|Home | About | Journals | Submit | Contact Us | Français|
Renin released by ischemia/reperfusion (I/R) from cardiac mast cells activates a local renin-angiotensin system (RAS). This exacerbates norepinephrine release and reperfusion arrhythmias (VT/VF), making RAS a new therapeutic target in myocardial ischemia.
We investigated whether ischemic preconditioning (IPC) prevents cardiac RAS activation in guinea-pig hearts ex-vivo. When I/R (20-min ischemia/30-min reperfusion) was preceded by IPC (2×5-min I/R cycles), renin and norepinephrine release and VT/VF duration were markedly decreased, a cardioprotective anti-RAS effect. Activation and blockade of adenosine A2b/A3-receptors, and activation and inhibition of PKCε, mimicked and prevented, respectively, the anti-RAS effects of IPC. Moreover, activation of A2b/A3-receptors, or activation of PKCε, prevented degranulation and renin release elicited by peroxide in cultured mast cells (HMC-1). Activation and inhibition of mitochondrial aldehyde dehydrogenase type-2 (ALDH2) also mimicked and prevented, respectively, the cardioprotective anti-RAS effects of IPC. Furthermore, ALDH2 activation inhibited degranulation and renin release by reactive aldehydes in HMC-1. Notably, PKCε and ALDH2 were both activated by A2b/A3-receptor stimulation in HMC-1, and PKCε inhibition prevented ALDH2 activation.
The results uncover a signaling cascade initiated by A2b/A3-receptors, which triggers PKCε-mediated ALDH2 activation in cardiac mast cells, contributing to IPC-induced cardioprotection by preventing mast-cell renin release and the dysfunctional consequences of local RAS activation. Thus, unlike classical IPC where cardiac myocytes are the main target, cardiac mast cells are the critical site at which the cardioprotective anti-RAS effects of IPC develop.
How critical mast cells are in cardiac pathophysiology is not well understood. Yet, numerous mast cells are present in the mammalian heart (~50,000 mast cells/g human heart tissue) in close proximity to vessels and nerves and their density markedly increases in heart failure, ischemic cardiomyopathy and experimental infarct models.1 Mast cells synthesize, store and release a variety of mediators. We recently reported that cardiac mast cells are also an important source of the aspartyl protease renin.2 When released by ischemia/reperfusion (I/R), this renin initiates the activation of a local renin-angiotensin system (RAS); the locally formed angiotensin II (ANG II) exacerbates norepinephrine (NE) release from cardiac sympathetic nerves and elicits reperfusion arrhythmias.3 When mast cells are depleted or pharmacologically stabilized, renin and NE release and reperfusion arrhythmias are markedly reduced.3 Thus, the release of mast cell-derived renin represents a new target in the prevention and treatment of ischemic cardiac dysfunction.
Brief ischemic exposures have repeatedly been shown to protect the heart both structurally and functionally from a subsequent prolonged exposure to I/R. Ischemic preconditioning (IPC), as this phenomenon is known, has been the object of wide attention since its discovery.4–6 Adenosine release and activation/translocation of protein kinase C (PKC) have been identified as necessary steps in the cardioprotection afforded by classical IPC (e.g., infarct size reduction).6–8 Of the PKC family of serine/threonine kinases, the PKCε isoform has been shown to play a key cardioprotective role against I/R9,10 and to exhibit anti-secretory activity in mast cells.11
Accordingly, we set out to determine whether IPC might also prevent the activation of a local cardiac RAS initiated by the release of mast cell renin and, if so, whether this novel IPC paradigm involves the activation of PKCε in mast cells. Given that phosphorylation of mitochondrial ALDH2 by PKCε correlates with cardioprotection,12 we hypothesized that IPC could promote PKCε-induced activation of ALDH2, which would then remove toxic aldehydes known to degranulate mast cells, such as acetaldehyde and 4-hydroxynonenal (4-HNE), formed by lipid peroxidation.13–15 Our study outlines a novel protective anti-RAS effect of IPC; we find that the sequential activation of adenosine A2b- and A3-receptors, PKCε and ALDH2 in cardiac mast cells diminishes the release of renin elicited by I/R and thus, curtails ANG II- and NE-induced arrhythmias.
See the on-line Supplement for the complete Materials and Methods section.
