Myoglobin-deficient mice are viable and have preserved cardiac function due to their ability to mount a complex compensatory response involving increased vascularization and the induction of the hypoxia gene program (hypoxia-inducible factor-1α, endothelial PAS, heat shock protein27, etc.). To further define and explore functional roles for myoglobin, we challenged age- and gender-matched wild-type and myoglobin-null mice to chronic hypoxia (10% oxygen for 1 day to 3 wk). We observed a 30% reduction in cardiac systolic function in the myoglobin mutant mice exposed to chronic hypoxia with no changes observed in the wild-type control hearts. The cardiac dysfunction observed in the hypoxic myoglobin-null mice was reversible with reexposure to normoxic conditions and could be prevented with treatment of an inhibitor of nitric oxide (NO) synthases. These results support the conclusion that hypoxia-induced cardiac dysfunction in myoglobin-null mice occurs via a NO-mediated mechanism. Utilizing enzymatic assays for NO synthases and immunohistochemical analyses, we observed a marked induction of inducible NO synthase in the hypoxic myoglobin mutant ventricle compared with the wild-type hypoxic control ventricle. These new data establish that myoglobin is an important cytoplasmic cardiac hemoprotein that functions in regulating NO homeostasis within cardiomyocytes.
- nitric oxide
- knockout mice
- cardiac systolic function
myoglobin (Mb) is a monomeric cytoplasmic hemoprotein that is tissue restricted to cardiomyocytes and oxidative skeletal myofibers. In general, studies support the hypothesis that Mb is an essential protein that facilitates oxygen delivery from the intracapillary erythrocyte to the mitochondria to maintain oxidative phosphorylation for myocardial contractility (16, 27, 57, 58, 63). Although several studies challenge the role of Mb in intracellular oxygen transport (21, 62), a number of studies utilizing pharmacological inhibitors of Mb support a functional role of Mb in facilitating oxygen transport in striated muscle (16, 27, 57, 58). As an alternative, we pursued a gene disruption strategy to generate mice that are viable and fertile despite a complete absence of Mb (20). Mice lacking Mb have preserved cardiac function and exercise capacity compared with their wild-type (WT) littermates under normoxic conditions. Additionally, Mb-deficient (Mb–/–) mice survive by mounting a complex cellular and molecular compensatory response that includes increased capillary density and an induction of the hypoxia gene program (hypoxia-inducible factor-1α, VEGF, heat shock protein27, etc.) (22, 23, 36). These findings have prompted our laboratory and others to reexamine the functional role of Mb in the heart.
Additional functions have been proposed for Mb, including the role(s) as a reservoir for oxygen, as a cytoprotective protein against reactive oxygen species (i.e., H2O2, , and ONOO–), and as a modulator of the signaling molecule nitric oxide (NO) (19, 28, 30, 59). NO is an important inter- and intracellular physiological messenger capable of evoking a number of cellular responses that are beneficial or potentially toxic. NO avidly binds to heme-containing proteins such as soluble guanylate cyclase (sGC), Mb, hemoglobin, and cytochrome c oxidase, and these hemoproteins are essential in the modulation of NO bioactivity and responsiveness to the concentration of NO (13, 15, 26, 35).
In the present study, we challenged the Mb-null mice with chronic hypoxia in an attempt to further define and explore alternative functional roles for Mb. We hypothesized that the cellular and molecular adaptations in Mb mutant mice would be insufficient to maintain cardiac function in response to chronic hypoxic conditions. Here, we present evidence to support our hypothesis and establish Mb as an important regulator of NO homeostasis in the heart.
MATERIALS AND METHODS
Animals. Homozygous Mb–/– mice were generated utilizing gene disruption technology and genotyped as previously described (20). The handling and use of these mice were in accordance with National Institutes of Health and University of Texas Southwestern Medical Center's institutional guidelines. All the mice used in this study were adult males (2–4 mo old) and were from a C57BL/6J strain background.
