OUP user menu

Cardiac chymase converts rat proAngiotensin-12 (PA12) to angiotensin II: effects of PA12 upon cardiac haemodynamics

Hamish C.G. Prosser, Malcolm E. Forster, A. Mark Richards, Chris J. Pemberton
DOI: http://dx.doi.org/10.1093/cvr/cvp003 40-50 First published online: 15 January 2009


Aims The aim of this study was to observe the direct physiological and biochemical cardiac effects in response to a newly identified putative component of the renin–angiotensin system, proangiotensin-12 (PA12); and investigate whether PA12 can serve as a substrate for Angiotensin II (AngII) generation.

Methods and results The direct cardiac actions of PA12 and its role as a substrate for chymase-dependent AngII generation were investigated in Sprague–Dawley rats using an isolated heart model of cardiac ischaemia–reperfusion injury. PA12 potently constricted coronary arteries with no significant effect on left-ventricular contractility. PA12 impaired recovery from global ischaemia, maintaining coronary constriction and markedly increasing release of creatine kinase and troponin I (TnI), indicating greater myocardial injury. Analysis of perfusate collected after transcardiac passage revealed a marked increase in AngII production from hearts infused with PA12. Cardiac AngII production was not blocked by angiotensin-converting enzyme inhibitors, whereas inhibition of chymase with chymostatin significantly reduced AngII production and attenuated PA12-induced vasoconstriction and myocardial damage following ischaemia. Furthermore, Angiotensin II type 1 receptor (AT1R) blockade abolished PA12 activity. In vitro, PA12 was efficiently and precisely converted to AngII as assessed on reverse phase-high performance liquid chromatography coupled to tandem mass spectrometry. This conversion was blocked by chymostatin.

Conclusion PA12 may act as a circulating substrate for cardiac chymase-mediated AngII production, in contrast to ACE-mediated AngII production from AngI.

  • ProAngiotensin-12
  • Langendorff isolated rat heart
  • Ischemia-reperfusion injury
  • Cardiac angiotensin II production
  • Chymase

1. Introduction

Knowledge of the renin–angiotensin system (RAS) began with the initial discovery of renin in 1898, an enzyme shown to significantly alter blood pressure.1 Since then exploration into the RAS has found it to be one of the most important homeostatic systems in pressure–volume homeostasis. Renin released from the kidney in response to reductions in renal perfusion, acts on the angiotensin-precursor peptide angiotensinogen, cleaving it to release the biologically inactive decapeptide Angiotensin1–10 (AngI). Angiotensin-converting-enzyme 1 (ACE1) cleaves a further two amino acids from the C-terminus of AngI to produce Angiotensin1–8 (AngII), a potent biologically active peptide triggering a broad range of effects throughout the body including aldosterone secretion, salt and water retention, and potent arteriolar vasoconstriction.1 Biologically active AngII-derived peptides have been isolated, including AngIII (Ang2–8) and AngIV (Ang3–8), suggesting angiotensinogen is a precursor to many functional peptides.2 AngII (and its active derivative peptides) binds to two known G-protein coupled receptors, AngII receptor Types 1 and 2 (AT1 and AT2). AT1 is expressed abundantly by most cell types throughout the body and is the primary receptor mediating the pressor response to AngII.3 AT2 is far less abundant in the adult and has been suggested to stimulate vasodilation, counteracting the pressor and other effects of AT1.3,4

Proangiotensin-12 (PA12) is a newly discovered peptide believed to be an immediate proteolytic fragment of angiotensinogen. PA12 was isolated from the rat small intestine where it was most abundant, but it is also present in the circulation and throughout many tissues and organs of the rat.5 The sequence of PA12 is identical to that of Angiotensin I (AngI) with an extended C-terminus, thus comprising Ang1-10-Leu11–Tyr12. To date, a single report has documented that PA12 can constrict rat aorta in vitro and elevate blood pressure in anaesthetized normotensive Wistar rats in vivo.5 Whether PA12 is present in human (or other mammalian) blood or tissue is unknown. It has been suggested the effects of PA12 are dependent upon both ACE1 and AT1R activities, and that PA12 levels are elevated in hypertensive rat hearts.6 However, whether PA12 has direct cardiac actions is not known, nor whether PA12 can serve as a substrate for generation of AngII or other AngII-related peptides.

