Reperfusion injury limits the benefits of revascularization in the treatment of myocardial infarction (MI). Breathing nitric oxide (NO) reduces cardiac ischemia-reperfusion injury in animal models; however, the signaling pathways by which inhaled NO confers cardioprotection remain uncertain. The objective of this study was to learn whether inhaled NO reduces cardiac ischemia-reperfusion injury by activating the cGMP-generating enzyme, soluble guanylate cyclase (sGC), and to investigate whether bone marrow (BM)-derived cells participate in the sGC-mediated cardioprotective effects of inhaled NO. Wild-type (WT) mice and mice deficient in the sGC α1-subunit (sGCα1−/− mice) were subjected to cardiac ischemia for 1 h, followed by 24 h of reperfusion. During ischemia and for the first 10 min of reperfusion, mice were ventilated with oxygen or with oxygen supplemented with NO (80 parts per million). The ratio of MI size to area at risk (MI/AAR) did not differ in WT and sGCα1−/− mice that did not breathe NO. Breathing NO decreased MI/AAR in WT mice (41%, P = 0.002) but not in sGCα1−/− mice (7%, P = not significant). BM transplantation was performed to restore WT BM-derived cells to sGCα1−/− mice. Breathing NO decreased MI/AAR in sGCα1−/− mice carrying WT BM (39%, P = 0.031). In conclusion, these results demonstrate that a global deficiency of sGCα1 does not alter the degree of cardiac ischemia-reperfusion injury in mice. The cardioprotective effects of inhaled NO require the presence of sGCα1. Moreover, our studies suggest that BM-derived cells are key mediators of the ability of NO to reduce cardiac ischemia-reperfusion injury.
- guanosine 3′,5′-cyclic monophosphate
- cardiac ischemia-reperfusion injury
coronary heart disease remains the single leading cause of death in the United States (32). The extent of cardiac damage is a major determinant of morbidity and mortality in patients presenting with an ST-elevation myocardial infarction (MI). Therapies that reduce MI size, such as percutaneous coronary intervention, markedly reduce mortality in these patients (47). However, reperfusion of ischemic myocardium following percutaneous coronary intervention may paradoxically exacerbate myocardial injury, compromising the recovery of left ventricular (LV) function (55).
Molecules that release/regenerate nitric oxide (NO), including NO-donor compounds (46) and nitrite (10), have been reported to reduce the extent of cardiac ischemia-reperfusion (I/R) injury in experimental models. Of note, nitrite is a metabolite and storage form of NO that must first undergo biological modification via nitrite reductases or a direct disproportionation to generate NO (reviewed in Ref. 28). The cardioprotective effects of NO may be exerted via S-nitrosylation of several proteins, including caspase 3, mitochondrial complex 1, and L-type calcium channels (31, 35, 45, 48). Alternatively, NO may elicit its effects by activating soluble guanylate cyclase (sGC), the enzyme responsible for NO-stimulated cGMP synthesis. sGC is a heterodimeric, heme-containing enzyme consisting of one α- and one β-subunit. There are two active sGC isoforms, sGCα1β1 and sGCα2β1 (41). Of these, sGCα1β1 is the principal sGC isoform found in the heart and vasculature (38). We and others have studied mice deficient in the sGCα1-subunit (6, 33). We reported that a deficiency of sGCα1 markedly decreases both basal and NO-stimulated LV sGC enzyme activity (6). However, blood pressure and LV contractile function did not differ in sGCα1−/− and wild-type (WT) mice on a C57BL/6 background (5).
Accumulating evidence suggests that increased cGMP levels in cardiac myocytes protect against cardiac I/R injury by modulating intracellular calcium handling and by opening the mitochondrial ATP-sensitive K+ channel (4, 17). cGMP may also exert cardioprotective effects in cells other than cardiomyocytes by reducing the I/R-induced activation of endothelial cells and circulating cells (4). The administration of drugs designed to increase cardiac cGMP concentrations, including membrane-permeable cGMP analogs (18), inhibitors of cGMP metabolism by phosphodiesterase 5 (13), natriuretic peptides that activate cGMP synthesis by particulate guanylate cyclase (23), and compounds that stimulate sGC activity in the absence of added NO (26), all reduced cardiac I/R injury.
