Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

The ETs and sarafotoxins are a family of potent vasoconstrictor peptides (Inoue et al., 1989; Kloog and Sokolovsky, 1989). Two subtypes of mammalian ET receptor have been cloned and sequenced. The first was denoted ET_A and demonstrates selectivity for ET-1 over ET-3 (Arai et al., 1990). The other receptor, ET_B, is non-isopeptide-selective (Sakurai et al., 1990). Both receptors have been shown to mediate contraction, but the ET_B receptor may also mediate vasodilation via endothelial release of NO (Masaki et al., 1991). The contractile and vasodilator ET_B receptors have been termed ET_B2 and ET_B1, respectively (Sokolovsky et al., 1992; Warner et al., 1993). A putative ET_C receptor, with high affinity for ET-3, has been cloned from the dermal melanophores of Xenopus laevis (Karne et al., 1993). Although there is pharmacological evidence for such a receptor in vascular tissue, a mammalian homolog has not yet been cloned (Masaki et al., 1992; Douglas et al., 1995).

ETs have been implicated in many pathophysiological conditions, including PHT. Elevated circulating ET-1 levels have been reported in patients with both primary and secondary PHT (Stewart et al., 1991). Increased plasma levels also occur secondary to left heart dysfunction, congenital heart defects and cardiac surgery, and they are positively correlated with the degree of PHT and negatively correlated with prognosis (Cody et al., 1992; Yoshibayashi et al., 1991). Kiowski et al., (1995) reported that the ET_A/ET_B receptor antagonist bosentan decreased pulmonary vascular resistance in heart failure patients. Recently, there has been much interest in ET receptor antagonists as possible therapeutic agents for the treatment of cardiovascular disease, including PHT. For this reason, we recently characterized the ET receptors that mediate vasoconstriction in human PRAs (McCulloch et al., 1996) as well as in rat PRAs (MacLean et al., 1994; McCulloch, et al., in press) and investigated how responses to ET-1 are altered by hypoxia-induced PHT (McCulloch and MacLean, 1995; MacLean et al., 1995; McCulloch et al., in press). These studies demonstrated that both ET_A and ET_B receptors mediate contraction in PRAs. ET_A and ET_B receptors also mediate contraction in large rabbit pulmonary arteries (LaDouceur et al., 1993; Hay et al., 1996). Because it is the PRAs that are thought to be functionally important in resistance changes observed in PHT, the main aim of this study was to characterize the ET receptors that mediate vasoconstriction in rabbit PRAs.

The development of PHT secondary to chronic heart failure, induced by coronary ligation in the rat, can be ameliorated by long-term treatment with an ET_A receptor antagonist (Sakai et al., 1996). We have recently demonstrated that LVD, induced by coronary ligation, in the rabbit causes an increase in right ventricular weight, lung weight and the pulmonary artery pressure and muscularization of prealveolar pulmonary arterioles (Deuchar et al., in press). A secondary aim of this study was therefore to determine whether there were any changes in ET-mediated vasoconstriction in the PRAs secondary to LVD in this model.

There is evidence that NOS may be up-regulated in patients with heart failure, and inhibition of NOS increased pulmonary vascular resistance in these patients (Habib et al., 1994). Therefore, we investigated the presence of basal and/or agonist-induced NO release in the sham-operated and coronary-ligated rabbit PRAs by examining the effect of the NOS inhibitor L-NAME on responses to ET-1 and S6c.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

The rabbit model. Lungs were obtained from the rabbit coronary-ligation model of LVD that has been extensively characterized (Pye et al., 1996; Deuchar et al., in press). Briefly, the left circumflex coronary artery of New Zealand White rabbits is fully ligated to produce an acceptable area of infarction. The infarct area was assessed histologically in a cohort of rabbits and was 16 ± 2% for n = 6 rabbits (for methodology, see Pye et al., 1996). Because of the lack of collateral circulation of the rabbit, there is no reduction in this infarct size with time (Pye et al., 1996). Age-matched animals undergo the same procedures as the experimental animals except that the ligatures placed around the coronary artery are not secured. These animals are subsequently referred to as "sham-operated" or control animals. The ejection fraction of the rabbits used in the studies described here was assessed by echocardiography as previously described (Pye et al., 1996). In a cohort of the rabbits used for the present study, the ligation procedure reduced the ejection fraction from 74.3 ± 1.3% to 44.3 ± 0.8% (n = 12, P < 0.0001). We have previously demonstrated changes in this model that are consistent with the onset of PHT (Deuchar et al., in press). In the rabbits used in this study, the ratio of right ventricular weight to body weight was increased from 0.42 ± 0.01 to 0.51 ± 0.02 (n = 25, P < 0.001) and lung weights increased from 12.0 ± 0.3 g to 13.1 ± 0.3 (n = 50, P < 0.5). We have also demonstrated that there is small pulmonary artery vascular remodeling in these rabbits that is consistent with the onset of PHT (Deuchar et al., 1997).