132 male Hartley guinea pigs (300–350 grams; Charles River Laboratories, Kingston, NY) were anesthetized with CO2 and euthanized by stunning with IACUC approval. Isolated hearts were perfused at constant pressure with oxygenated Ringer at 37°C in a Langendorff apparatus (Radnoti Glass Technology, Monrovia, CA).
Ischemia/reperfusion (I/R): after equilibration all hearts were subjected to 20-min global ischemia followed by 30-min reperfusion. Ischemic preconditioning (IPC): 2 × 5-min cycles of ischemia, each followed by 5-min reperfusion. Pharmacological prevention of IPC: antagonists were perfused for 20 min (Glyceryl trinitrate, GTN, 30 min) before and during IPC, and then washed out for 15 min before I/R. Pharmacological preconditioning: given agents were perfused for 2 × 5-min cycles before I/R, except for δV1-1 (PKCδ inhibitor) administered during the entire 30-min reperfusion following the 20-min ischemia. Prevention of pharmacological preconditioning: antagonists were perfused for 20 min (GTN, 30 min) before and during pharmacological preconditioning and then washed out for 15 min before I/R.
Coronary flow was measured every 2 min; samples were assayed for renin, norepinephrine, β-hexosaminidase and creatine phosphokinase (CPK). Surface ECG was obtained from left ventricle and right atrium, recorded in digital format, and analyzed using Power Lab/8SP (AdInstrument, Colorado Springs, CO).
The human mastocytoma cell line (HMC-1) was a gift of by Dr. I. Biaggioni (Vanderbilt University, Nashville, TN). Cells were maintained in suspension culture as previously described.2
β-hexosaminidase and renin coronary overflow was measured as previously described.3 HMC-1 cells were suspended in Ringer buffer and equal volumes aliquoted in Eppendorf tubes and incubated at 37°C with a given agent (i.e., Alda-1, ψεRACK or LUF5835 + IBMECA) for 10 min (preceded or not by a 30-min incubation with GTN). Acetaldehyde, H2O2 or 4-HNE was subsequently added for 20 minutes. All results were normalized and expressed as percent above control.
Coronary effluent was assayed for norepinephrine by HPLC with electrochemical detection as previously described.3
Coronary effluent was assayed for creatine phosphokinase release using a CPK assay kit (Genzyme Diagnostics, Charlottetown, PE, Canada).
RT-PCR: total RNA was extracted from HMC-1 cells using TRIzol reagent (Invitrogen, Carlsbad, CA), 1 μg of total RNA from each sample was reverse-transcribed and cDNA amplified by RT-PCR using a QIAGEN (Valencia, CA) One-step RT-PCR kit. PCR products were analyzed by agarose gel electrophoresis and ethidium bromide staining. Immunostaining: HMC-1 cells were fixed and permeabilized on glass slides and stained with the goat anti-A2b-receptor Ab (Santa Cruz Biotechnology, Santa Cruz, CA) conjugated to Alexa Fluor 488 donkey anti-goat IgG and with rabbit anti-A3-receptor Ab (Santa Cruz) conjugated to Alexa Fluor 488 donkey anti-rabbit IgG. Nuclei were stained with DAPI. For immunofluorescence, cells were examined with an inverted fluorescence microscope (Nikon Eclipse TE 2000-U, Morrell Instruments, Melville, NY) interfaced to an electron multiplying charge-coupled device (Hamamatsu, Photonics, Bridgewater, NJ) and processed with Metamorph software (version 6.2; Universal Imaging Corp.).
Cytosolic and membrane fractions of HMC-1 cells were separated and Western blot analysis was performed using a PKCε-specific antibody (Santa Cruz).
ALDH2 activity in HMC-1 cells was determined spectrophotometrically by monitoring the reductive reaction of NAD+ to NADH at 340 nm as previously described.16
Acetaldehyde, H2O2, IBMECA, MRS1754, MRS1523, DPCPX, chelerythrine, 5-hydroxydecanoate and cyanamide were purchased from Sigma-Aldrich (St. Louis, MO); 4-hydroxy Nonenal (4-HNE) in ethanol solution was purchased from Cayman Chemical. LUF5835 was a gift from Dr. M.W. Beukers (University of Leiden, Leiden, Netherlands); EXP3174 was a gift from Merck Sharp & Dohme Ltd (Whitehouse Station, NJ); ψεRACK, δV1-1 and Alda-1 were synthesized in the Mochly-Rosen lab (Stanford University School of Medicine, Palo Alto, CA). Phorbol 12-myristate 13-acetate was purchased from LC Laboratories (Woburn, MA). GTN was purchased from Hospira Inc. (Lake Forest, IL). Human plasma angiotensinogen was purchased from Calbiochem (San Diego, CA).
