INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE HEMODYNAMIC PROFILE OF patients with congestive heart failure (CHF) is characterized by increased systemic vascular resistance with reduced peripheral perfusion. The important role of the endothelium in regulating tissue perfusion was not recognized until recently. Fundamental and clinical studies indicate that impaired endothelium-dependent relaxation is, to a large extent, related to reduced nitric oxide (NO) activity. Restoration of the endothelial function is considered a desirable goal of heart failure therapy. Improvements in endothelium-dependent relaxation have been observed with a variety of interventions, such as supplementations of a precursor of NO (L-arginine), estrogen, angiotensin-converting enzyme inhibition, lipid lowering, radical scavenging by antioxidants, and physical activity (for review, see Ref. 10).

The effects of NO are, however, complex and related to at least three independent smooth muscle mechanisms: 1) activation and regulation of guanylate cyclase sensitivity (8, 19); 2) cGMP-dependent activation of intracellular relaxant mechanisms (1, 4, 7, 23); and 3) a direct activation of K⁺ channels, leading to relaxation (5). The efficacy of NO-induced relaxation derived either from the endothelium or from exogenous sources depends, therefore, on these three mechanisms. However, Münzel et al. (21) reported that nitrate tolerance was associated with a rise in circulating levels of endothelin (ET)-1. Furthermore, the decreased vascular responsiveness to nitrate could be reversed by ET-receptor antagonism (17). Regardless of whether there is a direct relationship between ET-1 and the regulation of smooth muscle sensitivity to nitrate, these data suggested to us that other endothelium-derived factors such as ET-1 and possibly endothelium-derived hyperpolarizing factors (EDHF) (3, 26, 30) may influence smooth muscle responses to NO and its derivatives. These regulatory mechanisms may directly or indirectly regulate guanylate cyclase sensitivity to NO and modify vascular reactivity.

Patients with CHF are known to be tolerant to nitrate therapy, thus explaining the considerably large doses necessary to achieve the desired hemodynamic effects (11, 12). However, the relationship between ET-1 and nitrate-induced dilation has never been investigated in CHF. Studies, however, have revealed the important influence of ET-1 in the pathogenesis of CHF (25). ET_A-receptor (ET receptor A) antagonism has been shown to improve long-term survival in rats with CHF (20, 24) and has beneficial effects on left ventricular and myocyte function (27). Acute administration of an ET_A-receptor antagonist or a specific endothelin-converting enzyme inhibitor has been shown to improve urinary flow rate and urinary sodium excretion in association with an increase in glomerular filtration rate and renal plasma flow (33, 34). Finally, in patients already on standard heart failure therapy, short-term oral endothelin-receptor antagonist therapy improved systemic and pulmonary hemodynamics (28) and produced a significant reduction in forearm vascular resistance (18).

The improvement of cardiac function and vascular resistance by ET_A-receptor blockade in CHF is most likely smooth muscle and cardiomyocyte related, because ET_A receptors are predominantly expressed in these cell types. However, an indirect effect of ET_A-receptor blockade may involve the endothelium and its regulatory role on smooth muscle reactivity (16). Chronic ET_A inhibition further increases circulating levels of ET-1 in CHF (22, 27), which may overstimulate endothelial ET_B receptors (ET receptor B) and consequently modify endothelial cell function. Although nitrate-induced relaxation is endothelium independent, its efficacy is increased by endothelial denudation in vitro (31). We undertook this study to investigate specifically the consequences of CHF on the endothelium-dependent regulation of smooth muscle reactivity to both sodium nitroprusside (SNP) and contractile agents. To investigate the role of ET-1 in this regulatory mechanism, the effects of chronic ET_A-receptor inhibition were investigated in rats with CHF.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Induction of CHF. Myocardial infarction was produced in male Wistar rats (200-250 g) by ligation of the proximal left anterior descending coronary artery (22). In brief, under anesthesia with ketamine-xylazine-water (5:2.5:1.5), the proximal left anterior descending coronary artery was identified and ligated with a 6-0 silk suture. Control rats underwent a similar operative procedure but without ligation of the coronary artery. Animals were maintained on standard rat chow with water ad libitum.

Chronic ET_A-receptor blockade. One group of CHF rats was treated for 4 wk with LU-135252 (LU; Knoll), a selective ET_A-receptor antagonist (15), by gavage once daily at a dose of 50 mg/kg (CHF+LU rats, n = 24), whereas groups of CHF rats (CHF, n = 24) and sham-operated rats (n = 22) received the vehicle (NaCl 0.9%). The treatment started 48 h after ligation of the coronary artery. Hemodynamic parameters were recorded in xylazine-ketamine-anesthetized rats at 4 wk. Treatment with LU was discontinued 48 h before anesthesia to avoid any residual influence of LU on vascular reactivity in vitro.

