Novel role of aminopeptidase-A in angiotensin-(1–7) metabolism post myocardial infarction

Abstract

Aminopeptidase-A (APA) is a less well-studied enzyme of the renin-angiotensin system. We propose that it is involved in cardiac angiotensin (ANG) metabolism and its pathologies. ANG-(1–7) can ameliorate remodeling after myocardial injury. The aims of this study are to 1) develop mass spectrometric (MS) approaches for the assessment of ANG processing by APA within the myocardium; and 2) investigate the role of APA in cardiac ANG-(1–7) metabolism after myocardial infarction (MI) using sensitive MS techniques. MI was induced in C57Bl/6 male mice by ligating the left anterior descending (LAD) artery. Frozen mouse heart sections (in situ assay) or myocardial homogenates (in vitro assay) were incubated with the endogenous APA substrate, ANG II. Results showed concentration- and time-dependent cardiac formation of ANG III from ANG II, which was inhibited by the specific APA inhibitor, 4-amino-4-phosphonobutyric acid. Myocardial APA activity was significantly increased 24 h after LAD ligation (0.82 ± 0.02 vs. 0.32 ± 0.02 ρmol·min⁻¹·μg⁻¹, MI vs. sham, P < 0.01). Both MS enzyme assays identified the presence of a new peptide, ANG-(2–7), m/z 784, which accumulated in the MI (146.45 ± 6.4 vs. 72.96 ± 7.0%, MI vs. sham, P < 0.05). Use of recombinant APA enzyme revealed that APA is responsible for ANG-(2–7) formation from ANG-(1–7). APA exhibited similar substrate affinity for ANG-(1–7) compared with ANG II {K_m (ANG II) = 14.67 ± 1.6 vs. K_m [ANG-(1–7)] = 6.07 ± 1.12 μmol/l, P < 0.05}. Results demonstrate a novel role of APA in ANG-(1–7) metabolism and suggest that the upregulation of APA, which occurs after MI, may deprive the heart of cardioprotective ANG-(1–7). Thus APA may serve as a potentially novel therapeutic target for management of tissue remodeling after MI.

aminopeptidase-a (APA) is a zinc metallopeptidase that hydrolyzes the NH₂-terminal aspartate of angiotensin (ANG) II to generate ANG III (9, 49). APA is expressed in several organs (25, 38), and its activity has been shown to be upregulated in pathology associated with increased ANG II levels (3, 20, 34, 50). The metabolic product resulting from this activity, ANG III, shares many physiological functions with ANG II such as stimulation of aldosterone release and tissue proliferation triggered through binding to the ANG II type 1 receptor (AT₁) (5, 42). ANG III has also been reported to be responsible for blood pressure regulation as well as renal damage in hypertensive rats (14, 29, 35). Moreover, ANG III has a more pronounced effect than ANG II in stimulating collagen protein synthesis in cardiac fibroblasts (42). A plethora of studies found that central inhibition of APA ameliorates the progression of cardiovascular (CV) diseases by lowering brain ANG III levels (2, 20, 44). Interestingly, it has been shown that oral administration of the centrally acting APA inhibitor RB150 markedly reduced blood pressure in hypertensive rats (3, 26). Therefore, APA is a critical mediator of CV complications, suggesting that a reduction in APA activity may be a potential therapeutic target.

ANG-(1–7) is a heptapeptide that is generated from ANG II via hydrolysis of the COOH-terminal phenylalanine catalyzed by angiotensin-converting enzyme 2 (ACE2) or prolyl carboxypeptidase (6, 40). ANG-(1–7) is a well-known vasodilator, antiproliferative, and cardioprotective peptide against myocardial remodeling after injury (11, 12, 16, 28, 30, 47). However, the clinical applications of ANG-(1–7) are limited due to its extremely short half-life (∼20 s) (22). Since the peptide sequence at the amino terminus of ANG-(1–7) is similar to that of ANG II, ANG-(1–7) could be a target for APA. Indeed, Schwacke and colleagues (36) recently identified a possible role of APA in ANG-(1–7) metabolism using a computational network model. As APA is ubiquitously expressed in the body and is activated in disease conditions (20, 37, 38, 50), it is predictable that APA is a potential pathway for ANG-(1–7) degradation, thereby worsening CV outcomes.