Inhaled NO is a selective pulmonary vasodilator that has been approved by the United States Food and Drug Administration and the European Medicine Evaluation Agency and European Commission for the treatment of neonatal hypoxemia with pulmonary hypertension (2). Since breathing NO decreases pulmonary vascular resistance without altering systemic blood pressure, its effects were initially considered to be limited to the lungs. However, it has become increasingly appreciated that breathing NO can elicit a wide spectrum of physiological effects in peripheral tissues, including the heart (37).
The overall goal of this study was to further characterize the molecular mechanisms responsible for the cardioprotective effects of inhaled NO. Our first objective was to determine whether reduced cardiac sGC enzyme activity caused by the deficiency of sGCα1 impacts cardiac I/R injury in mice. Our second objective was to test whether the ability of inhaled NO to reduce cardiac I/R injury requires the sGC-cGMP signaling pathway. Our third objective was to investigate whether bone marrow (BM)-derived cells participate in the sGC-mediated cardioprotective effects of inhaled NO. We report that the deficiency of sGCα1 does not alter the degree of cardiac I/R injury in mice. Breathing NO reduces cardiac I/R injury in WT but not in sGCα1−/− mice. Moreover, the transplantation of WT BM into sGCα1−/− mice restores the cardioprotective effects of inhaled NO.
sGCα1−/− mice with a targeted deletion of exon 6 of the sGCα1 were generated on a C57BL/6 background, as previously described (5, 6). Body weight-matched 8- to 12-wk-old WT (Jackson, Bar Harbor, ME) and sGCα1−/− male mice were studied. The animal experimental protocols were approved by the Subcommittee on Research Animal Care at Massachusetts General Hospital, which conforms to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH Publication No. 85-23, Revised 1996).
Casting of the coronary arteries.
Coronary arteries of WT and sGCα1−/− mice were perfused using a dye compound mixture (Microfil, MC-130 Red, Flow Tech, Carver, MA), as previously described (42). Briefly, mice were anesthetized with an intraperitoneal administration of ketamine (120 mg/kg) and xylazine (5 mg/kg) and ventilated (MiniVent, Hugo Sachs Elektronik, Harvard Apparatus, Holliston, MA). A thoracotomy was performed. Heparin was administered (500 units in 0.5 ml ip), and the mice were euthanized with pentobarbital sodium (200 mg/kg ip). The thoracic aorta and inferior vena cava were cannulated with a 20-gauge catheter (Angiocath, Becton Dickinson Infusion Therapy Systems, Franklin Lakes, NJ). Saline was infused via the thoracic aorta (20 cmH2O) and was drained from the inferior vena cava for 15 min, followed by an infusion of the dye compound mixture for 5 min. The specimens were placed in glycerin for tissue clearing according to the manufacturer's protocol. Photomicrographs were taken at ×16 magnification (Olympus DP71 microscope digital camera on Olympus SZX12 microscope, Olympus America, Center Valley, PA).
Cardiac I/R injury.
Male mice were anesthetized by intraperitoneal administration of ketamine (120 mg/kg) and xylazine (5 mg/kg) and ventilated at an inspired oxygen fraction near 1.0 without or with 80 parts per million (ppm) NO in nitrogen (Medical-Technical Gases, Medford, MA). Following thoracotomy, the left coronary artery branching pattern was recorded. Myocardial ischemia was induced by ligation of the left coronary artery at the level of the left atrial appendage for 60 min, followed by reperfusion for 24 h. NO gas was administered using a separate and gas-primed mechanical ventilator for 60 min, beginning 10 min after coronary ligation and continuing until 10 min after reperfusion. NO metabolites in blood and tissues rapidly increase and plateau within 5–15 min after commencing NO (37). NO was given during ischemia for 50 min to ensure the presence of stable levels of NO metabolites at the time of reperfusion. Because the first minutes of reperfusion serve as a window of opportunity for cardioprotection (39), mice breathed NO until 10 min after the reperfusion.
At 24 h after reperfusion, the left coronary artery was religated, and tissue marking dye (0.25 ml; TMD-BL, Triangle Biomedical Science, Durham, NC) was injected through a right carotid artery catheter (polyethylene-10 tubing; Intramedic, Becton Dickinson) to determine the area at risk (AAR) of infarction. The heart was harvested, and four consecutive 1-mm cardiac slices beginning at the apex were stained with 2,3,5-triphenyltetrazolium chloride (1% wt/vol; Sigma-Aldrich, St. Louis, MO) to measure the MI size. The ratio of AAR to LV area (AAR/LV) and the ratio of MI area to AAR (MI/AAR) were determined by blinded investigators, as previously described (37).