Data are presented as means ± SEM. Non-parametric tests were used throughout. For 2-group comparisons, Mann-Whitney test was used (Figs. 1 and and2).2). For comparisons among more than 2 groups, Kruskal-Wallis test followed by post-hoc Dunn’s test was used (Figs. 1, ,2,2, ,3,3, 4D & F, ,5,5, ,66 and and7).7). GraphPad Prism version 4.03 for Windows, GraphPad Software, San Diego, CA, was used. P<0.05 was considered statistically significant.
Spontaneously beating Langendorff-perfused guinea-pig hearts were subjected to 20-min global ischemia followed by 30-min reperfusion (I/R). This resulted in mast cell degranulation, demonstrated by a 202±31% (n=5; ±SEM) increase in β-hexosaminidase (β-HEX) overflow into the coronary effluent. I/R also caused large increases in renin and NE overflow (i.e., ~2.5- and ~75-fold, respectively), and severe ventricular arrhythmias (tachycardia and fibrillation, VT/VF) that lasted ~12 min (Fig. 1). We had previously shown that this enhanced NE overflow and arrhythmias result from the activation of a local RAS by renin released from cardiac mast cells.3
When I/R was preceded by IPC (i.e., 2 × 5-min cycles of ischemia, each followed by 5-min reperfusion), mast cell degranulation was only approximately half that occurring with I/R alone, as indicated by a marked decrease in β-HEX overflow (i.e., the overflow of β-HEX increased by 202±31% and 109±14% with I/R and I/R preceded by IPC, respectively; n=5 and 5; P<0.05). IPC also greatly reduced the overflow of renin and NE and the duration of VT/VF (i.e., an overall ~70–85% decrease) (Fig. 1), clearly indicating a cardioprotective anti-RAS effect of IPC.
Notably, the overflow of renin and NE during preconditioning was only slightly, but not significantly higher than that in non-conditioned hearts (renin overflow, i.e. pg/h/g of ANG I formed, was 7.10 ± 0.99 during IPC and 6.81 ± 2.12 in control conditions; NE overflow, was 3.82 ± 1.17 pmol/g during IPC vs. 4.02 ± 0.77 in control conditions; means ± SEM; n= 8 + 6; P=0.95 and P=0.75 for renin and NE, respectively). Thus, the finding that IPC markedly attenuated the I/R-induced release of renin and NE was not due to depletion of renin and NE pools during preconditioning.
Adenosine, protein kinase C (PKC) and mitochondrial ATP-sensitive potassium channels (mKATP) have been identified as necessary steps in the cardioprotection afforded by classical IPC (e.g., infarct size reduction).6–8 Further, although ANG II is widely perceived as a deleterious agent, it has also been found to mimic the cardioprotective effects of IPC.17–19 Thus, we investigated whether adenosine, mKATP channels, ANG II and PKC play a role in the cardioprotective anti-RAS IPC paradigm.
First, we assessed whether inhibition or activation of adenosine receptors prevents or mimics, respectively, the IPC-mediated attenuation of renin release in hearts subjected to I/R. We found that the combined blockade of adenosine A2b- and A3-receptors with compounds MRS1754 (50 nmol/L)20 and MRS1523 (100 nmol/L)21 prevented the IPC-induced attenuation of renin and NE release, and the alleviation of reperfusion arrhythmias (Fig. 1A). Conversely, the combined activation of A2b- and A3-receptors with compounds LUF5835 (50 nmol/L)22 and IBMECA (50 nmol/L),23 mimicked the cardioprotective anti-RAS effects of IPC (Fig. 1A). In contrast, activation of A2b- or A3-receptors alone failed to mimic the effects of IPC (renin overflow was 26.94 ± 2.80, 19.77 ± 3.9 and 24.37 ± 2.58 pg/hr/g of ANG I formed for I/R, I/R + LUF5835 and I/R + IBMECA, respectively; n=6 + 5 + 5, P>0.20). Blockade of adenosine A1-receptors with DPCPX (300 nmol/L; IC50=18 nmol/L)24 was also ineffective (Fig. 1A). Thus, only the combined activation of A2b- and A3-receptors appears to induce the cardioprotective anti-RAS effects of IPC.