PRAs. Animals were killed by sodium pentobarbitone 8 weeks after the procedure. The lungs were removed, and intralobar PRAs (I.D. ~ 150 µm) were dissected out and cleaned of surrounding parenchyma. These were mounted as ring preparations (~ 2 mm long) on a wire myograph, bathed in Krebs solution at 37°C and bubbled with 16%O₂/5%CO₂ balance N₂. This gave a final bath O₂ concentration (measured with an oxygen electrode and blood gas analyser) of approximately 120 mm Hg and CO₂ tension of approximately 35 mm Hg to yield values equivalent to those found in pulmonary arteriolar blood. Vessels were then tensioned to give a transmural pressure equivalent to approximately 16 mm Hg, which is similar to in vivo pressures of pulmonary arterioles.

Experimental protocol. After a 1-h equilibration period, the response of the PRAs to 50 mM KCl was determined twice. Cumulative concentration-response curves (CCRCs) were then constructed to ET-1, ET-3 or S6c (1 pM-0.3 µM). Some vessels were preincubated with ET receptor antagonists (FR 139317, BQ 788 or SB 209670) for 45 min before the construction of the CCRC, and some were preincubated with L-NAME for 30 min. For comparison, CCRCs were also constructed to 5-hydroxytryptamine (5-HT) and KCl. We analyzed endothelial-dependent vasodilation by preconstricting vessels with the thromboxane mimetic U46619 (30 nM) and constructing CCRCs to ACh (0.1 nM-1 µM), ET-1, ET-3, S6c (all 0.01 pM-0.01 µM), substance P (1 pmol-0.1 µM), bradykinin (0.1 nM-1 µM), the calcium ionophore A23187 (0.1 nM-1 µM), histamine (0.1 nM-1 µM), -methyl 5-HT (0.1 nM-1 µM) and ionomycin (0.1 nM-1 µM). Endothelium-independent vasodilation was assessed by preconstricting with U46619 and constructing CCRCs to SNP.

Drugs and solutions. The composition of the Krebs/bicarbonate saline (pH 7.4) was as follows (in mM): NaCl 118.4, NaHCO₃ 25, KCl 4.7, KH₂PO₄ 1.2, MgSO₄ 0.6, CaCL₂ 2.5, glucose 11, EDTA 23. The following drugs were used: ET-1 (Thistle Peptides, Glasgow, Scotland), ET-3 (Peninsula Laboratories, St. Helens, England), BQ788 (N-cis-2,6-dimethylpiperidinocarboxyl-L-g-methylleucyl-D-I-methocarbonyltrypophanyl-D-norleucine) (Peptide International, Louisville, KY), FR139317((R)2-[(R)-2-[(S)-2-[[1-(hexahydro-1H-azepinyl)]carbinyl]amino-4-methylpertanoyl]amino-3-[3-(1-methyl-1Hindoyl)]propionyl]amino-3-(2-pyridyl)propionic acid (Neosystems, France), SB209670 ([(+)-(1S, 2R, 3S)-3-()1-(3,4-methylenedioxyphenyl)-5-(prop-1-yloxy)indene-2-carboxylic acid] (SmithKline Beecham Pharmaceuticals, King of Prussia, PA), -methyl-5-HT (Semat, St. Albans, England). S6c, L-NAME, U46619 (9,1-dideoxy-11,9-epoxymethano-prostaglandin F₂), ACh chloride, substance P, sodium nitroprusside, bradykinin, histamine diphosphate, ionomycin and A23187 were supplied by Sigma, Poole, UK. Stock solutions of S6c were prepared in 0.1% acetic acid and those of BQ788 in 0.1% dimethyl sulfoxide. All other drugs and dilutions were prepared in distilled water.

