Journal Logo

Articles

Pharmacologic Characterization of Receptor Types Mediating Coronary Vasodilator Actions of Sensory Neuropeptides in the Guinea Pig

Lam, Francis F. Y.

Author Information
Journal of Cardiovascular Pharmacology: April 2000 - Volume 35 - Issue 4 - p 646-652
  • Free

Abstract

Sensory neurons containing neuropeptides such as substance P and calcitonin gene-related peptide (CGRP) are widely distributed in the body, and some of them are known to innervate the coronary vasculature. Thus immunohistochemical studies have revealed colocalization of substance P-like immunoreactivity (-LI) with CGRP-LI in cardiac C-fiber afferents in a variety of species including guinea pig (1) and humans (2). Many of these sensory nerve terminals are found in close proximity to blood vessels (3,4), implying that they may have an action on this structure.

Although the natural stimuli for cardiac C-fiber afferent is obscure, it is known that a variety of drugs as well as ischemia can activate sensory receptors in the heart (5,6). Because peripheral branches of sensory nerves possess dual "sensory-efferent" function (7), activation of cardiac sensory receptors can induce not only increased afferent nerve discharge centrally, but also could lead to local release of neuropeptides. The released neuropeptides are likely to have influence on blood flow of the heart. For example, one of the neuropeptides released from sensory neurons is CGRP. This is a well-known vasodilator (8) that has been shown to induce coronary vasodilatation in guinea pigs (1) and humans (9). CGRP is thought to produce its effect through interaction with two types of CGRP receptors (CGRP1 and CGRP2 receptors) based on the evidence that the C-terminal fragment of the peptide, hCGRP8-37, can readily antagonize the effects of CGRP on guinea pig atria, but is much less effective on rat vas deferens (10). Conversely, the linear CGRP analogue [Cys(acetomethoxy)2,7]CGRP is a much more potent agonist on vas deferens than on guinea pig atria (10-12).

Another neuropeptide that is often colocalized and coreleased with CGRP from sensory neurons is substance P. Like CGRP, it is a vasoactive peptide that has been reported to induce dilation of coronary vascular bed in the guinea pig (1,13). Substance P is a member of the tachykinin peptide family that includes neurokinin A and neurokinin B, all sharing the common C-terminal sequence of Phe-X-Gly-Met-NH2. Tachykinins are known to produce their effects by activation of three distinct neurokinin receptor types; the neurokinin-1 (NK1), neurokinin-2 (NK2), and neurokinin-3 (NK3) receptors. Substance P displays the highest affinity for the NK1 receptor, whereas neurokinin A and neurokinin B preferentially bind to NK2 and NK3 receptors, respectively. However, the selectivity of these endogenous neuropeptides for their preferred receptors is poor, and consequently, their physiologic actions may be mediated through interaction with all three receptor types (14,15).

In the last decade, the development of selective tachykinin receptor agonists has provided researchers with the ability of stimulating one tachykinin receptor type. Furthermore, selective tachykinin receptor antagonists also exist, and these are proving to be useful tools in receptor characterization. Three such antagonists are FK888 (N2-[(4R)-4-hydroxy-1-(1-methyl-1H-indol-3-yl)carbonyl-L-prolyl]-N-methyl-N-phenylmethyl-3-(2-naphthyl)-L-alaninamide) (16), SR48968 ((S)-N-methyl-N-[4-acetylamino-4-phenylpiperidino)-2-(3,4-dichlorophenyl)-butyl] benzamide) (17), and SR142801 ((S)-(N)-(1-(3-(1-benzoyl-3-(3,4-dichlorophenyl)piperidin-3-yl)propyl)-4-phenylpiperidin-4-yl)-N-methylacetamide) (18), which are selective for NK1, NK2, and NK3 receptors, respectively.

In this study, the receptor types mediating coronary vasodilator actions of substance P and CGRP were characterized using selective agonists and antagonists of their respective receptors.