A single dose of 10 Gy was administered to WT or sGCα1−/− male 6-wk-old-recipient mice. Twenty-four hours later, BM cells (1 × 107 cells in 0.25 ml saline) harvested from WT or sGCα1−/− donor mice were injected into the tail vein of each recipient mouse. Eight weeks after irradiation and BM transplantation, the successful reconstitution of the hematopoietic lineage was monitored using DNA from the ear and blood of BM-chimeric mice in two PCR reactions: one specific for the WT allele and the other for the sGCα1-deficient allele. DNA was prepared from blood and ears using the DNA Isolation Kit (DNeasy, Qiagen, Valencia, CA). For amplification of a 204-bp DNA fragment from the WT sGCα1 gene, the primer pair 5′ -GGGGGCTCTATCTGTCCGACAT-3′ and 5′-TGAGCAACCTCAGAGGGGAA-3′ was used. The amplification of a 1,216-bp fragment, specific for the sGCα1-deficient allele, was performed, as previously described (6).
Assessment of BM chimerization.
The ratio of the donor-to-recipient-derived peripheral leukocytes in WT mice that received sGCα1−/− BM was measured via flow cytometry. Six-week-old CD45.1 congenic mice (n = 3, Jackson) were irradiated (10 Gy), and one day later, BM cells from sGCα1−/− mice (CD45.2) were infused via a tail vein, as described in BM transplantation. After eight weeks, blood was withdrawn from CD45.1 and CD45.2 WT mice, sGCα1−/− mice, and irradiated CD45.1 congenic mice that had received sGCα1−/− BM. Blood samples were stained with allophycocyanin-labeled mouse anti-mouse CD45.2 antibody (catalog no. 558702, BD Biosciences, San Jose, CA) and phycoerythrin-labeled mouse anti-mouse CD45.1 antibody (catalog no. 553776, BD Biosciences). Flow cytometric analysis of stained cells was performed (Life Science Research II cytometer, BD Biosciences), and the data were analyzed using FlowJo 7.5.5 software (Tree Star, Ashland, OR) to determine the ratio of CD45.1 and CD45.2 in each mouse.
Data acquisition and statistical analysis.
All data are presented as means ± SE. For Fig. 2, data were analyzed using two-way ANOVA with Bonferroni post hoc test (GraphPad Prism 5, version 5.01; GraphPad Software, La Jolla, CA). For Fig. 3, data were analyzed using t-test between mice treated with oxygen and 80 ppm NO (Origin 7.0, Origin, Northampton, MA). Variations in coronary artery anatomy were analyzed using Fisher exact test (GraphPad Prism 5). P values < 0.05 were considered significant.
Variations in coronary artery anatomy in sGCα1−/− mice.
In most male C57BL/6 WT mice, we observed that the left coronary artery branches into a septal branch followed by the bifurcation of the left anterior descending coronary artery (LAD) and the left circumflex (LCx) coronary artery (42), distal to the level of left atrial appendage (low bifurcation; Fig. 1, top). In a minority of male WT mice (7 of 95), the LAD-LCx bifurcation is high and occurs proximal to the level of left atrial appendage (high bifurcation; Fig. 1, bottom). To minimize the variability in the AAR measurement, mice with a high bifurcation LADs were not subjected to cardiac I/R injury and were excluded from our study. Of note, high bifurcation was observed more frequently in male sGCα1−/− mice (45 of 129; P < 0.0001). These results suggest that sGCα1 regulates coronary artery development in mice.
Deficiency of sGCα1 does not worsen cardiac I/R injury.
In view of the known cardioprotective effects of cGMP, we sought to test the hypothesis that the presence of sGCα1 limits myocardial injury in mice subjected to 1 h of cardiac ischemia and 24 h of reperfusion. The fraction of the LV rendered ischemic by coronary artery occlusion (AAR/LV) did not differ between WT and sGCα1−/− mice (Fig. 2). Unexpectedly, we observed that MI/AAR did not differ in the two genotypes, arguing against a protective role of sGCα1β1-derived cGMP in cardiac I/R.
sGCα1 is required for the cardioprotective effect of inhaled NO against I/R injury.