To investigate the in situ metabolism of ANG peptides in the myocardium, we used a new enzyme assay that was developed based on matrix-assisted laser desorption ionization (MALDI) imaging for the identification of ANG II metabolism in the kidney (17). The merits of using MALDI enzyme assay over other conventional assays is that it allows the use of natural substrate. In addition, MALDI imaging is coupled with a software which allows visualization of the spatial distribution of enzymes in situ (17). Using this approach, we provide evidence, for the first time, that APA is expressed in myocardial segments from the ventricular area of the heart and it is an important mediator of myocardial ANG-(1–7) metabolism. Results were confirmed by MS-MALDI in vitro enzyme assays and enzyme kinetic studies demonstrating a similar affinity of APA to ANG-(1–7) compared with ANG II. Moreover, APA activity and expression were increased significantly after MI compared with sham. Taken together, inhibition of APA could be a potentially new therapeutic target for the amelioration of CV diseases associated with APA activation.

MATERIALS AND METHODS

Animals.

Mice (12-wk-old male; n = 5–6 per group) with C57Bl/6 background [wild type (WT)] were housed at 22°C under 12:12-h light-dark cycle with ad libitum access to water and standard mouse chow. ACE2 knockout (ACE2 KO) were littermates with C57Bl/6 background (1). All experimental protocols were approved by the Wright State University (WSU) Animal Care and Use Committee.

Surgical protocol.

To prepare the mice for surgery, mice were first injected intramuscularly (im) with atropine sulfate (0.04 mg/kg body wt; Vedco) as a preanesthetic to prevent respiratory secretions that may obstruct airways during surgery. Next, mice were anesthetized by intraperitoneal (ip) injection of ketamine/xylazine (100/10 mg/kg body wt; Vedco) and then placed in a cage under heating pads to maintain body temperature and left undisturbed for ∼10 min. Once animals reached the correct plane of anesthesia, mice were then moved to a surgical platform and secured in the supine position. A rectal probe was used to maintain body temperature between 36.5 to 37.5°C. ECG leads from radiotelemetry probe model TA11ETA-F10 (Data Science International) were inserted subcutaneously (sc) into the limbs of mice to monitor the heart rate and ECG of mice during surgery. A 5-mm incision was made over the trachea of mice to visualize the insertion of tracheal intubation [size 60 polyethylene (PE) tubing; Becton Dickenson] that was connected to a rodent ventilator (Minivent type 845; Hugo Sachs Electronik, Harvard Apparatus) with ambient room air at a tidal volume and frequency calculated using the following equations: Vt = 6.2 × Mb^1.01, where Vt is tidal volume (in ml) and Mb is animal body mass (in kg); and BPM = 53.5 × Mb^−0.26, where BPM is ventilation rate (breaths/min) (41). Mice were then carefully rotated radially and heart was exposed via left thoracotomy performed through the fourth intercostal space. A pericardial incision was then made by blunt dissection with forceps, exposing the anterior wall and apex of the LV. The LAD was then located and permanent occlusion of the LAD was performed below the left atrium to the midpapillary level using a 8–0 nylon suture (Ethicon, Somerville, NJ). Observation of ST-segment elevation on the ECG apparatus was used to confirm MI induction. Lidocaine (<0.1 ml, 1%; Vedco) was topically added to the surgical site as a local anesthetic and to prevent arrhythmia. The thoracic cavity was then closed using a 5–0 Vicryl (Ethicon) suture and air pneumatically siphoned with a syringe connected to a chest tube inserted into the chest incision. Following closure of the skin incision (6–0 nylon; Ethicon) and removal of the chest tube, mice were supplemented with subcutaneous bolus injection of saline solution (1% vol/wt) and administered analgesics [carprofen (5 mg/kg body wt sc; Pfizer Animal Health) and buprenorphine (0.1 mg/kg sc; Reckitt Benckiser Pharmaceuticals)] in an area away from the surgical site. Once the mouse was stabilized and respiratory reflexes had returned, the intubation was removed and the mice were then moved to a heated recovery area and supplemented with 100% oxygen until full recovery. Sham operations were performed on animals following the same methods; however, the suture was passed underneath the LAD without ligating the vessel.