METHODS

The isolated guinea pig heart preparation

All experiments were conducted in accordance with the Animal Research Ethics Committee, The Chinese University of Hong Kong. Male Hartley guinea pigs (350-450 g) were killed by stunning followed by removal of the heart. A metal cannula was inserted into the ascending aorta of the isolated heart. The cannula was then attached to a three-way stopcock at the end of the perfusion apparatus. A Masterflex peristaltic pump was used to perfuse the heart at a constant rate (6 ml/min) with Krebs buffer (37°C, pH 7.4) containing the following (mM): NaCl, 118; KCl, 4.7; CaCl2, 2.5; KH2PO4, 1.2; NaHCO3, 24.9; MgSO4, 1.2, and glucose, 12. The heart was surrounded by a water-heated glass jacket maintained at 37°C, and perfusion pressure was measured at the lateral arm of the three-way stopcock using a Statham (P23XL) pressure transducer (West Warwick, RI, U.S.A.) and recorded with a Graphtec MiniWriter (WTR 751) (Minato-ku, Tokyo, Japan). It has been observed that elevation of potassium concentration in the buffer could facilitate measurement of vascular responses and eliminate ventricular contractions (13); thus after a 30-min equilibration period with normal buffer, perfusion was switched to a modified buffer containing 20 mM KCl for the remainder of the experiment. The NaCl concentration was reduced accordingly to maintain isotonicity.

Studies of neuropeptide actions

Investigations of neuropeptide actions were commenced when the baseline perfusion pressure of the guinea pig heart had stabilized at the higher level after the switch to 20 mM KCl buffer. Neuropeptide agonists were then administered by bolus injections (0.1 ml) through a tubing that empties near the site at which the aortic cannula was attached to the perfusion apparatus. Dose-response curves to the agonists were achieved by allowing 15-min intervals between each application to avoid tachyphylaxis. The action of CGRP was long-lasting and the dose-response curve to this agent was therefore constructed by cumulative dosing. An additional infusion line on the perfusion apparatus allows perfusion of antagonist into the preparation by a Harvard infusion pump (model 975A) (South Natick, MA, U.S.A.). Antagonist perfusion was commenced 15 min before bolus administration of agonist and continued throughout the experiment.

Drug formulation

The following drugs were used: substance P, [Sar9, Met(O2)11]-substance P, [MePhe7]-neurokinin B, r-α-CGRP, [Cys(acetomethoxy)2,7]CGRP, and hCGRP8-37 dissolved in 0.9% NaCl; [Nle10]-neurokinin A4-10 dissolved in ammonium hydroxide (0.01%); and FK888, SR48968, and SR142801 dissolved in ethanol (15%). The percentage of solvent refers to that of stock solution, which was 10 μM. Subsequent dilution of all stock solution was made in Krebs buffer containing 20 mM KCl. All drugs were purchased from Cambridge Research Biochemicals Ltd. (U.K.) except for hCGRP8-37, which was purchased from Sigma (St. Louis, MO, U.S.A.), and FK888, which was kindly donated by Fujisawa Pharmaceuticals Ltd. (Japan), and SR48968 and SR142801 were gifts from Sanofi Recherche (France).

Calculations and statistical analysis

Data are reported as mean ± standard error of the mean (SEM). ED50 values were determined from individual sigmoidal fits of the dose-response curves using Kaleidagraph (Synergy Software, Reading, PA, U.S.A.). An unpaired one-tailed t test was used to analyze the differences between ED50 values. The pA2 values of antagonists were obtained by the method of Arunlakshana and Schild (19). Apparent affinity estimates (pKB) from single antagonist concentrations were calculated by use of the Gaddum-Schild equation pKB = −logKB = log(x − 1) − log[A]; where x is the dose-ratio, and [A], the antagonist concentration. Differences between dose-response curves were analyzed by repeated measures two-factor analysis of variance (ANOVA), followed where appropriate by comparisons of means by Planned Contrasts (SuperAnova, Abacus Concepts, Berkeley, CA, U.S.A.). The latter procedure is very efficient for comparing a limited subset of possible contrasts. This is useful for testing hypotheses about the data that are more specific than the hypotheses automatically tested for each term in the ANOVA model (20). The p values <0.05 were considered statistically significant.

RESULTS

Coronary vasorelaxant effects of tachykinins and CGRP

Bolus injections of substance P produced dose-dependent reductions in perfusion pressure as a result of dilation of the guinea pig coronary resistance vessels. The mean ED50 value calculated from the individual dose-response curves was 20.4 ± 7.0 pmol (n = 5). Maximal vasodilator response was achieved at 1 nmol substance P with 33.9 ± 2.8% reduction in basal perfusion pressure. Administration of the selective NK1 receptor agonist [Sar9, Met(O2)11]-substance P produced dose-dependent dilation of the guinea pig coronary blood vessels with an ED50 value of 14.4 ± 4.0 pmol (n = 5), which is not significantly different from that of substance P (p = 0.4783). Statistical analysis also showed no significant difference between the dose-response curves of the NK1 receptor agonist and substance P (p = 0.1052). However, the NK1 receptor agonist produced a maximal decrease in basal blood flow of 42.2 ± 2.7%, which is significantly higher than that produced by substance P (p = 0.0024; Fig. 1).