Data analysis. pEC₅₀ values were calculated by computer interpolation from individual CCRCs. The CCRCs were not subjected to any curve-fitting program, because they were neither typically sigmoidal nor biphasic and there are no suitable curve-fitting programs that accommodate such curves, i.e., the first component was less than 30% of the maximal response. In addition, responses to S6c were "dropped off" at high concentrations, a phenomenon that curve-fitting cannot accommodate.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Responses to KCl

LVD had no effect on the sensitivity of the PRAs to KCl (fig. 1). The maximal responses to KCl in the control animal vessels were not significantly different from those in the vessels removed from the coronary-ligated, LVD animals (336 ± 56 mg wt. (n = 12/7) vs. 358 ± 103 mg wt. (n = 9/7)).

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Cumulative concentration-response curves to KCl in PRAs from sham-operated ( n/n = 6/3) and coronary-ligated ( n/n = 10/5) rabbits. Data are expressed as a percentage of their own maximal response. Each point represents mean ± S.E.M. n/n: number of preparations/number of animals.

Responses to ET-1, S6c and ET-3

Figure 2A and B demonstrates CCRCs to ET-1, S6c and ET-3 in sham-operated and coronary-ligated rabbits, respectively. The CCRCs can be seen to contain a relatively shallow component at the lower concentrations of ET agonists. All three ET receptor agonists were potent vasoconstrictors of the rabbit PRAs; EC₅₀ values are summarized in table 1. The rank order of potency for these peptides was S6c > ET-1 = ET-3. No significant difference in agonist potency was noted between the sham-operated and coronary-ligated rabbit vessels. The values for the maximal contractions are shown in table 2. In the sham-operated rabbits, the maximal contraction to S6c was less than that of ET-3 and not quite statistically decreased compared with ET-1 (P < 0.06). However, in all vessels tested, the S6c CCRC exhibited a sudden "drop-off" at high concentrations. The maximal responses to S6c and ET-3 were less than that to ET-1 in the coronary-ligated rabbit PRAs, and the maximal responses to ET-3 in these rabbits were significantly reduced compared with the response in the sham-operated rabbit vessels.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   A) Responses to endothelin-1 (, n/n = 8/6), endothelin-3 (, n/n = 10/9) and sarafotoxin S6c (, n/n = 10/9) in PRAs from sham-operated rabbits. B) Responses to endothelin-1 (, n/n = 12/7), endothelin-3 (, n/n = 9/6) and sarafotoxin S6c (, n/n = 11/7) in PRAs from coronary-ligated rabbits. Data are expressed as a percentage of the response to 50 mM KCl. Each point represents mean ± S.E.M. n/n: number of preparations/number of animals.

Effect of Antagonists on ET Receptor-Mediated Contraction in Control Rabbits

FR139317 (vs. ET-1). The selective ET_A receptor antagonist FR139317 (1 µM) failed to inhibit the ET-1-evoked response (table 1; fig. 3A).

View larger version (16K): [in this window] [in a new window]   Fig. 3.   A) Cumulative concentration-response curves to endothelin-1 (, n/n = 8/6) and endothelin-1 in the presence of 1 µM FR139317 (, n/n = 7/6). B) Cumulative concentration-response curves to ET-1 (, n = 8/6), in the presence of 1 µM BQ788 (, n/n = 6/5) and in the presence of 1 µM FR139317 and 1 µM BQ788 (, n/n = 7/6). Data are expressed as a percentage of the response to 50 mM KCl. Each point represents mean ± S.E.M. n/n: number of preparations/number of animals.

BQ788 (vs. ET-1). BQ788 (1 µM) inhibited the response to ET-1 up to 1 nM in that it removed the shallow component of the CCRC (fig. 3B). Responses to higher concentrations were not inhibited by BQ788. There was no significant change in the magnitude of responses to any concentration of ET-1.

BQ788 + FR139417 (vs. ET-1). FR139317 (1 µM) did not influence the effect of BQ788 (1 µM, fig. 3B).