Accordingly, we provide the first documentation of direct haemodynamic and endocrine effects of PA12 on the isolated, perfused rat heart, and report its effects in a model of cardiac ischaemia–reperfusion (I–R) injury. We also provide the first evidence that PA12 is converted to AngII in vitro and ex vivo by enzymatic chymase activity, and provide a comparison with AngI under the same protocols.

2. Methods

2.1 Materials

Male Sprague–Dawley (SD) rats weighing 300–400 g were obtained from the Christchurch Animal Research Facility, University of Otago, New Zealand. Rats were housed under controlled temperature (21°C), humidity (∼40%), and natural day length with free access to standard rat chow and water.

Synthetic rat PA12 was obtained from Phoenix Pharmaceuticals (Belmont, CA, USA), while AngI and the ACE1 antagonists Captopril and Ramipril, as well as the chymase inhibitor Chymostatin were all obtained from Sigma-Aldrich (St Louis, MI, USA). The angiotensin II receptor type 1 blocker, CV-11974 (Candesartan, 2-ethoxy-1-[[2′-(1H-tetorazol-5-yl)biphenyl-4-yl]methyl]-1H-benzimidazole-7-carboxylic acid)h was a generous gift from Takeda Chemical Industries Ltd (Osaka, Japan). PA12 and captopril were diluted in distilled water, aliquoted, and stored at −20°C prior to use. Ramipril and chymostatin were dissolved in DMSO, while CV-11974 was dissolved in 1 M Na2CO3 solution, aliquoted, and stored at −20°C prior to use.

2.2 Langendorff isolated rat heart perfusion

Isolated rat heart perfusion was performed as previously described.7 Briefly, rats were anaesthetized with sodium pentobarbital (50 mg/kg, i.p.). The heart was rapidly excised and mounted on the Langendorff apparatus, cannulated above the aortic valve and perfused at 12 mL/min (constant retrograde flow) with perfusion buffer comprising (mmol/L): 123 NaCl, 22.0 NaHCO3, 4.7 KCl, 1.2 KH2PO4, 1.1 MgSO4·7H2O, 1.5 CaCl2·2H2O, and 11.0 glucose (final pH 7.40). Buffer was maintained at 37°C and oxygenated with 95% O2/5% CO2. The left atrium was removed allowing a 40% ethanol-filled balloon (attached to a pressure transducer) to be inserted through the mitral valve into the left ventricle (LV) enabling measurement of left ventricular haemodynamic contractile parameters. A side-arm cannula attached to a second pressure transducer was inserted into the aortic cannula above the heart to measure perfusion pressure (PP), an indirect measure of coronary arterial tone. Hearts were allowed to settle for 30 min before being paced at 320 b.p.m. using an electrode attached to a Digimeter DS2A-Mk. II stimulator placed on the right atrium. Hearts were allowed a further 30 min to resettle before any experimental protocol was started. All data were recorded using a Powerlab Chart 5 System (ADInstruments). Drugs were diluted using perfusion buffer to enable infusion of the drug or vehicle over 30 min at 0.5 mL/min using a syringe pump feeding directly into the perfusion line. This investigation was approved by the University of Otago Animal Ethics Committee and conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996).

2.3 Ischaemia–reperfusion protocols

All antagonist agents (100 µmol/L captopril, 100 nmol/L CV-11974, 100 nmol/L chymostatin, and ramipril) were co-infused with either 10 nmol/L PA12 or 10 nmol/L AngI. Experiments were run in pairs: one heart infused with both the antagonist and PA12 or AngI, the other simultaneously infused only with PA12 or AngI from the same stock solution and same buffer reservoir providing a parallel control for each antagonist experiment. The heart given the antagonist was chosen at random.

The doses of 10 nmol/L PA12, 10 nmol/L AngI, 100 nmol/L CV-11974, 100 µmol/L captopril, and 100 nmol/L chymostatin administered were based upon preliminary dose–response studies. After 30 min infusion of agents, perfusion was stopped and hearts underwent no-flow global ischaemia for 45 min with pacing halted. Hearts were then reperfused with accompanying pacing for 105 min following ischaemia (Figure 1).

Figure 1

Ischaemia–reperfusion experimental protocol. Following a settling period, rat hearts underwent preconditioning infusion (30 min), no-flow global ischaemia (45 min), and reperfusion with buffer only (105 min). Arrows indicate haemodynamic and hormonal sampling time points.