FIG. 1
FIG. 1:
Effects of substance P (○; n = 5), [Sar9 Met(O2)11]-substance P (□; n = 5), [Nle10]-neurokinin A4-11 (▵; n = 5), and [MePhe7]-neurokinin B (▿ n = 6) on coronary perfusion pressure of the isolated guinea pig heart perfused with isotonic buffer containing 20 mM KCl. Drugs were administered as bolus injections into the coronary circulation in a volume of 0.1 ml with 15-min intervals between each application. Data are shown as mean ± SEM (shown by vertical bars) expressed as percentage decrease from basal perfusion pressure. Dose-response curves for substance P and [Sar9 Met(O2)11]-substance P are not significantly different (p > 0.05; two-factor ANOVA). However, subsequent comparisons of means by Planned Contrasts (see Methods) showed significant differences between their means at 0.1 nmol (***p < 0.001) and 1 nmol (**p < 0.01).

At a dose range of 1 pmol to 1 nmol, neither the selective NK2 receptor agonist [Nle10]-neurokinin A4-10 nor the selective NK3 receptor agonist [MePhe7]-neurokinin B had an effect on the basal perfusion pressure. These results are illustrated in Fig. 1.

Cumulative administrations of CGRP produced dose-dependent relaxation of the coronary blood vessels with a mean ED50 value of 128.7 ± 16.8 pmol (n = 5). This ED50 value is significantly different from that of substance P and the selective NK1 receptor agonist (p = 0.0003 and p = 0.0001, respectively). The selective CGRP2 receptor agonist [Cys(acetomethoxy)2,7]CGRP produced no change on basal perfusion pressure at doses of 10 pmol to 1 nmol. These results are shown in Fig. 2.

FIG. 2
FIG. 2:
Effects of calcitonin gene-related peptide (CGRP; ○; n = 5) and the selective CGRP2 receptor agonist [Cys(acetomethoxy)2,7]CGRP (□; n = 6) on coronary perfusion pressure of the isolated guinea pig heart. CGRP was applied cumulatively as bolus injections into the coronary circulation, whereas [Cys(acetomethoxy)2,7]CGRP was administered with 15-min intervals between each application. The selective CGRP1 receptor antagonist hCGRP8-37 (•; n = 8) was perfused into the preparation 15 min before bolus administrations of CGRP and continued throughout the experiment. Dose-response curves of CGRP in the absence and in the presence of hCGRP8-37 are significantly different (*p < 0.05; two-factor ANOVA).

Effects of the selective neuropeptide receptor antagonists

The effects of the selective NK1 receptor antagonist FK888 on substance P-induced vasodilator responses are shown in Fig. 3. This antagonist produced concentration-dependent inhibition on the dose-response curves to substance P. At 300 nM, which was the highest concentration of FK888 used, the coronary vasorelaxant effect of substance P was abolished (p = 0.0002; n = 5). The pA2 value for FK888 determined from the Schild plot was 9.2.

FIG. 3
FIG. 3:
Effects of 3 (▴), 30 (▾), and 300 nM (•) of the selective NK1 receptor antagonist FK888 on substance P-induced coronary vasodilatation (○) in the isolated guinea pig heart. FK888 perfusion was commenced 15 min before bolus administrations of substance P and continued throughout the experiment; n = 5. Significant difference: *p < 0.05; **p < 0.01; ***p < 0.001 (two-factor ANOVA).

Two concentrations of the selective NK2 receptor antagonist SR48968 were tested on substance P. As shown in Fig. 4, at 30 nM, SR48968 produced no significant change on the substance P dose-response curve (p = 0.1484; n = 5), albeit a small reduction (16%) in the response to 1 nmol substance P (p = 0.0354). Conversely, 300 nM SR48968 produced significant partial suppression of the substance P dose-response curve (p = 0.0137; n = 5). The maximal response to 1 nmol substance P was reduced by 51% (p = 0.0001). SR48968 was estimated to have a pA2 value of 7.8.

