ACE phenotyping in human heart

Abstract

Aims: Angiotensin-converting enzyme (ACE), which metabolizes many peptides and plays a key role in blood pressure regulation and vascular remodeling, is expressed as a type-1 membrane glycoprotein on the surface of different cells, including endothelial cells of the heart. We hypothesized that the local conformation and, therefore, the properties of heart ACE could differ from lung ACE due to different microenvironment in these organs.

Methods and results: We performed ACE phenotyping (ACE levels, conformation and kinetic characteristics) in the human heart and compared it with that in the lung. ACE activity in heart tissues was 10-15 lower than that in lung. Various ACE effectors, LMW endogenous ACE inhibitors and HMW ACE-binding partners, were shown to be present in both heart and lung tissues. "Conformational fingerprint" of heart ACE (i.e., the pattern of 17 mAbs binding to different epitopes on the ACE surface) significantly differed from that of lung ACE, which reflects differences in the local conformations of these ACEs, likely controlled by different ACE glycosylation in these organs. Substrate specificity and pH-optima of the heart and lung ACEs also differed. Moreover, even within heart the apparent ACE activities, the local ACE conformations, and the content of ACE inhibitors differ in atria and ventricles.

Conclusions: Significant differences in the local conformations and kinetic properties of heart and lung ACEs demonstrate tissue specificity of ACE and provide a structural base for the development of mAbs able to distinguish heart and lung ACEs as a potential blood test for predicting atrial fibrillation risk.

Fig 1. ACE activity in human tissues.

ACE activity in tissue homogenates from 10 donors and citrated plasma was quantified using a spectrofluorimetric assay with 2 mM ZPHL and 5 mM HHL as substrates. Homogenates were prepared at 1:9 ratio (weight/volume) and further diluted 1/10 to minimize an effect of putative endogenous ACE inhibitors and LMW ACE effectors. Plasma was diluted 1:4 with PBS. Data are expressed as mU per gram of tissue (for homogenates) or per ml of undiluted plasma, p<0.01. Insert: ACE activity (mU/ml) in homogenates of different chambers of human hearts from 10 donors. V-ventricles; A-atria; L-left; R-right. Each value is a mean of several (2–3) experiments in duplicates, p<0.05.

Fig 2. Effect of dilution on the apparent ACE activity in the heart and lung homogenates.

ACE activity was measured in the heart and lung homogenates from 10 donors at different dilutions using two substrates, ZPHL and HHL (as in the legend to Fig 1). Data are expressed as % from the ACE activity in undiluted homogenates (A,B), as well as % of ZPHL/HHL ratio from that for undiluted homogenates (C,D). Each value is a mean of several (2–3) experiments in duplicates, p<0.01.

Fig 3. Conformational characteristics of different ACEs.

Conformational fingerprinting of the heart and lung ACEs was performed with a set of 17 mAbs to the two-domain ACE. Immunoprecipitated ACE activity from purified ACEs solutions (A), tissue homogenates from 10 donors (B), or ACEs after perfusion into rat blood circulation (C) are presented as % (“binding ratio”) for heart ACE from that of lung ACE. Ratios increased more than 20% are highlighted in orange, more than 50% in dark orange, and more than 200% in red, while decreased more than 20% are highlighted in yellow and more than 50% in deep blue. Data are mean ± SD of at least 3 experiments (each in duplicates), p<0.01.

Fig 4. ACE effectors in the heart and lung tissues.

Conformational fingerprinting of heart and lung ACEs was performed with a set of 17 mAbs to ACE as in the legend to Fig 3. Immunoprecipitated ACE activity after purification of heart and lung ACEs by anion-exchange and the affinity chromatography (A,B), after dialysis (C, D) and filtration through filter with 100 kDa limit (E, F) is presented as % (“binding ratio”) from immunoprecipitated ACE activity from the parent homogenates. Ratios increased more than 20% are highlighted in orange, more than 50% in dark orange, and more than 100% in red. Ratios decreased more than 20% are highlighted in yellow, more than 50% in deep blue. Data are mean ± SD of 2–3 experiments (each in duplicates), p<0.01.

Fig 5. Dependence of ACE activity on pH.

The assays of the activity of the purified heart and lung ACEs toward 0.5 mM Z-Phe-His-Leu (A) and 1.3 mM Hip-His-Leu (B) were performed in 25 mM acetate-MES-Tris-borate buffer, containing 0.15 M NaCl and 1 μM ZnCl₂. Each value is a mean of several (2–3) experiments in duplicates.

Fig 6. Effect of dilution on ACE activity in the homogenates of heart chambers.

ACE activity and ZPHL/HHL ratio were measured in the homogenates of human heart chambers at different dilutions using two substrates (as in the legend to Fig 1). Data are expressed as absolute values (A-C) and as % from homogenates at maximal dilution—1/30 (D-F). Each value is a mean of several (2–3) experiments on separate homogenates in duplicates.

Fig 7. Differences between ACE protein level and ACE activity in the heart chambers.

ACE activity was measured in the undiluted homogenates of human heart chambers and at 1/10 dilution as in Fig 1. ACE protein level was quantified after precipitation of ACE by a set of mAbs and washing out putative endogenous ACE inhibitors/effectors [26]. Data are expressed as a percentage from the corresponding data for left ventricle homogenate. Each value is a mean of several (3) experiments in duplicates. Bars highlighted in orange represent the samples with values higher than 20% of mean ± SD for control samples (left ventricle homogenate), * p<0.05, ** p<0.01.

Fig 8. Conformational fingerprinting of ACE in different heart chambers.

Conformational fingerprinting of ACE in the homogenates (1:9) of different heart chambers was performed with a set of 17 mAbs to ACE as in Fig 3. Immunoprecipitated ACE activity from these homogenates undiluted and further diluted 1/10 is presented as % (“binding ratio”) from the immunoprecipitated ACE activity from left ventricle homogenate (undiluted and diluted, correspondingly). Ratios increased more than 20% are highlighted in orange. Data are mean ± SD of at least 3 experiments (each in duplicates), p<0.01.

Fig 9. Hypothetical scheme of the assay for quantification of heart-derived ACE in the blood.

According to our estimation (based on heterogeneous ACE expression in capillary endothelial cells of different organs [15]) lung ACE provides about 75% of blood ACE, whereas heart-derived ACE activity in the blood could not be more than 1%. Therefore, overall increase in blood ACE as a result of 3-fold increase in the heart ACE due to atrial fibrillation [2] will not increase substantially blood ACE activity in these patients. However, precipitation and quantification from the blood of heart-derived ACE by mAbs, specific for heart ACE, may be diagnostically relevant.