Histology.

2,3,5-Triphenyl-2H-tetrazolium chloride (TTC) staining was performed to ensure the MI in mice. Hearts were washed with cold PBS and incubated with prewarmed 1% TTC (Sigma-Aldrich, St. Louis, MO) with shaking for 15–20 min. Sections were fixed with freshly prepared 4% paraformaldehyde.

Tissue samples.

Mice were decapitated 24 h after the LAD ligation, and hearts were quickly removed. Three-millimeter transverse segments within the ventricular areas were created using a prechilled heart block (Zivic Laboratories, Pittsburgh, PA) and immediately frozen in liquid nitrogen. Tissues were kept at −80°C until used for experiments.

Tissue sectioning.

Heart sections (12 μm) were cut on a cryostat at −20°C and thaw-mounted onto prechilled indium-tin-oxide-coated glass slides (Delta Technologies, Loveland, CO). Slides were desiccated at room temperature for 30 min prior to evaluation of enzyme activity.

In situ enzyme activity.

Heart sections were incubated with 10–1,000 μmol/l ANG II (Sigma-Aldrich) at 37°C for 5–15 min. The substrate concentration and incubation periods were determined based on the optimal ANG-(1–7) and ANG III signal formation. For inhibitory study, different concentration of inhibitors ranging from 1 to 1,000 μmol/l were added to 100 μmol/l ANG II solutions, and sections were incubated with the mixtures for 10 min. The specific APA inhibitor, 4-amino-4-phosphonobutyric acid (24) [4-APBA, gift from Dr. R. Speth, Nova Southeastern University, that corresponds to “glutamate phosphonate” used by Grobe et al. (17)], was added to the incubation mixtures to inhibit ANG III formation while ANG-(1–7) formation was inhibited using the specific ACE2 inhibitor, MLN-4760 (gift from the former Millennium Pharmaceuticals. Cambridge, MA).

In vitro enzyme assays.

Heart homogenate (50 μg protein) was incubated for 30 min at 37°C in buffer containing 5 nmol/l ANG II, 2 mmol/l PMSF, and 20 μmol/l bestatin. Reactions were acidified by adding 0.1% trifluoroacetic acid (TFA). Stable-isotope-labeled internal standards (New England Peptide, Gardner, MA) were added for absolute quantitation. The reaction mixture was dried overnight using a speed vacuum concentrator. The pellet was resuspended in 0.1% TFA, and peptides were purified using C₁₈ Ziptips (Millipore, Billerica, MA) according to a previously published procedure (18). Enzyme assays using 1 and 5 ng recombinant APA enzyme (rAPA, R & D, Minneapolis, MN) were performed according to the manufacturer recommendations. For kinetic study and K_m determination, rAPA was incubated with ANG II and ANG-(1–7) in concentrations ranging from 1 to 90 μmol/l for 30 min at 37°C and peptides were purified using C₁₈ Ziptips (Millipore, Billerica, MA). Inhibition of peptide formation was tested using 3–30 nmol/l APA inhibitor 4-APBA. Mass spectra were obtained using a Bruker Autoflex III smart beam MALDI TOF/TOF instrument. A total of 7,000 laser shots were acquired randomly for each spot at a laser frequency of 100 Hz.

Quantitation of MALDI images.

Signals were quantitated as integrated intensity and normalized to the area of the whole heart section by MetaMorph image analysis software (Ver. 7.7.3) (Molecular Devices, Sunnyvale, CA) using specific settings within the Integrated Morphometry Analysis toolkit.

Western blot.