BQ788 (vs. S6c). BQ788 (1 µM) inhibited the S6c-induced response (table 1; fig. 4A). The estimated pK_b value was 7.1 ± 0.2. The maximal response to S6c was increased by BQ788 (fig. 4A).

View larger version (16K): [in this window] [in a new window]   Fig. 4.   A) Cumulative concentration-response curves to S6c (, n/n = 10/9) and in the presence of 1 µM BQ788 (, n/n = 7/7). B) Cumulative concentration-response curves to endothelin-3 (, n/n = 10/9) and in the presence of 1 µM BQ788 (, n/n = 10/9). Data are expressed as a percentage of the response to 50 mM KCl. Each point represents mean ± S.E.M. n/n: number of preparations/number of animals.

BQ788 (vs. ET-3). BQ788 (1 µM) inhibited responses to ET-3 (table 1; fig. 4B). The estimated pK_b value was 6.6 ± 0.1.

SB209670 (vs. ET-1). 0.1 µM SB209670 had no effect on responses to ET-1 in the PRAs from the sham-operated rabbits. However, when present at 1 µM, it did inhibit responses (see table 1; fig. 5A). SB209670 inhibited ET-1 > 30 nM, and a shallow component of the CCRC to ET-1 was uncovered in the presence of SB209670 (fig. 5A). An estimated pK_b value for SB209670 was determined at the EC₅₀ level of the ET-1-induced response (6.8 ± 0.2). Maximal responses were not affected by SB209670.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   A) Cumulative concentration-response curves to ET-1 (, n/n = 8/6), in the presence of 0.1 µM SB 209670 (, n/n = 6/6) and in the presence of 1 µM SB 209670 (, n/n = 8/7). B) Cumulative concentration-response curves to sarafotoxin S6c (, n/n = 10/9), in the presence of 0.1 µM SB 209670 (, n/n = 7/6) and in the presence of 1 µM SB 209670 (, n/n = 7/6). Data are expressed as a percentage of the response to 50 mM KCl. Each point represents mean ± S.E.M. n/n: number of preparations/number of animals.

SB209670 (vs. S6c). SB209670 (1 µM) produced a concentration-dependent inhibition of responses to S6c, and the estimated pK_b value at the EC₅₀ value of the S6c-induced response was 7.8 ± 0.2 (fig. 5B).

Effect of Antagonists on ET Receptor-Mediated Contraction in Rabbits with LVD

The potency of SB209670 (1 µM) on the response to S6c in the coronary-ligated rabbits was not so profound as in the sham-operated rabbits (table 1). Indeed, the estimated pK_b value was significantly less at 6.7 ± 0.1 (1 µM SB209670, P < 0.0001). There were no other differences in the profile of antagonist effects.

Effect of LVD on 5-HT-Induced Contraction

The potency of 5-HT did not differ significantly between the control group (EC₅₀: 6.1 ± 0.2, n = 7 vessels from 6 rabbits) and the LVD group (EC₅₀: 6.1 ± 0.1, n = 7 vessels from 5 rabbits, fig. 6). The maximal response is reduced by ~30% (P < 0.05).

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Cumulative concentration-response curves to 5-hydroxytryptamine in PRAs from sham-operated (, n/n = 7/6) and coronary-ligated (, n/n = 7/5) rabbits. Data are expressed as a percentage of the response to 50 mM KCl. Each point represents mean ± S.E.M. n/n: number of preparations/number of animals.

L-NAME (vs. ET-1). i) Sham-operated control rabbits. Inhibition of NOS synthase with 100 µM L-NAME had no effect on responses to ET-1 (fig. 7A; table 3).

View larger version (16K): [in this window] [in a new window]   Fig. 7.   A) Sham-operated rabbits. Cumulative concentration-response curves to ET-1 (, n/n = 8/6) and in the presence of 100 µM L-NAME (, n/n = 9/7). B) LVD rabbits. Cumulative concentration-response curves to ET-1 (, n/n = 12/7) and in the presence of 100 µM L-NAME (, n/n = 7/6). Data are expressed as a percentage of the response to 50 mM KCl. Each point represents mean ± S.E.M. n/n: number of preparations/number of animals.