To determine the role of sGCα1 in the cardioprotective effects of inhaled NO, we subjected WT and sGCα1−/− mice to I/R and evaluated the ability of breathing 80 ppm NO (during ischemia and for the first 10 min of reperfusion) to reduce MI size. NO inhalation reduced MI/AAR by 41% in WT mice (P < 0.001; Fig. 2). In contrast, breathing NO did not alter the MI/AAR in sGCα1−/− mice (7%, P = not significant). These results provide evidence that sGCα1 is required for inhaled NO to protect against cardiac I/R injury.
Inhaled NO does not modulate cardiac survival protein kinases.
To evaluate whether cardiac survival protein kinases are involved in the cardioprotective effects of inhaled NO, we measured the phosphorylation of members of the mitogen-activated protein kinase family (Erk1/2) and the phosphatidylinositol 3-kinase (PI3K)-Akt cascade (AKT) in the LV of WT and sGCα1−/− mice subjected to 60 min of ischemia and 10 min of reperfusion. Phosphorylation of Erk1/2 and AKT (supplemental Fig. 2) was greater in ischemic than in nonischemic regions of the LV in both WT and sGCα1−/− mice.1 However, breathing NO did not alter the phosphorylation levels of Akt or ERK in either WT mice or sGCα1−/− mice.
We also examined the impact of breathing NO on cytosolic and mitochondrial PKCε phosphorylation in the hearts of WT mice subjected to 60 min of ischemia and 10 min of reperfusion. Breathing NO did not alter PKCε phosphorylation in either ischemic or nonischemic areas (supplemental Fig. 3).
sGCα1 in BM-derived cells mediates the cardioprotective effect of inhaled NO against cardiac I/R injury.
To examine the role of circulating cells in the cardioprotective effects of inhaled NO, we performed lethal irradiation and BM transplantation to generate WT mice with sGCα1-deficient (knockout) BM (WT-KO) and sGCα1-deficient mice with WT BM (KO-WT). As controls, WT mice were transplanted with WT BM (WT-WT), and sGCα1-deficient mice were transplanted with sGCα1-deficient BM (KO-KO). PCR analyses of genomic DNA from ear tissue and blood from chimeric mice were performed to confirm the efficacy of BM transplantation (supplemental Fig. 1, top). In addition, chimerization was assessed in WT-KO mice by characterizing the CD45 antigen in peripheral circulating hematopoietic cells. Congenic CD45.1 WT mice were irradiated and reconstituted with the BM from CD45.2 sGCα1−/− mice. In the resulting WT-KO mice, 97.1% of the peripheral hematopoietic cells were CD45.2 positive (supplemental Fig. 1, bottom).
WT-WT, KO-KO, WT-KO, and KO-WT mice were subjected to 1 h of ischemia and 24 h of reperfusion without or with inhaled NO (during ischemia and for the first 10 min of reperfusion). AAR/LV did not differ between groups whether or not the animals received NO (Fig. 3, top). There was no difference in MI/AAR in WT-WT, WT-KO, and KO-WT mice that were not treated with inhaled NO. However, MI/AAR was less in KO-KO mice than in WT-WT mice (P = 0.049). NO inhalation reduced MI/AAR by 31% in WT-WT mice (P = 0.012 vs. WT-WT mice that did not receive NO) and by 38% in KO-WT mice (P = 0.024 vs. KO-WT mice that did not receive NO). In contrast, breathing NO did not alter MI/AAR in KO-KO mice. Inhaled NO tended to reduce MI/AAR in WT-KO mice, but this trend did not achieve statistical significance despite studying 12 mice in each group (P = 0.059 vs. WT-KO mice that did not receive NO).
In the present study, we report that sGCα1 is required for inhaled NO to reduce MI size in a murine model of cardiac I/R injury. Moreover, we have shown that the presence of sGCα1 in BM-derived cells is sufficient for inhaled NO to exert a cardioprotective effect. These observations highlight the pivotal role of cGMP signaling by BM-derived circulating elements in producing the cardioprotective effects of NO inhalation.