Hypoxic exposure. A Plexiglas chamber (87 × 42 × 45 cm) was engineered to maintain a constant hypoxic environment of 10% O2 by infusing a gas mixture of 10% O2 and 90% N2 into the chamber at a rate of 1 l/min. It was continuously monitored for oxygen and carbon dioxide concentrations, temperature, and humidity. Within the hypoxic chamber, mice were individually housed in 15 × 32-cm cages and were maintained in a 12:12-h light-dark cycle. To ensure the mice were exposed to a significant hypoxic stimulus, we measured the hematocrit levels of these mice using conventional methods. Echocardiographic M-mode images (Hewlett-Packard Sonos 5500) were obtained from a parasternal short-axis view at the level of the left ventricular (LV) papillary muscle in lightly anesthetized mice (0.015 ml Avertin/g body wt) as previously described (36). As demonstrated by Roth el al. (47), Avertin has minimal adverse cardiodepressant effects after an equilibrium state has been reached. Thus, as recommended by Roth el al. (47), echocardiographic measurements were performed 15 min after the mice were anesthetized. The heart rate was also monitored during echocardiography.
In selected studies, WT and Mb–/– mice were treated with the NO synthase (NOS) inhibitor Nω-nitro-l-arginine methyl ester (l-NAME; Sigma; St. Louis, MO). Mice were pretreated for a 1-wk period under normoxic conditions with l-NAME (0.07 mg · g body wt–1 · day–1), which was dissolved in the drinking water (33, 34, 39, 42). Water consumption was monitored to ensure both the WT and Mb–/– mice received similar doses of l-NAME. The mice were then exposed to chronic 10% hypoxia, and treatment with l-NAME was continued during this hypoxic period. Echocardiography was used to evaluate the LV systolic function in mice before l-NAME treatment and 1 wk after l-NAME treatment under normoxic conditions and after exposure to hypoxic conditions.
Histological and immunohistochemical analyses. Tissues for histological and immunohistochemical studies were perfused via the LV with PBS, followed by 4% paraformaldehyde-PBS (pH 7.4), excised, postfixed for 14–16 h at 4°C, and embedded in paraffin. Selected tissue sections were stained with hematoxylin and eosin or Masson's trichrome for morphological analyses. Immunohistochemical analyses were undertaken using a primary rabbit anti-inducible NOS (iNOS) serum (1:200, Santa Cruz Biotechnology; San Diego, CA), which was detected by a secondary antiserum (1:100 goat anti-rabbit biotinylated serum; Vector Laboratory, Burlingame, CA) and horseradish peroxidase (HRP)-dimethyl-aminoazobenzene (DAB). Substitution controls revealed an absence of signal (data not shown). To limit the loss of immunohistochemical signal, no counterstain was utilized.
Vascular quantitation. Capillary density was quantified in WT and Mb–/– ventricles by staining paraffin-embedded methyl-Carnoy-fixed sections with biotinylated Bandiera simplicifolia lectin B4 (20 mg/ml, Vector Laboratory) and detected with DAB (36). Digital photographs were taken of the LV, and capillary density was quantified at four different levels (50 μm apart) for each heart. With the use of a computerized digital analysis program (Scion Image 1.62c), images were captured, and vascularity was quantitated from four different (random) regions in the lateral LV.
Western blot analysis and biochemical assays. Western blot analysis was performed according to a previously published protocol (20). Rabbit anti-Mb serum (1:3,000, DAKO; Carpenteria, CA) was used as the primary antiserum, which was detected using a HRP-conjugated secondary antiserum. Mouse monoclonal α-tubulin antibody (1:3,000, Sigma) was used as a standard to ensure equal protein loading of the lanes. Band intensity was quantitated using a computerized digital analysis program (Scion Image 1.62c). NOS enzymatic activity was determined in the supernatant of homogenized hearts from normoxic and hypoxic WT and Mb–/– mice by measuring the conversion of l-[3H]arginine to l-[3H]citrulline as previously described (43, 53). NOS activity was linear with time (up to 60 min) and with protein concentration (up to 60 μg/reaction). Furthermore, NOS activity was fully inhibited by 2 mM l-NAME. Calcium-dependent and -independent NOS activity were partitioned by the addition of 5 mM EGTA to the incubation mixture. Selective inhibition of neuronal NOS (nNOS) and endothelial NOS (eNOS) activities by Ca2+ chelation and the lack of effect on iNOS activity were confirmed in lysates of COS-7 cells singularly transfected with cDNA for each isoform.