L-NAME vs. S6c) i) Sham-operated control rabbits. Inhibition of NOS synthase with 100 µM L-NAME had no effect on the potency of S6c, and maximal contraction to S6c was increased (67.9 ± 8.2 vs. 101.8 ± 13.7, P < 0.05; fig. 8A; table 3).

View larger version (17K): [in this window] [in a new window]   Fig. 8.   A) Sham-operated rabbits. Cumulative concentration-response curves to sarafotoxin S6c (, n/n = 10/9) and in the presence of 100 µM L-NAME (, n/n = 9/7). B) LVD rabbits. Cumulative concentration-response curves to sarafotoxin S6c (, n/n = 12/7) and in the presence of 100 µM L-NAME (, n/n = 7/6). Data are expressed as a percentage of the response to 50 mM KCl. Each point represents mean ± S.E.M. n/n: number of preparations/number of animals.

Endothelium-Dependent Vasodilation

We found no evidence for endothelium-dependent vasodilation using any of the agents in either the sham-rabbit vessels or the coronary-ligated rabbit vessels.

Endothelium-Independent Relaxation: SNP

The thromboxane mimetic U46619 (30 nM) produced the same degree of preconstriction in the control group (54.5 ± 7% of response to 50 mM KCl, n = 14 vessels from six rabbits) and in the LVD group (48.6 ± 5% of response to 50 mM KCl, n = 14 vessels from seven rabbits). The rabbits with LVD were less sensitive to the relaxant effects of SNP (EC₅₀: 6.8 ± 0.1) than their controls (EC₅₀: 7.3 ± 0.2, P < 0.05) although the maximal response to SNP was unaffected (fig. 9).

View larger version (16K): [in this window] [in a new window]   Fig. 9.   Vasodilation induced by SNP in PRAs from sham-operated (, n/n = 14/6) and coronary-ligated (, n/n = 14/7) rabbits. Vasodilation is expressed as a percentage of the preconstricted tone induced by 30 nM U46619. Each point represents mean ± S.E.M. n/n: number of preparations/number of animals.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The rank order of potency for the ET agonists in all vessels studied was S6c > ET-1 = ET-3 rabbit PRAs. This is indicative of vasoconstriction being mediated by ET_B-like receptors, as has been previously shown in the larger pulmonary artery of the rabbit (Fukuroda et al., 1994a; Hay et al., 1996). The results show that SB209670 has estimated pK_b values of 6.8 and 7.8 against ET-1 and S6c, respectively, in the sham-operated rabbit pulmonary arterioles. Hay et al. (1996) have shown that SB209670 has similar pK_b values (6.7 and 7.7) against responses to ET-1 and S6c, respectively, in large rabbit pulmonary arteries. Ohlstein et al. (1994b) also showed that SB209670 has a pK_b value of 9.4 against ET_A-mediated responses in the rat aorta. This confirms that, in our study, SB209670 is acting as an antagonist against an ET_B receptor in the rabbit PRAs and this SB209670-sensitive receptor has a pharmacological profile similar to that in the larger pulmonary arteries. As described in "Results," the shallow component of the CCRC to ET-1 was resistant to SB209670 and was more obvious in the presence of SB209670. This may suggest that responses to ET-1 are not mediated by a homogeneous ET_B population, an interpretation consistent with the observations and conclusions of Hay et al. in the large rabbit pulmonary artery.

The ET_B receptor antagonist BQ788 removed the initial "shallow component" of the ET-1 CCRC but failed to affect the rest of the CCRC. It did, however, inhibit responses to both S6c and ET-3, with estimated pK_b values of 7.1 and 6.6, respectively. Again, this phenomenon was observed by Hay et al. (1996) in the larger rabbit pulmonary arteries, where the pK_b values were 6.2 and 5.1 for S6c and ET-3, respectively, whereas there was no effect on ET-1. These authors concluded that the ET_B receptors in rabbit large pulmonary arteries are not sensitive to BQ788. It would appear, therefore, on the basis of both agonist and antagonist interactions, that the ET_B receptor in the large and small pulmonary arteries are pharmacologically similar.

The response to S6c in the rabbit PRAs demonstrated a "drop-off" at higher concentrations. We see a similar phenomenon in the responses to S6c in rat PRAs (MacLean et al., 1994; McCulloch and MacLean, 1995). This effect is not affected by administration of L-NAME in either species and hence is not due to NO production. We attribute this to a desensitization of the receptor, because if we raise the tone of these vessels with U46619 and administer one dose of a high concentration of S6c, we obtain a contraction and not a vasodilation (C.C. Docherty and M.R. MacLean, unpublished observation).

