INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

NATRIURETIC PEPTIDES ARE a family of compounds that help modulate cardiovascular responses. Atrial (ANP) and brain natriuretic peptides (BNP) are expressed in high concentrations in the heart, whereas C-type natriuretic peptide (CNP) is found primarily in vascular endothelial cells and the central nervous system (7). Three natriuretic peptide receptors (NPRs) have been described (22). Two of these receptors, NPR-A and NPR-B, are linked to a particulate guanylyl cyclase and appear to mediate most of the biological effects of the natriuretic peptides by increasing intracellular levels of cGMP (22). NPR-A has high affinity for both ANP and BNP and little affinity for CNP, whereas the reverse is true for NPR-B (22). The third receptor, NPR-C, is devoid of any guanylyl cyclase activity and is believed to act primarily as a clearance receptor (24, 26).

Both ANP and BNP are potent diuretics that have vasorelaxant and antimitogenic effects on pulmonary and systemic vascular smooth muscle (1, 2, 9, 13). In addition, they have inhibitory effects on the renin-angiotensin system. Recent studies suggest that ANP and perhaps BNP play important physiological roles in limiting the severity of pulmonary hypertension and right ventricular hypertrophy that develops during exposure to chronic hypoxia. Cardiac and plasma levels of both peptides are increased during chronic hypoxia (1, 9, 27), most likely as the result of increased cardiac synthesis (9) and decreased pulmonary clearance (16). The administration of exogenous ANP or BNP during chronic hypoxia mitigates the development of pulmonary hypertension, pulmonary vascular remodeling, and right ventricular hypertrophy (14, 17). Conversely, disruption of ANP-NPR-A signaling by the administration of monoclonal antibodies against ANP or targeted disruption of the genes for ANP or NPR-A worsens pulmonary hypertensive and right ventricular hypertrophic responses to chronic hypoxia (20, 35, 41).

Unlike pulmonary circulation, systemic arterial pressure does not increase in rodents during exposure to chronic hypoxia (36). As a result, hypertrophy of the left ventricle is not normally seen during adaptation to hypoxia, or is limited to the right ventricular portion of the interventricular septum (10). However, in a preliminary study of mice with gene-targeted disruption of the gene for NPR-A (19), we observed increases in left ventricular as well as right ventricular mass after exposure to chronic hypoxia. Although systemic arterial pressures were not measured in that study, we wondered whether hypoxia-induced left ventricular hypertrophy was caused by the lack of an inhibitory effect of NPR-A on cardiac growth as opposed to an increase in left ventricular afterload. Recent studies suggest that ANP can inhibit cardiac growth independent of its hemodynamic effects. For example, an antimitogenic effect of ANP on cardiac myocytes has been demonstrated in vitro (5, 38), and in one study (25), left ventricular ANP expression was inversely proportional to left ventricular mass in two strains of rats with equal systemic arterial pressures.

In the present study, we hypothesized that gene-targeted disruption of NPR-A worsens right ventricular hypertrophy and causes left ventricular hypertrophy during adaptation to hypoxia. Furthermore, we hypothesized that the potentiation of hypoxia-induced cardiac hypertrophy caused by NPR-A disruption is independent of its antihypertensive effect on pulmonary and systemic arterial pressure. To test these hypotheses, we measured right and left ventricular free wall mass, interventricular septum mass, right ventricular pressure, and systemic arterial pressure in mice with 2 (wild type), 1, and 0 functional copies of the NPR-A gene after exposure to 3 wk of normoxia or hypobaric hypoxia.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Mice. Generation of mice with various copies of NPR-A has been reported previously (30). Briefly, homologous recombination was used for disruption of Npr1, the gene that encodes NPR-A in mice. The disrupted Npr1 was constructed by replacing exon 1, intron 1, and a portion of exon 2 with the neomycin resistance gene oriented opposite of Npr1. Male chimeras consisting of strain 129 embryonic stem cells containing the disrupted Npr1 (1/0) were mated with strain B6 wild-type females (1/1) to produce F1 129 × B6 hybrid offspring containing 2 copies (1/1, wild type) or 1 copy (1/0) of a functional Npr1 gene. Because the 129 and B6 parental mouse strains are inbred, the F1 129 × B6 hybrid mice are genetically identical except at the targeted locus. Breeding pairs of F1 mice heterozygous for the disrupted Npr1 (0/1) produced F2 mice with 0, 1, or 2 copies of the Npr1 gene (termed as 0, 1, or 2 copy mice).