FIG. 4
FIG. 4:
Effects of 30 (▴) and 300 nM (▾) of the selective NK2 receptor antagonist SR48968 on substance P-induced coronary vasodilatation (○) in the isolated guinea pig heart. SR48968 perfusion was commenced 15 min before bolus administrations of substance P and continued throughout the experiment. Dose-response curves of substance P in the absence and in the presence of 30 nM SR48968 are not significantly different (p > 0.05; two-factor ANOVA; n = 5 for each). However, subsequent comparisons of means by Planned Contrasts showed significant differences between their means at 0.1 nmol (**p < 0.01) and 1 nmol (*p < 0.05). Dose-response curves of substance P in the absence and in the presence of 300 nM SR48968 are significantly different (*p < 0.05; two-factor ANOVA; n = 5).

The selective NK3 receptor antagonist SR142801 produced similar partial suppressions of the substance P dose-response curves at 30 and 300 nM (p = 0.0011 and p = 0.0023, respectively; n = 5 for each). The maximal response to 1 nmol substance P was reduced by 60% (p = 0.0001) and 53% (p = 0.0001) in the presence of 30 and 300 nM SR142801, respectively (Fig. 5). The apparent pKB values calculated by the Gaddum-Schild equation were 8.5 ± 0.2 and 7.8 ± 0.4 (p = 0.1546; n = 5 for each) at 30 and 300 nM, respectively.

FIG. 5
FIG. 5:
Effects of 30 (▴) and 300 nM (▾) of the selective NK3 receptor antagonist SR142801 on substance P-induced coronary vasodilatation (○) in the isolated guinea pig heart. SR142801 perfusion was commenced 15 min before bolus administrations of substance P and continued throughout the experiment; n = 5. Significant difference: **p < 0.01 (two-factor ANOVA).

Only one concentration (300 nM) of each of the three selective tachykinin receptor antagonists was tested on the NK1 receptor agonist-induced vasodilatation. As shown in Fig. 6, FK888 abolished all responses to the NK1 receptor agonist at this concentration (n = 10). However, SR48968 and SR142801 produced only partial inhibitions on the NK1 receptor agonist-induced dose-response curves (p = 0.0242 and p = 0.0018, respectively; n = 5 for each) with 24% (p = 0.0025) and 40% (p = 0.0001) reductions on the maximal response, respectively. The pKB values calculated by the Gaddum-Schild equation were 7.6 ± 0.4 (n = 5) for SR48968 and 8.0 ± 0.5 (n = 5) for SR142801.

FIG. 6
FIG. 6:
Effects of 300 nM FK888 (•; n = 10), 300 nM SR48968 (▴; n = 5), and 300 nM SR142801 (▾; n = 5) on [Sar9 Met(O2)11]-substance P-induced coronary vasodilatation (○; n = 5) in the isolated guinea pig heart. Antagonist perfusion was commenced 15 min before bolus administrations of [Sar9 Met(O2)11]-substance P and continued throughout the experiment. Significant difference: *p < 0.05; **p < 0.01; ***p < 0.001 (two-factor ANOVA).

In the presence of 30 nM of the selective CGRP1 receptor antagonist hCGRP8-37, a significant rightward shift (p = 0.0067; n = 8) of the CGRP dose-response curve was produced (Fig. 2). The pKB value calculated for hCGRP8-37 was 7.7 ± 0.1 (n = 8).

DISCUSSION

This study has confirmed previous findings that the sensory neuropeptides substance P and CGRP are effective vasodilators of the guinea pig coronary vascular bed. Furthermore, the receptor types mediating their vasorelaxant effects were identified to be the NK1 receptors and CGRP1 receptors, respectively.

Immunohistochemical studies have demonstrated the colocalization of substance P and CGRP immunoreactivities in the same secretory vesicles in varicosities of perivascular sensory nerve fibers in the guinea pig heart (1,21). Such findings correlate well with a putative role for sensory neuropeptides in the regulation of coronary blood flow. In addition, this study has confirmed the observation of previous workers that substance P possesses vasorelaxant activity on the guinea pig coronary blood vessels (1,13). Moreover, our study showed that the selective NK1 receptor agonist [Sar9, Met(O2)11]-substance P was equally effective in dilating the coronary blood vessels, producing dose-response curves and ED50 values similar to that of substance P. These results indicate that NK1 receptors mediate tachykinin-induced vasodilator responses in the guinea pig heart.