cGMP assay. Normoxic and hypoxic mice were euthanized by cervical dislocation, and the hearts were excised within 15 s. The great vessels and atria were removed, and the heart was snap frozen in liquid N2 and stored at –80°C. The frozen ventricles were weighed, homogenized in 1 ml of ice-cold 0.5 N perchloric acid, and then centrifuged at 14,000 rpm at 4°C for 5 min. The supernatant contained cGMP, which was measured using a standard radioimmunoassay as previously described (9, 17).
Statistical analysis. Data analysis between WT and Mb–/– mice at any one hypoxic time period was performed using two-tailed paired Student's t-test. Comparison within a genotype to varying durations of hypoxia was performed utilizing ANOVA (single factor), followed by a Bonferroni post hoc test. Data values are reported as means ± SE.
Survival in the absence of Mb is dependent on cellular and molecular adaptations resulting in preserved cardiac function. To further define the role for Mb, we challenged Mb–/– and WT age- and gender-matched mice to 10% O2 for various time periods (1 day to 3 wk). At baseline (normoxic conditions), there was no difference in the hematocrit levels between WT and Mb–/– mice (51.31 ± 0.66% vs. 51.30 ± 0.72%, respectively, n = 21 for each group). We observed an increase in the hematocrit levels in both hypoxic WT and Mb–/– mice, and at any single hypoxic time period there was no difference in the hematocrit levels between WT and Mb–/– mice (Fig. 1A). For example, after 3 wk of hypoxic (10% O2) exposure, we observed a similar increase in the hematocrit levels in both WT and Mb–/– mice compared with their respective baseline levels (73.19 ± 2.55% and 74.75 ± 1.31%, respectively, P < 0.001). This polycythemic response in WT control mice was comparable to previously published results and served as a further confirmation that the mice were exposed to hypoxic conditions (64).
Echocardiography was utilized to evaluate the LV systolic function of WT and Mb–/– mice in response to chronic hypoxia by measuring the fractional shortening of the LV. As illustrated in Fig. 1, B and C, Mb–/– mice exposed to chronic 10% O2 for a 1-wk period developed an ∼30% reduction in systolic function compared with normoxic Mb–/– mice (0.43 ± 0.02 and 0.59 ± 0.02, respectively, P < 0.05, n = 6 for each group) and hypoxic WT mice (0.43 ± 0.02 and 0.59 ± 0.03, respectively, P < 0.001, n = 6 for each group). This degree of hypoxia-induced LV systolic dysfunction in the Mb–/– mice was constant and without further deterioration, regardless of the period of hypoxic exposure (1–3 wk). In contrast, hypoxic WT mice did not develop myocardial depression at any time period after exposure to hypoxia. Under normoxic conditions, the heart rate was similar between WT and Mb–/– anesthetized mice [615 ± 36 and 585 ± 70 beats/min, respectively, not significant (NS)]. Furthermore, no evidence of LV dysfunction was observed in normoxic WT and Mb–/– hearts using the anesthetic agent. Finally, under hypoxic conditions, no difference was observed in the heart rate between WT and Mb–/– anesthetized mice, indicating that the anesthetic agent did not have a differential effect on cardiac function between the two groups.