2.4 Analysis of cardiac AngII, atrial natriuretic peptide, myocardial creatine kinase, and troponin I release

Perfusate samples were collected at specific time points after passing through the heart for hormonal and biochemical analysis (Figure 1). Samples of infusion solution from the syringe were also taken prior to passing through the heart. Angiotensin II production was determined by extracting perfusate samples (collected into 500 IU Aprotinin/0.1% Triton X-100) through Bond Elut columns and measured using our previously described, specific radioimmunoassay (RIA).8 This assay shows no detectable cross-reactivity with PA12 or Angiotensin I.

Atrial natriuretic peptide (ANP) secretion was measured by RIA after extraction through SepPak columns as previously described.9 Perfusate creatine kinase (CK) and troponin I (TnI) concentrations were measured using the Abbot Aeroset platform (Canterbury Health Labs, Christchurch, New Zealand).

2.5 Ex-vivo cardiac conversion of PA12 analysis

To assess the ex vivo cardiac conversion of PA12, we employed reverse phase-high pressure liquid chromatography (RP-HPLC) coupled with specific RIA. RP-HPLC was carried out at 40°C, with a gradient of 0–60% CH3CN/0.1% TFA over 60 min, using a 22 cm Brownlee C18 RP-HPLC column. The gradient elution profile was calibrated with Ang(1–7), AngII, AngI, and PA12 standard peptides. To assess enzymatic cardiac conversion of PA12 after a single cardiac passage, 1 mL samples of perfusate (during infusion of 10 nmol/L PA12) were collected into Aprotinin/Triton and subjected to RP-HPLC as described, with fractions collected at 1 min intervals. Fractions were dried under air at 37°C, reconstituted in AngII RIA buffer and subjected to AngII RIA, as described.8

2.6 In vitro conversion of PA12 by chymase: tandem mass spectrometry analysis

To confirm that chymase could generate authentic AngII from PA12, we employed in vitro incubation of PA12 with recombinant chymase. PA12 (12.72 µmol/L) was incubated either: (a) alone; (b) in combination with recombinant human chymase (3.33 pmol/L); or (c) in combination with both 3.3 pmol/L chymase and 33.4 µmol/L chymostatin. Incubations proceeded for 30 min at 37°C in 50 µL PBS. Reactions were quenched by addition of 2 µL glacial acetic acid (final concentration of 0.6 mol/L), and the reaction products of each submitted to RP-HPLC using the gradient described earlier. Fractions were collected at 1 min intervals and portions submitted to AngII RIA analysis. The remainder of the RP-HPLC fractions were dried under air and those with AngII immunoreactivity and UV detection were submitted to tandem mass spectrometry (MS/MS).

2.7 Tandem mass spectrometry

Peptides were resuspended in 30% [v/v] acetonitrile and 0.1% [v/v] trifluoroacetic (TFA) acid in water. One microlitre of peptide solution was premixed with 2 µL of matrix (10 mg/mL αcyano-4-hydroxycinnamic acid dissolved in 65% [v/v] aqueous acetonitrile containing 0.1% TFA and 10 mmol/L ammonium dihydrogen phosphate). 0.8 µL of sample/matrix mixture were spotted onto a MALDI sample plate (Opti-TOF 384 well plate, Applied Biosystems, MA) and air dried.

Samples were analysed on a 4800 MALDI tandem Time-of-Flight Analyser (Applied Biosystems, MA). All MS spectra were acquired in positive-ion mode with 800–1000 laser pulses per sample spot. The six strongest precursor ions of each sample spot were used for MS/MS collision-induced fragmentation (CID) analysis. CID spectra were acquired with 2000–4000 laser pulses per sample spot using the 2 kV mode and air as the collision gas at a pressure of 1 × 10−7 torr.

For protein identification MS/MS data were searched against the SWISS-PROT amino acid sequence database using the Mascot search engine (matrix science). The search was set-up for full tryptic peptides with a maximum of four missed cleavage sites. Carboxyamidomethyl cysteine, oxidized methionine, and pyroglutamate (E, Q) were included as variable modifications. The precursor mass tolerance was 75 p.p.m. and the maximum fragment mass error 0.3 Da.

