009±157nM). Interestingly, these residues do not directly bind 33RE. All mutants were therefore subjected to molecular dynamics (MD) simulations in the presence and absence of 33RE in addition to co-crystallization of 33RE with the N-domain mutant having all S2 and S2’ residues mutated. Trajectory analyses highlighted the S2’ residues’ importance in formation of a favourable interface between the ACE subdomains and thus a closed, ligand-bound complex. This was supported by X-ray crystallography and provides a molecular basis for the inter-subsite synergism governing 33RE’s 1000-fold N-domain selectivity. Enzyme kinetics were also used to study the concentration-dependent competitive inhibition and time-dependent irreversible oxidative inactivation of ACE catalysed by the Cu-Gly-GlyHis-lisinopril (CuGGHLis) metallodrug. Although both domains displayed nanomolar affinity for metallodrug binding (N-domain Ki = 44.94±1.84nM and C-domain Ki = 15.57±1.30nM), rapid and complete CuGGHLis-mediated inactivation occurred exclusively in the N-domain upon incubation with ascorbate and H2O2 redox co-reactants (k2 = 59 710 M-1 min-1 ). Michaelis-Menten characterization of the residual activity after partial N-domain inactivation revealed a decreased rate for hydrolysis of a non-domain selective substrate. This suggests that although CuGGHLis binds with similar affinity to both domains, the metal-chelate is optimally orientated in the N- but not the C-domain to catalyze oxidation of residues involved in substrate hydrolysis. The C-domain, in contrast, showed increased susceptibility to oxidative inactivation by diffuse radicals. This is of physiological significance as C-domain inactivation in normotensive individuals could result in accumulation of pro-inflammatory peptides. Since the N-domain is more heavily glycosylated, the potential role of unique glycans in diffuse radical shielding was studied using glycoprotein MD simulations. Unique C-domain solvent tunnels were identified that could increase diffuse radical access and, additionally, the mechanism whereby glycosylation contributes to ACE thermal stability was described for each site. This has implications for future ACE crystallography studies and the design of ACE-modulating agents with potential anti-inflammatory activity. This study demonstrated the utility of combining in vitro and in silico approaches to reveal how subtle amino acid or glycosylation site differences between the highly homologous domains control dynamic behaviour. It furthermore elucidated how two inhibitors with different mechanisms of action selectively target the N-domain active site by exploiting these differences and provided valuable insight for future anti-fibrotic ACE inhibitor design.
Description

Reference:

Collections