As illustrated in Fig. 1D, histological evaluation of hypoxic WT and Mb–/– hearts showed similar structural differences. This was confirmed by echocardiographic measurements of ventricular dimensions (see Table 1). After 1 wk of hypoxic exposure, the anterior wall thickness was similar between WT and Mb–/– hearts (0.99 ± 0.04 vs. 1.00 ± 0.01 mm, respectively, NS, n = 6 in each group). Likewise, there was no difference in the posterior wall thickness between hypoxic WT and Mb–/– hearts (0.95 ± 0.02 vs. 0.98 ± 0.01 mm, respectively, NS, n = 6 in each group). Figure 1D and Table 1 also indicate the development of right ventricular and LV hypertrophy in both hypoxic WT and Mb–/– hearts. In addition, we did not observe any morphological evidence of either ischemic or necrotic myocardial injury in Mb–/– or WT hearts under hypoxic conditions, as assessed by Masson's trichrome staining (data not shown). Furthermore, we did not observe any differential expression of Mn-SOD and p38 in the WT and Mb-null hearts exposed to hypoxic conditions, suggesting that oxidative stress did not contribute to the LV dysfunction in the Mb-null heart (data not shown).
Hypoxia has been shown to be associated with an increase in capillary density (29, 60, 61). Using an immunohistochemical protocol, as described in previous studies, we quantitated the capillary density in the LV (23, 36). As demonstrated by our laboratory and others (22, 23, 36), under normoxic conditions there is a significant increase in capillary density in the LV of Mb–/– mice compared with the LV of WT mice (Fig. 2A; 4,729 ± 87 vs. 4,116 ± 88 capillaries/mm2, respectively, P < 0.001, n = 18 for each group). Upon exposure to chronic hypoxia, we observed an ∼35% increase in capillary density of the LV in both WT and Mb–/– mice (Fig. 2A; P < 0.001, n = 20 for each group). This increase in capillary density is finite in both hypoxic WT and Mb–/– ventricles, and no further increase occurs with increased exposure to hypoxic conditions.
Mb expression is markedly induced in response to hypoxic conditions in the WT ventricle, as observed by Western blot analysis (Fig. 3, A and B). After 2 wk of hypoxic exposure, the Mb content in WT hearts increased by 31% from normoxic levels (1.22 ± 0.08 vs. 0.97 ± 0.01 mm, respectively, P < 0.05, n = 3).
To determine whether the hypoxia-induced LV systolic dysfunction in hypoxic Mb–/– mice was a reversible process, we exposed age- and gender-matched mice to 2 wk of chronic hypoxia and then returned them to normoxic conditions (21% O2). We evaluated the LV function of these mice by echocardiography immediately after hypoxic exposure versus 1 and 7 days after reexposure to normoxic conditions. As previously shown, we observed a 30% reduction in systolic function in Mb–/– mice compared with WT mice after 2 wk of hypoxia (Fig. 4A; P < 0.001, n = 3 for each group). However, this hypoxia-induced LV systolic dysfunction was reversible in Mb–/– mice after they were returned to normoxic conditions for 1 and 7 days (Fig. 4A). The reversibility of the LV dysfunction of the hypoxic Mb-null mice was consistent with a nonstructural etiology.
Having determined that hypoxia-induced LV dysfunction in Mb mutant mice was a reversible process, we examined whether the LV dysfunction was a NO-mediated mechanism. Therefore, we treated WT and Mb–/– mice with l-NAME, a potent inhibitor of NOS, to determine whether we could prevent the cardiac dysfunction observed in hypoxic Mb–/– mice. As illustrated in Fig. 4B, l-NAME prevented the development of LV systolic dysfunction in hypoxic Mb–/– mice. In addition, l-NAME did not have any adverse effect on the systolic function of normoxic or hypoxic WT mice (Fig. 4B). To determine that both WT and Mb–/– mice were receiving similar amounts of l-NAME, we measured daily water consumption and observed no difference between the two groups (2.44 ± 0.36 vs. 2.51 ± 0.47 ml, respectively, NS, n = 6 in each group).