Although the dose-response curves to substance P and the selective NK1 receptor agonist were not significantly different, the latter agent, nevertheless, produced a significantly greater maximal response than substance P (44.2 ± 2.7% compared with 33.9 ± 2.8%). Substance P may therefore be perceived as a partial agonist in this preparation, but this is unlikely because substance P is not known to be a partial agonist on NK1 receptors. A more probable reason would be that in this study, the method of bolus administrations of drugs allows too short a contact time for agonists to accomplish dynamic equilibrium with receptors in the coronary vasculature. Therefore the selective NK1 receptor agonist, owing to its higher efficacy or greater initial rate of reaction with the NK1 receptors, is able to elicit a greater maximal response than substance P within the same limited time frame of interactions with the NK1 receptors. This speculation was confirmed by a subsequent study (results not shown), which showed that continuous perfusion of 10 μM substance P produced much greater reduction on the basal perfusion pressure (60.0 ± 4.3%; n = 6) compared with that produced by bolus administration of substance P or the NK1 receptor agonist (p = 0.0003 and 0.0085, respectively).

Studies with selective tachykinin receptor antagonists in this study have further consolidated the view that NK1 receptors mediate the vasorelaxant effects of tachykinins. Thus it was shown that the selective NK1 receptor antagonist FK888 is capable of abolishing both substance P-induced and the NK1 receptor agonist-induced vasodilator responses. Detailed analyses of the nature of the antagonism was not carried out because this method of drug administration does not permit agonists to achieve dynamic equilibrium with their receptors. Nevertheless, a pA2 value of 9.2 was estimated from the Schild plot for FK888, which correlates well with its pA2 value of 9.3 in the guinea pig ileum (16).

In contrast, at relatively high concentrations of the selective NK2 receptor antagonist SR48968 or the selective NK3 receptor antagonist SR142801, tachykinin-induced vasodilator responses were not abolished but were partially suppressed. SR48968 was estimated to have a pA2 value of 7.8. This is a relatively small pA2 value for SR48968 on NK2 receptors compared with those found by Emonds-Alt et al. (17) on isolated tissues of the rabbit pulmonary artery (pA2 = 10.3) and the human bronchus (pA2 = 9.4). In a like manner, SR142801 has apparent pKB values around 7.8 and 8.5. These values are small for SR142801 on NK3 receptors compared with those found by Croci et al. (22) on isolated guinea pig tissues of the ileum (apparent pKB = 9.2) and the Taenia cecum (apparent pKB = 9.5). Conversely, high concentrations of SR48968 have been shown to exhibit significant binding to NK1 receptors of the guinea pig lung in vivo (23). Likewise, in a NK1 receptor functional assay, SR142801 was shown to antagonize [Sar9, Met(O2)11]-substance P-induced relaxation of rabbit pulmonary artery significantly (18). These findings indicate that despite SR48968 and SR142801 being labeled as selective antagonists for NK2 and NK3 receptors, they also possess NK1 receptor-blocking action. Thus the partial inhibitions of the coronary vasorelaxant effects of the tachykinins by these antagonists could be due to cross-inhibition on the NK1 receptors. Nevertheless, it should be noted that the inhibitory action of SR142801 on substance P-induced vasodilatation was not dependent on the concentration of SR142801 used. Thus it is possible that SR142801 produced its inhibitory effect through a mechanism not involving tachykinin receptor blockade.

Although the mechanisms underlying the inhibitory effects of tachykinin receptor antagonists are still uncertain, this study with tachykinin agonists has provided strong evidence that NK2 and NK3 receptors do not play a part in mediating tachykinin-induced vasodilatation. This conclusion is derived from the findings that both the selective NK2 receptor agonist [Nle10]-neurokinin A4-10 and the selective NK3 receptor agonist [MePhe7]-neurokinin B were devoid of vasodilator action in the guinea pig heart. Furthermore, a previous study on naturally occurring tachykinins in the same preparation has shown that substance P- and neurokinin A-induced vasodilator responses were similar, and that desensitization of one tachykinin produced cross-tachyphylaxis to the other peptide (13). Such results added more support to the idea that tachykinin-induced coronary vasodilatation in the guinea pig is mediated by the same population of NK1 receptors.