Twenty four hours after MI or sham, the 3-mm myocardial segment in the left ventricular area (where infarction was induced) was excised and homogenized in phosphate-buffered saline (PBS) containing protease inhibitor (Complete Lysis M, Roche diagnostics, Mannheim, Germany). Tissue homogenates were centrifuged at 10,000 g for 10 min at 4°C to remove cellular debris. Total protein content was determined in supernatant using BSA as a standard and Bio-Rad reagent (Bio-Rad, Hercules, CA). For protein denaturation, 50 μl of tissue lysate was added to 50 μl sample loading buffer (8% SDS, 125 mmol/l Tris-HCl, pH 6.8, 20% glycerol, 0.02% bromophenol blue, 100 mmol/l dithiothreitol) (Bio-Rad) and boiled for 10 min. Protein (70 μg) was loaded into the wells of a 10% SDS-PAGE gel and separated by electrophoresis for 1 h. Proteins on gel were then transferred (Bio-Rad transfer apparatus) to a 0.2 μm PVDF membrane (Millipore). The membrane was blocked for 1 h with 10% nonfat milk in 10 mM Tris-buffered saline with Tween 20 (TBS-T) at room temperature (RT). Membrane was probed overnight at 4°C with goat anti-APA antibodies (1:250, Santa Cruz, CA) and then incubated with horse radish peroxidase conjugated to donkey anti-goat secondary antibodies (dilution 1:10,000 in TBS-T, Jackson Immunoresearch) at RT for 1 h. Blots were detected using supersignal chemiluminescent substrate and visualized by exposing the membrane for 1 min using a ChemiDoc system with Image Lab analyzer (Bio-Rad). The relative amounts of proteins of interest were determined by normalizing to GAPDH. Western blot images were quantified using Image Lab software (Bio-Rad).

Statistical analysis.

All data are expressed as means ± SE. Data were analyzed using STATISTICA 7.1 (StatSoft, Tulsa, OK). Significance was achieved when P < 0.05. MALDI data were analyzed using one/two-way ANOVA. Modified Tukey post hoc test was used to compare means.

RESULTS

Optimization of myocardial enzyme assay using MALDI imaging.

Mouse hearts were sectioned transversally in the ventricular area. The different anatomical regions and chambers in the transverse myocardial sections are shown in Fig. 1A. Optimal conditions for myocardial ANG II metabolism were determined using a combination of different concentrations and incubation time periods for the substrate (ANG II) (Fig. 1). Range of substrate concentration as well as the incubation time were chosen based on previous reports of the enzyme assay (17). Two peaks were predominant at the end of the incubation period: m/z 931 and m/z 899, which correspond to ANG III and ANG-(1–7), respectively (Fig. 1, B–E). The optimal peak intensities for the metabolites were seen after 10 min of incubation with 100 μmol/l substrate concentration (Fig. 1, F and G).

Fig. 1.

Matrix-assisted laser desorption ionization (MALDI) imaging of ANG II metabolism in the myocardium. A: transverse myocardial section stained with hematoxylin and eosin (H and E) showing different anatomical areas within the myocardium. B: photographic image of a myocardium section before incubation with ANG II for spatial visualization of the generated peptides. C: MALDI imaging of ANG III [mass-to-charge (m/z) 931] generated in myocardium. D: MALDI imaging of ANG-(1–7) m/z 899 generated in the myocardium. E: overlay of ANG III and ANG-(1–7) signals. F: ANG-(1–7) formation after incubation with ANG II (10–1,000 μmol/l) at 5, 10, and 15 min. G: ANG III formation after incubation with ANG II (10–1,000 μmol/l) at 5, 10, and 15 min. Data are means ± SE. LV, left ventricle; RV, right ventricle; LVW, left ventricular wall; IVW, interventricular septal wall.

Validation of the formed peptides by MS/MS.

To validate m/z 931 and m/z 899 generated in the myocardium after incubation with ANG II, the chemical nature of the peptides formed was confirmed by MALDI-TOF/TOF as described before (17). Results showed identical spectra between the standard ANG-(1–7) and the peptide m/z 899 formed in the enzyme assay (Fig. 2, A and B). Similar to m/z 899, the MS/MS for m/z 931 formed in the enzyme assay was identical to that obtained for the standard ANG III (Fig. 2, C and D). These results are consistent with previous data which verified ANG III and ANG-(1–7) as the two main metabolites formed in the in situ enzyme assay of ANG II (17).