Hypoxic exposure. Mice were placed in hypobaric chambers (0.5 atm) for 3 wk. An air intake valve was adjusted to maintain intrachamber pressure at 380 mmHg (0.5 atm) while allowing adequate airflow through the chamber (10-15 l/min) to prevent accumulation of CO₂ and NH₃. Intrachamber pressure was monitored via a pressure gauge in the wall of the chamber. Chambers were opened three times weekly to clean animal cages and to replace food and water. All mice were fed standard mouse chow and were allowed to take food and water ad libitum. Normoxic controls were kept in identical cages adjacent to the hypoxic chambers.

Hemodynamic measurements. After 3 wk of hypoxia or normoxia, mice were weighed and anesthetized under normoxic conditions with ketamine (60 mg/kg im) and xylazine (20 mg/kg ip). The trachea was cannulated with a blunted 20-gauge needle and the lungs were ventilated with room air using an inspiratory pressure of 9 cmH₂O and end-expiratory pressure of 2 cmH₂O. The right carotid artery was cannulated with a 25-gauge angiocatheter connected to a Statham P23-G pressure transducer with a short segment of polyurethane-50 (P-50) tubing. Carotid arterial pressure was recorded continuously on a polygraph. Carotid artery peak pressure (CAPP) was measured as the average PP recorded over a 1-3 min period during which pressure recordings remained stable. Right ventricular PP (RVPP) was measured as described previously by Fagan et al. (6). Briefly, the anterior chest wall was shaved and a 26-gauge needle attached to a Statham P23-G pressure transducer by a short segment of P-50 tubing was inserted directly into the right ventricle using a transthoracic approach. RVPP was recorded as described above for CAPP.

Lung histology. Lungs were fixed by intratracheal infusion of normal buffered formalin at a pressure of 23 cmH₂O. After fixation, lungs were embedded in paraffin, sectioned, and stained with trichrome. All slides were reviewed simultaneously by two investigators in blinded fashion. Pulmonary vessels >100 mm in diameter or fully muscularized vessels associated with bronchi were excluded because these vessels are uniformly muscularized. Each vessel was categorized as being nonmuscular (no smooth muscle identifiable), partially muscularized (smooth muscle identifiable in <3/4 of the vessel circumference) or fully muscularized (muscularization >3/4 of vessel circumference). The percentage of muscularized peripheral pulmonary vessels was determined by dividing the number of vessels in each category by the total number counted for that experimental group. Lung sections were examined from three mice in each experimental group and at least 35 vessels were examined per mouse.

Hydroxyproline levels. Hydroxyproline concentrations were measured by alkaline hydrolysis of tissue homogenates as described previously (37). Briefly, frozen samples of right and left ventricular free wall were homogenized in nine volumes (wt/vol) of distilled water. An aliquot of homogenate was mixed with NaOH (2 N final concentration) in a total volume of 50 µl. Samples were heated to 120°C in an autoclave for 20 min, mixed gently with 450 µl of chloramine-T, and incubated for 25 min at room temperature. Samples were then mixed with 500 µl of freshly prepared Erlich's aldehyde reagent and incubated for 20 min at 65°C. Samples were read in a spectrophotometer at 550 nm against a standard curve of 2-20 µg/ml of hydroxyproline.

ANP measurements. Plasma samples were acidified with three parts 4% acetic acid and prepared on C₁₈ extraction columns (Waters; Milford, MA) activated with 10 ml methanol and 10 ml deionized water. Samples were loaded on the columns, washed with 10 ml 4% acetic acid, and eluted with 2 ml of 90% ethanol, 0.4% acetic acid, and evaporated to dryness by vacuum centrifugation. Dried samples were reconstituted in assay buffer and ANP concentration was measured using a commercially available enzyme-linked immunoassay with rat -ANP as a standard (Cayman; Ann Arbor, MI). Interassay variability using this immunoassay is ~10% in our laboratory.

Statistics. Mean values were calculated for each group of mice. Differences in mean values between normoxic and hypoxia-adapted NPR-A genotypes were compared by two-way ANOVA. Where significant differences were observed, pairwise comparisons were done using Fisher's least-significant difference test. Differences in percent muscularization of pulmonary vessels were analyzed by proportion comparison and Yates correction. Data are expressed as means ± SE. Differences in mean values were considered significant at P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effect of NPR-A gene expression and hypoxia on body weight and hematocrit. Body weight was not affected by NPR-A genotype under normoxic or chronically hypoxic conditions (Table 1). Under normoxic conditions, hematocrit was lower in 0 copy than in wild-type mice, but there were no genotype-related differences in hematocrit in hypoxia-adapted mice. Three weeks of hypoxia suppressed body weight and increased hematocrit in each NPR-A genotype compared with their respective normoxic controls (Table 1).