CGRP produced coronary vasorelaxant effect that was longer lasting than those of the tachykinins and was not subject to tachyphylaxis. These features of CGRP-induced vasodilatation are common to those found on human coronary blood vessels (24) and in several other vascular sites including those in rat knee joints (25). At present, no selective CGRP2 receptor antagonist is available, and thus only the effect of the selective CGRP1 antagonist, hCGRP8-37, was tested. Most studies on guinea pigs have shown hCGRP8-37 to possess a pA2 value of around 7 (26). hCGRP8-37 is therefore not a very potent antagonist, but despite this, the present study showed that it produced significant rightward shift of the CGRP dose-response curve at a concentration of 30 nM. Perfusion of a larger quantity of hCGRP8-37 would be too costly for this preparation, and this made it not feasible to test for the effect of higher concentration of the antagonist. Nevertheless, an apparent pKB value of 7.7 was estimated for hCGRP8-37 with the single concentration used. This is a respectable pKB value in view of the limited affinity of hCGRP8-37 on CGRP1 receptors. Furthermore, taking into consideration that the selective CGRP2 receptor agonist [Cys(acetomethoxy)2,7]CGRP was not effective in altering the coronary blood flow in this preparation, these findings indicate that CGRP1 receptors rather than CGRP2 receptors mediate the coronary vasorelaxant effect of CGRP. The lack of CGRP2 receptor involvement in the vascular effect of CGRP also has been reported in rats in which intravenous injection of [Cys(acetomethoxy)2,7]CGRP produced no hypotensive effect (27).

CGRP has been shown to induce relaxation of coronary arteries in the rat by endothelium-dependent as well as endothelium-independent mechanisms (28). In contrast, the vasodilatory actions of CGRP were not affected after mechanical removal of the endothelium in isolated porcine (2), bovine (29), and human coronary arteries (9,24). Furthermore, specific receptors for CGRP have been identified in the smooth muscle layer of porcine, bovine, and human epicardial arteries (30,31). These data suggest that in most species, apart from the rat, CGRP acts directly on vascular smooth muscle rather than through the release of endothelium-derived relaxing factor (EDRF) in producing coronary vasodilatation. Conversely, substance P has been shown to induce an endothelium-dependent relaxation of the coronary blood vessels in the pig (2,32) and guinea pig (13). Thus much of the vasodilator response to substance P is mediated by EDRF, which is released because of stimulation of NK1 receptors on the endothelium. Conversely, tachykinin receptors that mediate vasoconstriction are suspected to be present on the smooth muscle cells of rabbit pulmonary artery (NK2) and rat portal vein (NK3) (33,34). However, neither substance P nor any of the selective tachykinin receptor agonists produced vasoconstriction in our preparation. Thus vasoconstrictor tachykinin receptors are probably absent in the guinea pig coronary vasculature.

CGRP has been consistently shown to be a more potent vasodilator than the tachykinins in many vascular sites (4,8,24,25). However, in our study, CGRP was much less effective than the tachykinins in producing coronary vasodilatation. It remains to be confirmed that receptor locations for CGRP and substance P in coronary blood vessels of the guinea pig follow the general pattern as described earlier. If so, this could be the explanation for the reversed potencies of CGRP and substance P seen in this study. If can be envisaged that for equal doses of CGRP and substance P administered via the intraluminal route, a smaller amount of CGRP can penetrate to the CGRP1 receptors located in the smooth muscle layer, as opposed to the higher proportion of substance P that can come into contact with the NK1 receptors located on the endothelium. When drugs are applied via the bolus route, a brief contact time of drugs with the blood vessels also is permitted. This can magnify the difference in end concentrations of CGRP and substance P at their respective target receptor sites. A reversed pattern of vasorelaxant potencies of CGRP and substance P in comparison with other studies is, therefore, conceivable in this study.

It has been shown that the neurotoxin capsaicin produced coronary vasodilatation in the guinea pig that was resistant to NK1 receptor blockade by FK888, but susceptible to inhibition by the CGRP1 receptor antagonist, hCGRP8-37(35). In human coronary arteries, capsaicin was able to relax the blood vessels after substance P tachyphylaxis (2,24), and CGRP but not substance P was found to mimic capsaicin-induced coronary vasodilatation in the pig (4). These authors suggested that CGRP, rather than substance P, mediates the relaxant effects seen on activation of cardiac sensory nerves. However, experiments with isolated guinea pig hearts have demonstrated that cardiac sensory nerve fibers can release substance P when stimulated with capsaicin (36). These seemingly discordant findings can be explained if the hypothesis is true that receptors for CGRP and substance P have different locations in the guinea pig coronary blood vessels. Thus capsaicin-evoked release of substance P from perivascular nerves, as a reversed situation to intraluminal drug administration, cannot reach the endothelial receptors in sufficient quantity to elicit a response. In contrast, CGRP receptors are located in close proximity to its site of release at the perivascular region, and therefore, it can serve as a mediator of capsaicin-evoked vasodilatation. It remains to be determined if this phenomenon also applies to neuropeptides released by other stimuli in the guinea pig heart. For the present, until there is evidence showing inadequate release of substance P gaining access to the endothelial receptors under physiologic or pathophysiologic conditions, it would be prudent to consider substance P with CGRP, as neurogenic mediators capable of producing coronary vasodilatation.