2.8 Statistical analysis

All data are presented as mean +SEM. Analysis of changes in cardiac hormones and haemodynamics were performed on SPSS using a two-way ANOVA with repeated measure, with Bonferroni’s multiple comparison test, post hoc. Individual and cumulative data comparisons were made using a Student’s t-test. In all statistical tests, a value of P < 0.05 was considered significant.

3. Results

3.1 PA12 increases coronary perfusion pressure pre- and post-cardiac ischaemia

PA12 dose-dependently increased PP indicating vasoconstriction of the coronary arteries. This effect was maximal at 10 nmol/L and sustained throughout the 30 min infusion period (Figure 2A). 1 and 10 nmol/L PA12 elevated PP by an average of 11 ± 0.7 and 32.3 ± 4.4% (P < 0.05, n = 6 and P < 0.0001, n = 8) respectively, compared with vehicle infusion. PA12 had no significant effect on left-ventricular developed pressure (DP), nor ± dP/dt (ΔDP from 0 to 30 min was 8 ± 2.27 and 9 ± 2.15% for vehicle and PA12, respectively).

Figure 2

PA12 elicited a dose-dependent increase in perfusion pressure (PP), significant at 1 and 10 nmol/L compared with vehicle (A, n = 8). PA12 activity was unaffected by the ACE1 antagonist captopril (B, n = 5), was abolished by the angiotensin II type 1 receptor blocker CV-11974 (C, n = 6), and was significantly attenuated by chymostatin (D, n = 5). Infusion of 10 nmol/L AngI elevated PP equal to that of 10 nmol/L PA12. AngI activity was significantly inhibited by captopril, while chymostatin caused minimal attenuation (E, n = 5). (**P < 0.05, ***P < 0.005, bar indicates infusion period).

Infusion of 10 nmol/L AngI significantly increased PP (P < 0.05, n = 5), comparable to that of 10 nmol/L PA12 (Figure 2E), and also showed no significant effects on DP or ± dp/dt (ΔDP from 0 to 30 min was 9 ± 1.30 and 9 ± 3.41% for vehicle and AngI, respectively).

Infusion of captopril alone (an ACE1 inhibitor) caused a significant drop in PP compared with vehicle (P < 0.005, n = 5). However, when co-infused with PA12, 100 µmol/L captopril had no effect on PA12-induced increase in PP with levels comparable to that of PA12 infused alone (Figure 2B). This lack of response of PA12 to ACE1 inhibition was confirmed by infusion of a second ACE1 inhibitor ramipril across a dose range of (0.01–1 µmol/L; data not shown). Captopril abolished the AngI-induced elevation in PP significantly below both vehicle and AngI levels (both P < 0.01, n = 5, Figure 2E). The gradual increase in PP observed during the co-infusion was equal to that of infusion of captopril alone (see Figure 2B).

The AT1R antagonist, CV-11974, completely abolished PA12-induced elevations in PP (P < 0.001, n = 6), suggesting the AT1 receptor can mediate PA12 activity (Figure 2C).

Infusion of the chymase inhibitor chymostatin at the dose employed had no effect on PP or LV contractility. However co-infusion of chymostatin with PA12 significantly attenuated PA12-induced increases in PP by an average of 63.9 ± 1.59% (P < 0.001, n = 5, Figure 2D). In contrast, co-infusion of chymostatin with AngI caused minimal attenuation of AngI-induced increases in PP during infusion, remaining significantly above vehicle (P < 0.003, Figure 2E).

3.2 Ischaemia–reperfusion

In ischaemia–reperfusion, hearts preconditioned with 10 nmol/L PA12 showed significantly elevated PP above vehicle levels throughout the reperfusion period (P < 0.05, Figure 3A). Hearts preconditioned with 10 nmol/L AngI also showed elevated PP above vehicle levels, significant at 120 and 180 min (both P < 0.05, n = 5, Figure 3D). PA12 and AngI had no significant effect on DP, ± dp/dt, systolic or diastolic pressure during reperfusion.

Figure 3

Perfusion pressure following global ischaemia in isolated hearts preconditioned with 10 nmol/L PA12 (n = 8) or 10 nmol/L AngI (n = 5). 100 µmol/L captopril had no effect on PA12-induced elevations in PP during reperfusion (A), however 100 nmol/L CV-11974 (B) and 100 nmol/L chymostatin (C) both attenuated the PA12-induced increase. AngI caused minimal elevation of PP immediately following ischaemia before rising above control levels (P < 0.05 at 120 and 180 min, D). Captopril significantly attenuated the AngI-induced elevation in PP throughout reperfusion; while chymostatin displayed minimal inhibitive potency, significant for the initial 10 min of reperfusion (D). +Individual data point comparison P < 0.05; **Data set comparison P < 0.05.