The intracellular second messenger, cGMP, has been demonstrated to protect against cardiac I/R injury in a variety of animal models (reviewed in Ref. 4). sGC is one of the two types of enzymes responsible for cGMP synthesis and is the enzyme that produces cGMP when stimulated by NO. Although there are two active isoforms of sGC, α1β1 and α2β1, the former is considered to be responsible for almost all of the sGC enzyme activity detected in the heart and blood vessels (6, 38). To examine the cardioprotective effects of sGC-derived cGMP, we compared MI/AAR in WT and sGCα1−/− mice subjected to cardiac I/R. We previously reported that cerebral I/R injury was less in WT mice than in sGCα1−/− mice (1). Unexpectedly, MI/AAR was not greater in sGCα1−/− than in WT mice. In fact, in mice that had undergone irradiation and BM transplantation, sGCα1-deficient mice appeared to be protected. These results suggest that basal cGMP levels generated by sGCα1β1 may negatively impact the outcome of cardiac I/R injury.
Low levels of cGMP, generated by sGCα2β1, can mediate, at least in part, some of the physiological effects of NO. Both we (38) and Mergia and colleagues (33) observed NO-induced, sGC-dependent vasodilation in sGCα1−/− mice in the absence of any measureable increases in cGMP. Previously, we reported that cardiac cGMP levels in mice breathing NO in WT mice were not different from the mice breathing air (21). However, we cannot exclude that, in sGCα1−/− mice, inhaled NO induces the remaining isoform (sGCα2β1) to generate low levels of cGMP in cardiac cells or other cell types. Our results do indicate that these low levels of cGMP generated by sGCα2β1 are not sufficient for inhaled NO to protect against cardiac I/R.
Multiple mechanisms, both cGMP dependent and independent, have been proposed to explain the protective effects of NO-donor compounds and inhaled NO on cardiac I/R injury (35). To identify the role for sGC and cGMP signaling in the cardioprotective effects of inhaled NO, we compared the efficacy of inhaled NO in WT and sGCα1−/− mice subjected to cardiac I/R injury. We previously reported (37) and have confirmed in the current study that breathing NO can reduce I/R injury in WT mice. However, in sGCα1−/− mice, the beneficial effects of inhaled NO were abolished. These results demonstrate that signaling via sGCα1 and cGMP is required for the cardioprotective effects of inhaled NO.
Activation of protein kinases including mitogen-activated protein kinase family (Erk1/2) and the PI3K-Akt cascade (AKT) have been reported to limited cardiac myocyte apoptosis in cardiac I/R injury (22). Moreover, cGMP-mediated activation of PKG may protect the heart from I/R injury through the activation of PKCε and/or the reduction of calcium overload, both of which may inhibit mitochondrial transition pore opening at reperfusion (4). However, breathing NO did not alter the phosphorylation of Akt, ERK, or JNK in the hearts of either WT or sGCα1−/− mice and did not modulate PKCε phosphorylation in WT mice after cardiac I/R injury. The time point 10 min after reperfusion was selected based on previous reports where cardiac I/R-induced phosphorylation of AKT and ERK1/2 was apparent at 10 and 30 min but not 120 min after the reperfusion (36). Similarly, we observed increased phosphorylation of Akt in LV extracts at 10 but not 60 min after reperfusion (data not shown). For the PKCε translocation measurements, we studied a time point previously determined to be relevant for the protective role of PKCε signaling in a rat I/R injury model (7). Taken together, at least at the time point we examined in our model, the modulation of survival kinases or the PKCε pathway does not appear to be critical for the cardioprotective effects of inhaled NO. We acknowledge that the exact downstream mechanisms remain to be elucidated.