In summary, this study with selective neuropeptide receptor agonists and antagonists has confirmed the involvement of NK1 receptors in mediating tachykinin-induced relaxation of the guinea pig coronary blood vessels. In addition, CGRP-induced coronary vasodilatation was found to be mediated by CGRP1 receptors. These results indicate that sensory neuropeptides present in cardiac C-fibers are capable of influencing coronary blood flow, and thus they are suspected to play a role in regulating the coronary vascular tone.

Acknowledgment: I am grateful to Fujisawa Pharmaceuticals Ltd. (Japan) for the gift of FK888 and Sanofi Recherche (France) for the gifts of SR48968 and SR142801. I thank Miss Ethel Ng for her invaluable technical assistance, and also Prof. Robert Jones for his helpful advice in the preparation of the manuscript.

REFERENCES

1. Gulbenkian S, Eavinsson L, Opgarrd OS, Wharton J, Polak JM, David-Ferreira JF. Peptide-containing nerve fibres in guinea-pig coronary arteries: immunohistochemistry, ultrastructure and vasomotility. J Auton Nerv Syst 1990;31:153-68.
2. Franco-Cereceda A. Calcitonin gene-related peptide and tachykinins in relation to local sensory control of cardiac contraction and coronary vascular tone. Acta Physiol Scand 1988;(suppl 569):1-63.
3. Wharton J, Polak JM, McGregor GP, Bishop AE, Bloom SR. The distribution of substance P-like immunoreactive nerves in the guinea-pig heart. Neuroscience 1981;6:2193-204.
4. Franco-Cereceda A, Rudehill A, Lundberg JM. Calcitonin gene-related peptide but not substance P mimics capsaicin-induced coronary vasodilation in the pig. Eur J Pharmacol 1987;142:235-43.
5. Cereceda AF, Saria A, Lundburg JM. Differential release of calcitonin gene-related peptide and neuropeptide Y from the isolated heart by capsaicin, ischaemia, nicotine, bradykinin and ouabain. Acta Physiol Scand 1989;135:173-87.
6. Mair J, Lechleitner P, Langle T, Wiedermann C, Dienstl F, Saria A. Plasma CGRP in acute myocardial infarction. Lancet 1990;335:168.
7. Maggi CA, Meli A. The sensory-efferent function of capsaicin-sensitive nerves. Gen Pharmacol 1986;19:1-43.
8. Brain SD, Williams TJ, Tippins JR, Morris HR, Macintyre I. Calcitonin gene-related peptide is a potent vasodilator. Nature 1985;313:54-6.
9. Franco-Cereceda A. Calcitonin gene-related peptide and human epicardial coronary arteries: presence, release and vasodilator effects. Br J Pharmacol 1991;102:506-10.
10. Dennis T, Fournier A, Cadieux A, et al. hCGRP8-37, a calcitonin gene-related peptide antagonist revealing calcitonin gene-related peptide receptor heterogeneity in brain and periphery. J Pharmacol Exp Ther 1990;254:123-8.
11. Dennis T, Fournier A, St. Pierre S, Quirion R. Structure-activity profile of calcitonin gene-related peptide in peripheral and brain tissues: evidence for receptor multiplicity. J Pharmacol Exp Ther 1989;251:718-25.
12. Mimeault M, Fournier A, Dumont Y, St. Pierre S, Quirion R. Comparative affinities and antagonistic potencies of various human calcitonin gene-related peptide fragments on calcitonin gene-related peptides in brain and periphery. J Pharmacol Exp Ther 1991;258:1084-90.
13. Hoover DB, Hossler FE. Vasoconstrictor and dilator responses to neurokinin A in isolated guinea-pig heart. Peptides 1993;14:29-36.
14. Burcher E, Mussap DJ, Geraghty DP, McClure-Sharp JM, Watkins DJ. Concepts in characterization of tachykinin receptors. Ann N Y Acad Sci 1991;632:123-36.
15. Mussap CJ, Geraghty DP, Burcher E. Tachykinin receptors: a radioligand binding perspective. J Neurochem 1993;60:1987-2009.
16. Fujii T, Murai M, Morimoto Y, et al. Pharmacological profile of a high affinity dipeptide NK1 receptor antagonist, FK888. Br J Pharmacol 1992;107:785-9.