Hearts preconditioned with captopril alone displayed significantly reduced PP throughout reperfusion (P < 0.05, Figure 3A). Captopril attenuated PA12-induced elevations in PP immediately following ischaemia, before antagonistic potency was lost and PP increased to equal PA12 alone (Figure 3A). In contrast, captopril significantly attenuated AngI-induced elevations in PP during reperfusion (P < 0.05 compared with AngI alone, Figure 3D).

Co-infusion of CV-11974 with PA12 abolished PA12-induced increases in PP during reperfusion (P < 0.005, Figure 3B).

Chymostatin attenuated the PA12-induced increase in PP following ischaemia, maintaining PP below that of hearts preconditioned with PA12 alone (P < 0.05 at 120 and 180 min, Figure 3C). Co-infusion of chymostatin with AngI saw a reduction in PP immediately following ischaemia when compared with hearts preconditioned with AngI alone (P < 0.05 at 77, 80, and 85 min, Figure 3D).

During reperfusion hearts preconditioned with 10 nmol/L PA12 or 10 nmol/L AngI released significantly more CK and TnI (both P < 0.05) than control hearts, indicating greater myocardial injury (Figures 4 and 5). The addition of captopril had no effect on PA12-induced CK or TnI release (Figures 4A and 5A); however, both CV-11974 and chymostatin significantly blocked PA12-induced elevations in both CK and TnI (Figures 4B and C and 5B and C). In contrast to PA12, captopril significantly attenuated AngI-induced elevations in CK and TnI release when compared with hearts preconditioned with AngI alone (both P < 0.05, Figures 4D and 5D); however chymostatin could not (both P > 0.05, n = 5, Figures 4D and 5D).

Figure 4

Cumulative creatine kinase release in PA12 preconditioned hearts was significantly higher compared with vehicle controls, an affect which was not attenuated by 100 µmol/L captopril (A), but was significantly abolished by 100 nmol/L CV-11974 (B) and 100 nmol/L chymostatin (C). Hearts preconditioned with 10 nmol/L AngI produced cumulative creatine kinase release significantly higher than vehicle controls (D); this increase was significantly attenuated by 100 µmol/L captopril, but not by 100 nmol/L chymostatin. **P < 0.05 compared with vehicle or PA12, as indicated by bars.

Figure 5

Cumulative myocardial TnI release in PA12 preconditioned hearts was significantly higher compared with vehicle controls, an affect which was not attenuated by 100 µmol/L captopril (A), but was significantly abolished with 100 nmol/L CV-11974 (B) and chymostatin (C). Hearts preconditioned with 10 nmol/L AngI produced cumulative TnI release significantly higher than control during reperfusion (D); this increase was significantly attenuated by 100 µmol/L captopril, but not by 100 nmol/L chymostatin. **P < 0.05 compared with vehicle or PA12, as indicated by bars.

3.3 Perfusate atrial natriuretic peptide levels during ischaemia–reperfusion

Hearts preconditioned with either 10 nmol/L PA12 or 10 nmol/L AngI showed no significant difference in perfusate ANP concentrations compared with vehicle throughout both infusion and reperfusion. Infusion of 100 µmol/L captopril alone reduced ANP secretion during infusion and significantly during reperfusion (P < 0.05), as well as attenuating both PA12- and AngI-induced ANP release (P < 0.05). CV-11974 and chymostatin had no influence on ANP secretion with or without the presence of PA12 or AngI.

3.4 PA12 is cleaved into Angiotensin II during cardiac passage: dependence on chymase, but not angiotensin-converting-enzyme 1

An average of 457.2 ± 113.2 pmol/L and 651.3 ± 79.73 pmol/L IR-AngII was measured from perfusate samples collected at 10 min into infusion from hearts preconditioned with 10 nmol/L PA12 and 10 nmol/L AngI, respectively, both significantly above that of vehicle (8.15 ± 7.75 pmol/L, P < 0.001, Figure 6A and B). PA12-derived AngII formation was not attenuated by captopril co-infusion (508.0 ± 169.5 pmol/L, P = NS vs. PA12 alone), but was almost completely abolished by chymostatin co-infusion (60.62 ± 8.71 pmol/L, Figure 6A), reducing IR-AngII by 87%. In contrast, both captopril and chymostatin significantly inhibited perfusate concentrations of IR-AngII collected from hearts infused with AngI by 92 and 81%, respectively (P < 0.001 vs. AngI alone, Figure 6B). (IR-AngII within the drug-infusion syringe was non-detectable).