The above experiment demonstrates that the hypoxia-induced LV systolic dysfunction we observed in hypoxic Mb–/– mice is a NO-mediated process. Because cardiac dysfunction first developed in the Mb mutant mice after 1 wk of hypoxic exposure, we used a conventional assay to measure the ventricular NOS activity level in these mice exposed to 1 wk of chronic hypoxia. Under normoxic conditions, the total NOS activity was similar in WT and Mb–/– hearts (0.50 ± 0.11 vs. 0.42 ± 0.08 pmol · mg protein–1 · min–1, respectively, NS, n = 10 for each group). As shown in Fig. 5A, we observed a more than a twofold increase in total ventricular NOS activity in hypoxic WT mice compared with normoxic WT controls (n = 3 for the hypoxic group). In the hypoxic WT heart, this increase in NOS activity was attributable solely to an increase in Ca2+-dependent NOS activity (eNOS and nNOS), and there was a complete absence of ventricular Ca2+-independent NOS activity (iNOS) in both normoxic and hypoxic WT mice (Fig. 5, B and C). In Mb–/– mice, there was a 1.5-fold increase in the total ventricular NOS activity under hypoxic conditions compared with under normoxic conditions (n = 3 for the hypoxic group). We observed a simultaneous decrease in the Ca2+-dependent NOS activity (Fig. 5B). In addition, there was an approximately sixfold increase in ventricular Ca2+-independent NOS activity (iNOS) in hypoxic Mb–/– mice compared with normoxic Mb–/– mice (Fig. 5C). These results were further corroborated as we observed a marked increase in iNOS expression in the hypoxic Mb-null LV (Fig. 5D) by immunohistochemistry compared with the controls.
To determine whether this NO-mediated LV dysfunction in the Mb–/– heart was a cGMP-dependent mechanism, we measured cGMP levels in ventricles from hypoxic WT and Mb–/– mice (Fig. 6). We observed no significant difference in cGMP levels in ventricles from hypoxic Mb–/– mice compared with hypoxic WT mice (Fig. 6). These results suggest that the LV dysfunction in the Mb-null mice exposed to chronic hypoxia is due to a NO-dependent, cGMP-independent pathway.
We have previously observed that adult mice lacking Mb are viable, fertile, and have preserved cardiac function. In the present study, a distinguishable cardiac phenotype was identified only in Mb mutant mice challenged with chronic hypoxic exposure and not in WT control mice. LV systolic dysfunction was observed in Mb-deficient mice after exposure to chronic hypoxic conditions, suggesting that the molecular and cellular adaptations are insufficient to maintain cardiac function in the absence of Mb. The LV dysfunction observed in Mb-null hypoxic mice was reversible with reexposure to normoxic conditions and could be completely abolished when mice were treated with l-NAME, a pharmacological inhibitor of NOSs. These pharmacological studies support the conclusion that the LV systolic dysfunction in hypoxic Mb–/– mice is via a NO-mediated mechanism.
Although our data suggest a NO-mediated mechanism for the hypoxia-induced LV dysfunction in Mb-null mice, it is possible that other genetic and physiological factors may also contribute to our experimental results. In the present study, we measured serum hemocrit levels as an indirect measure of the oxygen-carrying capacity in hypoxic WT and Mb–/– mice. We observed a similar rise in serum hematocrit levels in both hypoxic WT and Mb–/– mice compared with their respective controls. Therefore, the serum oxygen-carrying capacity in both groups of mice was the same.
We also analyzed the structural integrity of hypoxic WT and Mb–/– hearts utilizing both histological and echocardiographic techniques. We observed a similar degree of right ventricular and LV hypertrophy in the respective mouse models after the exposure to chronic hypoxia using morphological techniques. These observations were further extended and supported utilizing echocardiographic techniques to determine the LV wall thickness. Masson's trichrome staining did not reveal any myocardial necrosis with hypoxic exposure in either group. Therefore, alterations in the structural integrity of the hypoxic heart are unlikely to explain our data.
In the present study, we also measured the LV capillary density in response to hypoxia. Our data reveal a significant increase in LV capillary density in both genotype groups, and this increase was finite after 1 wk of hypoxic exposure. Although there was a significantly greater LV capillary density in the 1-wk hypoxic WT heart compared with the 1-wk hypoxic Mb–/– heart, no significant differences were observed between groups after 2 wk of hypoxic exposure. This finite increase in the capillary density in response to chronic hypoxia was consistent with the notion that the LV dysfunction observed in the Mb-null mouse exposed to chronic hypoxia was a result of limited oxygen delivery. The results of the l-NAME experiments argue against such a mechanism and further support a NO-mediated mechanism for the LV dysfunction observed in the Mb-null mouse exposed to chronic hypoxic conditions.