Effect of NPR-A gene expression on right ventricular pressure and mass. Under both normoxic and chronically hypoxic conditions, RVPP was greater in 0 copy than in 1 or 2 copy mice (Fig. 1A). Exposure to 3 wk of hypoxia increased RVPP in each genotype compared with its normoxic control. Although RVPP was greatest in the hypoxia-adapted 0 copy mice, absolute and percent increase in RVPP compared with normoxic controls were similar in all three genotypes (absolute increase: 11.1, 13.9, and 14.2 mmHg; percent increase 43, 61, and 41%; for 2, 1, and 0 copy mice, respectively).

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Right ventricular systolic pressure (RVPP, A) and right ventricle weight-to-body weight (RV/BW, B) ratio in mice with 2, 1, or 0 copies of the NPR-A gene under normoxic condition and after adaptation to 3 wk of hypobaric hypoxia. Values are means ± SE, *P < 0.05 vs. normoxic controls; P < 0.05 vs. 2 copy mice; P < 0.05 vs. 1 copy mice.

Effect of NPR-A gene expression on CAPP and left ventricular mass. No significant difference in CAPP was seen among 0, 1, or 2 copy mice kept under normoxic conditions (Fig. 2A). In hypoxia-adapted mice, CAPP was similar in 0 and 2 copy mice, but was lower in 1 copy mice than in the other genotypes (Fig. 2A). Three weeks of hypoxia did not increase CAPP in any NPR-A genotype compared with their normoxic controls.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Carotid artery peak pressure (CAPP) (A), left ventricle plus septum weight-to-body weight [(LV+S)/BW] ratio (B), and left ventricular free wall weight-to-body weight (LV/BW) ratio (C) in mice with 2, 1, or 0 copies of the NPR-A gene under normoxic condition and after adaptation to 3 wk of hypobaric hypoxia. Values are mean ± SE, *P < 0.05 vs. normoxic controls; P < 0.05 vs. 2 copy mice; P < 0.05 vs. 1 copy mice.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   RV/(LV+S) in mice with 2, 1, or 0 copies of the NPR-A gene under normoxic condition and after adaptation to 3 wk of hypobaric hypoxia. Values are means ± SE, *P < 0.05 vs. normoxic controls.

Effect of NPR-A gene expression and hypoxia on right and left ventricular collagen content. Collagen content of the right and left ventricles were measured to determine whether the increase in ventricular mass was due to increased myocardial fibrosis. No NPR-A genotype-related differences in right or left ventricular hydroxyproline concentration were seen under normoxic or chronically hypoxic conditions. (Fig. 4).

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Right (RV, A) and left ventricular (LV, B) hydroxyproline concentrations in mice with 2, 1, or 0 copies of the NPR-A gene under normoxic conditions and after adaptation to 3 wk of hypobaric hypoxia. Values are means ± SE.

Effect of NPR-A gene expression and hypoxia on muscularization of peripheral pulmonary vessels. Under normoxic conditions, mice with 0 copies of the NPR-A gene had a higher percentage of partially muscularized peripheral pulmonary vessels than two copy mice (Table 2). Hypoxia increased the percentage of partially muscularized vessels in wild-type mice. There was a trend toward a hypoxia-induced increase in fully muscularized vessels in 0 copy mice, but the difference was not quite statistically significant (P = 0.08). Under hypoxic conditions, 0 copy mice had a higher percentage of fully muscularized pulmonary vessels than 2 copy mice.

Effect of NPR-A gene expression on plasma ANP levels. No significant differences in plasma ANP levels were seen among NPR-A genotypes under normoxic or chronically hypoxic conditions; however, there were trends toward higher plasma ANP levels in 0 copy than in 2 copy mice under both conditions (Fig. 5). When data from normoxic and hypoxic mice were combined, plasma ANP levels were significantly higher in 0 copy than in 2 copy mice (P = 0.01).