Hemodynamic measurements and infarct-size determination. The right carotid artery was catheterized (Intramedic, PE 50) for measurement of left ventricular (LV) and systemic arterial pressures on a physiological recorder (model no. 2200; Gould Instrument, Cleveland, OH). Mean values were determined by electronic averaging. Three milliliters of blood were withdrawn from the catheter for measurement of immunoreactive ET-1 levels, as previously described in detail (22, 29). Perimeter tracings of the entire LV and of the infarct area were used to determine infarct size by planimetry.

Isometric recording of tension of isolated cerebral arteries. Rat middle cerebral arteries (MCA) were harvested 4 wk after initiation of treatment and placed in ice-cold physiological salt solution (PSS), containing indomethacin (10 µmol/l, an inhibitor of cyclooxygenase), of the following composition (in mmol/l): 130 NaCl, 4.7 KCl, 1.18 KH₂PO₄, 1.17 MgSO₄, 14.9 NaHCO₃, 1.6 CaCl₂, 0.026 EDTA, and 10 glucose, aerated with 12% O₂-5% CO₂-83% N₂ (pH 7.4). Two-millimeter-long segments were mounted on 20-µm tungsten wires in microvessel myographs (IMF, Univ. of Vermont) as previously described (29, 30, 32).

Chemicals. The following drugs were purchased from Sigma Chemical (St. Louis, MO): indomethacin, N-nitro-L-arginine (L-NNA), phenylephrine (PE), and SNP. ET-1 and sarafotoxin 6c (S6c) were purchased from American Peptide (Sunnyvale, CA), anti-ET-1 antibody was from Peninsula (Belmont, CA), and [¹²⁵I]-labeled ET-1 was from Amersham (Oakville, Ontario, Canada). All drugs for reactivity studies were dissolved in PSS except for indomethacin, which was dissolved in ethanol; the final concentration of ethanol in the bath was 0.1% (vol/vol). Solutions were prepared fresh every day and kept on ice. LU was a generous gift from Dr. Michael Kirchengast (Knoll, Ludwigshafen, Germany).

Statistical analysis. Results are expressed as means ± SE. In all experiments, n represents the number of rats. Statistical differences between means were determined by analysis of variance followed by a Scheffé's F-test. P < 0.05 was accepted as significant for differences between groups of data.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Serum ET-1 levels and hemodynamic and morphometric parameters. Infarct rats gavaged with saline and infarct rats gavaged with LU developed CHF of similar severity: they had lowered mean arterial pressure and first derivative of the change in LV systolic pressure over time (dP/dt) and increased LV end-diastolic pressure and heart rate compared with sham-operated rats (Table 1). There was cardiopulmonary congestion as revealed by the increased lung-to-body weight ratio. The CHF+LU group, however, had higher plasma ET-1 levels and a larger infarct size, although the ratio of ischemic tissue weight to total LV (including septum) weight was similar in both groups.

Baseline vascular parameters. There was no difference in the external diameter of cerebral arteries isolated from sham-operated (156 ± 1 µm, n = 96 segments) and CHF (154 ± 1 µm, n = 110 segments) rats, but there was an increase (P < 0.05) in CHF+LU (173 ± 4 µm, n = 79 segments) rats.

SNP-induced relaxation. SNP induced relaxation of isolated denuded arteries preconstricted with 10 µmol/l PE (Fig. 1). However, the pD₂ value for SNP was decreased (P < 0.05) in CHF rats (7.82 ± 0.06, n = 11) compared with sham-operated rats (8.28 ± 0.04, n = 7). The treatment normalized the decreased sensitivity to SNP in CHF rats (pD₂ = 8.36 ± 0.12, n = 6).

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Relaxation of isolated denuded cerebral arteries from sham-operated rats (Sham, n = 7), rats with congestive heart failure (CHF, n = 11), and CHF rats treated with LU-135252 (CHF+LU, n = 6) induced by cumulative addition of sodium nitroprusside (SNP). Results are expressed as means ± SE. * P < 0.05 vs. Sham rat arteries.