The present results show that the CCRCs to ET-1 include a "shallow component" at lower concentrations of ET-1. We have previously found similar CCRCs to ET-1 in both human and rat PRAs (McCulloch and MacLean, 1995; McCulloch et al., 1996). In the rat PRAs, the reasons for the biphasic response are still unclear, but it may be due to a heterogeneous population of ET_B receptors or to the presence of inhibitory ET_A receptors (McCulloch and MacLean, 1995; McCulloch et al., in press). In human PRAs, the first component of the response to ET-1 is clearly due to a population of ET_B receptors mediating vasoconstriction, whereas the second component is due to higher concentrations of ET-1 causing contraction by stimulating ET_A receptors (McCulloch et al., 1996). In the present study, the shallow component was resistant to the effects of SB209670 but sensitive to BQ788. Indeed, the effect of SB209670 was to "uncover" or exaggerate this component. The simplest explanation for the shallow component of the ET CCRC in the rabbit PRAs is that there is a heterogeneous population of ET_B-like receptors. However, this interpretation of the data remains speculative. Curve-fitting for a biphasic response is not possible where the first component of the curve is less than 30% of the maximal response. This theory cannot therefore be assessed mathematically.

What is clear is that ET_A receptors exert little or no overriding vasoconstrictor influence at any concentration of ET-1. FR139317 did not inhibit responses to ET-1 in either the sham-operated or the coronary-ligated rabbit vessels. Synergy between ET_A and ET_B receptors has, however, been reported in larger rabbit pulmonary arteries, where administration of both an ET_A and an ET_B receptor inhibitor is required to inhibit responses to ET-1 (Fukuroda et al., 1994b). In the PRAs, however, FR139317 had no effect on the ability of BQ788 to inhibit responses to ET-1, which suggests that such synergy could not be observed using FR139317 and BQ788 as antagonists. The combined effects of FR139317 and BQ788 might be expected to be the same as the effect of SB209670. The differential results obtained may be explained by SB209670 being considerably more potent than FR139317 at the ET_A receptor (Ohlstein et al., 1994a; 1994b; Sogabe et al., 1993); this would suggest that ET_A receptors modulate responses to ET-1. Alternatively, the ET_B receptor population present may be uniquely sensitive to SB209670, or an ET_A receptor population in this setting may be sensitive to SB209670 but insensitive to FR139317. All these suggestions remain speculative.

We examined the rabbit PRAs for endothelium-dependent vasodilation to determine whether this changed with LVD. Surprisingly, we could find no evidence for endothelium-dependent vasodilation, even though we tried a range of agonists known to induce endothelium-dependent vasodilation in PRAs. Endothelium-dependent relaxation has, however, been demonstrated in adult rabbit large pulmonary arteries (Johns et al., 1989). An absence of endothelium-dependent relaxation could be due to insensitivity of the vascular smooth muscle to NO, damage to the endothelium or an absence of endothelium-dependent NO. The absence of endothelium-dependent relaxation cannot be due to insensitivity to NO, because all vessels we tested relaxed in the presence of the NO donor SNP. It is also unlikely that the endothelium was damaged by the setting-up procedure, because we can demonstrate 100% endothelium-dependent vasodilation in fetal and newborn rabbit PRAs, which are considerably more delicate than the adult vessels (Docherty and MacLean, 1995). We have also used agonists other than U46619 to raise the vascular tone and failed to observe any endothelium-dependent relaxation (C.C. Docherty and M.R. MacLean, unpublished observations). The most likely explanation for the absence of endothelium-dependent relaxation is that it does not occur in the PRAs from the adult rabbit. These results are entirely consistent with previous studies that demonstrate endothelium-dependent vasodilation in large, but not small, pulmonary arteries of the rat and sheep (Leach et al., 1992; Kemp et al., 1997). The physiological relevance of these observations remains speculative. Curiously, however, in the rat, although endothelial NOS (eNOS) activity is normally absent from the endothelium, it can be observed in PRAs removed from rats with PHT (Isaacson et al., 1994; Xue et al., 1994). In addition, we recently demonstrated an increase in ACh-induced vasodilation in PRAs removed from chronic hypoxic PHT rats (MacLean and McCulloch, in press). This must be a compensatory event in the face of increased pulmonary pressure.