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Plasma atrial natriuretic peptide (ANP) levels in mice with 2, 1, or 0 copies of the NPR-A gene under normoxic conditions and after adaptation to 3 wk of hypobaric hypoxia. Values are means ± SE, *P < 0.05 vs. 2 copy mice.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the present study, right ventricular pressure was increased in NPR-A-deficient mice under normoxic and hypoxic conditions. These findings are in agreement with those of earlier studies in mice with targeted disruption of the genes for ANP and NPR-A (20, 41) and strongly suggest that ANP/NPR-A signaling plays an important role in modulating pulmonary arterial pressure. In addition, the present study suggests that NPR-A plays an important role in inhibiting cardiac hypertrophy, particularly under hypoxic conditions. The right ventricular hypertrophic response to hypoxia was nearly three times greater in mice with absent NPR-A expression than in mice with 1 or 2 copies of the functional NPR-A gene. Furthermore, 3 wk of hypoxia increased left ventricular free wall weight >40% in mice with absent expression of the gene for NPR-A, but had no effect on left ventricular mass in wild-type or 1 copy mice.

The cause of the heightened hypertrophic response to hypoxia in the absence of NPR-A is not clear. In a previous study, Zhao et al. (41) concluded that the hypertrophic effect of disrupted NPR-A expression on the right ventricle was secondary to the greater severity of hypoxic pulmonary hypertension found in these mice. They also saw a significant increase in left ventricular mass, but attributed this finding to an increase in the right ventricular portion of the interventricular septum, although interventricular septum and left ventricular free wall weight were not measured. In the present study, the heightened right ventricular hypertrophic response to hypoxia in NPR-A knockout mice could not be explained by a greater pulmonary hypertensive response. The RVPP was greater in 0 copy mice than in 1 or 2 copy mice, but absolute and percent increase in RVPP induced by hypoxia was the same in all three genotypes. Similarly, hypoxia had no effect on CAPP in any NPR-A genotype, but increased left ventricular mass in the 0 copy mice. These findings suggest that the cardiac hypertrophic response to hypoxia in mice that lack NPR-A expression is not due entirely to hemodynamic alterations.

Findings from other studies support the hypothesis that the inhibitory effect of ANP/NPR-A signaling on cardiac growth is not dependent on its antihypertensive effect. In cultured cardiac myocytes, cellular hypertrophy is inhibited by ANP and exaggerated by the NPR-A antagonist, HS-142 (5, 11). Recently, Knowles et al. (21) demonstrated that NPR-A deficient mice have greater left ventricular mass than wild-type mice even when systemic blood pressure is kept equal in both genotypes from birth by the administration of antihypertensive medication. Furthermore, they found that left ventricular hypertrophic responses to aortic banding were fivefold greater in NPR-A knockout than in wild-type mice. In another study, Kishimoto et al. (15) found that crossing NPR-A knockout mice with transgenic mice that have cardiac specific overexpression of NPR-A resulted in mice with decreased cardiac myocyte size but the same systemic arterial pressure as NPR-A knockout mice. These studies strongly suggest that ANP/NPR-A signaling has a direct inhibitory effect on cardiac hypertrophy.

Mechanism(s) by which NPR-A inhibits cardiac hypertrophy is not known. In vitro studies suggest that the antihypertrophic effect of ANP on cardiac myocytes is mediated by cGMP (5, 38). ANP-induced increases in intracellular cGMP have been shown to both inhibit and to increase mitogen-activating protein kinases (MAPKs) activation (33, 34, 38). Various MAPKs have been implicated in the signal transduction of hypertrophic stimuli. However, in the study by Knowles et al. (21), MAPKs activity measured in whole heart tissue was the same in NPR-A knockout and wild-type mice 7 days after aortic banding or sham operation.

Another possibility is that disrupted expression of NPR-A results in altered expression of other mediators of cardiac growth, such as the adrenergic and renal angiotensin systems and endothelin-1 (ET-1). ANP has been shown to antagonize many of the effects of angiotensin II and aldosterone and to inhibit synthesis of ET-1 (12). Although no effect of ANP genotype has been seen on baseline ET-1 expression or nitric oxide synthase activity (29), previous studies have shown increased catecholamine and angiotensin II levels in ANP-deficient mice (28). Studies of the expression and activity of neurohumoral systems that could effect cardiac growth have not been done in NPR-A-deficient mice and are needed to further elucidate mechanisms responsible for the heightened cardiac hypertrophic responses associated with absent NPR-A expression.

Marked interstitial fibrosis has been reported in the hearts of NPR-A-deficient mice and could contribute to the increase in cardiac mass (30). However, we found no significant genotype-related differences in collagen content as assessed by hydroxyproline concentration under normoxic or chronically hypoxic conditions, suggesting that the increase in cardiac mass was due to an increase in cardiac cell size or number. These findings are consistent with those of a recent study (15) that found that cardiac myocyte size was increased in the hypertrophied hearts of NPR-A knockout mice compared with wild-type mice.