View larger version (31K): [in this window] [in a new window]   Fig. 2.   Contractions [%maximal contraction (%Emax)] induced by a depolarizing solution containing 40 mmol/l KCl (A) and SNP (0.1 µmol/l)-induced relaxation (B) in Sham, CHF, and CHF+LU rats. Experiments were performed with endothelium (+E; n = 15, 11, and 17, respectively), after blockade of nitric oxide (NO) synthase (+L-NNA, 100 µmol/l; n = 8, 10, and 8, respectively), or without endothelium (E; n = 11, 10, and 8, respectively). Results are expressed as means ± SE. L-NNA, N-nitro-L-arginine. * P < 0.05 vs. Sham.  P < 0.05 vs. +E.  P < 0.05 vs. CHF.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Expression of the ratio of relaxation (Rel; SNP, 0.1 µmol/l) to contraction (Cont; 40 mmol/l KCl solution) as an index of the endothelium-dependent regulation of smooth muscle sensitivity to SNP: for a ratio <1, the efficacy of SNP-induced relaxation is low, the reference sensitivity being without endothelium (~1). Experiments were performed with endothelium (+E; n = 15, 11, and 17, respectively), after blockade of NO synthase (+L-NNA, 100 µmol/l; n = 8, 10, and 8, respectively), or without endothelium (E; n = 11, 10, and 8, respectively). Results are expressed as means ± SE. * P < 0.05 vs. Sham.  P < 0.05 vs. +E.  P < 0.05 vs. CHF.

ET-1- and S6c-induced contraction. ET-1 (300 nmol/l) induced a maximal contraction in all groups (Fig. 4), with similar pD₂ values in sham-operated rats, CHF rats, and CHF+LU rats (Table 2). NO synthase inhibition increased (P < 0.05) the sensitivity to ET-1 in CHF rats but had no effect in sham-operated and LU-treated CHF rat arteries (Table 2). In the absence of endothelium (Fig. 5), pD₂ values for ET-1 were similar in all groups (Table 2).

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Effects of cumulative addition of endothelin (ET)-1 on the reactivity of isolated cerebral arteries from Sham (n = 7), CHF (n = 8), and CHF+LU (n = 11) rats. Experiments were performed without (A) or with (B) NO synthase inhibition. Results are expressed as means ± SE.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Effects of cumulative addition of ET-1 after endothelial denudation on the reactivity of isolated cerebral arteries from Sham (n = 8), CHF (n = 10), and CHF+LU (n = 11) rats. Results are expressed as means ± SE.

View larger version (27K): [in this window] [in a new window]   Fig. 6.   Effects of cumulative addition of sarafotoxin 6c on the reactivity of isolated cerebral arteries from Sham (n = 10), CHF (n = 7), and CHF+LU (n = 9) rats. Experiments were performed after NO synthase inhibition. Results are expressed as means ± SE. * P < 0.05 vs. Sham.  P < 0.05 vs. CHF.

View larger version (26K): [in this window] [in a new window]   Fig. 7.   Effects of cumulative addition of sarafotoxin 6c after endothelial denudation on the reactivity of isolated cerebral arteries from Sham (n = 10), CHF (n = 7), and CHF+LU (n = 5) rats. Results are expressed as means ± SE. * P < 0.05 vs. Sham.  P < 0.05 vs. CHF.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the present study, we report that CHF alters vascular smooth muscle reactivity to SNP and the endothelium-dependent regulation of smooth muscle reactivity. Despite a lack of effect of a chronic inhibition of ET_A receptors on cardiac function, this treatment normalized almost all measured vascular responses. Surprisingly, however, it is the endothelium-dependent regulation of smooth muscle reactivity that was altered in CHF. Furthermore, ET_A-receptor antagonism modified the endothelium-dependent regulation of smooth muscle reactivity rather than smooth muscle reactivity per se.

Responses to SNP. Patients with CHF are known to be tolerant to nitrate therapy, thus explaining the considerably large doses necessary to achieve the desired hemodynamic effects (11, 12). As shown in Figs. 1 and 3, sensitivity to SNP is decreased in CHF, which may explain this tolerance. Most interestingly, the results suggest that smooth muscle sensitivity to SNP remains low and unaffected regardless of the experimental conditions (Fig. 3); compared with the sham-operated group, CHF is associated with a lack of endothelium-dependent regulation of smooth muscle sensitivity to SNP. It is therefore clear that CHF uncouples smooth muscle sensitivity to SNP from endothelial regulation.