Figure 6

Perfusate IR-AngII from rat hearts infused with 10 nmol/L PA12 (A) or 10 nmol/L AngI (B), and effects of supra-added 100 µmol/L captopril and 100 nmol/L chymostatin. (C) Perfusate samples collected from hearts infused with PA12 + captopril were subjected to reverse phase-high performance liquid chromatography [elution gradient 0–60% CH3CN over 60 min (1 mL/min)]. Eluted fractions were collected and measured for immunoreactive-AngII (IR-AngII) via specific RIA and overlapped with known elution times of angiotensin peptides (indicated by arrows; ***P < 0.005).

The presence of authentic AngII within the cardiac perfusate was confirmed by subjecting perfusate samples to RP-HPLC/RIA. Hearts preconditioned with PA12 alone and PA12 with captopril combined produced a peak of immunoreactive AngII that eluted consistent with synthetic standard (Figure 6C), indicating captopril was incapable of abolishing the conversion of PA12 to AngII in the heart. In contrast, co-infusion of PA12 with chymostatin significantly attenuated AngII generation (Figure 6A).

3.5 Chymase converts PA12 to Angiotensin II in vitro: high performance liquid chromatography tandem mass spectrometry analysis

In vitro incubation of PA12 with chymase confirmed chymase is capable of converting PA12 into authentic AngII. Thus, 30 min incubation of PA12 alone in PBS yielded on RP-HPLC/UV a single peak corresponding with PA12 standard (Figure 7A, solid line). Incubation of chymase with PA12 resulted in several UV peaks, two of which correlated exactly with AngII and PA12 standards (Figure 7B). A third minor peak also was observed at 27 min, consistent with Ang(1–7) (Figure 7B). The addition of chymostatin abolished all peaks bar PA12 on UV. RIA analysis of UV fractions from HPLC revealed that incubation of PA12 alone contained barely detectable IR-AngII (Figure 7A). In contrast, incubation of chymase with PA12 produced significant IR-AngII (Figure 7B, dashed line). Addition of chymostatin abolished all IR-AngII (data not shown). MS/MS analysis of the IR/UV AngII peak in Figure 7B confirmed that it was authentic AngII, with a confirmed Mr of 1046.4899 (Figure 7C). MS/MS analysis of a single UV peak in Figure 7A, gave a Mr of 1572.7407, identical with that of complete PA12 (Figure 7D).

Figure 7

Reverse phase-high performance liquid chromatography and tandem mass spectrometry analysis of PA12 incubated with chymase at 37°C for 30 min in PBS. Each treatment: PA12 alone (A), and PA12 + chymase (B), underwent reverse phase-high performance liquid chromatography with UV detection (solid line, right-hand axis), and specific AngII RIA for AngII (dotted line, left-hand axis). Arrow indicates AngII peak on the UV profile corresponding to the IR-AngII peak (B). Tandem mass spectrometry performed on treatment ‘PA12+chymase’ confirmed the presence of both AngII (C) and PA12 (D). Chymostatin abolished all detectable AngII when incubated with PA12 and chymase combined (data not shown).

4. Discussion

PA12 is a newly isolated peptide, identified as a potential component of the RAS by Nagata et al.5 in late 2006, who reported that PA12 dose-dependently constricted rat aorta segments and significantly elevated blood pressure in vivo, via ACE1- and AT1 receptor-dependent mechanisms. There have been no reports describing the direct cardiac actions of PA12, and only a single report has described the conversion of PA12 to AngII by the heart.10 Accordingly, we provide in this report the first evidence that:

  1. PA12 infusion induces potent, dose-dependent vasoconstriction in rat coronary arteries, without significant effect on LV contractility.