Finally, it is possible that the increase in left ventricular mass in hypoxia-adapted NPR-A-deficient mice was not the result of absent NPR-A signaling but instead was caused by diastolic ventricular interaction with the hypertrophied right ventricle. The greater pulmonary hypertension that develops in the NPR-A knockout mice may result in higher right ventricular end-diastolic volumes that compromise left ventricular filling and affect left ventricular stroke work (3). The present study cannot rule out ventricular interaction as a mechanism for hypoxia-induced left ventricular hypertrophy. However, this mechanism has not been known to cause left ventricular hypertrophy during chronic hypoxia in other animal models. Indeed, in a previous study (20), we found no evidence of hypoxia-induced left ventricular hypertrophy in ANP knockout mice, despite the development of severe right ventricular hypertrophy.

Greater left ventricular mass in NPR-A knockout mice has been described previously by Oliver et al. (30) under normoxic conditions and was also seen in the present study. However, Oliver et al. (30) found that mean systolic blood pressure was 16 mmHg greater in NPR-A knockout than in wild-type mice, unlike the present study where no genotype-related differences in CAPP were observed. The different findings between the present study and that of Oliver et al. (30) may be related to hemodynamic measurement techniques. Oliver et al. (30) measured systemic arterial pressure by tail-cuff in conscious animals, whereas we measured systemic pressure by catheterization of the carotid artery under anesthesia. Although it is possible that the animals in our study had genotype-related differences in blood pressure when conscious, it is unlikely that the relatively small increase in systemic arterial pressure observed by Oliver et al. (30) (~10% greater than wild-type mice) accounted for the marked increase in left ventricular mass. In fact, in another study, Oliver et al. (31) found no difference in left ventricular mass among mice with 1, 2, or 3 copies of a functional NPR-A gene despite a difference of 14 mmHg in mean systolic arterial pressure between 1 and 3 copy mice. Although we cannot exclude the possibility that the animals in our study would have manifested genotype-related differences in blood pressure had we measured them in conscious animals, previous studies have found no increase in systemic arterial pressure in rats (36) or humans (8) exposed to chronic hypoxia.

In the present study, right and left ventricular hypertrophic responses to hypoxia were greater in the NPR-A knockout mice than we had observed in an earlier study of ANP knockout mice (20). The less severe hypertrophic response in ANP knockout mice may be due, in part, to NPR-A activation by BNP. We have previously shown that cardiac BNP expression and plasma BNP levels are increased in rats adapted to chronic hypoxia (9) and that the administration of exogenous BNP inhibits hypoxia-induced increases in right ventricular pressure and mass (17). Expression of NPR-A, and BNP are elevated in mice with absent ANP expression (39, 40) and may serve to limit the cardiac hypertrophic response to hypoxia. Thus disrupted expression of NPR-A may be a better model than disrupted expression of ANP to study the hypertrophic effect of the natriuretic peptides.

Plasma ANP levels were increased in the NPR-A knockout mice. This finding is consistent with an earlier report of increased ventricular steady-state ANP mRNA levels in NPR-A-deficient mice (15). However, it is unlikely that altered expression of ANP or BNP affected cardiac hypertrophic responses in the present study. Both ANP and BNP act to limit cardiac hypertrophic responses instead of accentuating them. Furthermore, the lack of a functional NPR-A receptor should negate the effect of altered expression of either peptide. It is possible that ANP and BNP could act via the other NPRs, NPR-B and NPR-C. However, the dissociation constants for ANP and BNP are 100- to 1,000-fold greater for NPR-B than for NPR-A (4), and in preliminary studies (18), we found no evidence of increased cardiac hypertrophic responses in mice with disrupted expression of NPR-C.

In summary, our data support the hypothesis that NPR-A plays an important physiological role in modulating cardiac mass, pulmonary arterial pressure, and muscularization of pulmonary vessels, probably by mediating the biological effects of ANP and BNP. The inhibitory effect of NPR-A on cardiac hypertrophy appears to be especially important during exposure to chronic hypoxia and is not well explained by genotype-related differences in pulmonary or systemic hemodynamic responses. These findings suggest that NPR-A has a direct inhibitory effect on hypoxia-induced cardiac hypertrophy. Pharmacological manipulations to increase NPR-A signaling or intracellular cGMP levels may represent a new approach to limiting ventricular hypertrophic response in patients with chronic lung disease and ischemic cardiomyopathies.