Contractile responses to high K⁺. Contractions induced by a depolarizing solution appear to be independent of endothelial regulation in cerebral arteries of control rats, as shown by Fig. 2A. It is most likely that under normal conditions, cerebrovascular contractile responses are regulated by the endothelium to be homogeneous regardless of whether NO is present. We have previously reported that endothelial mechanisms compensate for a lack of NO in physiological conditions via the intervention of other factors such as EDHF (30). CHF, however, compromises this mechanism (Fig. 2A). Contractions are decreased under basal experimental conditions and potentiated by NO synthase inhibition. Smooth muscle responses, however, are similar to those obtained in sham-operated rats after endothelial denudation. Thus NO exerts a greater inhibitory influence on contractions induced by high K⁺. The mechanisms involved in endothelium-dependent regulation of smooth muscle contractility under depolarized conditions are unclear. First, blockade of NO synthase induced a similar increase in tension in the three groups. Thus basal NO-dependent regulation of tone is not affected. This does not mean that in the presence of stimuli, NO production is not affected: we previously reported that after endothelial injury, stimulated but not resting endothelial function was altered (31). Likewise, in CHF, high external K⁺ seems to facilitate NO release, because endothelial denudation did not reduce the contraction to the level of the control conditions, i.e., with endothelium. The mechanisms responsible for the changes in contractile response to high K⁺ may be related to a change in the mechanisms by which smooth muscle membrane potential regulates endothelial responses. Work by Dora et al. (9) suggests that smooth muscle depolarization increases endothelial intracellular Ca²⁺ via gap junctions in resistance arteries. More recently, it has been shown that contractions induced by high K⁺ and norepinephrine were increased by a gap junction inhibitor in an endothelium-dependent manner (14). Alternatively, Fleming et al. (13) reported that smooth muscle contraction increased endothelial NO synthase activity in a Ca²⁺-independent process associated with tyrosine kinase activation. These mechanisms are most likely essential for vascular control, and myoendothelial communications may be one of the vascular targets of CHF. Smooth muscle depolarization will trigger NO release from the endothelium. This may represent a compensatory mechanism for the lower guanylate cyclase sensitivity, possibly in relation to the change in ET-1 metabolism.

Contractile responses to S6c. CHF strongly increased S6c-dependent contractions in endothelium-denuded arteries (Fig. 6). This observation has been previously reported in heart failure (6). This contractile response was, however, efficiently prevented by the endothelium: S6c induced no contraction in its presence and a similar contraction after NO synthase inhibition in CHF and sham-operated rats (Fig. 5). Furthermore, the exaggerated contractile response observed in denuded arteries was normalized by LU in CHF rats except for the highest concentration of S6c tested (300 nmol/l). This suggests that blockade of ET_A receptors modifies ET_B-receptor number and/or sensitivity. It is unlikely that the decreased contractile response to S6c observed in CHF+LU is due to the blockade of ET_B receptors by the ET_A antagonist for two reasons: first, LU was withdrawn from the rat diet 48 h before anesthesia, and, second, it is increased endothelial ET_B-receptor activity that is observed (Fig. 5) rather than a decrease (see below). We know also that ET_A receptors are not expressed on endothelial cells (2). Therefore, chronic ET_A-receptor antagonism increases smooth muscle ET_B-receptor-mediated responses.

Contractile responses to ET-1. In sham-operated animals, the sensitivity of ET-1-induced contraction was affected neither by NO synthase inhibition nor by endothelium removal. pD₂ values were similar in the three groups of animals (Table 2). In CHF, however, the sensitivity of ET-1 was increased by NO synthase inhibition compared with the control experimental conditions (Table 2). These data are in agreement with the increased responses to high external K⁺ after NO-formation blockade (Fig. 2A), confirming that CHF is associated with an upregulation of NO-dependent inhibition of smooth muscle cell contraction. Interestingly, LU treatment prevented the changes in ET-1 sensitivity associated with CHF. Therefore, SNP-induced relaxation and S6c-, ET-1-, and high external K⁺-induced contraction all appear to be under a tight endothelium-dependent regulation. CHF seems to alter this regulatory mechanism, which is, however, unclear. The lack of increase in smooth muscle sensitivity to SNP after NO synthase inhibition in CHF rats could favor smooth muscle contractility, an effect reversed by the LU treatment and compensated by an over production of NO during smooth muscle stimuli. This decrease in smooth muscle sensitivity to ET-1 in LU-treated CHF rats could also be related to the upregulation of ET_B-receptor-mediated vasorelaxation. Altogether, LU therapy in CHF normalized ET-1 sensitivity: this suggests that the beneficial effects of ET_A-receptor blockade in CHF may be partly endothelial via an upregulation of the ET_B-dependent pathway and an improvement of the regulatory mechanisms of smooth muscle sensitivity to NO.

Clinical relevance. The reduced sensitivity to SNP may explain the reported greater tolerance to nitrates and the unreported but frequently observed reduced incidence of headaches in patients with CHF. Consequently, higher dosages of nitrates are necessary to obtain clinical benefit (11, 12) and may precipitate the development of nitrate tolerance in these patients. The observation that chronic ET_A-receptor inhibition increased smooth muscle sensitivity to SNP suggests that inhibition of ET_A receptors in CHF may reduce nitrate tolerance, as recently suggested (17), and improve peripheral perfusion. Furthermore, all of the beneficial effects of LU on cerebral arteries were not associated with improvement of cardiac function. In fact, CHF+LU rats even had a larger infarct size. One study has previously demonstrated that early initiation of ET_A-antagonist therapy after myocardial infarction was associated with deterioration of LV function (22), whereas it is associated with improvement when started after the acute phase (24). Our results further demonstrate the importance of independent effects of ET_A-antagonist therapy on the myocardium and the vascular wall.