Our data support the hypothesis that Mb has an additional function as an inactivator or scavenger of NO, which is distinct from its traditional role in oxygen flux. NO is an important inter- and intracellular physiological messenger capable of evoking a number of cellular responses that are either beneficial or potentially toxic to the cell (8, 14, 18, 45, 48, 51, 55). NO is generated by the oxidation of l-arginine and is catalyzed by one of three isoforms of NOS (3). Both nNOS and eNOS have Ca2+-dependent NOS activity and are constitutively expressed in the heart. In contrast, iNOS has Ca2+-independent NOS activity and is usually biosynthesized in a cell after it has been stimulated by a stressful event (i.e., ischemia or hypoxia).
Recent studies indicate that the type of NOS that generates NO determines whether NO will have a beneficial or toxic effect on the cell. For example, eNOS, which localizes to the caveolae, has been shown to generate NO that inhibits β-adrenergic-induced inotropy, whereas nNOS, which targets the cardiac sarcoplasmic reticulum, has been shown to facilitate cardiac contractility (2). In contrast, the induction of iNOS activity in hearts exposed to a stress stimulus (i.e., ischemia or hypoxia) is generally believed to result in detrimental effects (i.e., decreased contractile function, sudden death, etc.) (41, 50). However, recent studies have challenged the notion that iNOS expression has detrimental effects in the cardiomyopathic heart (7, 8, 14, 31, 45).
Many of the intracellular effects of NO are mediated by the ability of NO to bind to various heme-containing proteins (i.e., NO binding to sGC). Therefore, hemoproteins (Mb, hemoglobin, sGC, lipoxygenases, or free radicals such as superoxide) are essential in the modulation of NO bioactivity and responsiveness to the concentration of NO (13, 19, 26, 35). We have previously shown that NO generation is altered in skeletal muscle of Mb-null mice after a 30-Hz electrical stimulation (23). These results support the presence of NO-mediated mechanism(s) in Mb mutant skeletal muscle groups to presumably promote oxygen delivery through vasodilation and/or to mediate NO signaling pathways through adaptations that have not been completely defined. Additionally, Flögel et al. (19) demonstrated that Mb not only has a functional role in oxygen metabolism but also avidly binds NO within myocytes.
NO has a number of beneficial effects in various systems and importantly functions as a physiological vasodilator in the coronary artery vascular bed, a regulator of mitochondrial respiration, a modulator of myocardial contractility, and a key metabolite in blood pressure regulation (1, 30, 32, 59). Most of the functions for NO in the cardiovascular system are attributable to the enzyme eNOS, which is expressed in the endothelium as well as the cardiomyocyte. In contrast, iNOS is expressed in the cardiac myocyte primarily under stressful or pathological conditions. Recent data have challenged this notion that iNOS-derived NO is detrimental to the myocyte and may in fact have a beneficial effect on myocardial oxygen consumption in congestive heart failure (8).
The NOS activity in the WT heart, exposed to either normoxic or hypoxic conditions, is attributed solely to Ca2+-dependent NOS isoforms (nNOS and eNOS) and not to the Ca2+-independent NOS isoform (iNOS) (Fig. 5, A–C). In response to chronic hypoxia, Mb content is markedly increased in the WT heart and presumably scavenges excess NO, resulting in preserved systolic function (Fig. 3). However, in Mb-deficient mice, iNOS is the predominant isoform in the heart under hypoxic conditions (Fig. 5C). Low levels of iNOS activity were observed in the hearts of Mb-null mice exposed to normoxic conditions, indicating a low level of baseline stress; this finding is further supported as we have previously reported an induction of the hypoxia-inducible gene program in these mice (Fig. 5, C and D) (36). Under normoxic conditions, the Mb mutant mice are able to maintain cardiac function as a result of cellular and molecular adaptations that steepen the Po2 gradient within the myocyte as well as decrease the distance of O2 diffusion from the capillary to cardiac mitochondria (22, 23, 36). However, these adaptations are insufficient to maintain cardiac systolic function in Mb-null mice exposed to chronic hypoxia as a result of altered NO homeostasis. This conclusion is based on the rescue of cardiac function in hypoxic Mb–/– mice treated with l-NAME. Thus, in the absence of Mb, perturbed NO homeostasis results in cardiac systolic dysfunction. Similar mechanisms may be associated with clinical heart failure (15, 24, 25, 49, 50). The cardiomyopathic heart is characterized by decreased Mb content, altered NO homeostasis, and ventricular dysfunction (24, 25, 44). In this clinical setting, a component of cardiac dysfunction may be a result of altered NO homeostasis due to impaired inactivation of NO by decreased amounts of Mb.