  2. PA12 significantly worsens ischaemia–reperfusion injury.

  3. PA12 actions are blocked by AT1R and chymase inhibition, but not ACE1.

  4. Chymase effectively processes PA12 to authentic AngII ex vivo and in vitro.

The cardiac vasoconstrictive actions of PA12 were approximately equal to that of AngII, and lasted the entire period of infusion and reperfusion. We observed no effects of PA12 upon LV contractility, nor any effect with infusion of 10 nmol/L AngI. This may suggest that PA12 actions are restricted to the vascular surface, and be minimal or absent in cardiac myocytes; or, that its effects were blunted in our single pass Langendorff setup. However, AngI also had no effect upon LV contractility and coupled with previous reports suggesting AngI and AngII exert inotropic effects in atrial but not ventricular myocardium,11,12 it is possible AngI and PA12 possess limited ability to contract LV tissue.

In a model of ischaemia–reperfusion injury, we found preconditioning the heart with PA12 significantly elevated coronary PP following ischaemia, and increased both CK and TnI release. PA12 also maintained cardiac ANP production slightly above that of hearts preconditioned with vehicle during both infusion and reperfusion. Taken together these results indicate PA12 impairs recovery from global ischaemia in the rat heart by reducing coronary flow and causing increased myocardial damage. These results were comparative with hearts preconditioned with AngI, however, differed in that AngI exhibited slightly reduced potency immediately following ischaemia when compared with PA12.

Captopril, chymostatin, and CV-11974 specifically inhibit ACE1, chymase, and the AT1 receptor, respectively.1315 Blockade of the AT1 receptor abolished the PA12-induced vasoconstriction, revealing the AT1 receptor can mediate the response to PA12, indicating that either PA12 itself binds and activates AT1R, or PA12 is initially cleaved into fragments capable of binding and activating AT1R. To examine whether PA12 activity is dependent upon ACE1, captopril (and ramipril) was co-infused directly into the rat heart with PA12. Neither antagonist inhibited PA12-induced vasoconstriction in the isolated rat heart, nor had any effect on CK or TnI levels following ischaemia. Analysis of perfusate samples collected after passing through the heart revealed an elevated presence of perfusate AngII from hearts infused with PA12, indicating PA12 was converted to AngII, not by ACE1, but by another cardiac enzyme. The concentrations of captopril and ramipril used in this study were relatively high (100 µmol/L), newly obtained and capable of causing significant vasodilation when infused alone, indicating potency and viability. These results contrast those of Nagata et al.,5 which suggested PA12 constriction of aortic strips and change in mean arterial pressure in vivo were dependent on ACE1. Differential tissue expression of chymase and ACE1 may underlie these discrepancies, as well as ex vivo vs. in vivo methodological differences, but also the relative activities of chymase and ACE1 in the circulation to produce AngII from PA12 need to be clarified.

The ‘normalized’ results observing PP would suggest that inhibition of ACE may, in fact, potentiate PA12 activity. This fits with the observed minor elevation in perfusate IR-AngII levels from hearts infused with PA12+captopril compared with infusion of PA12 alone. There is little, if any, literature observing this phenomenon where inhibition of ACE stimulates upregulation of chymase and/or other enzymes capable of generating elevated vasoconstriction and AngII formation; as in all studies observing ACE vs. chymase-induced AngII generation and vasoconstriction, AngI has been the substrate, which we and others16,17 have observed to have little dependence upon chymase activity. Wei et al.18 suggested an interaction between mast cells and Bradykinin (BK) in the cardiac interstitium, whereby antagonism of the BK2 receptor significantly reduced mast cell density. Combining this latter report with knowledge that ACE1 inhibition increases BK2 activity,19 leads to the hypothesis that ACE inhibition could theoretically increase mast cell density, providing a greater amount of chymase and other ACE-independent AngII-forming enzymes capable of generating AngII from PA12.

Boucher et al.20 were the first to show enzyme(s) other than ACE are capable of converting AngI into AngII. Indeed enzymes other than ACE1 and ACE2 have been shown to convert angiotensinogen or AngI into smaller active peptides and/or inactive fragments, including carboxypeptidases and chymase.2 ACE1 inhibitors do not completely block AngII production and ACE1-independent pathways are present in many mammalian species.2125 Chymase, a serine protease, is synthesized and stored in granules of mast, endothelial, and mesenchymal cells widely expressed throughout mammalian tissues.24,26 Indeed, chymase is suggested to be the enzyme primarily responsible for AngII formation in tissue, while ACE1 is the dominant conversion mechanism within the circulation.27,28 Accordingly, we employed chymostatin, a chymase inhibitor, to test whether chymase may be responsible for cardiac PA12 conversion to AngII in our heart preparations. Chymostatin significantly inhibited PA12-induced vasoconstriction in the isolated rat heart during both infusion and reperfusion, as well as reducing the perfusate AngII concentrations by ∼80%. In vitro analysis confirmed that chymase alone is capable of converting PA12 into AngII, and overall these results suggest that the cardiac conversion of PA12 to AngII is at least partially chymase-dependent.