BM-derived cells have been implicated in the pathogenesis in cardiac I/R injury (29, 52, 53). The cardioprotective effects of inhaled NO were evident in sGCα1−/− mice in which sGCα1 was restored in BM cells via transplantation. This observation suggests that sGCα1 in BM-derived cells is sufficient for inhaled NO to exert cardioprotection against I/R injury. Inhaled NO may activate sGC in BM-derived cells as they transit the lung, causing them to produce cGMP. Several mechanisms exist by which sGC can modulate the function of various BM-derived cell types. sGC expression in BM-derived cells is well established. For example, sGCα1β1 was found to be present in platelets (33), neutrophils (8), and B lymphocytes (40). NO-cGMP signaling was reported to be involved in inhibition of platelet aggregation (16, 50), neutrophil inhibition and activation (43, 51), neutrophil degranulation (51), neutrophil chemotaxis (25), neutrophil migration (11), eosinophil migration (49), eosinophil adhesion (9), and activation and proliferation of T cells (15). Any and all of these mechanisms may be involved in the ability of inhaled NO to attenuate the cardiac injury associated with cardiac I/R. Additional studies are required to determine which type of BM-derived cells contribute to the protective effects of inhaled NO and how sGC affects the function of these BM-derived cells. On the other hand, breathing NO leads to the generation of NO metabolites in the plasma and erythrocytes (37). NO metabolites, including nitrite (56), nitrate (30), S-nitrosothiols (20), and NO-heme (37), may reach the periphery, regenerate NO, and exert effects locally on sGC-containing BM-derived cells.
The observation that breathing NO tended to reduce MI/AAR in WT mice transplanted with BM from sGCα1−/− mice (WT-KO) raises the possibility that sGCα1 in cells other than those derived from the BM (i.e., endothelial cells, cardiac myocytes, etc.) may also contribute to the cardioprotective effects of inhaled NO. Furthermore, because 3% of circulating leukocytes in chimeric WT-KO mice were derived from the WT recipient, we cannot exclude the possibility that residual WT BM-derived cells were sufficient to exert some of the beneficial effect of inhaled NO.
During the course of our studies of cardiac I/R injury, we observed that an abnormal coronary artery anatomical pattern (proximal left coronary artery bifurcation) was much more frequently present in sGCα1−/− mice than in WT mice. We do not believe that this observation is attributable to subtle strain differences between the two genotypes because the same abnormal pattern in coronary artery anatomy was observed more frequently in female sGCα1−/− mice than in female WT mice on an SV129 background (see supplemental material). Aberrations in coronary artery anatomy, including variations in the branching of the left coronary artery, were previously reported in C57BL/6 mice (14, 27, 34, 42). To our knowledge, no major variations in the coronary vasculature were reported among mice carrying genetic alterations in the NO-sGC-cGMP pathway [i.e., targeted deletions of neuronal NO synthase (3, 54) and endothelial NO synthase (19, 44, 54) or transgenic mice with cardiac (12) or systemic (24) overexpression of endothelial NO synthase]. Our findings suggest that sGC and its product, cGMP, play an important role in the specification of coronary artery anatomy.
In summary, we found that a deficiency of sGCα1 does not alter the degree of cardiac I/R injury in mice. On the other hand, sGCα1 is required for inhaled NO to reduce MI size in mice subjected to cardiac I/R injury. Moreover, our observations highlight a critical role for sGCα1 in BM-derived cells in mediating the cardioprotective effects of inhaled NO. These results demonstrate the importance of sGC- and cGMP-dependent signaling in the cardioprotective effects of inhaled NO and possibly of other strategies designed to increase cardiac NO levels.
This study was supported by Eleanor & Miles Shore Fellowship grants from Harvard Medical School (to Y. Nagasaka and E. S. Buys) and a Scientist Development Grant from the American Heart Association (to E. S. Buys). Research of P. Brouckaert was supported by grants from the Fonds Wetenschappelijk Onderzoek-Vlaanderen and the University of Gent Special Research Fund (Bijzonder Onderzoeksfonds-Geconcerteerde Onderzoeksacties).
W. M. Zapol and K. D. Bloch have obtained patents relating to the use of inhaled nitric oxide. These patents are assigned to Massachusetts General Hospital, which has licensed them to Ikaria and Linde Gas Therapeutics (Lidingo, Sweden). W. M. Zapol receives royalties, and K. D. Bloch has received research grants from Ikaria.
We thank Dr. Fumito Ichinose and Dr. Stephan Janssens for helpful discussions and Dr. Hui Zheng for the statistical advice. We acknowledge Drs. Yang Feng, Wei Chao, Megan Sykes, Rajeev Malhotra, Elena Egorina, Michael Sovershaev, Matthias Derwall, and Gian-Paolo Volpato Martin for technical advice during the course of this study.
↵1 Supplemental Material for this article is available at the American Journal of Physiology-Heart and Circulatory Physiology website.
- Copyright © 2011 the American Physiological Society