Fig. 2.

MALDI TOF/TOF of m/z 931 and m/z 899 formed in situ in the myocardium after incubation with ANG II. A: MS/MS of ANG-(1–7) standard. B: MS/MS of the product m/z 899 generated in the myocardium. C: MS/MS of ANG III standard. D: MS/MS of the generated product m/z 931. E: reproducibility of ANG III and ANG-(1–7) formation within the myocardium. There was little variation for ANG III and ANG-(1–7) generation in the myocardial sections within the same animal (intra-animal) or between animals (interanimal).

Reproducibility of cardiac ANG II metabolism.

The incubation conditions for the ANG II in situ enzyme assay were applied to myocardial sections obtained from different animals within the same plane of sectioning (interanimal variation) or from the heart of one animal within different sectioning planes (intrasectioning variation). A 3-mm segment within the ventricular area of the heart was excised as described in materials and methods. Results showed consistent peptides intensities with minimal variation, which indicates the reproducibility of the myocardial MALDI imaging enzyme assay (Fig. 2E).

Inhibitor studies.

To confirm that the produced peptides are from enzymatic reactions, we tested the peptide formation using specific enzyme inhibitors. ANG-(1–7) formation from ANG II was inhibited in a dose-dependent manner using MLN-4760, a specific ACE2 inhibitor, (Fig. 3, A and B). In addition, ANG III formation from ANG II was inhibited in a concentration-dependent manner using the selective APA inhibitor, 4-APBA (Fig. 3, C and D). Interestingly, the MALDI imaging of ANG II studies showed that inhibition of myocardial APA resulted in a significant increase of ANG-(1–7) signal at a concentration of 10 and 100 μmol/l (Fig. 4, A and B). Similarly, MS-MALDI in vitro enzyme assay using ANG II as the substrate and myocardial homogenate revealed an accumulation of ANG-(1–7) when APA was inhibited with 4-APBA (Fig. 4C).

Fig. 3.

Inhibition of ANG-(1–7) and ANG III formation by specific enzyme inhibitors. A: ANG-(1–7) formation was inhibited by angiotensin-converting enzyme 2 (ACE2) inhibitor MLN-4760 in a concentration-dependent manner. B: quantitation of generated ANG-(1–7) after treatment with MLN-4760. C: ANG III formation was inhibited by the selective aminopeptidase-A (APA) inhibitor, 4-amino-phosphonobutyric acid (4-APBA). D: quantitation of ANG III formation after treatment with 4-APBA. Data were analyzed using one-way ANOVA. Modified Tukey post hoc test was used to compare means. Values are means ± SE (*P < 0.05 vs. 0 μmol/l; n = 3).

Fig. 4.

Increase of ANG-(1–7) formation after APA inhibition. A: ANG-(1–7) formation after incubation of myocardial sections with 100 μmol/l ANG II and 0, 1, 10, 100 μmol/l APA inhibitor 4-APBA. B: quantitation of ANG-(1–7) formation after APA inhibition using the MALDI imaging approach. Data were analyzed using one-way ANOVA. Modified Tukey post hoc test was used to compare means (*P < 0.05 vs. 0 μmol/l; n = 3 per group). C: quantitation of ANG-(1–7) formation before and after treatment with 50 μmol/l 4-APBA using the in vitro MALDI approach. Myocardial homogenates were incubated with 50 μmol/l ANG II for 30 min in MES buffer (pH = 6.5) at 37°C. The formed peptide was quantitated based on ANG-(1–7) stable-isotope-labeled standard. Data were analyzed using Student's t-test. Values are means ± SE (*P < 0.05; n = 3/group).

Role of ACE2 in the APA-mediated increase of ANG-(1–7).