Because the opening mKATP channels is known to be involved in the mediation of classical IPC-induced cardioprotection,7,25,26 we questioned whether these channels may also play a role in the anti-RAS effects of IPC. Thus, we induced IPC in the presence of the mKATP antagonist, 5-hydroxydecanoate (100 μmol/L; 5-HD; IC50=30 μmol/L).27 5-HD failed to affect the cardioprotective anti-RAS effects of IPC. In fact, the IPC-induced attenuation of renin and NE release and the abbreviation of reperfusion arrhythmias were the same in the presence and absence of 5-HD (Figure IA in the on-line only Data Supplement). Hence, the cardioprotective anti-RAS effects of IPC do not appear to depend on the opening of mKATP channels.
According to early reports in rabbits and rats, ANG II mimicked the cardioprotective effects of classical IPC.17–19 Thus, we determined whether ANG II, which is locally produced in the heart subjected to I/R,3 contributes to the anti-RAS effects of IPC. For this, we induced IPC in hearts perfused with the AT1-receptor antagonist EXP3174. EXP3174 (300 nmol/L; IC50=6 nmol/L)28 did not prevent the cardioprotective anti-RAS effects of IPC (Figure IA in the on-line only Data Supplement). In fact, the IPC-induced attenuation of renin and NE release and the abbreviation of reperfusion arrhythmias were the same in the presence and absence of EXP3174 (Figure IA in the on-line only Data Supplement). Therefore, AT1-receptors are probably not involved in the mediation of the cardioprotective anti-RAS effects of IPC.
Because PKC activation/translocation is likely to be involved in the cardioprotective effects of classical IPC,8,29 we next investigated the role of PKC in the cardioprotective anti-RAS effects of IPC. Treatment of hearts with the general, non-isoform selective, PKC activator phorbol 12-myristate 13-acetate (PMA, 0.05 nmol/L, 2 × 5-min cycles before I/R) mimicked the protective effects of IPC on renin and NE release, and reperfusion arrhythmia duration (Figure IB in the on-line only Data Supplement). Moreover, inhibition of PKC with the specific, but non-isoform selective, chelerythrine (2.8 μmol/L) prevented IPC’s effects on the same parameters (Figure IB in the on-line only Data Supplement). Thus, general PKC activation appears to mediate the cardioprotective anti-RAS effects of IPC.
Of the PKC family of serine/threonine kinases, the PKCε isoform has been shown to play a key cardioprotective role against I/R.9,10,30,31 Thus, we tested whether ψεRACK, a selective activator peptide of PKCε,32 mimics the anti-RAS effects of IPC. ψεRACK (500 nmol/L), perfused for 2 × 5-min cycles followed by a 5-min washout before I/R, mimicked the protective anti-RAS effects of IPC. Indeed, the overflow of renin and NE, and the duration of VT/VF were reduced by ~55–90% as compared with I/R hearts (Fig. 1B). Moreover, selective inhibition of PKCε with εV1-2 (1 μmol/L)33 prevented IPC’s effects on the same parameters (Fig. 1B). Thus, PKCε activation appears to be required and sufficient for the genesis of the cardioprotective anti-RAS effects of IPC.
Stimulation of adenosine A2b- and A3-receptors mimicked the anti-RAS effects of IPC (see Fig. 1A); selective activation of the PKCε isoform also had anti-RAS effects similar to IPC (see Fig. 1B). Given that adenosine is known to activate PKC, thus initiating the traditional preconditioning cascade,6–8 we determined whether the anti-RAS IPC-like effects of A2b- and A3-receptors rely on the consequent activation of PKCε. To verify this notion, we assessed whether PKCε blockade would prevent the IPC-like effects of A2b- and A3-receptor agonists. We found that selective inhibition of the PKCε isozyme with εV1-2 (1 μmol/L)33 prevented the anti-RAS IPC-like effects resulting from the combined activation of A2b- and A3-receptors (Fig. 2A). Thus, A2b- and A3-receptor-mediated activation of PKCε appears to be the first significant step in the anti-RAS preconditioning pathway.