The cytotoxic effects of NO can be mediated either via a cGMP-dependent or -independent pathway. The cGMP-dependent deleterious effects of NO are mediated by either the stimulation of protein kinase G proteins resulting in decreased myofilament Ca2+ sensitivity or by the inhibition of L-type Ca2+ channels (37, 38, 40, 52). Alternatively, the cytotoxic effects of NO that are cGMP independent are mediated either by the nitrosylation of the sarcomeric proteins or by the direct effect of NO on mitochondrial respiration (4–6, 11, 12, 46, 51, 54–56, 65). Because the hypoxia-induced LV systolic dysfunction observed in the Mb–/– mouse is reversible within 24 h upon exposure to normoxic conditions, we hypothesize that this is the result of a metabolic disturbance rather than a structural alteration. Therefore, one possibility is that the deleterious effects of NO in our murine model is acting via a interaction of NO with the electron transport chain.
A previous study by Chung et al. (10) has demonstrated that NO regulation of mitochondrial respiration was observed only in response to pharmacological doses of nitrite (>10 mM) in the isolated, perfused heart preparation. In contrast, in the present study, we hypothesize that hypoxia-induced LV dysfunction in Mb-null mice is due to an interaction of NO with the electron transport chain. The rationale for this hypothesis is supported by a number of published studies. Taylor et al. (58) demonstrated that under hypoxic conditions, ATP and phosphocreatine are depleted significantly more in rat hearts with decreased concentration of functional Mb compared with hearts with normal levels of Mb. Several additional studies (1, 32, 59) have demonstrated that NO modulates oxygen mitochondrial respiration in the heart. NO has been shown to directly inhibit mitochondrial oxidative phosphorylation through reversible binding to cytochrome c oxidase (complex IV), where NO competes with dioxygen to bind at the binuclear site of cytochrome c oxidase (4–6, 11, 12, 46, 54, 56, 65). As reviewed by Trochu et al. (59) and recently demonstrated by both Clement et al. (12) and Shiva et al. (54), this inhibition of cytochrome c oxidase by NO is enhanced under low oxygen tension.
In our present study, we did not observe any difference in cGMP levels in hypoxic WT and Mb–/– hearts. While we cannot exclude the possibility of a direct nitrosylation-mediated process or a NO-guanylate cyclase-cGMP-mediated mechanism, the current data support the hypothesis that the depressed LV systolic function observed in hypoxic Mb–/– mice is a NO-dependent, cGMP-independent mechanism.
In summary, we demonstrated that Mb is induced in the WT heart in response to a hypoxic challenge. In the absence of Mb, altered NO homeostasis is observed under chronic hypoxic conditions, resulting in reversible LV dysfunction via a cGMP-independent mechanism presumably involving the reversible inhibition of cytochrome c oxidase by NO. These results collectively support the paradigm that Mb functions to facilitate oxygen flux and maintain NO concentration gradients in the cardiomyocyte.
This study was supported by National Heart, Lung, and Blood Institute Grant HL-63788. P. P. A. Mammen is a Pfizer Postdoctoral Fellow in Cardiovascular Medicine.
We acknowledge the assistance of Drs. David Graber and Ted Chrisman with the cGMP measurements presented in this study.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2003 by the American Physiological Society