The role of ACE2 in myocardial ANG-(1–7) formation upon APA inhibition was tested. Results showed that APA inhibition blocked ANG II metabolism in WT and ACE2 KO mice (ACE2 KO) (Fig. 5A). ANG III formation was lowered to the same extent in both strains (Fig. 5B). Although ACE2 deletion decreased the generation of ANG-(1–7) compared with WT at baseline, it did not curtail the elevation of ANG-(1–7) signal after inhibition of APA (Fig. 5C). These results implied that APA plays an essential role in ANG-(1–7) degradation regardless of ANG-(1–7) generated by ACE2 and suggests that inhibition of APA would significantly prolong the half-life of ANG-(1–7).

Fig. 5.

Role of ACE2 in APA-mediated increase of ANG-(1–7). A: intensity of the substrate ANG II after incubation with 100 μmol/l ANG II (baseline) or with 100 μmol/l ANG II and 100 μmol/l 4-APBA (4-APBA). B: ANG III generation in wild-type (WT) and ACE2 knockout (KO) after incubation with 100 μmol/l ANG II (baseline) or with 100 μmol/l ANG II and 100 μmol/l 4-APBA (4-APBA) using the in situ MALDI assay. C: ANG-(1–7) generation in WT and ACE2 KO after incubation with 100 μmol/l ANG II (baseline) or with 100 μmol/l ANG II and 100 μmol/l 4-APBA (4-APBA) using the in situ MALDI assay. Data were analyzed using two-way ANOVA. Modified Tukey post hoc test was used to compare means. Values are means± SE (*P < 0.05 vs. WT, †P < 0.05 vs. baseline; n = 4–5 per group).

ANG-(1–7) is metabolized by APA to ANG-(2–7).

The activity of APA on ANG-(1–7) and ANG II was tested using an MS enzyme assay as described earlier with some modifications (7). Recombinant APA (rAPA) was used to test APA activity on ANG-(1–7) as a substrate (Fig. 6, A–E) compared with ANG II (Fig. 6, F–J). Results showed that ANG-(1–7) and ANG II were both metabolized with high specificity by APA. The metabolic products that resulted from the enzyme reactions (m/z 784 and m/z 931) were identified as ANG-(2–7) and ANG III, respectively (Fig. 6, B and G). In both cases, the products were not detected in the absence of rAPA (Fig. 6, A and F) but accumulated with increasing concentrations of the recombinant enzyme in the reaction mixture (Fig. 6,C and H). Furthermore, both ANG-(2–7) and ANG III formations were inhibited by 4-APBA in a dose-dependent manner (Fig. 6, D, E, I, J). These results show that the formed ANG-(2–7) is the metabolic product from the actions of APA on ANG-(1–7).

Fig. 6.

In vitro metabolism of ANG-(1–7) and ANG II by APA. Recombinant APA (rAPA) enzyme was incubated with ANG-(1–7) or ANG II for 30 min at 37°C in assay buffer. A: ANG-(1–7) in assay buffer. B: ANG-(1–7)+ 1 ng rAPA. C: ANG-(1–7)+ 5 ng rAPA. D: ANG-(1–7) + 1 ng APA + 0.3 nmol/l 4-APBA. E: ANG-(1–7) + 1 ng APA + 3 nmol/l 4-APBA. F: ANG II in assay buffer. G: ANG II + 1 ng rAPA. H: ANG II + 5 ng rAPA. I: ANG II + 1 ng APA + 0.3 nmol/l 4-APBA. J: ANG II + 1 ng APA + 3 nmol/l 4-APBA.

ANG-(1–7) is metabolized by APA with similar affinity as ANG II.

The affinity of APA to ANG-(1–7) and to ANG II was tested using a fixed amount of the enzyme with different substrate concentrations to determine Michaelis-Menten enzyme kinetics. Results showed a similar affinity constant for APA to ANG-(1–7) (K_m = 6.07 ± 1.12) as to ANG II (K_m = 14.67 ± 1.6) (Fig. 7, A and B). These results suggested that APA has a similar tendency to metabolize ANG-(1–7). This newly discovered enzymatic characteristic of APA could be further exploited toward approaches prolonging the half-life of ANG-(1–7) in vivo.

Fig. 7.