Since the cardioprotective infarct-sparing effects of PKCε activation30 have been found to depend on phosphorylation of mitochondrial ALDH2,12,34 we next assessed whether the anti-RAS effect of IPC are also determined by ALDH2 activation. For this, we assessed whether inhibition/inactivation of ALDH2 would abolish the anti-RAS effects of IPC, and whether activation of ALDH2 would mimic them. We found that GTN, perfused for 30 min at a concentration (2 μmol/L) which is known to inactivate ALDH2,12 prevented the cardioprotective anti-RAS effects of ψεRACK (i.e., GTN abolished the ψεRACK-induced inhibition of renin and NE release as well as the alleviation of VT/VF; Fig. 2B). We also found that the general ALDH inhibitors, cyanamide (CYA; 5 mmol/L)12 and GTN prevented the anti-RAS effects of IPC (Fig. 3A and B). Conversely, selective activation of ALDH2 with Alda-112,34 (20 μmol/L, 2 × 5 min) reproduced all of the anti-RAS effects of IPC, an action that was also prevented by cyanamide (Fig. 3A) and by pretreatment with GTN (Fig. 3B). Collectively, these findings suggest that ALDH2 activation by PKCε is a crucial mechanistic step in the development of the anti-RAS effects of IPC.
Given the pivotal role that mast cells play in the activation of RAS in the heart,3 cardiac mast cells are likely to be the site at which the anti-RAS effects of IPC develop. Since combined activation of adenosine A2b- and A3-receptors mimics the cardioprotective anti-RAS effects of IPC, whereas the combined blockade of the same receptors prevents the anti-RAS effects (see Fig. 1A), we first ascertained the presence of A2b- and A3-receptors on mast cells. For this, we used human mast cells in culture (HMC-1 cells). Total RNA (1 μg) was extracted from HMC-1 cells, reverse-transcribed and amplified by PCR using sense and antisense primers specific for human A2b- and A3-receptor genes. Figure 4A is an ethidium bromide-stained gel showing that the HMC-1 PCR products for these adenosine receptor subtypes are consistent with those reported by others.35 HMC-1 cells were also immuno-positive for the two adenosine receptor subtypes (Fig. 4B).
We next made certain that mast cell PKCε can be activated. Utilizing Western analysis in cytosolic and membrane fractions of HMC-1 cells, we found that the phorbol ester PMA (positive control) markedly increased the translocation of PKCε from cytosol to membrane (i.e., a hallmark of PKCε activation)(Fig. 4C-F). Incubation of HMC-1 cells with the PKCε activator ψεRACK (500 nmol/L; 30 min), prior to a 7-min incubation with a below-threshold concentration of PMA (3 nmol/L),36 also significantly translocated PKCε from cytosol to membrane (Fig. 4C and D). Moreover, incubating HMC-1 cells with the adenosine A2b- and A3-receptor agonists in combination (LUF5835 and IBMECA, 50 nmol/L each for 1 hr) also translocated PKCε (Fig. 4E and F).
Our findings in isolated guinea-pig hearts and cultured mast cells suggested that IPC may result from the activation of adenosine A2b- and A3-receptors expressed by cardiac mast cells and consequent PKCε-dependent activation of mitochondrial ALDH2. Thus, we next investigated the role of ALDH2 in mast cell degranulation and renin release elicited by prototypic toxic compounds formed in I/R. For this, we measured mast cell degranulation in response to acetaldehyde, 4-hydroxynonenal (4-HNE), another toxic aldehyde that accumulates during cardiac ischemia and hydrogen peroxide (H2O2), which triggers toxic aldehydes formation by membrane lipid peroxidation.13,15 Incubation of HMC-1 cells with acetaldehyde (300–700 μmol/L), H2O2 (0.1–1 mmol/L) or 4-HNE (3–30 μmol/L) elicited a concentration-dependent increase in the release of β-HEX (~15–42%; an indication of mast cell degranulation) and renin (~15–80%) (Fig. 5). Notably, pre-incubation of HMC-1 cells with the ALDH2 activator Alda-1 (20 μmol/L)12,34 prevented the degranulating effects of each acetaldehyde, H2O2 and 4-HNE (Fig. 5A-F). Moreover, pretreatment of HMC-1 cells with the ALDH2 desensitizer GTN (2 μmol/L for 30 min) prevented the anti-degranulating effects of Alda-1 (Fig. 5). These findings suggested that ALDH2 activation in mast cells prevents their degranulation by toxic aldehydes produced in the I/R heart and that this represent a crucial mechanistic step in the anti-RAS effects of IPC.