With regard to the effects of LVD, there was no marked change in the ET receptor profile in terms of agonist sensitivity or antagonist activity. Curiously, however, there was a decrease in the estimated pK_b value for SB209670 against the effect of S6c in the LVD group. It has been noted previously that SB209670 has a similarly reduced pK_b value against S6c in the rabbit bronchus compared with the rabbit pulmonary artery (Hay et al., 1996). Hay et al. proposed that such differences in antagonist potency could be explained by differences in the ET_B receptor, differences in the regional distribution of receptors or differences in the affinity of the antagonist for different binding domains within a single population of ET_B receptors. Any of these phenomena may explain our results, although they are less likely to be explained by a change in the ET_B receptor itself, because agonist potencies do not change and the effect of BQ788 is not altered.

There was a decrease in the maximal response to ET-3 in the vessels removed from the LVD rabbits. The reason is unclear, but it may be related to the possible effect of increased NO production. There was not a general depressant effect, though, because responses to ET-1 and KCl were not altered. The maximal response to 5-HT, however, was decreased in rabbits with LVD, so some vasoconstrictors such as 5-HT and ET-3 may be influenced to a greater extent by changes in NO activity. The results with KCl show that overall smooth muscle contractility was not affected by LVD.

The major difference between the sham-operated and LVD rabbits was observed with L-NAME. L-NAME had little effect on responses to ET-1 and S6c in vessels removed from the sham-operated rabbits but markedly potentiated responses in the LVD rabbits. This suggests that basal NOS activity was increased in the PRAs removed from the LVD rabbits. We have previously shown that there is increased endogenous tone in pulmonary arteries and arterioles from rats with PHT and that this may stimulate, and be countered by, endogenous NO release (MacLean et al., 1995; MacLean et al., 1996). Therefore, this phenomenon probably accounts for the increase in basal NO production in the PRAs from the rabbits with LVD. These results are compatible with the results of other studies that have demonstrated an increase in NO production associated with PHT. For example, there is evidence that NOS may be up-regulated in patients with heart failure, and the inhibition of NOS increased pulmonary vascular resistance in these patients (Habib et al., 1994). This must also be a compensatory response to an increase in pulmonary pressure. The results presented here demonstrate that this serves to maintain sensitivity to ET-1 in that the EC₅₀ for ET-1 was the same in the sham-operated and LVD rabbit vessels. However, if the influence of NO is removed, this uncovers a 2- to 3-fold increase in sensitivity to ET-1. Hence the increased influence of NO has indeed compensated for increased vasoconstrictor influences. It is thought that basal and agonist-stimulated NO activity can be regulated differentially, so it is not surprising that basal NO production can be influenced despite the absence of agonist-induced release of NO (Mian and Martin, 1995).

The PRAs removed from the rabbits with LVD demonstrated a decreased sensitivity to SNP. In the human, an increase in stimulation by NO by long-term nitrate administration is also followed by a tolerance to the effects of nitrates (Needleman and Johnson, 1973). Curiously, we have observed previously that in rats chronically treated with L-NAME such that eNOS is depleted, the pulmonary arteries are hypersensitive to SNP (MacLean and Macmillan, 1993). Hence there is evidence that when there is increased stimulation by NO either by up-regulation of eNOS or by the administration of nitrates, there is a reduction in sensitivity to NO. When eNOS is inhibited, the reverse case holds, and the smooth muscle becomes hypersensitive to NO.

In conclusion, this study shows that the pharmacology of the ET-1 receptor in rabbit PRAs is complex. There is a vasoconstrictor ET_B-like receptor that is BQ788-insensitive but is sensitive to SB209670 and mediates responses to ET-1 with agonist potencies in the rank order S6c > ET-1 = ET-3. Rabbit PRAs do not demonstrate endothelium-dependent relaxation. An increase in basal NO production may be an early physiological compensatory mechanism in response to the early elevation in pulmonary pressure with LVD to circumvent an increase in potency to ET-1.