Michaelis-Menten enzyme kinetics for APA and the substrates ANG II and ANG-(1–7) using the in vitro MALDI enzyme assay. A: rAPA (1 ng) was incubated with 1–90 μmol/l ANG II for 30 min at 37°C. B: rAPA (1 ng) was incubated with 1–90 μmol/l ANG-(1–7) for 30 min at 37°C. Results show similar binding affinities for APA to ANG-(1–7) (K_m = 6.07 ± 1.12) and ANG II (K_m = 14.67 ± 1.6). Values are means ± SE; n = 3 per group.

Increased APA activity and expression in MI.

To investigate the role of APA in ANG-(1–7) metabolism in pathology, we conducted MALDI imaging enzyme assays in mice myocardium after MI (24 h) or sham surgery. The infarction was confirmed using TTC staining (Fig. 8A). Higher APA protein level was detected in the mice hearts with LAD ligation compared with heart from sham-operated animals (Fig. 8B). Using the in situ MALDI assay with ANG II or ANG-(1–7) as the substrate, a significantly increased APA activity was detected in MI sections compared with sham (Fig. 8, C–G). Overall, ANG III and ANG-(2–7) were significantly higher in the infarct and remote area after MI although elevated product formation was mainly seen in the infarct areas (Fig. 8, C–F). This increase in APA activity was confirmed using the in vitro assay on myocardial homogenate obtained after MI or sham surgery (Fig. 8, H and I).

Fig. 8.

MALDI imaging after myocardial infarction (MI) and sham operation. A: representative 2,3,5-triphenyltetrazolium chloride (TTC) staining performed 24 h post-MI. Scale bar, 1 cm. (*infarct region). B: Western blot of APA of myocardial segments (a 3-mm myocardial segment was excised from the ventricular area of the heart at the plane of the papillary muscles) after MI or sham operation. APA expression was increased in MI compared with sham. (*P < 0.05; n = 3/group). C: in situ MALDI imaging assay of ANG III formed on a myocardial section after sham surgery using ANG II as a substrate. D: in situ MALDI imaging assay of ANG III on a myocardial section after MI surgery using ANG II as a substrate. The infarct area (*) is traced in red. E: in situ MALDI imaging assay of ANG-(2–7) formed on a myocardial section after sham surgery using ANG-(1–7) as a substrate. F: in situ MALDI APA imaging assay of ANG-(2–7) formed on a myocardial section after MI using ANG-(1–7) as a substrate The infarct area (*) is traced in red. G: quantitation of formed ANG III and ANG-(2–7). Data were analyzed using two-way ANOVA. Modified Tukey post hoc test was used to compare means. Values are means ± SE (*P < 0.05 vs. sham; n = 6/group). Signals were quantitated as integrated intensity and normalized to the area of the whole heart section. H: quantitation of the in vitro MALDI enzyme assay using homogenates from sham and MI hearts with ANG II as substrate. I: quantitation of the in vitro MALDI enzyme assay using homogenates from sham and MI hearts with ANG-(1–7) as substrate. Data were analyzed using Student's t-test. Values are means ± SE (*P < 0.05; n = 3–4/group).

DISCUSSION

This study investigated the role of APA in the myocardium using novel mass spectrometry-based strategies developed in our laboratory. Utilizing the merits of MALDI imaging, we report for the first time that APA is expressed in the myocardium. More importantly, this is the first study to reveal a previously unknown effect of APA in ANG-(1–7) metabolism. Our results highlighted catalytic efficiency of APA to metabolize ANG-(1–7) with a similar efficiency to that of ANG II. In addition, our results showed a marked increase in APA activity in the myocardium of mice after MI compared with sham. Therefore, we propose that an elevation of APA activity in myocardial injury could be responsible for enhanced ANG-(1–7) degradation which could be culminated into myocardial remodeling.

MS is one of the most accurate techniques for protein and peptide analysis. MS has been used to assess peptides as well as the enzymatic components of the renin-angiotensin system (7, 8). The main advantage of MS over other analytical techniques is that it allows for the use of the natural substrate and accurate peptide quantitation using isotopic peptide standards. In our previous studies, we have shown selective spatial distribution patterns of ANG II metabolites in the kidney, suggesting regionally localized formation of ANG III and ANG-(1–7) (17, 18). For the heart, our data showed colocalization of both metabolites in most of the myocardial regions.