We next measured ALDH2 activity in HMC-1 cells in response to the specific ALDH2 activator Alda-1,37 the non-isoform specific PKC activator PMA, the PKCε-selective agonist ψεRACK and the adenosine A2b- and A3-receptor agonists in combination. We found that the ALDH2 enzymatic activity (i.e., NADH production) was enhanced by Alda-1 (100 μmol/L), PMA (300 nmol/L), ψεRACK (0.5 and 1 μmol/L) and LUF5835 (50 nmol/L) and IBMECA (50 nmol/L) in combination (Fig. 6). Notably, selective PKCε inhibition with εV1-2 (1 μmol/L) prevented the increase in ALDH2 activity elicited by A2b- and A3-receptor agonists in combination (Fig. 6).
Given that activation of A2b- and A3-receptors in HMC-1 cells increased ALDH2 activity and this was inhibited by the PKCε antagonist (see Fig. 6), we next examined whether activation of A2b- and A3-receptors and PKCε would each protect mast cell from degranulation and renin release. We found that the large concentration-dependent increase in β-HEX and renin release elicited by incubation with H2O2 (0.1–1 mmol/L) was markedly inhibited by LUF5835 (50 nmol/L) and IBMECA (50 nmol/L) in combination (Fig. 7A and C) and by ψεRACK (500 nmol/L)(Fig. 7B and D). These effects were prevented by prior ALDH2 desensitization with GTN pre-treatment (2 μmol/L, for 30 min)(Fig. 7A-D). Collectively these findings indicate that activation of adenosine A2b- and A3-receptors on the mast cell membrane leads to an increase in PKCε activity and thus, ALDH2 activation which prevents the degranulating effects of toxic aldehydes such as those produced in I/R.
We finally sought to establish whether the cardioprotective anti-RAS effects of IPC, which most likely result from an action at the mast cell level, are independent of the IPC-induced reduction of myocyte damage. Since both PKCε activation10 and PKCε inhibition32 protect cardiac myocytes from I/R-induced damage, we compared PKCε activation with PKCδ inhibition in terms of creatine phosphokinase (CPK) release, as well as renin and NE release and VT/VF duration. We found that I/R (i.e., 20-min global ischemia followed by 30-min reperfusion) caused a characteristic increase in CPK overflow into the coronary effluent of isolated guinea-pig hearts which peaked between the 4th and 10th minute of reperfusion (Figure IIA in the on-line only Data Supplement). IPC, ψεRACK pre-treatment (500 nmol/L, 2 × 5-min cycles) and reperfusion with the PKCδ antagonist δV1-1 (500 nmol/L) each and all reduced total CPK overflow (0–20 min reperfusion) by ~50% (Figure IIB in the on-line only Data Supplement). However, whereas ψεRACK pre-treatment markedly reduced the activation of RAS, thus displaying cardioprotective anti-RAS effects, reperfusion with δV1-1 did not affect renin and NE release nor did alleviate VT/VF, thus failing to protect the heart from the consequences of mast cell degranulation, including induction of VT/VF (Figure IIC in the on-line only Data Supplement). These findings indicate that the cardioprotective anti-RAS effects of IPC are mediated by a separate pathway, which is independent of cardiac myocyte salvage, but directly dependent on the modulation of renin release from cardiac mast cells.
The notion of a local tissue-specific RAS, in addition to the classic circulating RAS, has now gained general recognition.38,39 Renin, the rate-limiting step in the RAS activation, has been found in cardiac myocytes 38,40 and renin mRNA has also been identified in heart fibroblasts, as well as in endothelial and smooth muscle cells of coronary vessels.38 Our laboratories recently demonstrated that mast cells can synthesize and secrete renin.2,3 I/R causes the release of renin from cardiac mast cells, activating a local RAS which results in severe arrhythmic dysfunction.3
We have now uncovered a novel cardioprotective anti-RAS paradigm of IPC and delineated its transductional pathway. IPC prevents I/R-induced renin release from cardiac mast cells via an adenosine-mediated activation of PKCε in these cells, followed by activation of mitochondrial ALDH2 which effectively prevents mast cell-degranulation. Given the pivotal role played by mast-cell renin in local RAS activation,1–3 we propose that the cardioprotective anti-RAS effect of IPC, typified by a marked decrease in the overflow of β-HEX, renin and NE, and the curtailing of VT/VF, are based on an inhibitory action at the mast-cell level. This novel IPC paradigm is unlike classical IPC, where myocytes are the main target of cardioprotection, and where IPC is characterized by infarct-size reduction and improved recovery of contractility.4–6
Although the adenosine A1-receptor has been often associated with IPC-induced protection of cardiac myocytes,6,8 we found here that the anti-RAS effects of IPC do not involve A1-receptors, since A1-receptor blockade failed to modify the anti-RAS effects of IPC. This agrees with the reported lack of A1-receptors in mast cells41 and with our proposal that mast cells are the critical site at which the cardioprotective anti-RAS effects of IPC develop.