17. Emonds-Alt X, Vilain P, Goulaouic P, et al. A potent and selective non-peptide antagonist of the neurokinin A (NK2) receptor. Life Sci 1992;50:PL-101-6.
18. Emonds-Alt X, Bichon D, Ducoux JP, et al. SR142801, the first potent non-peptide antagonist of the tachykinin NK3 receptor. Life Sci 1995;56:PL-27-32.
19. Arunlakshana O, Schild HO. Some quantitative use of drug antagonists. Br J Pharmacol Chemother 1959;14:48-58.
20. Gagnon J, Roth JM, Finzer WF, Haycock K, Feldman DS, Simpson J. Contrasts. In: Abacus Concepts, SuperANOVA. Berkeley, CA: Abacus Concepts, 1989:199-205.
21. Gulbenkian S, Merighi A, Wharton J, Varndell IM, Polak JM. Ultrastructural evidence for the coexistence of calcitonin gene-related peptide and substance P in secretory vesicles of peripheral nerves in the guinea-pig. J Neurocytol 1986;15:535-42.
22. Croci T, Landi M, Emonds-Alt X, Le Fur G, Manara L. Neuronal NK3 receptors in guinea-pig ileum and Taenia caeci: in vitro characterization by their first non-peptide antagonists, SR142801. Life Sci 1995;57:PL361-6.
23. Kudlacz EM, Logan DE, Shatzer SA, Farrell AM, Baugh LE. Tachykinin-induced respiratory effects in conscious guinea-pigs: modulation by NK1 and NK2 receptor antagonists. Eur J Pharmacol 1993;241:17-25.
24. Franco-Cereceda A, Rudehill A. Capsaicin-induced vasodilation of human coronary arteries in vitro is mediated by calcitonin gene-related peptide rather than substance P or neurokinin A. Acta Physiol Scand 1989;136:575-80.
25. Lam FY, Ferrell WR. Effects of interactions of naturally occurring neuropeptides on blood flow in the rat knee joint. Br J Pharmacol 1993;108:694-9.
26. Poyner DR. Pharmacology of receptors for CGRP and amylin. Trends Pharmacol Sci 1995;16:424-8.
27. Donoso MV, Fournier A, St. Pierre S, Huidobro-Toro JP. Pharmacological characterization of CGRP1 receptor subtype in the vascular system of the rat: studies with hCGRP fragments and analogs. Peptides 1990;11:885-9.
28. Preito D, Benedito S, Nyberg NCB. Heterogeneous involvement of endothelium in calcitonin gene-related peptide-induced relaxation in coronary arteries from the rat. Br J Pharmacol 1991;103:1764-8.
29. Greenberg B, Rhoden K, Barns P. Calcitonin gene-related peptide (CGRP) is a potent non-endothelium-dependent inhibitor of coronary vasomotor tone. Br J Pharmacol 1987;92:789-94.
30. Coupe MO, Mak JCW, Yacoub M, Oldershaw PJ, Barnes PJ. Autoradiographic mapping of calcitonin gene-related peptide receptors in human and guinea-pig hearts. Circulation 1990;81:741-7.
31. Knock GA, Wharton J, Gaer JAR, Yacoub MH, Taylor KM, Polak JM. Regional distribution of [125I]calcitonin gene related peptide binding sites in coronary arteries. Eur J Pharmacol 1992;219:415-25.
32. Gulati N, Mathison R, Huggel H, Regoli D, Bény JL. Effects of neurokinins on the isolated pig coronary artery. Eur J Pharmacol 1987;137:149-54.
33. D'Orléans-Juste P, Dion S, Drapeau G, Regoli D. Different receptors are involved in the endothelium-mediated relaxation and the smooth muscle contraction of the rabbit pulmonary artery in response to substance P and related neurokinins. Eur J Pharmacol 1985;125:37-44.
34. Regoli D, Drapeau G, Dion S, Couture R. New selective agonists for neurokinin receptors: pharmacological tools for receptor characterization. Trend Pharmacol Sci 1988;9:290-5.
35. Lam FY. Neurogenic mediators of capsaicin-induced vasodilatation in coronary blood vessels of guinea-pigs. Can J Physiol Pharmacol 1994;72:33.
36. Hoover DB. Effects of capsaicin on release of substance P-like immunoreactivity and physiological parameters in isolated perfused guinea-pig heart. Eur J Pharmacol 1991;141:489-92.
Keywords:

Tachykinins; Calcitonin gene-related peptide; Coronary vasodilatation; Guinea pig

© 2000 Lippincott Williams & Wilkins, Inc.