In contrast to PA12, generation of AngII from AngI was shown to be dependent upon ACE1 availability, consistent with prior reports.16 The addition of chymostatin caused significant attenuation of AngI-induced AngII generation (as previously reported29), however exhibited minimal antagonism of AngI-induced vasoconstriction. This indicates chymase can convert AngI to AngII, and suggests that chymase (or other cardiac enzymes) may further cleave AngI or AngII into AngIII [Angiotensin-(2–8)] and/or AngIV [Angiotensin-(3–8)], both shown to be potent vasoconstrictors.30 If so, the current study could only observe perfusate AngII levels via RIA, explaining the discrepancy between the non-significant reduction in PP and the significant reduction in AngII production from hearts receiving chymostatin + AngI.

We cannot exclude ACE2-mediated conversion of PA12 to AngII and other related active peptides. For example, our HPLC profiles were consistent with some Ang(1–7) formation, and either chymase or ACE1 may further cleave AngII to AngIII or AngIV. The data also cannot rule out PA12 itself binding to AT1R, causing the haemodynamic effects observed. However, as chymostatin abolished PA12-induced vasoconstriction and severely reduced generation of AngII, it is likely that AngII is responsible for most of the observed bioactivity.

PA12 is expressed in its highest levels within the small intestine,5 where a close analogue of chymase, chymotrypsin, is also expressed in high amounts and secreted from the pancreas.31 Bovine α-chymotrypsin is capable of hydrolysing AngI into AngII, before degrading AngII into biologically inactive fragments.32 PA12 may also be converted into active AngII and have a specific role within the small intestine, possibly linking the digestive and RAS. Thus, we extend these reports to identify PA12 as a potentially new precursor peptide specific for tissue-based generation of AngII. AngII blockade (by ACE or AT1R blockade) is a mainstay of treatment in hypertension, after ischaemic cardiac events and in human heart failure. Breakthrough of AngII generation is known to occur over the weeks following initiation of ACE1 treatment and may contribute to alleviation of ACE1-related cardiovascular protection.2 Elucidation of alternative pathways of AngII generation enhances the understanding of cardiovascular pathophysiology and may lay the foundation for improved therapy. Furthermore, the high expression of PA12 within the gut combined with its potent vasoconstrictive properties observed in the heart raises the possibility that PA12 links gut processes with concurrent circulatory responses. It is yet to be reported whether PA12 levels in the gut and circulation correlate with each other and/or the effects of eating.

While it is yet to be established whether ACE2 has any role in PA12 activity, (or whether PA12 is in fact present in humans or other mammalian species), we found this newly discovered peptide PA12 infused in isolated perfused rat hearts caused potent, sustained vasoconstriction and impaired recovery from ischaemia. IR-AngII was significantly elevated in perfusate collected from hearts infused with PA12, indicating PA12 is converted into AngII. One enzyme responsible for the conversion of PA12 to AngII appears to be chymase, and the AT1 receptor was found to mediate PA12 activity, most likely binding PA12-derived AngII. Further studies identifying PA12 in humans and its role in control of the circulation are clearly required.


This work was supported by the National Heart Foundation of New Zealand [1218 to H.C.G.P.] and the Health Research Council of New Zealand [Sir Charles Hercus Senior Research Fellowship to C.J.P]; A.M.R. holds the National Heart Foundation of New Zealand Chair of Cardiovascular Studies.


Thanks are extended to the staff of Endolab, Christchurch School of Medicine for assistance with RIA, and to staff of Canterbury Medical Laboratory for assistance with creatine kinase and troponin I analysis. Also to the Centre for Protein Research, University of Otago, Dunedin, New Zealand for their assistance with tandem mass spectrometry analysis.

Conflict of interest: none declared.


View Abstract