Mast cells are known to express both adenosine A2b- and A3-receptors42,43(see also Fig. 4). We found that the combined activation of A2b- and A3-receptors displayed IPC-like effects: it attenuated the release of renin and NE, and alleviated reperfusion arrhythmias associated with I/R. These cardioprotective anti-RAS effects were similar to those afforded by IPC, which in fact was abolished by blockade of adenosine A2b- and A3-receptors in combination. Although A2b- and A3-receptors are known as low-affinity receptors (Ki ≥ 5 and 1 μmol/L for A2b and A3, respectively),44 both were likely activated by endogenous adenosine during IPC. Indeed, interstitial adenosine was shown to reach a 7 μmol/L level after 6 min of IPC in the isolated heart.45
Significantly, the combined activation of A2b- and A3-receptors in HMC-1 cells in culture also prevented peroxide-induced degranulation and renin release, supporting our conclusion that these mast cell receptors play a major role in the anti-RAS effects of IPC. Activation of either A2b- or A3-receptor alone failed to mimic the cardioprotective effects of IPC, demonstrating the necessity that both receptors be activated for the initiation of IPC. Other actions of adenosine, such as coronary dilatation, have also been shown to require a concomitant activation of both A2b- and A3-receptors.46
That the activation of A2b- and A3-receptors in HMC-1 cells prevents degranulation and renin release concurs with the protective anti-secretory effect of A2b-receptors, demonstrated by enhanced mast cell activation when A2b-receptors are deleted in mice.47 Yet, other investigators have shown that activation of A2b- and A3-receptors promotes the release of mediators and cytokines from human lung fragments, rat RBL-2H3 cells, HMC-1 cells and macrophages from A2b-receptor-deleted mice.42,43,48 The discrepancy between these and our findings most likely depends on differences in cells and animal species, and on the different stimuli used to degranulate mast-cells.
Opening of KATP channels in the inner membrane of mitochondria in cardiac myocytes has been found to contribute to the protective effects of classical IPC.6–8 Yet, the mitochondrial KATP channel antagonist 5-hydroxydecanoate27 failed to modify the cardioprotective anti-RAS effects of IPC, indicating that these mitochondrial channels are not involved in the mast cell-dependent anti-RAS effects of IPC.
Having established that activation of A2b- and A3-receptors contributes to the anti-RAS effects of IPC, we asked whether A2b- and A3-receptors might signal via PKCε, given that this isoform has both cardioprotective9,10 and anti-secretory properties in mast cell-like RBL-H3 cells.11 Indeed, we found that selective blockade of PKCε with εV1-233 abolished the anti-RAS effects of IPC, while selective activation of PKCε with ψεRACK10 mimicked them. Notably, the IPC-like anti-RAS effects of A2b- and A3-receptor agonists in combination were prevented by selective PKCε inhibition. Collectively, at this point our findings suggested that the initial step in the anti-RAS signaling sequence of IPC could be an adenosine-induced stimulation of A2b- and A3-receptors resulting in PKCε translocation in cardiac mast cells. In fact, A2b- and A3-receptor activation in HMC-1 cells caused the translocation of PKCε from cytosol to membrane, the hallmark of PKCε activation.
Acetaldehyde and 4-HNE are formed during I/R,14,15 in part by lipid peroxidation caused by reactive oxygen species such as hydrogen peroxide.13 These toxic aldehydes, which are known to elicit mast cell degranulation,49 can be removed by mitochondrial ALDH2, a phosphorylation target of PKCε.12 Indeed, we found that Alda-1, a selective ALDH2-activating drug,12,34,37 not only mimicked the cardioprotective anti-RAS effects of IPC in the guinea-pig heart ex vivo, but also prevented the degranulating effects of acetaldehyde, H2O2 and 4-HNE in HMC-1 cells in culture. These effects were abolished by selective inactivation of ALDH2 with GTN. Similarly, pretreatment of HMC-1 cells with GTN prevented the anti-degranulating effects due to activation of A2b- and A3-receptors or PKCε. Hence, these findings add further support to the proposal that activation of ALDH2 in cardiac mast cells is the final crucial step of the protective anti-RAS pathway.