APA is expressed in several organs and has been shown to be activated in CV diseases (3, 19, 31, 43, 49). High APA activity was found in rat brain after MI (20). In addition, it has been reported that spontaneous hypertensive rats exhibit higher APA activity in the brain and kidney (13, 19, 50). Our results from the in vitro APA assays showed that MI hearts exhibit a more than twofold increase in APA activity. The in situ assay confirmed that ANG III formation and APA activity were more pronounced in the ischemic region of the heart compared with sham. Increased peptide formation was also observed in remote areas, which could be due to postinfarct remodeling and infiltration of cardiac fibroblasts (4). Indeed, a contribution of APA activity from cardiac fibroblasts to the overall activity of APA in the heart has been shown in previous studies localizing APA activity in both cardiac fibroblasts and cardiomyocytes (42).

The upregulation of APA after myocardial pathology suggests that it could be a major pathway for ANG-(1–7) degradation. It is worth noting that most of the previous studies attributed the enhancement in hemodynamic outcome upon APA inhibition to the decreased levels of ANG III as the major effector peptide of the brain (2, 3, 13, 26, 33, 48). Presented herein is evidence that APA is also capable of degrading ANG-(1–7) to ANG-(2–7) with an affinity comparable to the substrate ANG II. ANG-(1–7) has been shown to ameliorate neurogenic (10) and ANG II-induced hypertension (39), as well as baroreflex sensitivity in hypertensive rat models (21). In the periphery, ANG-(1–7) facilitates cardiac repair and ventricular function after MI (27, 32, 46, 47). Therefore, it is predictable that the beneficial effects of APA inhibition are not solely due to decreased ANG III, but rather a combination of reduced ANG III and increased ANG-(1–7) levels. Until now, there has been little known about the in vivo effects of ANG-(2–7), the metabolic product resulting from ANG-(1–7) degradation by APA. One study documents a mild pressor effect of ANG-(2–7) in humans (23), which would add another beneficial action of APA inhibition.

ACE2 has been reported as the main enzyme responsible for ANG-(1–7) formation in the heart (1, 15, 51). Recently, Huang et al. (20) showed that central treatment with the APA inhibitor (RB150) in rats post-MI during 28 days attenuates cardiac dysfunction and suggested that this effect is not only due to the blockade of the formation of brain ANG III but also due to an increased metabolism of brain ANG II to ANG-(1–7) by ACE2. To test the possibility that inhibition of APA might cause an increase in ANG-(1–7) levels indirectly by accelerating its generation form ANG II via ACE2, we used heart sections from ACE2 KO mice. As expected, there was a lower ANG-(1–7) generation in myocardium tissue from ACE2 KO mice at the basal level. However, after APA inhibition, ANG-(1–7) formation was not significantly different in hearts of ACE2 KO compared with WT animals. Collectively, these data suggest that accumulation of ANG-(1–7) in the presence of 4-APBA is mainly due to blocking APA activity. Alternatively, other compensatory mechanisms may be activated during inhibition of APA and ACE2, such as ACE2-independent metabolic pathways for ANG-(1–7) formation via prolyl carboxypeptidase (18) or neprilysin (45).

In conclusion, APA inhibitors could constitute a novel class of therapeutics for the treatment of heart failure patients by acting in the brain as well as directly in the heart.

GRANTS

This work was supported by Fulbright Scholarship (M. S. Alghamri) and National Institutes of Health Grant R01-HL-093567 (K. M. Elased and M. Morris) and fellowship F32-DK-093226 (N. Grobe).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: M.S.A. conception and design of research; M.S.A. and G.M. performed experiments; M.S.A. and N.G. analyzed data; M.S.A., K.M.E., and N.G. interpreted results of experiments; M.S.A. and N.G. prepared figures; M.S.A., M.M., and N.G. drafted manuscript; M.S.A., M.M., G.M., K.M.E., and N.G. edited and revised manuscript; M.S.A., M.M., G.M., K.M.E., and N.G. approved final version of manuscript.