Since A2b- and A3-receptors, PKCε and ALDH2 are also present in cardiac myocytes,37,50 it is conceivable that in addition to mast cells, activation of ALDH2 in cardiac myocytes mediates cardioprotection from I/R. To rule out this possibility we used the PKCδ inhibitor, δV1-1, which protects cardiac myocytes from damage by I/R,51 but does not affect mast cell degranulation (Palandiyani S.S. and Mochly-Rosen D., unpublished results). When we compared the cardioprotective effects resulting from PKCε activation with those due to PKCδ inhibition,32 we found that the mast cell-dependent anti-RAS effects of IPC were clearly separable from the protective effects on cardiac myocytes. One would have expected that the cardioprotective effects of PKCδ inhibition might reduce reperfusion arrhythmias, if these were to originate mostly from myocytes. On the contrary, PKCδ inhibition, which reduced myocytes damage by I/R to a similar extent as PKCε activation, did not abolish reperfusion arrhythmias. Thus, mast cell renin release and RAS activation are the predominant cause of arrhythmic dysfunction in I/R, and activation of ALDH2 in mast cells is an important cardioprotective mechanism in ischemic conditions.
Cardiac myocytes have been reported to express all of the RAS components, including renin.38,40 Renin mRNA has also been found in heart fibroblasts, as well as in endothelial and smooth muscle cells of coronary vessels.38 Thus, it is conceivable that in addition to cardiac mast cells, myocytes and other cells in the heart may contribute to renin production and RAS activation in I/R. In fact, the incidence of reperfusion arrhythmias was markedly reduced, but not completely abolished, in hearts isolated from mice lacking mast cells or in guinea-pig hearts perfused with mast cell-stabilizing agents.3 This suggests that other cells in the heart could release renin in response to ischemia/reperfusion.
In conclusion (see Fig. 8), we propose that adenosine released from various cells during I/R and IPC activates A2b- and A3-receptors on the surface of cardiac mast cells and this is followed by activation/translocation of PKCε, which then increases the catalytic activity of mitochondrial ALDH2. By eliminating reactive aldehydes and their mast cell-degranulating effects, ALDH2 prevents renin release and RAS activation. This reduces ANG II formation, inhibits excessive NE release and prevents the generation of reperfusion arrhythmias. Notably, the relevance of our findings extends beyond the disclosure of a new IPC mechanism. Indeed, although the discovery of IPC has generated a wealth of studies, their clinical translation has yet to come to fruition.52 As the search for cardioprotective drugs continues unabated, our findings elucidate novel basic mechanisms of pharmacological cardioprotection, i.e. the detoxification of reactive aldehydes and ROS in I/R by increasing the catalytic activity of mitochondrial ALDH2, thus alleviating the dysfunctional consequences of RAS activation in the heart. This new finding suggests that in addition to protecting cardiac myocytes from ischemia/reperfusion-induced injury, drugs that prevent cardiac mast cell degranulation may prevent the activation of a local renin-angiotensin system, thus providing an additional benefit to patients with myocardial infarction and perhaps in other cardiac oxidative-stress conditions, such as those occurring in heart failure.
This work was supported by NIH grants HL34215 and HL73400 (K.K., M.S.-R., F.C., R.E., R.B.S. and R.L.), HL47073 and HL46403 (R.L.), AA11147 (D.M.-R.), by CDCH B-09-11-3999-2005 of Central University of Venezuela (M.S.-R.) and a Pharmaceutical Research and Manufacturers of America Foundation pre-doctoral fellowship (N.Y.-K.C). We thank Paul J. Christos, Dr. P.H. for statistical help, partially supported by the Clinical Translational Science Center (CTSC) grant UL1-RR024996.
See the on-line Supplement for the complete Figure Legend section.
Conflict of Interest Disclosures
Dr. Daria Mochly-Rosen is the founder of KAI Pharmaceuticals, Inc., a company that plans to bring PKC regulators to the clinic. However, none of the work in her laboratory is in collaboration with or supported by the company. There are no other disclosures.